Lecture PDF IIT Bombay

Structural Analysis - II
Prof. P. Banerjee
Department of Civil Engineering
Indian Institute of Technology, Bombay
Lecture 01
Good morning. Today we are going to start off on a course Structural Analysis II, which is the
second in the series on courses of Structural Analysis. During this course, over a period of fortyfive odd lectures, we are going to be covering the various methods that are used for analyzing
structures which are known as statically indeterminate. Before I start off on the methods, I am
going to spend at least two to three lectures on reviewing the basic tenets on which structural
analysis are based and the methods that I am going to be using over and over again during this
course. Let us start off with what I had called as the basic tenets of structural analysis.
(Refer Slide Time: 02:31)
There are three. One - Equilibrium it essentially means that when a structure is subjected to
loads, the loads and the internal forces developed in the structure are in equilibrium. In other
words, the internal forces equilibrate the external loads. Without equilibrium what you would
have is that the structure would start moving and obviously, you do not want a civil engineering
structure moving in space. Therefore the body has to satisfy equilibrium. Second - Kinematics.
Kinematics is the fact that when a structure is subjected to loads the structure deforms. In other
words, under the loads the structure deforms and takes up an equilibrium state. This equilibrium
state is different from what is known as the undeformed shape of the structure which is the
structure when it was not subjected to any loads. This displaced shape, the displacements in the
structure and the strains that are developed due to the deformation in the structure have to be
compatible with each other. This is the basic concept of Kinematics.
The third and final tenet is very basic and that is stress strain compatibility. In other words, when
you have loads acting on a structure, the body deforms. Due to the deformation of the body
strains are set up various points in the structure and because of the strains stresses are developed
and these stresses are what are essentially the internal forces developed in the body. The stress
and strain have to satisfy the compatibility condition or the stress strain relationship. For
example, you know the elastic stress strain relationship which is governed by the Youngs
Modulus. The stress is given in terms of Youngs Modulus into strain. This is the simplest kind
of stress stain compatibility that we can discuss. In this course I shall be essentially referring to
only linear structures. What do linear structures mean? The relationship between the stress and
strain is linear which if you really look at it, is something in this form. Sigma, the stress, is
proportional to the strain. This proportionality is linear. This is what I am going to be looking at
during this course.
Therefore to review - a structure has to satisfy Equilibrium, it has to satisfy Kinematics and it has
to satisfy stress strain compatibility. Without these three your analysis will never be correct. You
have to satisfy all three. We shall see later on that when we deal with, what are known as
statically indeterminate structures, I will spend a little bit of time, in the next lecture talking
about, what do I mean by a statically indeterminate structure, but let us assume that we know
what a statically indeterminate structure is. When you want to analyze a statically indeterminate
structure you have to satisfy equilibrium, kinematics and the stress strain compatibility.
Stress strain compatibility is fundamental because this is something that is held at the basic tenet
or the basic building block on which we develop the entire edifice of structural analysis. This
stress strain compatibility can never be violated. However, we shall see later on, that when we
talk about statically indeterminate structures, some methods satisfy equilibrium exactly and
kinematics in a global sense. Other methods satisfy kinematics exactly and satisfy equilibrium in
a global sense. Understand that when I say equilibrium and kinematics, they have to be satisfied
everywhere in the structure and when I say that kinematics is exactly satisfied what I mean is
kinematics is satisfied everywhere in the structure and when I say equilibrium is satisfied only in
a global sense I mean that equilibrium at the local level may not be satisfied, however overall
equilibrium of the structure is satisfied.
In other methods they satisfy equilibrium exactly. In other words equilibrium is satisfied at the
local level and when I say kinematics at the global level, it means kinematics may be violated at
the local level but at the global level the displacements and the overall strengths in the structure
are satisfied. There is so much for the basic tenets of structural analysis. I want to discuss a little
bit about what I mean by indeterminacy. How do I analyze the structure? Normal structures are
always what we know as statically indeterminate. What do we mean by that? Remember that in a
statically determinate structure you could satisfy and obtain all the internal forces from the loads
by just dealing with equilibrium equations. Remember that in a planar structure for example, a
simple truss.
What is a truss? A truss is essentially a structure which is made up of members, where each
member is what is known as a two force member. What is a two force member? If I isolate the
member, this member will only be subjected to axial forces - that is a two force member. In other
words, in a truss, each member is only subjected to axial forces. Therefore, let us look at this
particular problem. How many unknowns do I have in this structure? It is a simply supported
truss. This is a hinge; a hinge restrains the lateral motion of a structure, however, it does not stop
the rotation at the particular point. Since it restrains lateral motion, the movement in the plane,
therefore it is possible that it can generate two reactions. These are my unknown reactions. Let us
say I am subjecting this structure to some loads. The hinge has two unknown reactions. This is a
roller support; a roller support only restraints this direction motion of the structure. Therefore, it
can only generate a reaction perpendicular to the movement direction. So we have three
unknown reactions. How many members do we have? 1, 2, 3 so three members we have. Each
member has only the actual force in it.
In other words, for each member we have one unknown internal force in the member and
therefore since we have three members we have three unknown member forces. How many
unknowns do you have in this structure? We have what are known as six unknown forces. How
many equations of equilibrium do we have? Let us look at it. Remember a point does not have
any dimension; a point is in a plane. How many equilibrium equations can we generate? We can
generate one which is Sigma Fx equal to 0 and the other one is Sigma Fy equal to 0. Therefore, at
a point we can only generate two equilibrium equations Sigma Fx equal to 0 and Sigma Fy equal
to 0. Therefore, for each point in the structure we can write two equations of equilibrium for
every point in the structure. How many points are there in this particular structure? A point or
more commonly, joint, are where two or more members meet or a member meets a support.
Therefore how many joints do we have in this structure? 1, 2 and 3 we have three joints and we
can write two equilibrium equations for every joint. In other words how many independent
equations of equilibrium do we have? 2 power 3 is equal to 6. We have six equations of
equilibrium that we can write for this structure and how many unknown forces do we have? 6.
Therefore, just by solving the equations of equilibrium, we can find out the reactions in this
structure and we can find out the internal forces in each of the three members. This structure is
known as a statically determinate structure. A statically determinate structure is one, in which
you can find out all the support reactions and the internal forces in the structure without using
anything other than the equations of equilibrium. What happens, if we have a situation, where the
number of unknown forces are greater than the number of equations of equilibrium available?
Let us look at the simple structure.
Again, I am looking at the same truss. How many unknown reactions? 1, 2, 3, 4 there are four
unknown reactions and of course the same three unknown member forces, so we have total of
seven unknown forces. Remember this is exactly the same structure so we only have 6 equations
of equilibrium. There is no way that you can find out all the unknown forces with just the
equations of equilibrium that you have. So how do you solve this problem? This problem is now
what is known as a statically indeterminate structure problem and this course, is where we are
going to be looking at all the different methods of being able to solve such a problem.
Let us review. We are saying that a statically determinate structure is one, in which the number
of unknown forces that are in the structure (member forces, support reactions etc.) are equal to
the number of independent equations of equilibrium. Remember, it is always very easy to
develop equations of equilibrium and you will find sometimes for example, if in a plane you take
moments at two different points on the plane are they independent, they will give you two
different equations are they independent? No. Why not? Because moment everywhere on a
plane, you should remember, moment is a free vector and therefore the two moment equations
that you write are not independent of each other. So, be very careful; when we say equations of
equilibrium we actually mean independent equations of equilibrium. Therefore, if the number of
unknown forces is equal to the number of independent equations of equilibrium that you can
write for this structure that structure is known as a statically determinate structure.
If, however, the number of unknown forces in the structure is more than the number of
independent equations of equilibrium that you can develop, then, it is a statically indeterminate
structure. How do you solve such a problem? We are going to spend the next forty-odd lectures
looking at that. Let us not try to solve that problem today itself.
Let us just look at what do I mean by Degree of Static Indeterminacy. You must have come
across this; you have done it in school, you have also been introduced to the concept in your
strength of materials or solid mechanics course and you have seen at many a time, degree of
Static Indeterminacy. We shall see later that for certain methods we will be looking at, the
Degree of Static Indeterminacy in the structure determines the complexity of the problem in
using those methods. This is a very important concept.
What do we mean by Degree of Static Indeterminacy? Let us go back to this problem. How
many unknown forces? 7. How many equations of equilibrium? 6. What is the static
indeterminacy? Subtract the equations of equilibrium from the number of unknown forces and
you have the Degree of Static Indeterminacy. What is the static indeterminacy of this particular
structure? 1. It is what is known as a single Degree of Static Indeterminacy. The single Degree of
Static Indeterminacy is 1. This is very simple for truss type structures to determine the static
determinacy. How do you determine the static indeterminacy? Let me define this for a truss
problem. How many unknown forces?
Let r be equal to the number of unknown reactions in the structure at the supports. Let m be the
number of truss members in the structure. Note, if m is the number of members, each member
has one unknown member force. Therefore when you have m number of members it implies that
you have m unknown member forces. Therefore, total number of unknown forces is equal to m
plus r. In a truss- planar truss (I am discussing only planar truss), what does a planar truss mean?
Let us look at it.
If you look at this, a planar truss is one where the truss and the loading on the truss are all in one
plane. In this particular case it is in the plane of the paper. So that is a planar truss. Now I am
looking at planar trusses, so you have the total number of unknown forces equal to m plus r.
What is the total number of equilibrium equations? They have to be equal to 2 into the number of
joints that you have in the structure. I am defining j as the number of joints that you have in the
structure. The total number of equilibrium, since you have two independent equations of
equilibrium at each joint, the total number of independent equilibrium equation is 2j. Then what
is the Degree of Static Indeterminacy? Let me write this down very carefully:
This is fundamental: the Degree of Static Indeterminacy of a truss structure is equal to m plus r
which are the total number of unknown forces in the structure minus the number of independent
equilibrium equations which are given by 2j i.e. This is the formula for determining the static
indeterminacy of a planar truss; m plus r m is the number of members r are the total number of
reaction forces. (m plus r) minus 2j, where j is the number of joints in the structure. This is the
way that you can determine the static indeterminacy of a planar truss structure. Now let us see
how we determine the static indeterminacy of other structures.
What other kind of structures do you have? A truss structure, remember, is a structure where
each member is subjected to just one unknown member force, an axial force. You have already
been exposed to a beam. What is a beam? A beam is a member which is subjected to both axial
deformations as well as flexure.
Of course, we know these more commonly as beam columns because a beam is one which is
primarily subjected to flexure. What is flexure? It is bending whereas a column is one, which is
primarily subjected to axial deformations. There is a difference between a column and a truss
member because a truss member is only subjected to axial forces; it cannot be subjected to any
other kind of forces and it only deforms axially. Whereas, a column is primarily subjected to
axial deformations that means that a column can be subjected to bending. However it primarily is
subjected to axial deformation and axial forces whereas, a beam can be subjected to axial forces
and axial deformation but primarily it carries the loads through bending or flexure. Let us look at
a structure that is made up of beam columns. What kind of structure is made out of beam
columns? A planar frame.
Let me draw a planar frame; the simplest kind of planar frame that we can have. It has two
columns and one beam. This is a plane frame, where if you look at it the connection or the joint
between the beam and column, if I draw it in reality it looks like this. This is the column, this is
the beam and this joint is a rigid joint. In a truss what does this joint look like? It looks like this,
a pin joint and that is the reason why in a truss, since this joint cannot transfer any flexural forces
the members become axial members whereas, in a plane frame this joint is a rigid continuous
joint and because of that you can transfer both forces as well as movements at the joint. We were
discussing the Degree of Static Indeterminacy. How do I determine the Degree of Static
Indeterminacy of a planar frame structure? Let us go back and let us look at a simple plane frame
member and I am going to draw a generalized member.
Let me draw a ring. Let us say that this is subjected to loads because this ring is a structure it is
subjected to loads it deforms, because of the deformation it is subjected to internal forces. What
are the internal forces? I can find out what the internal forces are by making a cut and see what
happens in this zone. If we look at this, if I make a cut here what are the forces that I develop at
this point? If, I look at this at the cut, I am taking now the cut out here, making the cut out first
and foremost I can generate an axial force, I can generate a shear force and I can generate a
bending movement. This is the notation that I am going to be using during this course: P - actual
force, V - shear force, M - bending movement.
In other words, if I look at this, these are my internal forces developed at a cut. Note that in this
particular problem, when I make this cut and I generate these three, can I find them out? I cannot
find them out. These three are unknowns which I cannot use any equilibrium equation. Take any
free bodies that you wish to and we can make two cuts and separate out another free body this is
a free body too. Every time I make a cut how many forces do I release? 3. If I make a cut and I
take this free body I have three unknown forces here. So how many forces do I have? 6. 6
unknown forces and for one free body in a plane how many equations of equilibrium can you
generate? 3. What are they?
Sigma Fx equal to 0, Sigma Fy equal to 0, and Sigma Mo equal to 0 (Mo is movement about any
point). These are the three equations of equilibrium. As soon as I release this and I take a free
body we have six unknowns and three equations. What is the Degree of Static Indeterminacy? It
is the number of unknowns minus the number of equations of equilibrium. What is the static
indeterminacy of this ring? It is 3. Therefore, this is fundamental. This is how we generate it. A
ring has three degrees of static indeterminacy. This is what I am going to be using, to find out the
static indeterminacy of a plane frame structure. How do I create a ring? Can I take a structure
and make it into a ring? Actually I cannot. I have to think a little bit and do it. Let us come back
to the problem that I was discussing a little bit earlier.
This is the plane frame. How do I find out the static indeterminacy of the plane frame? Can I use
(m plus r) minus 2j? What does that mean? Let us look at how many reactions I have; 1, 2, 3 so
three unknown reactions and how many members? It is 1, 2, 3 so three members. How many
unknown equations do I have? m plus r is 3, r is 3, m is 3 so m plus r is 6. How many joints 1, 2,
3, 4. So, if I use my formula (m plus r) minus 2j where j is equal to 4 m plus r is equal to 6 what
does that give me. The Degree of Static Indeterminacy is minus 2. Does that make sense?
Obviously it does not. What do you mean by Degree of Static Indeterminacy? Actually,
whenever the Degree of Static Indeterminacy is less than zero it essentially means that the
structure is unstable in a mechanism. Is this structure unstable in a mechanism? It is not. It is a
perfectly valid frame. If I subjected to loads; let's say I subjected to load here, is this structure
going to move or collapse? No it is not. Obviously, the formula that I generated for a truss
structure (m plus r) minus 2j is not valid for plane frame structures. Please understand this. This
equation is valid only for a truss structure not a plane frame structure. Then, how do I find out
the Degree of Static Indeterminacy of a plane frame structure? Let us go back to the ring. The
ring if you remember had three degrees of static indeterminacy. Can I take any plane frame
structure and make it into a ring? Actually it turns out I can.
Let us look at the single storey, single bay plane frame and now instead of having a hinge and a
roller support I am going to fix it. This in reality is the most common single bay single storey
plane frame and it is fixed at the foundation. This is the most standard thing. If you look at this
particular structure think about it, it is fixed over here, it is fixed over here. For all practical
purposes, I could have a situation where this could have something like this because it is fixed
here fixed here so I could have a member like this here. Now, look at this, what does this
resemble? Does it not resemble your ring? What is the static indeterminacy of this structure?
Since it is a ring it is 3. 3 is the Degree of Static Indeterminacy. Therefore a fixed, single bay
single storey frame has a static indeterminacy of three.
What happens if I put another storey on top of it? How many rings do you have in this structure?
You have one ring here and a second ring here. Each ring has three degrees of Indeterminacy so
a two storey single bay frame has, how many Degrees of Indeterminacy? 6. Now, do I always
have to take a fixed support? No. What do you mean by reducing the indeterminacy? How do
you reduce the indeterminacy? I have this single bay single frame which is fixed.
We just discussed that this particular frame has three degrees of static indeterminacy. Now let me
start releasing some of them. What does the fixed end have? A fixed end stops this point from
moving in this direction. It has two unknown support reactions plus at this point the joint cannot
rotate. In other words it can generate a movement. So a fixed support has three unknown
reactions. Now, what I am going to do is, I am going to release the restraint against rotation.
What happens?. When I release the restraint against rotation m has to disappear this m has to be
here. As soon as I have taken a fixed end and released its rotational capability what does it
become? It becomes exactly the same as a hinge support. Let me do something more; let me do
the same thing over here. Now I have made two releases in the three Degrees of Indeterminacy,
one I have released the hinge and two, I have released the rotation. I have made two releases and
I am going to make a third release. I am going to allow it to move in this direction, which means
I have taken this restraint and put it equal to 0. In other words, how many restraints am I left with
at this point? It is a single one.
Look at the structure. What does this resemble? It resembles this (Refer Slide Time: 46:04) hinge
supported at the left end roller supported at the right end.
How many releases did I make in the fixed structure to get to the hinge roller? I made one release
here, the rotational release here, one release here which is the rotational release here and one
release here which is the horizontal movement here. How many releases? There are three. When
you take a three Degree of Static Indeterminacy structure and put three of the forces equal to 0,
that is what I have done, here I have taken this moment put it equal to zero, taken this put it equal
to zero, taken this moment put it equal to zero. What have I done? I have removed three
unknown forces. When you remove three unknown forces what do you do? You essentially make
it into a 0 Degree of Static Indeterminacy. What is the 0 Degree of Static Indeterminacy? A
statically determinate structure. Therefore what do you do in planar frames? You essentially start
off with the ring. Try to create a ring. Let us now look at another structure, you might say well
you are always taking a plane frame structure and trying to make it into a ring.
Give me a multi-span beam and show me how you make it into a ring. This is a typical multi
span beam. What is the Degree of Static Indeterminacy? The first thing I am going to do is to
start fixing. This is a roller support; I am going to add two restraints to this. If I add two restraints
to a roller support I am going to stop it moving in this direction. What happens? It develops an
unknown reaction here. I am also going to stop the rotation here. So, what have I added? I have
added two unknown forces to the structure. Remember that I have added two unknown forces in
the structure. Furthermore, I am going to stop this also. I am going to fix this so I have added two
more unknown forces. So I have added four unknown forces.
In other words, I had a structure which was fixed at this end roller here and roller here. Now, to
make it into a fixed I have added two unknown forces here and I have added two unknown forces
here. Let us look at this problem. How many rings does it have? This is one ring, this is a second
independent rings. How many Degrees of Indeterminacy does this fixed structure have?
Obviously two rings 6. Now, to generate these two rings, how many unknown forces did I add?
It is 4 that is 2 plus 2 is equal to 4. What was the original structure how many Degrees of
Indeterminacy does it have? I have to take off these two unknown forces because these are
fictitious unknown forces that I added to make them into rings. I have to subtract 4 and how
many indeterminacies do I get? 2.
Let me see if I can get this same in another fashion. Suppose I remove this one unknown force
that was this, and also remove this. How many unknown forces I have removed? I have removed
two unknown forces and what does it become? A cantilever. And we know that a cantilever is a
statically indeterminate structure. So, by removing two unknown forces I have made it into a
statically indeterminate structure. So what is the static indeterminacy? 2. Either way you get that
this is a 2 degree of statically indeterminate structure. Today I have discussed how to generate
degrees of static indeterminacy and how to determine the Degree of Static Indeterminacy in a
structure.
Of course I have talked only of planar structures. In this particular course I am going to be only
talking about planar structures because once you can solve for planar structures you can solve for
space structures also. So, today I have looked at how to determine static indeterminacy of planar
structures be it trusses: (m plus r) minus 2j where m number of members r number of reactions j
number of joints. For a plane frame structure, add forces to make them into ring and then you
always remember that the number of rings that you have in a structure, the Degrees of
Indeterminacy per ring is three so suppose you have two rings, 2 power 3 is 6. How many
unknown forces did you add to make them rings? Subtract that and the answer will always give
you the static indeterminacy.
Thank you very much for today. Tomorrow I am going to be taking up on another kind of
indeterminacy that is Kinematic Indeterminacy.
Prof. P. Banerjee
Lecture - 02
Good morning. Today is the second lecture in the series of lectures on structural analysis. In the
previous lecture we discussed how to calculate the Static Indeterminacy of a structure. Today, we
are going to be discussing how to compute the Kinematic Indeterminacy of the structure.
Kinematic Indeterminacy, what do we mean by that? The Static Indeterminacy you saw last time
was essentially how to compute the number of redundant forces that you have in the system
which, if you removed would make the structure statically determinate and you could then solve
and find out all the forces in the structure purely by using equations of equilibrium. Kinematic
Indeterminacy is a completely different concept altogether. Although, the word Kinematic
Indeterminacy is actually a misnomer, it is more commonly known as the Degrees of Freedom
that a structure has. What are Degrees of Freedom? If you were to describe it in words, it is the
minimum number of displacement quantities that you require, to know before you can define the
displaced geometry of the structure. If you look at this, this is completely different. Although we
are using the same word, indeterminacy, it is a completely different concept. The previous one
referred to how to find out the forces in the structure, this one is referring to how to find out the
displaced geometry of the structure when it is subjected to loads. Note: the Kinematic
Indeterminacy is the minimum number of independent displacement quantities that you need to
define to be able to get the complete displaced geometry of the structure.
I will look at two different types of structures. One set of structures which are made up of only
actually loaded members if you remember those are trusses. I am going to start off with trusses
because even trusses and beams and frames which are essentially structures made up of members
which can deform flexurally, the treatment is different. Let us look at a truss and again I go back
to the simple truss that I had last time. We want to find out what the Kinematic Indeterminacy of
this truss structure is.
Before that I need to define what a Kinematically determinate remember, it is very easy for you
to know how to get the statically determinate structure. You know when you look at it what a
statically determinate structure looks like. How do you see a Kinematically indeterminate
structure? Let us take this same structure; I want to make this Kinematically indeterminate. How
will I make it Kinematically indeterminate? This is a Kinematically determinate structure. Why
is this Kinematically determinate? Let us look at this structure. Understand what Kinematic
Indeterminacy is.
It is the minimum number of displacement quantities that you need to define, the displaced shape
of a structure. In this structure, how will it displace under loads? How will it deform? Can this
point go anywhere? This point cannot go anywhere because it is restrained both in this direction
as well as this direction; it cannot move anywhere on the plane that means this point cannot go
anywhere. Can this point go anywhere? Neither can this point go anywhere. Can this point go
anywhere?
Again, it cannot, simply because every point has a hinge which restraints the displacements in
the plane of the structure. So what would be the displaced shape of the structure if it was
subjected to loads? Let us put some loads on the structure. These are the possible loads on a truss
structure because a truss structure can only be loaded at the joints. I put all possible loads that
this structure can be subjected to. What will be the displaced shape under these loads?
Remember, all these loads would get directly transferred to the supports and the structure will
not be subjected to any stress. Therefore there is going to be zero strain and since this going to be
zero strain there is going to be no displacement. What is a Kinematically Determinate structure?
It is a structure that does not displace. Is that interesting? Currently, it is not interesting. Later on,
we will see that we need to define a Kinematically Determinate structure before we can use what
is known as the displacement method of analyzing statically indeterminate structures.
Once we have a Kinematically determinate structure, it is very easy to look at this structure and
see what is going to be the Kinematic Indeterminacy or the Degrees of Freedom of this structure.
Let us look at it. Can this point go anywhere? No. Again there is a hinge which is restraining it
from moving anywhere. Can this point go anywhere? What is the restraint? The only restraint is
it cannot move vertically. But is there any restrain for moving it horizontally? No. So, this is a
possible displacement. This node can displace in this direction and therefore this horizontal
displacement is a degree of freedom of the structure. Let us look at this point now. Can this point
go anywhere? Note that this point can go anywhere on this plane and in a plane how many
displacements do you require to define? If you define two orthogonal displacement quantities,
they completely define the displacement of this point anywhere in space.
2
Now, if you look at this structure, a simple truss has three nodes: one node completely restrained
and cannot go anywhere, this node vertically restraint and can move horizontally, this node is
free to displace in any direction in the plane. We can define two displacement quantities which
will define the motion of this point anywhere in space. Now how many Degrees of Freedom or
what is the Kinematic Indeterminacy of this structure? 1, 2, 3 the Kinematic Indeterminacy (K.I)
of this structure is equal to 3. Therefore if you have a truss, any joint in the truss can move in the
plane and what you need to look at is what the restraints are. Let me redefine this problem. I am
not going to do an algorithmic way of getting the Kinematic Indeterminacy. How many nodes
does this structure have? It has three nodes. Now for the movement of a node or a joint in a plane
is given by two independent displacement quantities. In other words, if I know those two
displacements quantities I know where the joint is in the space. Let us look at it this way. How
many joints? 3. How many Degrees of Freedom per joint? 2. Therefore if this plane truss were to
have no supports what will be the Degrees of Freedom? It would be 2 into the number of joints
so let it be 6. The Degrees of Freedom 2 per joint multiplied by 3 is equal to 6. 6 Degrees of
Freedom since this truss has three joints. This is the unrestrained Degrees of Freedom.
In other words, the truss did not have any support, then the number of degree of freedom would
be 6. However, you do have two supports in the structure. What are the supports? At one joint,
you have a hinge which has two restraints; it stops it from moving vertically and it also stops it
from moving horizontally. It does not allow this joint to go anywhere so it gives two restraints to
this joint. What about this joint? This is a kind of roller joint. How many restraints does it give?
It gives exactly 1 it restrains the vertical motion of this joint. 1 meaning 2, plus 1. The total
number of restraints in this is 3. What is the restrained structure? How many Degrees of Freedom
does it have? 3.
Kinematic Indeterminacy by finding out displacements and also the algorithmic way: Find out
the number of joints in a truss, multiply that by 2. That gives you the unrestrained Degrees of
Freedom. You find out the number of restraints that the body has and then subtract those
restraints and you have the Degrees of Freedom or the Kinematic Indeterminacy of the truss. Let
us look at one another problem. Let us look at the slightly more complicated truss.
This is the truss; we have to find out the Kinematic Indeterminacy. How would I go about it? We
are going to use only the algorithmic way of getting it, and then I will show you which are those
displacements. How many joints? 4. What is the number of Degrees of Freedom? Note that
Kinematic Indeterminacy and Degrees of Freedom are used interchangeably. What are the
Degrees of Freedom per joint for a planar truss? 2 Degrees of Freedom. What is the total number
of unrestrained Degrees of Freedom? 2 into 4 is equal to 8. We need to define what the restraints
are? Restraints here are one hinge support, two restraints - one roller support and one restraint.
So a total of 3. The total actual Degrees of Freedom is 8 minus 3 is 5. The Kinematic
Indeterminacy is 5, for this structure. Once I have defined the kinematic Degrees of Freedom,
you need to understand what is Kinematic Indeterminacy, it is the minimum number of
displacement quantities, that you need to define, to be able to get the displaced geometry of the
structure. I need to find out and define those displacement Degrees of Freedom. Note that I
should have only 5 independent displacement Degrees of Freedom. Let us look at this. Are there
any Degrees of Freedom at this node? No. Both of them are restrained. Let us go to joint 2;
restraint none. What are the Degrees of Freedom? 1 2. Let us go to 3. Any restraint? No.
What are the Degrees of Freedom 1 2 the vertical displacement and the horizontal displacement.
What about 4? It is restrained in this direction and is not restrained in this direction, so how
many do I have? I have 1 2 3 4 and 5. 5 Degrees of Freedom; 5 Kinematic Indeterminacy. It
should be fairly easy for you now, to get the Kinematic Indeterminacy of a truss type structure.
Planar truss - 2 Degrees of Freedom, space truss - 3 Degrees of Freedom because you need 3
displacement quantities to define the position of a point in space. 3 independent: x displacement,
y displacement and z displacement. What will be the unrestrained Degrees of Freedom for a
space truss? 3 into the number of joints; that will be the unrestrained and then you find out the
number of restraints that you have which are the support conditions or the supports. Subtract the
restraints and you got the Degrees of Freedom for a truss and that defines the Kinematic
Indeterminacy for a truss structure. Let us move on to that for a beam.
This is a structure that you are all familiar with; a simply supported beam. What is the Static
Indeterminacy of the simply supported beam? 0. You know that it is a statically determinate
structure but now we are not looking at Static Indeterminacy we are looking at Kinematic
Indeterminacy. Again there are 2 joints 2 nodes in the structure and you can ask why we do not
use the same method as a truss. For every node to move in the plane, there will be two Degrees
of Freedom. 2 nodes, 2 Degrees of Freedom and three restraints. This is kinematical so if you use
the truss model 2 into 2 so 2 nodes 2 Degrees of Freedom per node that is 4 and there are 3
restraints; 2 here and 1 here one degree of freedom. What is the one degree of freedom this one
(Refer Slide Time: 20:43). Is that correct? Is this single kinematically indeterminate structure
single degree of freedom? No.
Understand that members in a beam and frame behaved differently from a truss; a truss is only
an actually loaded member. All we need to know is where the two joints at the end of a truss
member is and that gives us elongation of the truss and from that we can find out the force in the
truss. Can we do that for a beam? No, because the behaviour of a beam is in the way it is loaded,
it goes like this. The entire concept of displacement, Degrees of Freedom for a beam and a frame
would be different from a truss and that is why we need to look at it differently. How do we look
at it differently? An important point is that a joint that connects beam members or frame
members in addition to the two Degrees of Freedom which define its movement also has another
additional degree of freedom. What does that do?
That degree of freedom actually defines the movement of the members, relative to each other.
What is that? That degree of freedom is the rotation degree of freedom. Whenever you look at a
beam or a frame every joint when it is unrestrained has 3 Degrees of Freedom. So a joint has 3
Degrees of Freedom. What does this degree of freedom do? If I fix this and this, this disappears
because this cannot go up or down.
Remember, that if I have a Kinematically determinate structure I can define the displaced
geometry of the structure completely. If I restrain this and I restrain this, both the displacements
are restrained which means no displacement. But is that true? No. Under this load it displaces in
this manner. Understand that I required additional displacement quantities to define the displaced
shape of the structure and what are those? Look at this rotation and this rotation; these two
rotations gives me the displaced shape of the structure. That is that additional degree of freedom
that I need to define. Whenever you have a beam or a frame you have 3 Degrees of Freedom per
joint. Let us come back to this structure; the simply supported beam. How many Degrees of
Freedom does it have? How many joints? 2. How many unrestrained Degrees of Freedom? 3 per
joint, so a total of 6. Note that this hinge support still restraints only 2 which is this displacement
and this displacement. It has 2 restraints and this one. What was the total number of restraints? 3.
What are the Degrees of Freedom? 3. This simply supported beam has 3 Degrees of Freedom or
its Kinematic Indeterminacy is equal to 3.
What is Kinematic Indeterminacy? Number of unknown displacement quantities. Let us find
those out for this structure just like we did in the truss. Note that it restraints this but allows this,
so this is my first degree of freedom. This is a pin roller; it allows free rotation so this is another
degree of freedom. This joint restraints both the motions but allows it to rotate and that is my
third degree of freedom. In other words, this simply supported beam, if I knew the horizontal
displacement of this point, the rotation at this point and the rotation at this point, I would be able
to give you how this structure has displaced. If I knew the displacement quantities I could draw
the displaced geometry of the structure. This, in essence, defines the Kinematic Indeterminacy
for beams and frames.
I want to introduce one more joint that you are all familiar with; another statically determinate
structure; this is a cantilever beam. What is the Kinematic Indeterminacy of this cantilever
beam? How many joints? 1 and 2 so 2 joints. How many unrestrained Degrees of Freedom? 6.
You need to find out the number of restraints. How many restraints? Let us look at this joint, this
is a fixed support; a fixed support is one which restrains the point from displacing in any
direction and it also restrains the rotation of that point. How will this displace? This will displace
something like this. At this point, the slope would be 0, the displacements both vertically and
horizontally would be 0, so completely restrained. At this point, it is not restrained; it does not
have any support. What does this restrain? It restrains vertical, horizontal and rotation, so there
are three restraints. So the Kinematic Indeterminacy is 3. What are those 3? It is 1, 2, 3. So, for a
beam you should be able to get the Kinematic Indeterminacy or the number of Degrees of
Freedom for the structure. Let us move on to Kinematic Indeterminacy for a frame.
Note I just said that, the Kinematic Indeterminacy for both a beam and a frame are computed in
exactly the same fashion. I can use the same algorithm that I developed for the beam. Look at
this; how many nodes 1, 2, 3, 4 there are four joints in this structure. Following my procedure,
each joint unrestrained has three Degrees of Freedom. The number of unrestrained Degrees of
Freedom equal to 3 into 4 is equal to 12. Unrestrained Degrees of Freedom is 12. Now, we need
to find out what are the restraints in the structure. How many supports? Two supports, 1 here and
1 here. What are they? fixed supports. How many restraints does the fixed support give? 3. How
many fixed support? 2 What are the restraints 6 is equal to 2 into 3. Two supports with 3
restrains each. How many degrees of freedom? 6. Kinematic Indeterminacy, remember, is a same
as Degrees of Freedom 6. I should be able to draw them 1, 2, 3, 4, 5, 6. The vertical, horizontal
and rotation of this joint and the vertical, horizontal and rotation of this joint; 3 per node; total
six Degrees of Freedom. If I know these 6 displacement quantities I can draw the displaced
shape of this frame.
Now, I want to bring in the concept of constraints. Typically what we do, when we consider
beams and frames we tend to neglect, the axial deformations because by and large in a beam and
a frame, the main way that the loads are transferred is through flexural deformations. In other
words, the structure bends to take the loads. While flexural deformations are very large the axial
deformation are not zero, but are very small. We tend to neglect the axial deformations. What
happens then?
Let me go back to a simply supported beam; I neglect the axial deformation of this member,
therefore I am saying that it is axially rigid. Note, in general, for a statically determinate beam,
the forces that you normally find out are the sheer force bending movement not the axial force.
Implicitly, without knowing, you are actually neglecting axial deformation; you never explicitly
mention that. When you are analyzing a frame or a beam the member is axially rigid. What
happens when you have a member which is axially rigid? What happens here is that you are
constraining the structure. It has no effect on the Static Indeterminacy of the structure.
Remember, that it has a tremendous effect on the Kinematic Indeterminacy or the number of
Degrees of Freedom that you get from a structure. You are constraining a structure into one
actually rigid. When I say that this member is actually rigid, this cannot go anywhere. It can
move horizontally. Forget about the rotational Degrees of Freedom; I am looking only at the
axial deformation. It can go this way so this goes this way. What effectively happens to this
member when this goes this way? This joint cannot go anywhere, this joint can move
horizontally. The member is either axially shortened or if the displacement is in this direction it
is elongated. In other words, this degree of freedom actually corresponds to the axial deformation
of the body. If this member is axially rigid, since this point is not going anywhere; it cannot
deform axially. That means this displacement has to be zero.
By making the structure axially rigid I have eliminated a degree of freedom. If I say that this
beam is axially rigid how many Degrees of Freedom does it have? It has 2 rotations. It is a 2
degree of freedom structure. How do I put it into my algorithm? I truly believe that unless you
have an algorithm way of computing Static Indeterminacy and Kinematic Indeterminacy, you are
always going to flounder because you would not know how many indeterminacies there are in
the structure both static and kinematic. Let us go back to algorithm way. How do I include this
effect in the algorithmic way? What we define is two joints unrestrained Degrees of Freedom: 2
8
into 3 is equal to 6, restraints -3. Now, what have I done by making it axially rigid? I am
constraining the structure.
So, a new thing: constraints. Constraints, also reduce the number of Degrees of Freedom in the
structure. In this case how many constraints? It is axially rigid; whatever I have done I have
made this into 0. How many constraints do you think axial rigidity in a member brings in? 1.
Because a single degree of freedom defines the axial deformation of a member and when I make
it axially rigid that is the additional constraint that I am giving; that this body cannot deform
axially. I am making one constraint per member. How many members are there in this structure?
There is 1. How many Degrees of Freedom? 2. I have already shown you the Degrees of
Freedom in the structure. How about the algorithm for getting the Kinematic Indeterminacy of a
beam or a frame? Start off by finding out how many joints there are in the structure. Multiply the
number of joints by 3 that gives you the total number of unrestrained Degrees of Freedom. Then
find out the restraints. What are the restraints that are provided always by supports? You go to
every support and find out what kind of restraint it gives add up all the restraints from all the
supports. Those are the total number of restraints those are to be subtracted from the unrestrained
Degrees of Freedom. In addition to that, you need to look at the constraints. If, a member is
axially rigid how many constraints? Single; it is 1 per member. So, you subtract the total number
of constraints and you got your Degrees of Freedom or the Kinematic Indeterminacy of the
structure. Let us go to the frame.
In the frame that we were looking at, all the members are axially rigid members. Let us now see
how I compute the Degrees of Freedom. How many joints? 1, 2, 3, 4. How many members? 1, 2,
3. Unrestrained Degrees of Freedom: 4 joint 3 into 4 is equal to 12. Restraints: 2 fixed supports,
so 2 into 3 is 6 restraints. Constraints: how many members? 1; all axially rigid so constraints -3.
So, how many Degrees of Freedom does this structure have? 3. So a frame where all the
members are axially rigid has three Degrees of Freedom. What are those three Degrees of
Freedom? What did I have in the previous case? Let us compare the previous case to this case.
9
When I did not have the constraint of axial rigidity, these were the 6 Degrees of Freedom that I
had. Now let us look at what happens. 3 of those have to disappear because my Kinematic
Indeterminacy for this structure is 3; so I have to make three displacements disappear.
Remember, it is not 3 displacements that I am making disappear or equal to zero.
It may so happen that two displacements are equal to each other in which case it is only one
degree of freedom because remember the Kinematic Indeterminacy is the minimum number of
independent displacement quantities that I need to define to be able to give the displaced
geometry. If one degree of freedom is equal to another degree of freedom then what happens? I
need only one of them. Let us look at this degree of freedom. What does this entail? This joint is
not going anywhere vertically. If this joint moves vertically what happens? It means this member
is being axially deformed but it is axially rigid. So what happens? I know that this displacement
is equal to zero. Similarly, I know that this displacement equal to zero because this displacement
would entail this column deform. What else do I know? This member cannot axially deform.
What if this joint displaces by 1? How much will this joint displace by horizontally? Since this
member is axially rigid if this displaces horizontally by one so will this.
That means that these two displacement quantities are the same; they are going to have equal
value. So, I need to only consider one of them. Remember, I am not saying that this is equal to
zero. No. It is not equal to zero. However, I am eliminating this degree of freedom because this
and this are the same. I could do either one of them; either this or this you could eliminate. Then
what are my independent Degrees of Freedom for this axially rigid frame? 1 rotation, 2 rotations
and 3 horizontal displacement. Those are my three Degrees of Freedom for this axially rigid
frame. So you see it is fairly simple. You should now have no problems, in computing the
Kinematic Indeterminacy of a planar structure, be it a truss, a beam or a frame. How do you find
out the number of Degrees of Freedom per joint depending on whether you have a truss structure
or a beam structure?
If it is a truss structure, number of Degrees of Freedom per joint is two. Find out the number of
joints multiply it by two, then find out the restraints and you have got it. Remember that an axial
member cannot be constrained. Because if you stop the axial deformation of a truss member,
what else is left? Nothing; it cannot deform at all. Typically, in a truss structure you never have
constraints; you only have restraints provided by the supports. When you come to a beam and a
frame, if you do not stop any of the deformations, then you have three Degrees of Freedom per
joint. Number of joints gives you the total unconstrained Degrees of Freedom. You find out the
restraints and subtract them. Furthermore, if you want to neglect the deformation of a particular
member then you have to consider that member as a constraint. If you neglect the axial
deformation for each member, you have one constraint and if you make all of them then you
have that many numbers of constraints. This is relatively simple. I hope that at the end of this
tryou should be able to find out the Kinematic Indeterminacy and Static Indeterminacy of
frames. I am now going to show one more constraint where this beam is totally rigid.
10
This beam cannot deform axially or flexurally. What happens in such a situation? Axial
constraint I have already shown it as 1. How many for flexural constraint? You are making it
flexurally rigid also now. Let us look back at this simple beam. How many Degrees of Freedom
does this beam have? It is axially rigid, so how many of Degrees of Freedom is 1 2. Now let me
make this beam totally, flexurally rigid. It cannot bend. If it cannot bend, this is 0 and this is 0. If
you have flexural rigidity then how many constraints per member are there? 2. When this was
not flexurally rigid we had 3 Degrees of Freedom. I am going to start from there; three Degrees
of Freedom. How many constraints? Additional constraints due to flexural rigidity is 2. How
many Degrees of Freedom? 1. Now I want to show you some structures and we will go through
and compute the Kinematic and Static Indeterminacy of these frames.
11
This is a propped cantilever. It is a statically indeterminate structure, because it has 4 support

reactions and 3 equations of equilibrium. It is statically indeterminate structure.
Let us look at another one. Here, what you have is this left joint which you have allowed to move
horizontally. Suppose, I apply a horizontal load to this beam what is going to happen? This beam
is going to move because there is no horizontal restraint. It only has two vertical restraints and a
movement restraint to the left hand. This structure is unstable. Therefore understand one thing, it
is very important before you start looking at Static Indeterminacy, you need to know whether it
12
is unstable or stable and you can only find out the indeterminacy of stable structures. Look at
this. This is stable, it is a cantilever.
Now let us look at this particular member. Look at the top. How many Degrees of Freedom do
you think it has? There is one frame and one member how many joints? 2 joints. There are 2
joints here; one here and one here. How many unconstrained Degrees of Freedom? Two joints so
you have 2 into 3 is equal to 6. You have six unconstrained Degrees of Freedom. Let us look at
the restraints. This joint is fixed, how many restraints? 3. What about this restraint? 1. So how
many restraints in total? 4. That means unconstrained 6 minus 4 is equal to 2 Degrees of
Freedom for this structure. What are the two Degrees of Freedom? One is this displacement and
13
one is this rotation. In addition to that I am going to say that this is actually rigid. This is implicit,
I do not want to state it. Typically, for a beam, unless you explicitly allow axial deformation, you
assume it to be axially rigid. So if it is axially rigid constraint is 1, Degrees of Freedom is 1
which is this rotation. So Kinematic Indeterminacy is 1. How do I find out Static Indeterminacy?
3 plus 1 is equal to 4. How many equations of equilibrium? 3. So what is that Static
Indeterminacy? 1. This structure has a single Kinematic Indeterminacy and a single Static
Indeterminacy.
Let us look at a few other structures. This is a multispan beam. I need to find out its number of
Degrees of Freedom and the Static Indeterminacy. If I remove this it becomes statically
determinate so there is only one Static Indeterminacy. How many kinematic? It is 1, 2, 3 so 3
into 3 is 9. How many restraints? It is 1, 2, 3, 4 so totally 4 restraints. Then 9 minus 4 is 5. Static
Indeterminacy is 1 and Kinematic Indeterminacy is 5.
14
Let us look at another structure. In this structure how many redundant forces? I am not going to
go into this; if I am release this and I release this, totally I will get a simply supported beam. It
has 4 redundant forces, so Static Indeterminacy is 4. Kinematic: it has three joints 1, 2, 3 so 3
into 3 is equal to 9. Restraints: 3 here, 3 here and 1 here. So 2 Degrees of Freedom and you have
shown the 2 Degrees of Freedom.
For a truss, remember that each one is triangulated so it is internally determinate and if I remove
one of these it becomes 3 and becomes externally determinate also. This is the only redundant
15
force. The Static Indeterminacy is 1 and for the kinematic you have 1, 2, 3, 4, 5, 6: 6 into 2 so 12.
12 - 4 restraints 2 1 1: 4 restrains gives you Kinematic Indeterminacy 8.
Look at the frame structure; we have already seen this. It is 3 and 6. Static Indeterminacy is 3,
Kinematic Indeterminacy is 6 and that brings us to the end of this lecture. At the end of this
lecture, I am going to assume from here that given any structure you should be able to find out its
Static Indeterminacy and Kinematic Indeterminacy or Kinematic Indeterminacy I replace
sometimes by degrees of freedom.
Thank you. We will continue next time by looking at some methods, from which we can use to
do our structural analysis.
16
Prof. P. Banerjee
Lecture 03
Good morning. We have moved on to our third lecture in this course. In the first lecture, we
looked at static indeterminacy and how to obtain the static indeterminacy of a structure. In the
second lecture, we looked at how to get the kinematic indeterminacy of the structure and then,
we quickly looked at various types of structures and looked at what their static indeterminacy
and kinematic indeterminacy were.
In this lecture, I am going to be looking at the method of virtual work. This sequence of three
lectures that we have are essentially to review things which you have already done, just to put
them into the perspective of what we are going to be talking about later. The method of virtual
work is something that you people have known for a long time. You may have known it as a unit
force method to get displacements or you may have known it as displacement approach to
finding out forces.
Let us just review the method of virtual work; it is actually not a method, it is a principle. What
does the principle of virtual work say? What it says is this: if you have a body that is in
equilibrium under a system of forces and if you displace the body or give a virtual displacement
to the body, then the work done by all the forces acting on the structure is equal to 0. What are
the important tenets? The body has to be in equilibrium under a set of forces and then, if you
apply a virtual displacement, the work done by all the forces is actually equal to 0 this is the
principle of virtual work. However, understand one thing: as soon as it says, it is no longer the
principle of virtual work because the principle of virtual work has two methods.
There is the virtual displacement method, which is the common one that we always refer to, that
you have a system of forces and you subject it to a set of virtual displacements and then the total
work done is equal to 0 (Refer Slide Time: 04:29). However, you have another method called the
virtual force method where if you look at the body under a set of forces and it has a displaced
shape. If I apply a set of virtual forces on this structure in equilibrium then the work done by
these virtual forces undergoing the real displacements of the structure are also going to be equal
to 0. This is known as the virtual force method.
What are these methods? Why do we use these methods? It is very important to understand that
it is the principle of virtual work, method of virtual displacement, the method of virtual force and
so on. It is very interesting to study about these methods. The whole principle of virtual work is
actually a very interesting principle which changes a geometry problem into an equilibrium
problem and an equilibrium problem into a geometry problem. In other words, the principle of
virtual work provides a kind of taking a real structure into a virtual zone in which geometry is
reflected by equilibrium and equilibrium in the real structure is represented by geometry in the
virtual structure. You will see how.
What do we use the principle method of virtual force for? What we use it for is to actually find
out displacements this is what we are trying to find out (Refer Slide Time: 06:36). In other
words, finding out displacements means we are trying to define geometry the displaced
geometry of the structure, but what do we do? We use the virtual force method. What does the
virtual force method do? You apply a set of virtual forces, find out all the forces in the structure
due to this action of virtual forces and then multiply the real displacements and deformation in
the structure and the total work done is 0. What this gives me is the displacements by actually
solving an equilibrium problem (Refer Slide Time: 07:19); so I use equilibrium to define
geometry and in the next method, we want to actually find out the forces by applying
displacements.
What do we do? We actually apply a virtual displacement pattern, find out the displacements and
deformation in all the members, then find out the work done by all the real forces undergoing
these virtual displacements and deformations and equate it to 0. In other words, what do we do?
With this, we find out forces in the structure; we define forces in the structure and what do we
use? We actually use geometry. Because given a virtual displacement to find out the
displacements and deformations of all the joints and members, you are required to solve
geometry. You see, it is just the flip side, to define geometry we use equilibrium and to define
forces we use geometry; that is, in essence, the method of virtual work. It is all very fine to say
this, but let us actually use these methods and that will become much more understandable when
I actually give you this. Let us use the virtual force method to find out the displacements in a
structure, let us do that.
Let me go back to my base truss that I have and now, let me say that I subject it to force F1 and
F2 (Refer Slide Time: 09:19). My question over here to you is to find out the horizontal
displacement of this joint (Refer Slide Time: 09:32) that is my question to you. These are the
real forces acting on the structure and I want to find out what is the horizontal displacement of
this point under these forces. How would I do that? Let us go through the steps and that will
actually illustrate the concept of how the method of virtual force is used to define the geometry.
Simple problem; I am not going to make it very complicated; I am not even trying to analyze the
structure.
Given these loads, find out the displacement that is my question. How do I do that is by using
the method of virtual force. What is the first step? Note that this is a statically determinate
structure I have made it purposely that. The first step is to find out all the forces in all the
members and supports; I can easily get it. Let me give some numbers, let me put that this is 5
meters (Refer Slide Time: 10:49), this is equal to 5 meters and 5 meters all of them have the
same E and A, all of them are identical. I need to find out this one.
First step: forget this, take the real structure. Under these set of forces, I find out the reactions.
Let me call this A (Refer Slide Time: 11:57), B, C. I will find out RAY, RAX and RCY. Note that
these are all the support reactions that you develop in the structure. Can I find these out? I can
find these out. I do not want to go into it. You have three unknowns, three equations one is for
horizontal, one is for vertical and one is taking movements about any point. You can find out
RAX, RAY, and RCY you know these now. Once you know these, can you find out the forces in
these members? I can analyze the truss and I can find out the forces in these members. You use
any method: method of joints or method of sections; I do not care how you do it but given the
fact that this is a statically determinate structure, I should be able to find out the force in member
AB, the force in member AC and the force in member BC these also I can find out. What are
these? These are for these applied loads (Refer Slide Time: 13:28) and I can find these out.
Once you know these, what else do you know? These are the axial forces in these members; I
can find out the axial deformation in member AB, in member AC and then in member BC. How
can I find these out? I can find these through. I am not even going into this, it is something that
you should know: axial deformation is given in terms of the axial force, the axial force divided
by A gives you stress, stress by E gives you the constant strain, strain into L is equal to the
deformation these are things that you have picked up a long time ago and I am just writing
them down. Of course, EA of AB, EA of AC (Refer Slide Time: 14:47) and similarly, for this,
FBC into LBC upon EA of BC.
So, in other words, since I am given the EAs, I am given the length (I can find out the lengths), I
can find out the forces, I can evaluate these three. Therefore, these are the steps for the real
structure you find these out. Now, the question is these are the real support reactions, these are
the real forces and these are the real deformations in each of the members I can compute these;
given this and the undisplaced geometry of the structure, I can always find out all of these. Let us
move on.
What is the next step?

I have found out all the real forces, real reactions and real deformations in the body; I know R, I
know axial and I know real all of them are due to the loads F1 and F2 applied on the truss. What
is the next step? I am going to apply the method of virtual force to get my r1 displacement. What
do I do? The method of virtual force says that I need to apply a set of virtual forces on the body.
I can choose what I want and the way I choose is, I apply only this, I do not apply any forces
here there is a reason behind it, which we will see. I apply a virtual force; this is a virtual force
(Refer Slide Time: 17:01).
How many degrees of freedom does this structure have? The structure, if you remember, has
three degrees of freedom, I have already looked at them. What are they? This displacement, this
horizontal (Refer Slide Time: 17:34) and this vertical displacement. What I am going to be doing
4
here is, I only need to find out this displacement. Therefore, what is the work done by all the
external displacements? They are r1 (the displacement that I am interested in) into dF. I have r2
and r3 over here, but they do not have forces attached with them, so r2 is equal to 0, r3 is equal to
0, which gives me r2 into 0 plus r3 multiplied by 0, which is essentially r1 into dF.
My work done by the external displacements is actually just r1 into dF; that is the reason why I
only apply a force here (Refer Slide Time: 18:20); that is the work done by this virtual force
undergoing this real displacement. What are the other things? The other things are that the
internal virtual forces will also do work because you have real deformations in the body under
the action of forces. What I need to find out is under this force, I need to find out what are the
forces in the body. Again, I find out RAY, RAX and RCY I find those out. I find also out the
virtual forces developed in all the members. In this particular case, it is very simple. You can say
that this is equal to delta F (Refer Slide Time: 19:33), this is equal to 0, this is equal to 0, you can
compute it and you will see that this is equal to 0, this is equal to 0 and this is equal to delta F I
have analyzed it; it is very simple, so I can analyze it; it does not matter, it is not relevant in this
particular case; however, I have done it, you can get it. Now, the one aspect of it is that if you
remember, the equations say total virtual work is equal to 0. We can also write this in another
way: the work done by the external displacements is equal to the work done by the internal
displacements.
What are the internal displacements? How can we write it this way? All the internal forces in the
structure are essentially reactions to the loads they actually oppose the loads, so the work done
by the external displacements and forces are actually opposite the work done by the internal
forces. If you look at this equation (Refer Slide Time: 20:54), I can actually write it in this
fashion; if I separate out the internal and external, this becomes this. Why is this minus? If you
really look at it, all the internal forces are reactions they oppose the loads and that is why there
is always a negative sign. When I take this, this goes into this, so this is also another way of
writing the virtual work equation.
What is the external virtual work in this particular case? You will see that this is the external
virtual work. What is the internal virtual work? The internal virtual work is the work done by
each deformation multiplied by subjected to the forces. In this particular case, the internal virtual
work is going to be equal to. What is the virtual force in each member? This is A (Refer Slide
Time: 21:58), this is B, this is C. In member AC, it is going to be delta F and what is the real
deformation? deltaAC. This becomes deltaAC, which we have computed earlier, multiplied by
delta F (that is the work done; also note that by definition, this is tensile (Refer Slide Time:
22:20), so this deltaAC is elongation and elongation multiplied by tensile is always given in this
fashion) plus the work done by deltaBC multiplied by the virtual force in it, which is 0, plus
deltaAB multiplied by the virtual force, which happens to be 0. What is the internal virtual work?
It is deltaAC into delta F that is the internal virtual work. What is the external virtual work? If I
equate the two, what do I get? You will see that r1 is equal to deltaAC this is what you get and
this is not surprising. Let us go back here. This point is not moving anywhere. How much does
this move? It moves by whatever is the axial deformation in member AC, so that is what you
have got. The method of virtual work has given me a geometry by actually solving two
equilibrium problems.
The first one was the real equilibrium problem (Refer Slide Time: 23:37) where you subjected it
to the real loads; the second one is the real structure subjected to the virtual load. This in essence
is how to use the virtual force method to get the geometry of a structure. I am just illustrating
these points. I am not going to be solving anything right now but I am just looking at it purely
from the point of view of the principle.
Once you understand the principle, we can then use it very easily; we will be using it for the rest
of the semester; in all the lectures that I give you, you will be using these methods. We will
continuously be using these methods so that you will be absolutely certain of what to do when
you come to it. Now, let us look at the virtual method of virtual displacement.
What is the virtual method of virtual displacement? Let me look at this: it is the same structure.
The whole point here is that I am going to get you so involved in this structure that. It is the
same truss structure and I have a set of forces, again the same set of forces and I am going to use
the method of virtual displacement. What is the method of virtual displacement used for? It is
used to solve equilibrium problems. What does that mean? In this particular case, due to these set
of forces, I need to find out what this is, what this reaction is (Refer Slide Time: 25:34).
Since this is a statically determinate structure, you can find it easily using equilibrium; but I want
to illustrate the method of virtual displacement to you, so we are going to assume that you do not
know how to solve equilibrium problems let us assume that; without solving the equilibrium
problems, can I get this reaction (Refer Slide Time: 26:03) given these set of loads? That is the
question here and I have to use the method virtual displacement. How do I do that?
Here, I say that under these set of forces, the structure is in equilibrium; so what I am going to do
is, I am going to apply a virtual displacement (note that I want to find out this reaction (Refer
Slide Time: 26:38)). What I do is this: corresponding to the reaction that I want to find out, I
release the restraint and I give it some displacement delta delta. Let us be very clear about it all
6
I am interested in is to ensure that this structure. Now, note that once I have removed the
restraint, this is a mechanism; since this is a mechanism and is a single degree of freedom
mechanism, I can give it a displacement and I can find out how every other point moves. Is that
clear? I can do that.
Now what? Note also another thing: since it is a mechanism, this displacement is not going to
lead to any deformation in any of these members; in other words, I can move this body as a rigid
body, because it is a one degree of freedom mechanism. I give it a displacement and given this
displacement, I need to find out the displacements of all other points this is a geometry
problem; let us see how I can do that. Let me again go back to the particular problem that we
were looking at. I need to know the geometry to be able to give you the geometry. I give this a
displacement d delta. Will this move in this direction? Note that this joint is here (Refer Slide
Time: 28:50), this point cannot go anywhere and I need to move this way. It does not mean that
it only moves in this direction; you have to ensure that every point moves the way it does in a
rigid body; in other words, you cannot have deformation on any of the members. Remember that
the displacement that I give, although I am showing it to be very large, is actually very small.
Small displacements still hold true; the virtual displacement that I give has to be a small
displacement; I might show it as very large, but I am just essentially exploding it to show you; it
is not as if it is going to be moving in that direction.
Note that for this member not to axially deform, it is free to move in this direction tangentially. If
it moves in any other direction, you have an axial deformation, which you cannot and so that
means this point can only move up here. I am going to show this, it may look longer, but
remember this displacement is very small compared to this length. Although it looks longer, it
really is not; it is just that I have exploded the view. This one goes this way (Refer Slide Time:
30:15).
Now, I need to know where this point goes. How do I do that? That is the geometry problem.
What I need to do is understand that this member (Refer Slide Time: 30:33) is connected over
here and since it cannot deform axially, the only way that it can move is along this direction.
How do I get where it has gone? Has it come here or has it gone there? That is what I need to
find out that is the geometry. For the geometry, if you notice, this point has come here (Refer
Slide Time: 31:04).
Let us suppose that this joint was not constrained by this member let us say that it is not
constrained here, it is free to move. How would it move? If this joint moved here, to ensure that
this member remains undeformed, this would have to go up because that would ensure that this
was undeformed.
Therefore, this point would move here (Refer Slide Time: 31:38) if it was not restrained, but it is
restrained. Obviously, since this joint has moved up this way, it cannot move up this way, so
what does it have to do? This joint has to move in this direction, but note that this member is this
way. This point has come here now. Now, for this member not to axially deform, how will it
have to go? It will have to go perpendicular to this member; the point where this and this connect
is where this joint actually is, to ensure that neither this member nor this member has axially
deformed.
7
Therefore, this is the point where this member is. Where is that point? If you note the way I have
drawn it with this as 5, this as 5 and this as 5, this is 90 degree, this is 45 and 45 degree that is
how I have managed the whole thing. Now if this is delta, when this has to move perpendicular
to this, what does that mean? Perpendicular to this is this, so that means it moves parallel to this
member. This part is d delta (Refer Slide Time: 33:20). How much has it moved horizontally?
Let us see how much it has moved horizontally. You have to understand how much this motion
is. Think about it. When you are going this way, how much is this motion? This is also 45, 45, so
it is going to be d delta by root 2. Similarly, you will see that this is equal to d delta upon root 2.
If you look at the total horizontal motion, it is going to be d delta by (root 2 into root 2), which is
equal to d delta by 2. This point has moved vertically by d delta and horizontally by d delta by 2.
Now, I can write down my virtual work equations.
The real forces undergoing the virtual displacements. Note that internal virtual work is 0,
because the internal deformations (the virtual deformations are 0) because I have ensured that
you do not have them; so internal virtual work even though under this you have real forces, the
virtual deformations are 0; internal virtual work is 0. What is the external virtual work? First is
this: RCY multiplied by d delta. Note that this is up and this is up, so the work done is positive.
Next is this force F2. How much is that undergoing? d delta, but F2 is downwards, displacement
is upwards, so it is negative work and so, it is minus F2 into delta by 2 (actually, this is the
point, so this is the work done and that is d delta by 2). Then comes the work done by F1.
If you look at F1, the work done by F1 is negative because this is facing in this direction and the
force is in this direction, so minus F1 into d delta by 2 and that is equal to 0. You get that RCY is
equal to F2 by 2 plus F1 by 2 this is what the principle of virtual displacement gives you. Let us
see whether that is true by actually looking at equilibrium; because this is a statically determinate
structure, I can find out from equilibrium. Let me take moments about this point (Refer Slide
Time: 36:34). Note that this is RAY and this is RAX, let me take moments about this point. If I
take moments about A, what do I have? Let me look back at that particular problem; let me go
back.
This is 5, this is 5 (Refer Slide Time: 37:09) and this is 5, this is F2 and this is F1 and I need to
find out RCY. Take moments about A is equal to 0. What do I get? RCY into 10, this is a
counterclockwise movement and both of these give rise to clockwise movement, so the
clockwise movements are F1 into 5 minus F2 into 5 is equal to 0, and RCY is equal to (F1 plus F2)
by 2.
Both the methods of virtual displacement and equilibrium give me the same value of RCY. This
in essence is the method of virtual displacement. What I have done is that I have reviewed with
you the method of virtual work, which actually has two different virtual work methods. One is
the method of virtual force, which is used to find out displacements. Remember? Let us review
that. What did I use the virtual force method for? I used the virtual force method to get the
displacement; this is the virtual force. The virtual force method is used to find out displacements
in the structure; the method of virtual displacement is used to find out the forces in the structure
using geometry. That is the point that I was trying to make right at the beginning: the virtual
force method is used to define geometry using equilibrium and the virtual displacement method
is used to define forces using geometry.
We are going to continuously use these methods over and over again during the entire course that
we have over here and you will see how useful this method is. Note that some people tend to use
the method of virtual work and some people use the method of stationary potential energy or
minimum potential energy all these methods are used to analyze a structure, these are just
methods, all of them really give you the same. You might have used the method of minimum
potential energy to analyze the displacements of a structure; you have used the Castigliano
principle to get the displacements of a structure. What is so good about the method of virtual
work? The reason why I am going to use the method of virtual work while I teach you structural
analysis of statically determinate structures is because all the other methods are essentially
mathematical principles where you need to differentiate something, you need to integrate
something it is all mathematical, whereas I like the method of virtual work essentially because
it is very very physical.
I apply a force, I find out those forces and then I use the virtual work principle which says that
the total work done by displacements and forces in a structure under equilibrium is 0, the virtual
work done is 0. In other words, when I use the method of virtual displacement, I actually
displace the structure (physically displace the structure), find out all the displacements and then I
multiply all the virtual displacements by the corresponding real forces acting at those points and
I have got my work. It is a very physical way of getting what you have here.
Similarly, the method of virtual force to define geometry; you have so many different methods:
the Castigliano principle, the conjugate beam, so many different ways. Why are we using the
method of virtual force? We are using the method of virtual force to essentially use equilibrium.
All the other methods that you see are either geometry or mathematics. In this one, you actually
apply a virtual force, find out all the forces in the structure and once you have found out the
forces in the structure, you multiply all the virtual forces by the real displacements and
deformations and put that equal to 0 and you have the unknown displacement that you need. It is
a very physical way of obtaining the forces and that is the reason why I am going to be using the
method of virtual work for the rest of the semester to solve all problems and that is the reason
why I have spent one lecture looking at how to use the method of virtual work. Of course, today,
I just used it for a very very simple problem; we will use the same methods to solve very very
complicated problems later. Today was just an illustration.
This brings me to the end of the review section of this particular course. I have reviewed how to
get the static indeterminacy and I have reviewed how to get the kinematic indeterminacy of a
structure. Therefore, given a structure, you should now be very comfortable in finding out the
static indeterminacy and kinematic indeterminacy of a structure. I recommend that you go back
and look at any book on structural analysis, pick up a few problems, look at them and try to get
the static and kinematic indeterminacy for those structures. Once you gain confidence, you will
see that it is really trivial. Please remember: use the algorithmic way of getting the static and
kinematic indeterminacy, which I have discussed over the last two lectures.
Today, I reviewed the method of virtual work which all of you are familiar with and all of you
have used it is just one method that you have used to get the displacements and forces in the
structure. I am using it here because I am going to be using it over and over again for the rest of
the semester. I think I am going to be ending here today, because I would like to end the review
module here. From next lecture onwards, we are going to be getting into the basic methods of
analysis. Since I have a little bit of time today, I am going to look at the methods that I am going
to be using.
First, I am going to be looking at the force method. Initially, I will be looking at the basic force
method, which by now you should be familiar with, but I will still look at it because I am going
to be using the method of virtual work to solve the force method problems that will be the new
way. Then, I am going to be introducing to you the matrix approach. Ultimately, the matrix
approach has several advantages with the standard approach of the force method; so I am going
to introduce you to the matrix approach of the force method. I will be using the force method to
10
solve truss problems, to solve beam problems and to solve frame problems. We are going to
restrict most of the problems to planar problems because ultimately, I will be using them for
illustration purposes but all the things that I talk about are going to remain exactly valid even for
space problems or space structures.
I am going to first look at the basic force method. Then, I am going to be looking at the matrix
approach of the force method; in that, I am going to be using the method of virtual forces to
develop the equations in the force method. After that, I am going to be looking at the
displacement method for analyzing structures. In the displacement method, I am going to be
using the method of virtual displacement to set up the equations that we are going to be solving
in the displacement method. This is the reason why I have spent today on the method of virtual
work: it is going to be used over and over again.
Again, in the displacement method, I am going to look at the basic formulation and then I am
going into the matrix approach, which is essentially referred to as the stiffness method; the
matrix approach for the displacement method is known as the stiffness method. We are going to
be looking first at the basic force method, matrix analysis of the force method, then we are going
to be looking at the basic displacement method, then we are going to be looking at the matrix
approach for the displacement method (the stiffness method) and that in essence will give you
the overview of the basic methods that can be coded in the computer to solve complicated
structural analysis problems. After that, of course, I will be looking at a simpler approach known
as moment distribution which is on how to solve complicated structures using hand calculations.
Finally, as I said, we will be looking at influence lines, how to obtain the maximum loads, design
loads and structures under moving loads.
Thank you very much.
11
Prof. P. Banerjee
Lecture - 04
Good morning everybody. In the last three lectures we have looked at the tools that we would
require to do structural analysis and that is, namely, finding out the static indeterminacy, the
kinematic indeterminacy or degrees of freedom and then we looked at the virtual work method
and the two principles that I used to solve various analysis problems, for example, finding out
displacements and forces in a structure. Today, we are going to be starting off with what is
known as the forced method for analyzing structures. I am going to start off by looking at a truss
structure to begin with and then we will move on to beams and later frames. So let us look at a
simple truss.
Let us say that the structure is subjected to this kind of loading. This is the loading on the
structure (Refer Slide Time: 2:52) and this is the truss structure we have. Understand that in the
force method, the starting point is to find out the static indeterminacy of the structure. Let us go
through the steps; number of members 1, 2, 3, 4, 5, 6 so 6 members, number of reactions 1, 2, 3;
number of joints 1, 2, 3, 4 . Static indeterminacy it is equal to number of unknowns; one for each
member, plus the number of reactions minus the number of equations of equilibrium that you
have which is 2 per joint and this is equal to 1. So this is a single indeterminate structure. It is a
truss which has a single degree of indeterminacy. Now we go into the steps of how to use the
force method to analyze. Let us state the problem. The problem is, I want to find out the forces in
all the members plus I also want to find out the horizontal displacement of this point under this
loading (Refer Slide Time: 4:48). So thats my problem statement: Evaluate the forces in all the
members and find out this displacement. We will find out the displacement later, first let us find
1
out all the forces in all the members. Obviously, since it is a statically indeterminate structure
you cannot find out all the forces in all the members using just the pure equations of equilibrium
since it is statically indeterminate to the first degree what you have is, the number of equations of
equilibrium is one less than the number of unknowns that you have. So essentially if we can
establish one other independent equation then we can find out all the unknown forces. That is
essentially the crux of the force method.
How do we go about finding it out? The first thing that you do in the force method is, you find
out the statically determinate base structure. In other words, you take the structure, eliminate one
of the unknown forces because it is a single degree of indeterminacy, so you have to eliminate
one of the unknown forces then what happens is you have a Statically determinate structure. If
you look at it in this particular case we have cut this member. Once we cut this member, we
know that the force in that member is zero. How many equations of equilibrium do we have? We
still have the same number of joints, so the number of equilibrium equations is still going to be 2
into 4.
Only thing is that the number of members has been reduced by one because one of the members
we know is going to have zero force so it becomes a statically determinate structure. Let us take
this problem, let us solve this problem and to that we take the same statically determinate
structure. In reality this problem is really just a super position of these two problems, let me just
number them here; a b c d, a b c d, a b c d. What we are saying is; that from the original structure
we get the statically determinate base structure where the member bd has been cut so that there is
no force in the member. And we say that this original structure actually can be solved. You can
find out the forces in these structures, by finding out the force in these structures plus the force in
all the members and reactions due to the unknown force in member bd.
What is the unknown force? We dont know what the unknown force is. So we assume that it is
an unknown X1. If you look at this, there is no force here, we have the force that bd actually has
so if you add all of them up you will see that this is exactly this. In other words the redundant
force, for this structure is force in member bd which is an unknown value X1. But suppose we
knew X1, then we could solve this problem. This becomes a statically determinate structure
because we know the force in one member. This is the basic idea behind the force method. You
take the actual statically indeterminate structure; make it into a statically determinate base
structure by, identifying certain redundant forces and putting them equal to zero in the base
structure. Once you have the base structure, then you add to the base structure the structure being
subjected to the redundant force. If you add the two of them together you got the original
structure. That is the basis of the force method. However this is still an unknown, how do you
find out this? Let us see how we solve that. Let us look at the base structure firstly and analyze it.
Assume that all six members have the same EA values.
I am looking at only the base structure. The statically determinate, where the redundant force is
zero; we will analyze that structure. Note that I am going to keep this member there, because the
members still exists. It is very important to understand that the member still exists. It is just that
in a statically determinate structure, we have assumed that the member force is zero. Let us now
see the steps. This is a, b, c, d. I have 10, 10, 10, and today I am going to actually analyze all the
structures. We are going to actually find out all the member forces by actually solving it.
The first step is to find forces in members using equilibrium. I am going to using the method of
joints. Let us first find out the unknown reactions. I am going to just find out the reactions
without actually writing down all the equations; I leave it up to you to write down the equations
because by now you will already have taken a course in basic structural analysis and finding out
forces in trusses, beams and frames are things that you are expected to know. So, I shall not
actually go around specifically solving statically determinate structure and how to find out the
forces in them; I will make an assumption that you know all of those. However in this particular
case I will actually go through the steps without actually writing the steps down. The first and
foremost, these are the unknown reactions. Sigma Fx is equal to 0 of the entire structure; that
gives me that this is equal to -10. The direction that I am showing, is actually in the opposite
direction 10; I am putting down -10 because I have shown it in this direction. That gives me
Sigma Fx equal to 0.
Sigma Fy is equal to 0 gives me that this plus this is equal to 20. However, that still does not help
us. I am going to take moments about point A of all the forces and from that I can find out this
value. What are the forces? This does not have any moment, this has 10 into (by the way
remember that this is four and this is three meters) what you have over here, this is going to be
10x4 clockwise plus 10 into 3 anti-clockwise divided by 4, because this clockwise, this
clockwise, this nothing, this nothing, this one anti-clockwise. This is going to be equal to 70 by 4
value because 10 into 3 plus10 into 4 is equal to 70, 70 divided by 4 is 70 by 4. Since this plus
this is equal to 20, this is going to be equal to 10 by 4 which is equal to 2.5 Kilonewton and this
3
is equal to 17.5 Kilonewton. Now we have got the reactions for the support. Now I can use the
method of joints to find out all the forces in all the members.
Let me start of with the simple one; member B. Note that this member although it exists I know
this force is zero in this. If I take Sigma Fx is equal to 0 of member B you will see that this is
going to be is equal to 10 Kilonewtons and similarly if I take Fy I will get this is equal to 10
Kilonewtons. I have found out the forces in these two members and member bc I have a
compressive force of 10 Kilonewtons, in member ab I have a compressive force of 10
Kilonewtons. Now, I go to member joint A. What is the force in this member? The force in this
member is going to be equal to if I take Sigma Fx is equal to 0, I am going to get force in ad
plus the horizontal component of the force in ac plus the member in ad is going to be is equal to
10. Similarly, you will see that the vertical component of this member ac plus this 10 (Refer
Slide Time: 17:58) which is downwards, minus this 2.5 which is this way.
Once you put it all together, here you will see that Sigma Fx is equal to 0 gives me this as more
force and if I find out Sigma Fy is equal to 0, you will see that this is equal to 70 by 4
Kilonewtons. This is zero and from that we can find out that this becomes 30 by 4 and 40 by 4
and so this is equal to a tension force of 50 by 4 Kilonewtons. If you look at this particular
situation, you will see that the equilibrium of this joint is satisfied automatically and you will see
also the vertical of this is satisfied. The forces Fab is equal to10 Kilonewtons; this is compressive.
Fbc is equal to10 Kilonewtons compressive. Fac is equal to12.5 Kilonewtons tensile. Fcd is equal
to17.5 Kilonewtons compressive and member ad is 0 Kilonewtons, no force. Note that I also
have member Fbd since I have cut it I know that is going to be 0 Kilonewtons. This is not the true
forces in the Original structure, it is in this structure (Refer Slide Time: 20:36) .
These are the forces that I found out for this particular structure which is the statically
determinate base structure. I have found out the forces in all the members. What is the next step?
The next step is actually to find out the deformations in all the members. What are the
4
deformations in each member? Please look at it; these are all axial forces. So the deformations in
the member i you can find out, is going to be equal to Fi - Force divided by A; that is Stress
divided by E which gives me the strain. The strain is uniform, therefore integrated over the
whole length gives me the axial deformation. If I rewrite this, this comes out as FiLi (EA)i. Force
in the member, axial force in the member, length of the member, the axial rigidity of the
member. This way I can find out for each one what the corresponding delta is going to be equal
to.
Let me write down those values. Given these forces I can write down that: deltaab is going ot
equal to 10 (Note I am going to put minus because tension is taken to be positive and since its
compressive is going to be 10) multiplied by Li what is Li; Lab? Lets, look at the length of Lab,
Lab is equal to 3 meters divided by (Note I have said all EA are the same ) So deltaab is equal to
minus 30 upon EA. Let us look at deltabc this is also is 10 Ton bc, length is 4 meters, so this is
going to be equal to minus 10 into 4 minus 40 upon EA.
Let us look at deltaac. Deltaac Force is 12.5 tension, so its 12.5 tension, what is the length? Length
is ac. So this 4, and this 3, and this is 90 degree, and this is a hypotenuse 5. Total length of ac is
5. So it is going to be 12.5 EA this is plus, so it is going to be plus 62.5 upon EA that is deltaac.
Deltacd, the force is 17.5 compression and the length is 3, so we can put it down as minus 17.5
multiplied by 3 divided by EA, so this is minus 52.5 EA. Finally deltaad and deltabd what are they
equal to? Since the Forces are 0, 0 so the deformations are going to be zero.
Let me ask you this question, this is something that we always tend to forget about when doing
structural analysis - Note that you have to have dimensional consistency. Here, I have said that
all the EA values are the same; I have not explicitly stated the EA value. What are the units of E?
Units of E are Newton per meter squared. They are normally given in Newton per meter squared;
force per area; that is the value of E because that is stress by strain. Stress is Newton per meter
square, strain does not have any units, then the units of area is in units of meter squared. What is
5
EA in terms of? It is in terms of Newton. What is force in terms of? Well I have given in
Kilonewtons over here but kilonewtons is nothing but 10 to the power of 3 Newtons. So the basic
units is Newton and the units of length is meter. When I write down delta is equal to FL upon EA
let us see what I am doing actually Newton meter upon Newton meter units consistent. Delta
which is the displacement or deformation has to be in meters.
We have dimensional consistency and that is very important to know when you actually solve a
problem. Suppose you are given E in the standard form in SI units which is GPA or giga Pascals
which is 10 to the power of 9 Pascals, and Pascals is Newton per meter squared. When you say
that some value is given in giga Pascals, you have to multiply by 10 to the power of 9 to bring it
into Newton per meter squared. If you are given a force which is in kilonewtons you have to
multiply by 10 to the power of 3 to bring it into Newtons. You need to bring all the parameters to
the same basic units otherwise you are going to always have a problem.
However, here I am just trying to state that you need to bring everything down if you want to
make F in Kilonewtons. Then EA also should be in kilo Newtons. If you make this area into
millimeter squared then E also needs to be in Newton per millimeter squared and your delta will
be in millimeter. As long as you are consistent you should not have a problem with units. I have
found out the deformations in all the members. What are these deformations due to? These
deformations are due to the loads that are applied on the structure. What is the next step? Let us
first go through all the steps and then I will put it all together. The next step is actually finding
out the forces in all the members due to the redundant force.
I know that it will always be in terms of X1, so I can find out the coefficients by actually putting
X1is equal to1 and then solving the problem. I am going to solve this problem now. This is
(Refer Slide Time: 29:10) actually this plus this so for this I have found out the forces and
deformations, for this the second one I need to now find out the forces. X1 is equal to 1 and due
to this I need to find out what the forces are. Remember that this is still 4 meters, it is still the
original structure. What about the reactions? This is a force that I am applying in a member. It is
an internal force so it is equal and opposite. Therefore there is no net force acting on the structure
and therefore all the support reactions are actually 0. Sigma Fx gives me this is equal to zero;
sigma Fy gives me this plus is this equal to zero. Then, I take moments about this point. Since
these two are acting equal and opposite at the same point this is going to be 0 and this is going to
be 0.
Now I need to find out, what the forces in all the members are? Let me first take joint b. In joint
b what is the force in this member? The horizontal component of this plus the force in this
member is going to be equal to 0. Then the vertical component of this plus the force in ab is
going to be equal to 0. From this I can find out that these are going to be equal to 4 upon 5, this is
going to be equal to 3 upon 5. Let us look at the other force; if you look at this particular
situation, what is the force? I know this force. What are the two unknowns? I do not know the
force in this member and this member.
Sigma Fx is equal to 0, Sigma Fy is equal to 0. Sigma Fx is equal to 0 will give me that the force
in this member, the horizontal component of this plus this is going to be equal to zero. If you do
that you will see that this works out to 1. Once you do that then you can find out what is the
force in this member because the vertical component of this is going to be equal to this. I am just
writing down the values, you should be able to find them out. What about the force in this
member? You will see that this will turn out to be 4 by 5. I found out the forces in all the
members so let me write them down.
Force in ab is equal 3 by 5 X1(X1 is equal to one and this is compressive; assuming X1 to be

tensile)
Fbc is equal to 4 by 5 X1 (compression)
Fac equal to X1 (tension)
memebrcd equal to 3 by 5 X1(compression)
mamberad equal to 4 by 5 X1 (compression)
memberbd equal to X1 (the one we will apply)
We have found out the forces in all the members due to an unknown force X1 . What would be
the force in the original structure?
In the original structure

Fab would be equal to 10 plus 3 by 5 X1
Fbc equal to 10 plus 4 by 5 X1
Fac would be 12.5 plus X1
Fcd would be 7.5 plus 3 by 5 X1
Fad would be 4 by 5 X1
Fbd would be X1 with the corresponding signs.
Essentially you see, I can find out the forces provided I know the value of X1. How do I find out
the value of X1? I have found out the forces. In the force method there is the additional equation
to establish the value of X1. Let us look at the original problem.
If you look at this, the original problem, what we have done is that we say that this is equal to
this plus this. The only problem here is that everything is fine except that we do not know the
value of X1. We need to an equation find out the value of X1. I made a cut here; under the action
of these forces what is going to happen to the cut? You see, all the joints will displace, the
members will deform and because of all of that what is going to happen? If this member moves
along this direction relative to this, which is bound to happen because they all move independent
of each other, there is going to become a separation. There is going to be an opening or an
overlap depending on how b and d move. If d moves closer to b there would be an overlap if d
moves away from b there is going to be an opening. But the major point is that I cannot stop
once I made a cut from these two points which actually in the real structure are the same point.
But once I made a cut and I let the structure deform, these two points are going to get separated.
Does that happen in the real structure? No, in the real structure any point over here is not going
to separate they are going to move whichever way, they are not going to separate. What is going
to happen? Let us have a look at the second point. When I apply X1 what do I do? Note that these
two joints, I have applied a force X1 across the joint. Since these two are independent points
when I apply X1, they are likely to overlap (assuming X1 is positive). When X1 is negative they
will pull apart. In both these structures what is happening? The points which I have over here,
are overlapping and it is opening up, depending on which is positive and which is negative. But
in the real structure does it happen? Do the points overlap? Does any point overlap with itself? It
cannot since it is a continuous member. In reality, in this structure, there is no overlap or
shortening. Do you understand what we get? We get what is known as a compatibility condition.
What we do is, we find out how much this has opened by and then add the two and the two
together have to add up to zero because the point cannot go anywhere. The point is stuck to each
other. So that is the independent equation.
We need to find out how much displacement I get at this point and we need to find out how
much displacement I get this point; add the two things up and put it equal to zero because this
9
point cannot go anywhere; they cannot separate or overlap in the real structure and therefore that
gives us the additional equation that we are interested in. Essentially, we have to find out how
much this has opened by under this situation and how much this closes by under this situation.
How do I do this? I need to find out a displacement. How do I find out a displacement using the
method of virtual work, use the method of virtual forces. Actually, I am trying to find out the
displacement at this point relative to each other. I actually apply a virtual force at this point and
compute the work done and the total work done has to be equal to zero; that gives me the
displacement. What is that force? Note that the force is exactly the same as a redundant force.
When I actually solved this, I can assume, if I take this to be the virtual force, what will be the
forces in all the members? I know the forces I am going to have, so if you look at it using the
method of virtual force, what is the displacement at this point? What is the displacement at this
point? The displacement at this point can be given by the virtual work equation.
External virtual work done by all the external forces:
10
This is the real structure, I am trying to find out the displacement at this point because of all
these forces. I need to find out the real displacements and real deformations. What are the
displacements? Real displacement corresponding to any force here. Real displacement is going
to be equal to all these forces. What is the work done by these forces? The work done by these
forces are equal to (Lets go back).
11
The virtual forces undergoing the real displacements; what are the virtual forces here? The only
virtual external forces are these reactions which are all zero. There are no external virtual forces,
so the work done by the external virtual forces is equal to zero because there are no external
virtual forces. Let us look at the internal virtual work done. The internal virtual work done is
how much? The virtual force here is this reaction, ab so I need to find out what is that ab. What
is the force in ab? 3 by 5. So 3 by 5 multiplied by the real deformation in member ab. What is the
real deformation in member ab? Remember we computed the real deformation. So I am going to
put 3 by 5 multiplied by minus 30 by EA that gives me the work done by the virtual force in ab
which is 3 by 5. Note that this is compressive, so the compressive force is going to be given by
minus. Undergoing the real deformation in member ab which is minus 30 by 9.
The internal virtual work contribution by ab is going to be minus 3 by 5 into minus 30 by 90.
That is the virtual work done by the real deformation in ab due to the loads undergoing the
virtual force in the system. This negative implies that there is a shortening, this negative implies
there is a compressive force. Let us go from ab to the next member which is bc. What is the
virtual force in bc? The virtual force in bc is 4 by 5 X1 compressive. If we put a unit force it is
going to be 4 by 5. That is going to be compressive minus 4 by 5 multiplied by the real
displacement which is minus 40 upon EA. Then I am going to add the next Force ac tension 1. It
is going to be plus, multiplied by the real deformation in ac which is 62 point 5 plus so this is 62
point 5 upon EA. That is member ac. Next is cd; what is the force? 3 by 5 into compression, so
the virtual force is minus 3 by 5. What is the real deformation in cd? The real deformation in cd
is 52 point 5 so that is minus 52 point 5 by EA.
Then we go to next, I have done ab, bc, cd, next is ad. What is it? The virtual force is minus 4 by
5. What is the real deformation? 0; so multiplied by 0. Then I need to look at bd which is 1
tension, so it is going to be plus1. What is the real deformation in bd? 0. What is the total internal
virtual work? You will see that is going to be equal to plus 18 upon EA plus 32 upon EA plus 62
point 5 upon EA plus 10 point 5 into 3 is 31 point 5EA plus 0 upon EA plus 0 upon EA so tha
12
total virtual internal work done is equal to 50 plus 94 144 upon EA; That is the internal virtual
work done by all the forces in all the members.
Note that the external virtual work done by all the forces is zero but we have neglected one, what
is that? This is also an external work and what is the external work done by this force? This force
is equal to 1 into. Note, what we are trying to find out here, we are trying to find this overlap
the delta. That is the reason why we applied the force X1. Therefore 1 into this delta is the
external virtual work. Once we put external virtual work is equal to internal virtual work; that is
our principle of virtual work because the total work done is zero we get delta equal to 144 upon
EA.
Note that since we have applied this force what we are actually computing is the opening up of
the system and therefore the opening up is equal to 144 upon EA. The next step now is to find
out what. We found out how much this has opened up by. We need to find out what is the
displacement at this point. What is the real load? The real load is this. To find out the overlap
what do we need? We need to apply the same virtual force. Once we have found out the delta
here due to force X1. Note that these are now the real forces, so for these forces we need to find
out what is the displacement. Note I have already done this; the real deformation is equal to
deltaab is equal to 3 into 3 9 by 5EA X1,deltabc is equal to 16 by 5EA X1,deltaac is equal to 5 upon
EA X1. Note this is negative, and negative because this is compression and this is tension, so this
is positive. deltacd is equal to 3 by 5 X1; into the length of cd which is 3 meters so this is going to
be equal to minus 9 by 5EA X1 deltaad is equal to 4 by 5 compression, so this is going to be
equal to minus 16 by 5EA X1.
And finally deltabd is equal to 5 upon EA X1. These are the deformations due to these. Now to
find out the displacement using the same virtual work principle you will get displacement is
equal to external virtual work all outside excepting for this, is going to be equal to one into delta
due to x. This original one (Refer Slide Time: 50:45) is delta due to load. This is delta equal to x.
The internal virtual work is going to be equal to; if you look at this the displacement multiplied
by this, we are going to get
13
27 upon 25EA X1, plus delta bc is going to be 64 upon 25EA X1, plus 5 upon EA X1 , into 1 that
is going to be 5 into X1, then CD is going to be equal to 27 upon 25EA X1, AD is going to be 16
by 6 into 4 by 5 64 upon 25 X1, plus 5 upon EA X1 into 1; this is by internal virtual work.
If I add all of those up I get 27 plus 64 that is 91, 91 and there are 2, 182, 182 upon 25 182 plus
250 is 432 upon 25EA X1. So, if you equate the two you get deltax is equal to 432 upon 25EA X1.
Now I have computed deltaL, deltax. What is the actual delta?
14
This going to be super position of the two because if you look at this, it is going to be the
displacement in this plus the displacement in this and delta is going to be equal to 144 upon EA
plus 432 upon 25EA X1. What is the actual delta? In the real thing delta is equal to 0. From this
you can find out X1 is equal to minus144 into 25 upon 432 equal to minus 25 by 3 Kilonewtons. I
have found out the value of X1, the force and once I find that out I can find out the force in all the
members because all I need to do is this plus this now I know X1 so I just need to find out add of
all of these up and I have got my X1.
15
This in essence is the force method and I hope that with this simple illustration I have been able
to illustrate the basic concept of the force method as applied to a single static indeterminacy
truss. Thank you.
16
Prof. P. Banerjee
Lecture 05
Good morning. In the last lecture, we looked at the application of the force method to analyze
statically indeterminate truss. It was a simple truss with a single degree of indeterminacy; as we
go along, we will run into more and more complicated problems. Therefore, I am always going
to start with a simple problem. Last time, we looked at a simple problem at how to use the force
method and the basic concept of the force method is what I established. Let us write that down
again. What was the basic idea behind the force method?
The basic idea behind the force method can be written down in the following manner: one,
determine static indeterminacy; two, identify redundant forces. Now, how many redundant
forces? Whatever is your static indeterminacy, you need to identify that many redundant forces.
What do you mean by redundant forces? In other words, we are going to define a base statically
determinate structure in which the forces corresponding to the redundants is equal to 0. That is
the reason why we identify the redundant forces, so that we can define a base static determinate
structure in which all the forces corresponding to the redundants is equal to 0. What is the next
step? The next step is to find the displacements corresponding to redundant forces in the base
structure. Define the base statically determinate structure; find the displacements corresponding
to the redundant forces in the base structure. What you do is, you define the base statically
determinate structure, find the displacements corresponding to the redundant forces in the base
structure, find the displacements due to redundant forces only and step six apply compatibility
condition. These six steps totally define the force method and that is what we have done in the
previous lecture. If you remember, let me quickly review the previous lecture.
This was our base structure. We identified the static indeterminacy and we identified that the
redundant force was the force in member bd, which we called as X1. Therefore, this was the base
structure where we had the loads and the force on bd was equal to 0. Then, we added the
structure where the force due to X1 was given and we found out the forces; we essentially found
the displacements corresponding to the redundant forces in the base structure. We found out the
displacement here due to these forces; find the displacements due to redundant forces only. We
found out the displacements in this structure due to the redundant force and then we applied the
compatibility condition (Refer Slide Time: 06:27). What was the compatibility condition?
The compatibility condition was that the actual delta due to the loads alone, due to the redundant,
was equal to 0. From that, we solved and we got the redundant forces. That is the force method.
In this lecture, I am going to be applying the force method to a beam. Understand that you have
to be able to find out the displacements; to find out the displacements in this structural analysis
that I am going to be doing, remember I said to find displacements I will always use the virtual
force method. Therefore, the whole process over here is based on the virtual work principle.
Remember I said that I was consistently going to use only virtual work. Where do we use virtual
work? We use virtual work to find out the displacements. That is the overall concept that I am
going to be using over and over again.
Let us look at a statically indeterminate beam. Let me put a loading here; let us say that these are
10 meters and this load is 10 Kilonewtons. This is a two-span beam; each span is 10 meters and
the load is applied at the center of the span; this is 5 meters; this is a, b, c. Think of it as a twospan bridge supported on two abutments and a central pier. We have a static load of 10
Kilonewtons applied at the center of span ab and we have to find out the bending moment
diagram for this structure. That is what we have to do in this particular case. We have to find out
the bending moment diagram for this.
What is the static indeterminacy of this? We have done this over and over again. The static
indeterminacy, if you look at it, there are four reactions, three equations; so static indeterminacy
is 1. What is the base structure? I will make this the base structure. I will identify this force as
my redundant. Since it is only single indeterminacy, I need to identify one and I will take this
one. You could have taken any one of them; it is up to you. The only thing is that you cannot
take a redundant that will make the structure unstable; that is all. You have to maintain the
structural stability; as long as your structural stability is maintained, it does not matter how you
define it. You could have taken this (Refer Slide Time: 10:24), you could have taken this, you
could have taken this, any one of the verticals, reactions could have been taken as redundant.
You can do that; it is not a problem.
However, in this particular case, I opt to take the reaction at c as my redundant. Therefore, what
is my base structure? My base structure becomes this because this X1is equal to 0. Remember
that in the base structure, this reaction X1 is equal to 0; of course in this I have the load. This is equal
to this is equal to this plus this (Refer Slide Time: 11:24). Do you agree that this plus this is
equal to this? It is obvious. Therefore, what are the steps here? Understand that this is the
redundant; so, what do I have to do? In this case, I have to find out the displacement X1 due to
the loading; then, I need to find out X1 due to the redundant. What is the actual X1 in this
particular case?
Since the displacement of this point is vertically restrained and is going to be equal to 0, all we
are saying is that our compatibility condition is X1L plus X1X is equal to 0; this is our
compatibility condition. The whole point over here is to find out these two displacements. Once
we find out these two displacements then we add them equal to 0 and we get our equation. That
single equation we can then solve for X1; so, let us go back. We are going to now start solving
the problem. Let us now solve this problem.
Let us look at the base structure. The objective of this problem is to find out the vertical
displacement of this point that is the whole idea behind this. This is a statically determinate
structure and you have to find out the displacement. Simple; I am going to be using the principle
of virtual force. Since I am finding out displacement, I have to use the principle of virtual force.
What are the steps that I go through? Note that the basic assumption over here is that only
flexural deformations; in other words, only flexure is considered, shear deformations are
neglected. Now you know how to compute flexural deformations using the virtual work
principle. We will go through it step by step.
When I have a member which is subjected to a moment, what happens under this effect? Under
this effect, this structure, plane sections remaining plane, it goes something like this; at the
neutral access, it remains the same dimension as it was here (Refer Slide Time: 14:53). The top
part compresses and the bottom part expands; the plane section, this section, remains plane;
original is this. After deformation, it becomes something like this; so, plane section is remaining
plane, the simple beam hypothesis.
What we are interested in essentially is the movement of this plane relative to this plane; we can
actually draw it in this fashion. This plane has not moved; only this plane has moved (Refer Slide
Time: 15:42). Deformation is all relative; this is what we are interested in. If you look at this,
what is the curvature? By definition, curvature is the rate of change of theta per unit length.
Curvature in a lot of places is given as kappa. In simple beam theory, this is given by M upon EI;
curvature is given as M upon EI; that is your curvature. Therefore, if my curvature is this, what is
this? (Refer Slide Time: 16:38) This is length dx. What is d theta? d theta is given by M upon EI
into dx. Given an M and given an EI value, the d theta is equal to M upon EI into dx. That is the
real deformation due to an applied moment M.
If you want to find out the virtual work done by this deformation when undergoing a virtual
moment small m, then the total virtual work done, internal virtual work done over this
infinitesimal area is going to be equal to small m into d theta which is for the infinitesimal length
5
dx; that integrated over the entire length is going to be my work done (Refer Slide Time: 18:01).
Note that since m is in this direction, theta is in this direction, they are positive and d theta is
given this way so this becomes 0 by L mM upon EI into dx. The internal virtual work done due
to flexure is given by the moment at a particular point multiplied by the real curvature, integrated
over the whole length.
If we do that, what do we have to do then? If we are using the virtual force principle, we find out
the bending moment due to the actual load, we find out the bending moment due to the virtual
force applied corresponding to the displacement that you are interested in finding out. Then, you
are going to integrate it over the whole length and that will give you the work done; that you can
equate to whatever external work that you need to do. This is due to flexural deformations only
and in a beam, we only consider flexural deformations; it is fundamental. What we need to do
here is find out the displacement in this particular case. To find out this displacement, what we
do is find out the bending moment due to this force (Refer Slide Time: 19:43). Then, we find out
the bending moment diagram due to
We are trying to find out this displacement and because we are trying to find out this
displacement (Refer Slide Time: 20:00), you have to apply a virtual force corresponding to this
displacement. Once you find out the force due to this displacement, you need to apply a virtual
force, find out the bending moment due to that; then, you need to go through the virtual force
method.
Let us do this process find out the bending moment due to the applied load. The applied load is
10 Kilonewtons. How do you find out the bending moment? In this particular course, I am not
going to actually go through all the steps of finding out bending moment, shear force, axial force
diagrams; never. I am going to assume that you know all of this. If you find it difficult
understanding how I am doing all these finding out the bending moments and shear force
diagrams I recommend that you go back to your first course in structural analysis and review
6
how to get bending moment diagrams. During this lecture, I am going to be just going through
the steps as if you know how to find out the bending moment diagram; I am not going to go
through the steps of finding out the bending moment diagram.
Here, 10; this is going to be 5, 5 (Refer Slide Time: 21:46); and the bending moment diagram is
going to look like this. Note that this is 10 meters; this is going to be equal to 25 Kilonewton
meter and the sense of curvature is going to be in this way. It is my point never to say positive
moment or negative moment; I always draw the bending moment diagram and then show the
sense of the curvature, so that you know what kind of moments are being applied at that
particular point. This is the bending moment diagram for your real load.
What is the next one? We need to find out the bending moment diagram for a virtual load
because we need to find out the displacement at this point. When you find this out, you will see
that this is equal to 1, this is equal to 2 and the bending moment diagram is going to look like
this. This is my M diagram; this is my small m diagram. Now, the internal work done is going to
be equal to integral over ab small mM dx plus integral over small m capital M upon EI into dx.
We will assume that the EI for both spans are the same, but if you look at bc, what is M in bc? M
is equal to 0. This essentially becomes integral; since this M is equal to 0, this becomes only over
ab this integral (Refer Slide Time: 24:31). You already know how to do these integrals when
you have straight lines; you have already done this in your solid mechanics course.
In this particular case, this is going to be 25; this is over ab. Since we are only integrating over
ab, this is my capital M diagram; this is at 5 meters; then, I have my small m diagram. One way
of doing this is finding out the expression from a to midpoint, then another expression for M
from midpoint to b and then taking this also as a function of X, integrating from 0 to 10 and you
can actually get it analytically.
I am going to use the simpler method which says that the area under this curve. Look at this
this is essentially an area under the curve. When you are doing integration of any function over
dx, what are you doing? You are actually finding out the area under this fx curve (Refer Slide
Time: 26:22).
Now, this I am going to divide by EI; so, I will call it the M upon EI diagram. If you look at this
M upon EI diagram, this is the small m diagram and what you are doing is, over m, M upon EI
which is.... This multiplied by this into dx is equal to the area under any one of these curves.
Note, of course, that this has to be separated from 0 to 5 because the expression for m is
different. It is going to be this plus 5 to 10 mM upon EI into dx. This part and this part have to be
done separately; this integral I am going to use is given by the area under any one of these curves
multiplied by the . Let us say for example the area under the M upon EI curve, multiplied by
the m bar value, the m value at the cg of this area. This is how integration is done and if we use
that, you will see that the m upon EI is given in this particular case. This is 0 to upon 25; the area
is 5 into 25 upon EI into half length into base, multiplied by cg; cg is two-thirds of the distance
from here (Refer Slide Time: 28:27); that is at 10 by 3. At 10 by 3, what is the value? Over here,
the value is going to be 10 upon 3; that is for this one.
Next, let us take this area. Note that both are of the same sense; that is why it is positive.
Similarly, if you take this one (Refer Slide Time: 29:00), that is also going to be 5 into 25 upon
EI into half multiplied by cg. The cg is at this point and that is 10 by 3 from here; this value will
be 20 by 3. If you add all of them up, you will see that this is equal to 10 by 3, 10, 5; so, you
are going to get 625 upon EI. That is the work done by the internal flexural deformations. What
is the work done by the external? External is equal to 1 into x1L; that implies that x1L is equal to
625 upon EI. What does that mean? This means that due to this load (Refer Slide Time: 30:14),
this displacement is equal to 625 upon EI.
If you look back at the original problem, we have found this out (Refer Slide Time: 30:28). The
next step is to find this out. How do you find that out? Simple. Due to this load, we find out the
bending moment diagram; then we apply a unit load here and find out the virtual bending
moment diagram. Note that all of it involves this load itself for which we have already found out
the bending moment diagram. I am going to now utilize that.
I am going to apply X1 is equal to 1, find out the work done and then the total displacement is
going to be multiplied by X1, the unknown. For this, we have already got our bending moment
diagram; we have already drawn it. It is like this and this is equal to 10. Again, this is the real
capital M diagram; the M diagram will have 10 into X1 upon EI. That is the bending moment M
upon EI diagram for the actual load X1. Now, for the unit load X1, which is my small m diagram,
this is going to be equal to 10. This is my small m diagram because I am trying to find out the
displacement at the same point.
Once I have that, I am just going to go ahead and calculate the internal virtual work. Internal
virtual work is nothing but This is 10 meters, this is 10 meters; so the internal virtual work is
10X1 upon EI into 10 into half multiplied by the value at this cross section, which is equal to 20
by 3. Again, we do the same thing on this side. You will see that you will get the same 10X1
upon EI into 10; that is the area and this is the moment at cg. What you get here is 2000 by 3EI
into X1 that is the internal virtual work. What is the external virtual work? You have already
got it 1 into X1X. This implies that X1X is equal to 2000 upon 3EI into X1. Therefore, we have
computed this and now we apply the compatibility condition.
What is the compatibility condition going to give us? X1L is equal to 625 upon EI, X1X is equal to
2000 upon 3EI into X1, compatibility condition x1L plus x1X is equal to 0. This implies that 625
by EI plus 2000 upon 3EI into X1 is equal to 0. This implies that X1 is equal to minus 1875 upon
2000, which is equal to minus 0.9375 Kilonewtons. Once I have this, note something very
interesting. How do I get the bending moment diagram for the entire thing?
Note that once I have X1, now I have this structure. I now know X1 is equal to minus 0.9375
Kilonewtons. I have this load. Now, can I find out my reactions? Of course, I can find out; I can
find out everything; I can draw the bending moment diagram. Actually, let me draw this. Due to
10
this load, it is going to be 5, 5 here. When this is minus, it essentially means that the load is in
this direction plus (Refer Slide Time: 36:16).
The load at this point is acting downwards; hat makes sense because when you do this, this tries
to pull it out and this load acts downwards to pull it back there; that is what this means.
Therefore, this is going to be equal to 1.875; this is going to be 6.875 Kilonewtons; this is going
to be 4.0625 Kilonewtons. This going to be the total loading that you have, reactions; once you
have the reactions, you can of course draw the bending moment diagram.
You will see that in this particular case, the bending moment diagram will look like this, with
this equal to 9.375 Kilonewton meters at this point (Refer Slide Time: 37:21). The value over
here is going to be equal to this into 5. That is going to be equal to 20.3125 Kilonewton meters;
this is going to be in this way; this is going to be sagging and this part is going to be hogging.
You have got the bending moment diagram using the force method.
I know that you probably have studied the force method to begin with. However, I have spent
two lectures looking at the basic concepts of the force method just to kind of review if you have
studied; if you have not studied, then there is sufficient detail in the last two lectures that I have
talked about, to give you an overview of how to use the force method to analyze a structure.
When you have flexural deformations, the basic concept remains this if you have the statically
indeterminate structure, define redundant forces. Once you have redundant forces, the base
structure has redundant forces equal to 0. Find out the displacement corresponding to the
redundant force in the base structure, apply this redundant force, find out the displacement of
that particular point and then superpose the two displacements, because after all, the two
structures together form the actual original structure.
Superpose the two displacements and find out what the displacement in the real structure is and
apply the compatibility condition to find out the value of the redundant force. Once you have
found out the value of the redundant force, you have essentially a statically determinate structure
for which you can find out the axial force in the members in a truss or the bending moment
diagram and shear force diagram for a beam structure. Now, I would like to spend a little bit of
time on how to take this forward for a frame. I will just spend 10 minutes talking to you about
how to use this for a frame. Let us take a simple frame.
11
I am not going to solve this problem today because during the next few weeks, we are going to
solve enough number of problems; I want to just introduce the concept to you. Let us say that I
have gravity loading and some wind loading on this structure. What is the redundancy of this
structure? If you look at it, you will get static indeterminacy is 1.
Which one would I take in this particular case? You can take any one of them; it does not matter;
but, I will take this as my redundant, in which case my base structure automatically becomes
this, because this is equal to 0 in the base structure. Let me say F1, F2; so, I have F2, F1. This is
equal to this plus this (Refer Slide Time: 41:59). Therefore, how do we solve the problem? Again
in this case, due to these loads, we find out what is the displacement corresponding to the
redundant. Since this is free to move, it will move. Then, we find out due to this how much this
moves by; that is x1X. In the real structure, how much does this point move by? Nothing; so, x1L
plus x1X is equal to 0.
The point I would like to make here is how do I find out this x1L and x1X in a frame? We have
already done that for a beam. In a frame, how do we do it? Understand that the only difference
between a beam and a frame is that it has a different way of computing bending moment
diagrams; that is all. Otherwise, even in the frame we only consider flexural deformations. Why?
We neglect axial deformations, they are very small; we also neglect shear deformations because
they are small.
After a few lectures, I am going to actually look at the relative value of the axial deformations
and shear deformations in normal situations and show to you that in normal situations, you can
neglect axial and shear deformations; but in some specific kinds of situations, you may not be
able to not consider axial and shear deformations. In those cases, you might have to consider all
three; but then, this is the beauty of the virtual work method because all you need to do is find
out the work done due to the flexural deformations, find out the work done due to the shear
deformations and find out the work done due to axial deformations. Remember work is a scalar;
12
you can add up all the work done by all the deformations and get the total work done. Let us look
at the flexural deformations in the frame; I am just going to go one step. Let us look at what it is
that we are interested in; we are interested in finding out this displacement due to these loads
(Refer Slide Time: 44:51) and this displacement due to this load. Let us see what happens.
This is a statically determinate frame; you should be able to find out the bending moment
diagram for this. How would you find out the bending moment? Find out these reactions. This
reaction sigma Fx is going to be equal to F1 and this, due to this load, is going to be F2 by 2, F2 by
2. Due to F1, let us say that this is 5 meters and so is this (Refer Slide Time: 45:40). Then, this
will become plusF1 and this will become minus F1. These are the reactions; this is this (Refer
Slide Time: 45:54). How will the bending moment diagram look? It is going to look in this
fashion. I am going to draw it any which way I want because I am always going to show the
sense; the sense is going to be in this particular case in this fashion.
What is the value over here? It is going to be equal to 5F1. What is the value over here by the way
of continuity? You have to have 5F1. What is going to happen over here? That depends on the
relative values. What is the bending moment in this section? Note that there is no shear; if there
is no shear, it is here. Therefore, in this particular case, it will probably go something like this
and this. Here, it is going to be in this fashion; here, it is going to be in this fashion (Refer Slide
Time: 46:47). The particular values will depend but you have got your M diagram for the frame,
for the given loading.
Now what is your small m? Put unit load here and find out the bending moment diagram. Note
that in this particular case, this is going to be 0, this is going to be 0 and this is going to be equal
to 1. If you look at the bending moment diagram for the small m diagram, it is going to be 5, 5, 5
with this in this fashion; this is going to be in this fashion and this is going to be in this fashion.
That is your small m diagram. Then, you follow the same procedure to compute mM upon EI dx
13
for each member, add them up and you have your total work done and the external work is going
to be 1 into x1L; from that, you can find out your x1L.
The basic point is that there is no difference between a beam and a frame as long as you consider
only flexural deformations. The only thing that happens is that in a beam, everything is along one
line. In a frame, it is spaced across the plane; so, you have a difference. The only thing is that the
equilibrium conditions look different. Otherwise, once you have the bending moment diagrams,
whether it is a beam or whether it is a frame, it does not really matter. Therefore, beams and
frames are really going to be the same problem and I am not going to treat them in any different
way. It is only that the equilibrium conditions give rise to different sets of equations in a frame as
opposed to a beam; there is nothing else.
For example, in a beam you may not have horizontal forces at a hinge because you normally only
have vertical forces on a beam, whereas in a frame you have both gravity loads as well as
horizontal environmental loads due to wind or earthquake. Therefore, you would have horizontal
forces, reactions at hinges; it actually does not matter one way or the other. After all, if you know
how to compute the bending moment diagram for a beam or a frame, that is all that is required
for you to be able to successfully apply the force method. We shall continue solving problems
etc; but, I want to now introduce the fact that if you look at it, you see this (Refer Slide Time:
50:28). I have a situation where I have a reaction and the displacement corresponding to the
reaction is always 0 because it is fully restrained.
Next lecture onwards, I am going to look at a situation where you have a partial restraint. In
other words, you can develop a reaction, but the displacement in the structure need not
necessarily be equal to 0. How would you treat those kinds of problems? Those are essentially
flexible support conditions that you are aware of. Next time, we are going to look at something
new flexible support. We are going to look at other kinds of situations what happens if a
member is not exactly the size that in its geometry it is supposed to be; that is called a lack of
fit. Then, you may have temperature stresses, erection stresses. How do you analyze the
structure for all those kinds of situations? We shall see how we can apply the force method for
all those kinds of situations over the next few lectures.
Thank you very much and look forward to seeing you in the next lecture.
14
Structural Analysis II
Prof. P. Banerjee
Lecture 06
Good morning. Today, we are going to continue with looking at the application of the force
method for analyzing statically indeterminate structures. Last time, if you remember, I had said
that today we are going to consider flexible supports and see how that can be included in the
force method procedure to be able to analyze the structure. Let us look at what we have.
This is the same truss that we had considered last time, excepting that the roller here has been
replaced by a flexible spring support where the flexibility is given as 1 over EA. EA is the value
that we have for all the members, actual rigidity for all the members; all of them are EA. If you
remember, that is what we had considered last time. I am just taking a flexibility value; I am
taking it to be 1 upon EA. It is not necessary that this has to be 1 upon EA; it can be any
flexibility that I define for this spring. In this particular case, I am defining it as 1 upon EA
because I am going to solve it using 1 upon EA. However, you can use any value of f and you
will see that whatever procedure I develop will be equally valid for that procedure.
How do you solve this problem? The last time we saw how to get the support reactions, what did
we do? By the way, this is a single degree of static indeterminacy problem. This is the same
problem as we dealt with last time; I am not going to go into all those details again. This time, I
am just trying to see how we can consider the effect of additional flexibility in the supports. I
have considered this support to be flexible. When would your support be flexible? In real life,
when would this happen? This would happen in case you had a pier where you considered the
axial deformation of the pier. If you consider the axial deformation of the pier, then you see that
the pier becomes axially flexible since the pier is essentially subjected to vertical axial forces
1
because that is how the load is transferred through the pier; what happens is essentially that we
are now considering the pier to be axially deformable. What we are saying in this particular case
is that the pier flexibility is given as 1 upon EA; that is all we are saying in this particular case.
How do we solve this problem? The problem goes in the same procedure as last time; you have
to develop your statically determinate base structure. I am going to have exactly the same thing
as last time. (Refer Slide Time: 05:07) I am cutting this member so that I get a statically
determinate structure. Then, I need to continue having this and these are the loads. This is my
base structure.
This plus this is equal to this. In essence, there is no difference in the solution procedure just
because you have flexibility.
What you have in this particular case is the following if I look at the base structure. This is my
base statically determinate structure. If I have a statically determinate structure, remember that I
am taking exactly the same thing as last time. The first step in analyzing this structure is to find
out the reactions; and as we saw last time, the reactions are identical. This is going to be 10 and
10 and then due to this, 30 by 4; this is going to have 17.5 Kilonewtons, 2.5 Kilonewtons and
this is going to be 10 Kilonewtons.
If you look at this, this is identical to. Therefore, till now we have not considered any aspect of
this flexibility; we have not included this flexibility into the entire system. However, the most
important point here is that whereas in the previous case this deflection was 0 because it was a
roller support, in this particular case there is going to be a deflection; but, the overall point is that
this deflection does not affect the forces in the structure at all. What it may do is that it may
affect the displacements in the structure. Therefore, the whole point is that even if I solve this
structure and I were to find out what the value of X1 would be, you would see that as far as the
forces in the members are concerned, whether I have a flexible support here or not will not
matter at all. In this particular problem, I am going to try to evaluate something else. Remember I
said last time that we could find out this displacement in addition to finding out all the forces in
all the members. Now if I do that, you see it is not just a question of finding out the forces in the
members and the deformation in the members; I also need to find out what this displacement is
going to be due to these forces (Refer Slide Time: 09:41).
How would I do that? Remember, to find out the displacement at this point due to all these
forces, I need to apply a force there. Now under this system, I am trying to find out two things I
am trying to find out the separation at this point and I am trying to find out the displacement at
this point. Both these are things that I need to find out together and I have the flexible support.
How do I do this? In terms of this particular structure, all I need to do is to find out the forces.
I do not want to go back and do the same thing all over again; I am going to just put the previous
structure that we had done. If you look at this, these are the forces in all the members; what we
need to do now is find out the forces due to these two.
To do that, one is to apply a unit force here and find it out. That is the unit virtual force so that I
can find out what the displacement at this point is in the statically determinate structure (Refer
Slide Time: 11:56). I am applying a unit force and this is identical to what we had solved over
here.
If I put X1 is equal to 1, then Fab is equal to 3 by 5, 4 by 5, 1, 3 by 5, 4 by 5 and 1.We have

already looked at all of these last time; therefore, there is no need to solve this particular
problem; it is identical. To find out the other displacement, what do I need to do? Note that again
in this particular case what I am doing is to find out this displacement, I need to apply a force at
this point. Now let us look at what happens. This is my flexible support. As far as the forces are
concerned, let us see what happens here. If I apply a unit force here, then the reactions, this is
going to be 1; if you look at these reactions, these reactions are going to be equal to 0. What does
that mean? That means that due to this displacement all that we have is a unit force in this and all
others are 0.
This is the virtual force that I am applying here; I am finding out the virtual forces in all the
members and the virtual support reactions that we have. The point that I am trying to make over
here is that you just need to go through the same step procedures to get your displacement. What
do you do? There are two steps; let me write down. This is deltaL. How do you find out deltaL?
deltaL is due to the load alone, which is due to this (Refer Slide Time: 14:15). How do you find
out the deltaL? We have already gone through the steps last time; I am not to go back and do the
same thing. You can find out this deflection and this deflection in this structure by just writing
down the virtual work equations where the real deformations at this point are actually applied.
The real deformations are multiplied by the virtual forces developed due to this and this (Refer
Slide Time: 14:46). Understand that this deltaL that I am finding out is the separation of member
bd due to loading in the base structure; this is the separation of member bd due to loading in the
base structure. How do I find this out? I find out the forces in all the members due to this, the
virtual forces, and multiply them with the real deformations due to the loading. We have already
got this earlier; there is no difference in this particular case.
The next thing that I am going to find out is r1 due to L that is the horizontal displacement due
to loading in the base structure. How do I find this out? I find out the deformations in the
member and then find out all the virtual forces in all the members. If you look at it, what really
5
happens? You will have r1 in the real structure, that is, r1L into 1; that is, the work done by the
external forces is equal to 1; that is, member ad 1 into the deformation of this member deltaad in
the original structure that is due to this loading. We have already found that those out earlier.
What is this delta? This delta is equal to 0. If it is equal to 0, then what is the delta in this? 0.
What is the deflection at this point due to the loading that we have? 0. We can find that out; this
is very easy to find out. This I have found out in the base structure due to the loading.
Now what is the next step? I also need to find out the deltaX. What is deltaX? It is the separation
of member bd due to X1; I can find out this also. What is the real structure? (Refer Slide Time:
17:58) The real structure is the one with X1 applied here. Find out all the displacements and
deformation forces in the members and the deformation in all the members; then take this as the
virtual force and I can find out this deltaX. I have already found this out the last time also. If you
remember, the deltaL that I got last time was this (Refer Slide Time: 18:21), the deltaX that I got
last time was this (Refer Slide Time: 18:27).
Therefore there is another aspect I need to consider, that is r1 due to X which is the horizontal
displacement due to X1. How do I find that out? In that case, my real structure is this structure; I
can find out the forces and the real deformation in all the members and my virtual structure is
this structure (Refer Slide Time: 19:04). I can find out the virtual forces which I have found out
and then I can do the virtual work to get r1X. Therefore, what is the final solution? The final
solution you will see is delta that is the deformation the separation in the original structure is
going to be equal to deltaL plus deltaX which by the way is equal to 0 because there is no
separation. r1 is equal to r1L plus r1X. Note that this gives me like last time, I found out that X1
is equal to minus 25 by 3 Kilonewtons; there is no difference in the flexibility at all.
Once I find out X1, I can find out r1X because r1X will always be in terms of X1 and since I know
X1, I can find this out, add it to this one, which is a number, and I have got my displacement at
this point. Now you may ask in this particular case, how does having the flexible support affect
this entire process? It actually does not affect this at all; it actually does not affect this at all; it
does not affect this at all (Refer Slide Time: 20:27). Does this affect this one? I am not going into
the details; you can actually solve it yourself.
You will see that this one is not going to give anything primarily because if you look at this
problem which we solved, the primary problem is that since this is equal to 0, this is equal to 0;
you see that r1L turns out to be 0 (Refer Slide Time: 21:03). Since r1L turns out to be 0, r1 is only
due to r1X and that is not affected by the flexible support. You may well ask what does the
flexible support do for this particular problem. Actually, the flexible support does not affect all
the things that we have found out. What would the flexible support affect? It would affect, for
example, if you were to find out this displacement.
Exercise: Find out the displacement in this structure due to these loads considering two
situations, one where you have the roller support.
Do you understand the problem? I want you to find out the deflection at this point in this
statically indeterminate structure due to this loading, where this is 4 meters, 3 meters; all of these
members have a flexural rigidity of EA. I also want you to find out this displacement due to the
same set of loads, excepting that the roller support is replaced by a spring whose flexibility
constant is 1 upon EA where EA is the flexibility of all the members; find these displacements
out. I am giving this as an exercise to you so that you can actually apply the principles that we
have been talking about. Find out the displacement at this point due to these loads. This is your
first problem, problem a; then find out this displacement, which is problem b, due to this.
I am going to tell you right upfront that in this particular case, you will see that this displacement
and this displacement are two different values. That means that the flexibility of this support has
an effect on it. Therefore, there are certain responses that are affected by the flexible support;
there are other responses that are not affected by the flexible support. I just gave you two
examples where the flexibility of the support does not affect the response. However, I have also
given you one exercise where you will see that the flexibility of the support does have an effect
on the displacement quantity that I am trying to find out.
Where do we go from here? What we have to do is we have to find out what other effects you
can have in a structure which may affect the static analysis of the structure. Let me look at
another classical situation; I am going to remain with trusses for this particular lecture because I
want to look at all the effects that you may have to consider in addition to the effect of the static
indeterminacy.
Let us look at what I call as lack of fit. What is lack of fit? Let us say that this is the same
structure. I am going to go back to the roller support and I have these loads a, b, c, d; this is
exactly the same structure. This is 4 meters, this is 3 meters; so, member ab is 3 meters, member
cd is 3 meters, member ad is 4 meters, member bc is 4 meters and member ac and bd are 5
meters each; 3, 4, 5 triangle. These are the actual lengths; let me write it down. Length of ab is
9
equal to length of cd is equal to 3 meters; length of ad is equal to length of bc is equal to 4

meters; length of ac is equal to length of bd is equal to 5 meters.
This is the geometry that I have (Refer Slide Time: 27:18). When I give this geometry to
someone, what are we saying? Please make the length of ab 3 meters, length of cd 3 meters,
length of ad 4 meters, bc 4 meters, ac and bd 5 meters; this is what we have given. However,
whenever you are putting together members, how do you do it? You never have a member of
size 3 meters exactly. What you have is probably two members of 10 meters each. What you
have to do is you have to cut them up so that you get exactly the sizes that you have. There is
also the factor that what we call as 10 meter length may not be exactly 10 meter length.
Therefore, there is this aspect that even though we want ab and cd to be 3 meters, ad and bc to be
4 meters and ac and bd to be exactly 5 meters, in reality, when someone actually cuts the
members and puts them, you might have a situation where some members are not exactly the
size that they are supposed to be.
In this particular case, let me take the situation where let us say when we cut Lac, it actually turns
out to be 5.005 meters instead of 5 meters. What does that mean? That essentially means that
there is a plus 5 millimeters lack of fit in member ac. In other words, the member ac is actually 5
millimeters longer than it is supposed to be. What effect does that have on the structure? If it was
a statically determinate structure, it would have no effect on the forces in the members; all that
would happen is that the structure would no longer be exactly a 3, 4, 5 kind of triangle; there
would be some small displacements and the nodes would be shifted slightly. However, there
would be no additional forces in the members due to the lack of fit alone.
However, in a statically indeterminate truss, that is not true. I am now going to show you that
due to this lack of fit there are going to be additional forces in the members. To do that, what I
am going to do is I am going to forget about the fact that you have a loading on the structure.
What I am doing is I am taking the problem as two parts. One part is that I am finding out the
forces in all the members due to the loading alone; I have already done that last time; I am not
going to repeat that here. What I am going to do is. What is the additional effect of this lack of
fit that I have because of the fact that member ac is 5 millimeters longer? That is the only
problem that I am going to be solving here right now.
10
Therefore, we have this situation that member ac is longer. What is going to happen? You have
this structure; no loading on it at all, this is a, b, c, d. The only thing that we are trying to find out
over here is that ac actually has a positive lack of fit. In other words, what we have is, member ac
is 5 millimeters longer than it should be. I can look at this problem in this manner that in the base
structure.
I am going to still solve this problem as two problems; it is still two problems plus. Only thing
to note is that there is no loading; there is no loading. The second problem is . This is the same
problem excepting that in this structure, there are no forces in any member; and this is how you
solve a lack of fit problem. I can say that in this particular structure, deltaac, that is, the lack of fit
turns out to be a deformation in member ac. Do you understand what I am saying? I am saying
that in this structure, which is the base structure, none of the members have any forces in them,
none. However, the lack of fit essentially means that I am treating this problem. Due to some
reason, deltaac, the ac which is supposed to be 5 meters, is actually 5 millimeters longer. This is
the original structure and I am saying that this member ac is longer by 5 millimeters; I can take
that longer as a deformation in the real structure. This is the deformation (Refer Slide Time:
34:24).
Now, I need to get the effect of this deformation. Note that I still have this 3, 4, 5 and this cut
that I have made was in the 3, 4, 5 triangle. Certainly, ac is longer; so, what is going to happen?
Actually, this point cannot go anywhere; to ensure that this is longer, this will have to move
somewhere along this direction. Then it will move 5 millimeters in this direction; because of
that, there is going to be a separation at this point induced due to this. I need to find out what the
separation over here is. How do I find that out? Again, use virtual displacement. If I use virtual
displacement, what do I have? Let us go back.
11
Apply a unit virtual force here. I have already solved this particular problem. What did we get?
We got this.
I am going to just copy that down without actually doing the analysis. This is 4 by 5, this is 3 by
5, this is 4 by 5; all the support reactions are 0, 3 by 5 and 1. These are the virtual forces set up
due to applying this. Now, what do I have to do? I am going to use the virtual work principle; I
am going to say that external virtual work is equal to internal virtual work. What is the external
virtual work? DeltaL. Note that this deltaL that I am writing down is not due to loading; so, I will
call it delta0; it is the separation at this point due to the base structure. The real displacement
12
delta0 multiplied by the virtual force gives me the external virtual work is equal to the work done
by all the real deformations. What are the real deformations? (Refer Slide Time: 37:12) ab is 0,
bc is 0, cd is 0, ad is 0, bd is obviously 0 and ac is plus 5 millimeters.
If you see, what I have to do is . All these virtual forces are multiplied by their corresponding
real deformations; all deformations are 0 excepting in member ac. What is the member ac? plus 5
millimeters. If I want to put it into units of meters, it is 5 into 10 to the power of 3 multiplied
by. What is the force virtual force in the member? 1. This gives me the separation in the real
structure due to. It is plus 5 into 10 to the power of 3 meters or plus 5 millimeters.
What does that mean? (Refer Slide Time: 38:20) Since this is longer, what has happened is that
this has overlapped. We have computed that; I have computed the delta0. Now, what do I need to
compute? I need to compute the delta due to x1. I have already have done that last time; I am
going to just take that down. Due to x1, I need to find out the displacement at this point; this I
have already done last time. If you remember, I had already done it and that deltax turned out to
be 5 upon EA into X1 (Refer Slide Time: 39:13). These were all due to the internal forces.
What did I have? I had the situation where. Just one second, I will compute that for you. The
deltaX is equal to 432 upon 25; this still remains the same (Refer Slide Time: 39:42). Therefore,
the deltaX is equal to 432 upon 25 EA into X1. Therefore, what is the total delta? Total delta is
equal to delta0 plus deltaX equal to 0, which implies that 432 upon 25 EA into X1 plus 5
millimeters is equal to 0. From that, I can find out the X1 value depending on what the value of
E1 is. The main point to note is that X1 is not equal to 0. Do you understand? Once X1 is not
equal to 0, all the member forces will depend on X1 because given X1, all the forces in all the
members can be found out. Understand this I had this situation where I did not have any load;
the only thing that happened was that ac was 5 millimeters longer than it should be; because of
that, you have forces in all the members. Therefore, a lack of fit actually gives rise to additional
forces in the members in a statically indeterminate structure.
What have we considered till now? We have considered how to find out forces in a truss due to
loading; we have done that. We found out the effect of the flexibility of a support on the truss.
We saw that the flexibility of a support does not directly have any effect on the forces in the
members; however, it can affect some of the displacement quantities. The third thing that we
looked at was lack of fit what effect does lack of fit has on a member? We saw by an
illustrative example that a lack of fit in a statically indeterminate truss which is not subjected to
any load whatsoever gives rise to member forces. We have found out using the virtual work
principle, the method of virtual force, how to get the particular values of all the forces due to a
given lack of fit.
Remember, since we are dealing I want to establish this once and for all. Since we are dealing
with linear structures, any effect on the structure can be considered separately. Since the method
of superposition is valid for linear structures, you can just find out everything due to one effect,
then everything due to another effect and just keep adding it up. The sum total of all the effects is
just going to be solving the problem for each individual effect and then adding up to get the final
effect.
13
Therefore, even if I had this situation where I had this case You see this? In this case where I
have a loading in addition to which I have a lack of fit, I can solve this problem in two ways. I
can say forget the lack of fit; I am going to first consider the effect of the loads, find out all the
member forces due to the loads alone. Then, I am going to forget about the loads completely and
I am going to take only the lack of fit. I am going to find out all the member forces due to the
lack of fit alone.
What if I have a situation where all the loads and the lack of fit are there together? If I have a
structure where I have all the loads as well as a lack of fit, I can treat this problem as a
superposition of two problems. One problem is where there is no lack of fit; the member ac is 5
meters and we consider the effect of the loads and analyze the structure for that. Then, I can
solve another problem where there are no loads; there is only the lack of fit; solve that problem.
Then, superpose the effect of the loads and the effect of the lack of fit onto each other. In other
words, I add up all the analysis results that I get for the load alone to all the analysis results that I
get for the lack of fit alone, add the two up and I have got the effect due to both the loading and
the lack of fit together.
This principle of superposition is valid; therefore, you will see that I will put situations where I
consider only one effect. The reason behind that is that if I consider each effect separately and
add up all of them, I am going to get the combined effect anyway. This is of course true only for
linear structures. In this course, we are only looking at linear structures; therefore, we can use the
principle of superposition.
Coming back to where I was, I have considered the case of loading alone. How to find out the
response of a statically indeterminate truss for loading alone? We have done that for flexible
support, we have done that for lack of fit; we have done that. Now, we have to consider another
aspect; if we consider that aspect, we will have considered all the cases that could possibly arise
in a truss.
14
That is called as thermal stress. What are thermal stresses? Thermal stresses are forces developed
in the structure due to a temperature variation in the structure. Let us take this situation that I put
together. In this structure, there is no loading; all the members are perfect. In other words in this
particular case, member ab and cd are all 3 meters, ac and bd are 5 meters, bc and ad are 4
meters; perfect. We have put it all together; we have constructed this truss; we have prepared the
supports and put the truss up. Now, when we built this truss during the night, the temperature
was 10 degree C; we put the truss up; and in the middle of the day, the temperature is 30 degrees.
What effect is this going to have?
First of all, there is this aspect that these are ambient temperatures. There is this entire concept
on how ambient temperatures actually give rise to temperature distributions in the structure. I am
not going to go into how to treat ambient temperatures and go to obtaining what are the actual
temperature variations that the members are subjected to. In this case, all I am going to do is I am
going to consider that due to whatever ambient temperature, member bd is thermally insulated.
What does thermally insulated mean? It means that no temperature change has any effect on
member ab. Member bc is also thermally insulated, member ab is thermally insulated, member
cd is thermally insulated, member ad is thermally insulated. Unfortunately, member ac is not
thermally insulated and it is seeing this additional 20 degrees change. What happens? That
means that the deltat for member ac is plus 20 degrees.
What happens in this case? Think of what happens when a member is subjected to an increased
temperature. What happens? The member elongates. How much does it elongate by? That is
given by the coefficient of thermal expansion. What is the coefficient of thermal expansion? The
coefficient of thermal expansion is alpha. Alpha is the strain that the member is subjected to due
to a 1 degree rise in the temperature. Alpha is the strain in the member due to 1 degree rise in the
temperature, the additional strain, axial strain. (Refer Slide Text: 50:44) This into the total
temperature rise will give me the strain in member ac. What is the deformation on member ac? It
is the strain into L. In other words, deltaac is alpha deltaac Lac.
15
What does thermal stress do? Thermal stress leads to an elongation, a real elongation of the
structure. How will I solve this problem? Do you see any difference between this and the lack of
fit? None. Again, the real deformation in the member due to thermal stresses gives rise to deltaac,
which I can compute given alpha, deltaac and Lac. Once I have this problem, you can treat this
problem exactly as we have done the lack of fit problem. That in essence is your analysis of a
truss for flexible support, lack of fit as well as thermal stresses.
Thank you.
16
Prof. P. Banerjee
Lecture - 07
Good morning. In the last lecture, we looked at how to tackle various kinds of conditions other
than loading in the analysis of statically indeterminate trusses. Today, we are going to
concentrate on the same thing for beams and one of the additional points which I had forgotten to
mention in my last lecture. I looked at support flexibility, lack of fit, thermal stresses; there is
also the consideration of support settlement.
In this particular lecture when I am looking at beams, I will also look at what affects support
flexibility as well as support settlement. These are two different things. A member can be axially
rigid and can displace rigidly; that is settlement. When does that happen? Support flexibility is
where we are considering the fact that a pier can deform axially. However, let us consider the
situation of support settlement. When do we get the situation of support settlement? The pier is
axially rigid; in other words, there is no deformation; it is axially rigid. The only problem is that
the pier support, which is the foundation, is not rigid; the foundation deforms under the load that
the pier puts on the foundation and it settles. If the foundation settles, then the top of the pier also
settles. If the top of the pier settles, then the support of the beam on top of the pier will also move
and that is support settlement. Today, we are going to be looking at beams.
This is the problem that we had looked at last time which was a two-span beam with a load in the
center span of one of them; and we solved how to get the forces in all the members as well as all
the supports. This time, I am going to look at support flexibility. What happens when I have a
flexible support? Let me not consider both of them flexible; let me just consider one of them
flexible. Let me take that this continues to be a roller support, but this is flexible. I have this load
and this is flexible with a spring constant K, which is equal to 1 upon f, the flexibility constant.
This is the problem.
Now, how would the problem change because of this flexibility? Let us see. The problem still
remains the same; in other words, I can solve this problem by considering this as my base
structure; so, I am removing this plus this. Now, I have solved this problem. How is the entire
thing going to be different? The difference to be noted is that here we computed X1L, if you
remember; this is the X1 due to load. Then we computed X1X; and in the last case where this was
this way, we took the compatibility condition is equal to X1L plus X1X is equal to 0, because in
the real original structure, what was the displacement? 0.
However, when you have a flexible support, what happens? This is no longer true (Refer Slide
Time: 06:38); so, what do you have? Let us see what happens. This is under this load. This one
and this one are identical to what we had computed for this case. There, we had put it equal to 0
and computed the value of X1. Here, in the actual structure if I have a load X1 in the support,
what is going to happen to the force in this member? Think about it. You will see that if there is a
force X1 here, this spring is going to be subjected to compression; if it is subjected to
compression, then what happens? This member is going to compress; this point cannot go
anywhere; so, if the member compresses, what is going to happen? This point is going to come
down.
How much is it going to come down by? That depends on what is the deformation in the
member. How do we compute the deformation in the member? If you look at it, the spring is
subjected to a compressive force of X1; and if it is subjected to a compressive force of X1, then
what is the deformation of this? The deformation of the spring is going to be equal to X1 by K
and minus. Why minus? Because if X1 is this way, the deformation is compression; and if
deformation is compression, then what happens to this point is it actually goes down by the value
of deltas.
Therefore, for this particular situation when we have the flexible support, the equation becomes
x1L plus x1X is equal to minus deltas. Why minus? Because x1L and x1X are taken as positive in
this direction; and if you look at deltas, it is actually downwards. Now, I am going to put deltas to
be positive when the spring expands. When the spring expands, it is positive; if the spring
compresses, this is negative. If deltas was positive, this would also be in this direction; so, it is
going to happen that x1L plus x1X will be equal to plus deltas.
If I plug that in and if I take the values that I got from the previous case, what was my x1L equal
to? 625 upon EI. So, this was x1L is equal to . Let me put it on another sheet.
x1L was equal to 625 by EI, x1X was equal to 2000 upon 3 EI into X1 and what is deltas equal to?
minus X1 upon K. Therefore, by putting that into the compatibility condition, which is x1L plus
x1X equal to deltas, what we get over here is 625 upon EI plus 2000 upon 3 EI into X1 is equal to
minus X1 upon K; K is a known factor because it is a spring constant. We know the flexibility;
therefore, we can find out X1 directly because this will become nothing but 625 upon EI plus
2000 upon 3 EI into X1 plus X1 by K is equal to 0 and you can solve this for X1.
3
Note that because of the foundation flexibility, the value of X1 changes. If you look at this, the
value of X1 actually goes down due to the foundation flexibility. That makes sense because if
you really look at it, what is happening here? If you really look at this, this one, if you look at the
deformation pattern, what is going to happen? This is going to go like this and this. In reality,
because of the. Due to the loading, you are going to have a compression in the spring, but due
to the deflection you are going to have an extension in the spring. Therefore, the compression in
the spring is negated a little by the extension and therefore the value of X1 goes down due to the
foundation flexibility. This is how you consider the effect of foundation flexibility. Note that in
this particular case, as opposed to the truss, the foundation flexibility directly has an effect on the
internal forces in the member. The bending moment diagram will change because the support
reaction has changed. Now, let us look at the situation of foundation settlement, how foundation
settlement has an effect.
Let us take the same case. I have looked at foundation flexibility; now we are looking at
foundation settlement. How will foundation settlement come in? Let us look at this. Let us say
that this foundation settles by a delta value and we want to know what that effect is going to be
equal to. How would I solve this problem? This would be equal to. Again, you have to
understand that the entire procedure is based on the same premise plus. Here, what are we
getting? We are getting a fact that. we find out x1L. Note that this x1L is going to be the same as
before; this is going to be 625 upon EI. We are going to find out x1X. This again is going to be
the same; it is going to be 2000 by 3 EI because there is no effect over here. However, when you
write the compatibility condition, what will it be? x1L plus x1X is equal to. Now, instead of 0
what is.? The displacement is this and if you look at this, what is this? (Refer Slide Time:
15:24) This is going to be minus delta because remember I am taking upwards as positive, so this
delta is actually a negative value. It becomes a negative delta and if you really look at it, all that
does is x1L plus x1X plus delta is equal to 0. If you really look at it, this is going to be equal to .
This is going to be 625 by EI plus 2000 by 3 EI into X1 plus delta is equal to 0.
The value of X1 goes up because of the support settlement and that makes sense because due to
the support settlement, this is going to take up a much larger value of reaction because support
settlement causes a larger reaction to come up. So, you see, foundation flexibility and foundation
settlement actually does not affect the solutions of the statically determinate problem themselves.
All that foundation flexibility and foundation settlement both do is that they affect directly the
compatibility conditions that you have.
In the original case, you would have X1L the displacement due to load plus the displacement
due to the redundant force X1 is equal to 0. When you have foundation flexibility, it is not equal
to 0; it is equal to the displacement due to the flexibility of the foundation and that is related to
the force in the foundation itself. When you have foundation settlement, all you have is that the
value of the foundation settlement comes in directly and you can compute it from there
obviously. Therefore, the only aspect of foundation flexibility and foundation settlement or in
support flexibility and support settlement is that your compatibility conditions are affected; the
analysis of the statically indeterminate are not affected. Hence there is so much for foundationrelated matters.
Now, let us look at the other kind of situation where. Today, if you remember, I am always
looking at other kinds of loads, in other words, the effect of what is known as non-load effects.
What are non-load effects? Foundation flexibility or support flexibility, foundation settlement or
support settlement, thermal stresses, lack of fit. I have already talked about lack of fit in trusses
and I am not going to talk about lack of fit in beams because it involves a little bit of a
difference.
For example, a beam being longer does not really affect the stresses in the beam. Think about it.
In this particular case, if this beam instead of being 10 meters was 10.1 meters, what would
happen? The only thing that would happen is that these rollers would just move to the right and
give no forces in the member. In other words, lack of fit does not normally have any effect on
beams because lack of fit has an effect on axial forces; and remember that in beams, axial forces
are not of major consideration normally. That is why lack of fit is not a major issue as far as
beams are concerned; that is why I am not considering that right now.
However, thermal stresses can have a large effect. The only difference is that. Again, uniform
rise in temperature of the member gives rise to axial deformations; the uniform rise in the
member does not really have a tremendous effect on beams, it does have in certain situations; we
will look at such problems. We have got the entire course where we are going to be solving a lot
of problems. What I am trying to do in the beginning is try to lay the theory so that later on we
are going to be solving a lot of problems; and when we solve the problems, that is when you
actually know that when even a lack of fit or uniform rise in temperature could have an effect in
beams and frame structures.
However, there are certain situations where you cannot really say beam, frame, truss; it may be a
combination. So, there are going to be various but those I leave for solution of example
problems directly. Right now, we are looking at, in essence, the theory or how to consider the
various effects. Later on, we will see by solving actual problems how these have an effect in
reality. Let us now look at a situation where you have thermal stresses.
5
I just said that uniform temperature rise has no effect on beams because they essentially lead to
axial deformations and axial deformations directly do not have a large role to play in beams,
because beams essentially are subjected to flexure. So, what kind of thermal stresses would have
an effect on beams? Let us look at that.
Let me start by looking at a simply supported beam. This is a statically determinate beam. In a
statically determinate beam, what kind of thermal stresses would give rise to problems? Let us
see. Let us have a situation where the top of the beam is heated. When can this happen? Let us
think of this as a room (Refer Slide Time: 23:04). A room is typically colder than the outside,
assuming Indian condition summer. If it was winter, then probably the outside temperature
would be less than the inside temperature. However, the point that I am trying to make is that the
delta t at the top and the delta t at the bottom are not equal to each other. In other words, the
temperature differential exists; there is a thermal gradient across the beam. Thermal gradient
what does that mean? The temperature here and the temperature here are not the same. When
there is a thermal gradient setup, what actually happens? Let us look at a particular phase (Refer
Slide Time: 24:11). What happens? If you look at it, before the temperature went up, this was the
phase. Now, the temperature at the top goes up by delta t; so, what will happen? Note that I am
looking at a cross section, so this is the beam (Refer Slide Time: 24:27).
What is going to happen to that beam? This point, because it has been subjected to delta t, it is
going to go alpha delta t top. This is going to be the strain at this point; and what is going to be
the strain at this point? (Refer Slide Time: 24:53) It is going to be alpha delta t bottom. Since
these two are not the same, these two are not going to displace the same; as long as the
temperature gradient is uniform, in other words it is linear from the top to the bottom, you will
see that each one will deform separately and if you look at the neutral axis, the neutral axis has
gone in this fashion, but if you look at the section, it has actually rotated; rotated means there is
a flexure.
If there is a flexure, what is going to happen? If you look at this, delta t was higher than delta b;
this is how the member would go in flexure. The bottom would go in compression, the top in
tension. If the inside temperature was higher than the outside temperature, then the member
would deform in this fashion. So, a differential temperature actually leads to flexure in a simply
supported beam. Not only does loading lead to flexure, differential temperatures also lead to
flexure; and how much is that flexure? Let us look at that. Remember that this is equal to d theta
(Refer Slide Time: 26:54). If this is deltax, then this goes dx, dx; and d theta is going to be equal
to alpha delta tb dx minus because remember that d theta is positive in this direction, that is
why it becomes negative; alpha into delta t top into dx all upon the depth of the beam.
Therefore, d theta by dx the curvature is given by alpha into delta t bottom minus delta t top upon
d. This is the real curvature due to a differential temperature and once you have the real
curvature then how do you find out anything? This is the real curvature. What we did was that
we also evolved that the real curvature was M upon EI when you have loading. When you have
loading, this is M where this is the moment due to the loading; and this is kappa due to that.
Therefore, what does virtual work give us? Virtual work gives us m Kappa dx. Whether that K is
due to M upon EI or whether it is due to differential temperature, it makes not one bit of
difference. This is the virtual moment established due to the virtual force system that we develop
in the structure. Let me now go back to the problem that we were looking at originally.
Let us take this situation. This is the structure and I have a situation where the delta t top is equal
to minus 20 degrees and delta t bottom is plus 20 degrees. Let me put this kind of a situation.
Why have I put this situation? Because if we look at this, what is the delta t at the neutral axis?
You will see it is 0 degrees. In other words, this temperature differential leads to pure flexure
because there is no strain in the neutral axis; that is why I am setting up a pure flexure kind of a
situation. There is no other load and I have to find out the bending moment in this indeterminate
truss due to this temperature differential. Let us see what happens. Did this temperature
differential have any effect as far as forces were concerned? You will see that in a statically
7
determinate, temperature only gives rise to flexure without giving rise to any forces; there are no
forces or bending moments in the structure. However, you will see that in a statically determinate
situation you do have this. Let us try to solve this problem.
Let me put a few things in here. I am going to say that alpha is equal to 1.2 into 10 to the power
of minus 5 millimeter per millimeter per degree centigrade. Note that this is the strain per unit
temperature, that is alpha; this is all I need right now. Let us say that the beam depth is equal to
50 centimeter; that is going to be 500 millimeter, which is equal to 0.5 meter. As long as you use
units consistently, you will never have a problem in getting any solution. I am still keeping the
same thing, so this is 10 meters, 10 meters. Therefore, the problem statement here is due to this
differential temperature where the top is minus 20 degrees and the bottom rises by 20 degrees; in
this case, there is a temperature differential of 40 degrees between the top and the bottom. What
would be the forces in the supports and the bending moment diagram for this statically
indeterminate truss? I am given that the depth of the beam is uniform. In other words, this is the
beam where the beam is 50 centimeter deep. This is my problem statement. How do I solve this
problem?
Again, we go back to the basics. This is my structure. What I need to do is I need to find out how
much this has gone up due to the differential temperature. Note that if this does not go up or
down due to the differential temperature, you will not have any support reaction here, you will
not have any x1; if you do not have any x1, the temperature will not have any effect on the
structure. Therefore, we have to first see whether there is a displacement here due to the
temperature. Then of course, we add up this and put it equal to 0 and you know, I mean all those.
Find out this displacement x1 and the compatibility condition is x1t plus x1x is equal to 0. Note
that if this is equal to 0, this is equal to 0 and this is equal to 0 (Refer Slide Time: 34:10), the
next one is equal to 0; if X1 is equal to 0, there are no forces in the members and there is no
effect of the differential temperature. So, the whole thing hinges on whether there is an x1t due to
the differential temperature; that is our whole focus over here. Let us see what happens. Now, I
have already done this here; what is the effect of the differential temperature?
This is the curvature that happens due to differential temperature. So, if I put that, what is ?
Note that the differential temperature is the same; in other words, all along the bottom you have
plus20 degrees, all along the top you have minus 20 degrees. So, what is the curvature? The
curvature, if you look at this, is uniform from here to here; and what is that curvature value equal
to in this particular case? It is equal to alpha, which is 1.2 into 10 to the power of minus 5
divided by 0.5 meters; that is alpha into delta t. If you look at this, what is this? Delta t bottom is
plus 20 degrees, minus delta t at top, so minus 20 degrees. If you look at this, this is equal to 1.2
into 10 to the power of minus 5 into 40 upon 0.5. If you put that together, you get 96 into 10 to
the power of 5 radians per meter. You can see that the units of alpha are millimeter per
millimeter per degree Centigrade; since millimeter per millimeter does not really have any units,
we can say per degree Centigrade.
Here, we have 40 degree Centigrade; that multiplies, the top becomes nothing, no units, the
bottom is in meters, so what we have is per meter. What is kappa? Kappa is the rate of change of
theta with length; so, kappa is d theta by dx. What is theta given in terms of? Radiance. What is
length given in terms of in this particular case? Meters. So, it is radiance per meters, 96 into 10 to
the power of minus 5.
This is uniform. In other words, if I were to draw the real kappa, it would be uniform across the
board and it would be equal to 96 into 10 to the power of minus 5 radians per meter. If I put a, b,
c, a, b, c, due to the temperature, I have a constant flexure across the entire beam of 96 into 10 to
the power of minus 5 radians per meter. So, does that give an effect? I need to find out what that
leads to in terms of this. What do I do?
Again, using the principle of virtual work, which is really to find out a displacement, I give a
unit; corresponding to the displacement, I give a unit virtual force and I find out the internal
forces. If I look at the internal forces, what do I get? If you look at the curvature, the curvature is
also this way; because the bottom actually increases and therefore, you have a situation where
the top compresses making the flexure positive in this way (Refer Slide Time: 39:02). So, this
one is going to be 10 units; this is my m diagram.
What I need to do is I need to find out x1t. It is going to be 1 into x1t that is the work done.
External virtual work is equal to this and the internal virtual work is going to be over from a to c
m into kappa dx. Remember this m M upon EI; M upon EI is nothing but the real. So, this is
the virtual moment, this is the real flexural deformation and this is integrated over the entire
length. If you put that, this is going to be equal to . What is the moment? Again, integral area
of the moment. if you look at this particular one, the area would be equal to 10 into 10 into
half into 2 multiplied by kappa, which is this value; so, 96 into 10 to the power of minus 5. What
happens here? x1t turns out to be equal to. This 10 plus 10 is 100, you get 96 into 10 to the
power of minus 3 meters, which is equal to 96 millimeters. Note that x1t turns out to be 96
millimeters, which is not a very small displacement. You see, because of the temperature
differential of 40 millimeters, x1t is equal to 96 millimeter. Now how do I find out the forces? I
have to find out x1.
10
Note that I have found out x1t, I need to find out x1x. I have already found out X1x earlier.
Remember? What was the x1x?
x1x is this; this does not change. If I put that in, if I plug that in here, what do I get?
11
I get x1t plus x1x is equal to 0. x1t is 96 into 10 to the power of 5 meters and this is 2000 upon 3 EI
into X1 is equal to 0. From that, you can find out the value of x1. Once you find out the value of
X1, it is a statically determinate structure and if it is a statically determinate structure you can
find out the bending moment diagram. I am not going into that detail because you know you
need to put in the value of EI, etc. to actually solve for it. In essence, I am just coming down to
the fact of how to compute the effects of thermal stresses on beams.
You may say that you have not considered any loading, but I go back again to the same point that
I made earlier: If I have a temperature as well as the loading, how would I do that with both of
them together? I would consider only the loading, then I would only consider only the
temperature, find out all the effects, add the two up, totally add it up algebraic addition and
then I have the effect of all the things together. So, it boils down to the entire fact that you have a
situation where you can find out temperature thermal stresses in beams.
Note that exactly in the same way you can consider frames too. Remember last time, two lectures
ago, I talked to you about beams and frames in terms of new thought processes; there are no new
thought processes because the only difference between a beam and a frame is geometry. A beam
lies along one direction; a frame lies in a plane. That is the only difference between all these.
What I would like to do today is I have developed the theory for the following things.
12
Force method: I have looked at the force method, I have applied the force method to both trusses
and beams; the thought processes are the same. It is: Trusses/Axial - main load transfer - beams
and frames/flexure - main load transfer. This is the essential difference between the two. We
have looked at how to consider the force method for trusses as well as beams and frames. What
are the kinds of things we have looked at? We have looked at external loading which is the most
common situation. Our analysis is normally always for external loading. However, over the last
two lectures, we have looked at a variety of other effects, we have looked at foundation
flexibility.
Again, I use the word support and foundation together. Then, we have looked at support
foundation settlement. We have looked at lack of fit, which is essentially for trusses. We have
looked at thermal stresses and this was separate, uniform for trusses. Why? Because you have to
understand that uniform gives rise to axial. Differential for beams and frames. Why differential?
Because differential is what gives rise to flexure. We have looked at all of these things for the
analysis of statically indeterminate structures.
After these lectures, we should be able to solve using the force method. We should be able to
analyze the forces and displacements in trusses, beams and frames for external loading, support
flexibility, support settlement, lack of fit in a truss and thermal stresses that are uniform in
trusses and differential in beams and frames. This is what we should be able to do. Of course,
you may ask how I solve a particular problem. I am going to spend the next few lectures looking
at a variety of problems so that you will be able to confidently use the force method to solve all
these problems. Over the last few lectures what I have tried to do, in essence, is to lay out the
theoretical basis. From here and out, I am going to be purely applying for example problems.
This is the basic force method. Now I am going to do a couple of problems over the next couple
of lectures.
13
Then, we are going to move into the matrix approach for force method. Why the matrix
approach? Essentially what you are trying to do is, if you are writing down the compatibility
equation, it becomes something like x1L plus x1X is equal to 0. If you really look at this, this is
more like a1 x1 plus a2 x2 is equal to 0; and then you can put a3 x1 plus a4 x2 is equal to 0. If you
really look at this, this can be written in terms of a matrix as a1, a2, a3, a4 into x1, x2 is equal to 0,
0. This is nothing but a matrix that we can solve. What do we need to do? We need to solve for
x1 and x2 and we can solve that using the matrix methods.
That is not the only aspect of the matrix approach. The matrix approach not only uses matrix
methods but it also actually develops the entire procedure using a completely different
theoretical premise. Of course, the basic concepts that we have looked at are on how to compute
deformation, how to use the virtual force system; they all remain the same; it is just that they are
set up in a different way. For trusses, you will see that they are really not very different, but for
beams and frames, you will see the entire basis become quite different. I just wanted to introduce
you today to this particular concept just to lay out the next few lectures that I am going to be
talking about; I am going to spend the next couple of lectures actually looking at some example
problems. We will look at one truss, a more rational truss; up till now what we have been doing
is, we were looking at simple structures to essentially illustrate the method. You may say all this
is fine but when I look at the actual truss, how do I solve that? We will look at that problem. We
will take up a truss problem and incorporate as many things as possible into it so that you can
gain confidence in solving the problems.
Then, I am going to take up a beam problem for an example beam problem and solve it all the
way through. Then, I am going to take a frame problem and solve it all the way through so that at
the end of the next couple of lectures, you are ready to use the force method to solve any truss or
beam frame problems totally by yourself. Then, I am going to move into the matrix method to
introduce you to a different way of solving and we will see quickly that the matrix method
actually is a very very good way of using the force method and very quick way of solving, but
14
the only thing is that it gets algorithmic and you lose sight of the basics. That is the reason why I
have used the force method, to start off with, it in its basic form so that you can look at the
physics of the problem. Later on, in the matrix method we will make it an algorithmic way of
analyzing structures. Thank you very much. I hope you have enjoyed listening to my lecture as
much I have enjoyed delivering it. Look forward to seeing you next time.
Thank you.
15
Prof. P. Banerjee
Lecture 08
In the last few lectures we have looked at how to get the static and kinematic indeterminacy
of a structure. Then we have reviewed the basic concepts of the principle of virtual work.
After that, we have actually looked at applying the virtual force method in getting the
response of statically indeterminate structures using the force method. The whole point is that
you should have already been exposed to the force method. I just spend some time looking,
reviewing and especially using the principle of virtual work to set up the equations.
Today, we are taking up a truss example and going through the entire procedure. I promised
last time that I should be looking at realistic example problems and today that is what I am
going to be doing. I am going to take a realistic problem: a truss problem, then I shall go
through with the entire analysis procedure and get the answers so that you would be able to
understand all the steps that go through in using the force method for a truss example. Next
lecture, I am going to be taking up a beam and a frame example so that we can walk through
not only the procedure itself, but the steps in the procedure, so that you can get whatever it is
that you have been asked to find out. Let us now look at the truss example.
This is a typical truss bridge that you are likely to see if you ever travel by train, on a railway
bridge. Of course what I have done is: typically you have many panels in a railway bridge
and I have just taken four panels 1, 2, 3, 4; each panel is 5 meters length and therefore the
total span of the bridge is 20 meters. If you look at the height of the bridge this is 5 meters;
this 5 meter length is a typical length that you would see in most bridges in India. The only
thing that I have done is have taken the panel the BCFG panel and I have put two diagonals
in it and taken the CDGH panel and I have put two diagonals in it. These are the end portals,
these are the verticals, these are the diagonals and this is the bottom chord and the top chord
(Refer Slide Time: 04:18 min). So BCD is the top chord, AFGHE is the bottom chord and
1
BFCGDH are the verticals. AB and DE are the end rakers and the diagonals are BG, FC, CH
and GD.
You are given that there is a loading. 150 Kilonewtons at F, 200 Kilonewtons at G, 150
Kilonewtons at H. So this is the loading and you are also given that all the members have an
area of 2000 millimeter squared and the E value - this is steel - so it is 200 GPa. The question
is: find the forces in members due to the loading alone and the second part is find the forces
in the members due to a temperature effect only. The temperature affects the top members
AB, BC, CD and DE; these are subjected to plus 25 degree Celsius increase in temperature
and alpha is given as 1.2 into10 to the power of 5 which is a coefficient of thermal expansion
for a steel member. This is the entire problem and this is what you have to find out.
If you look at this again, let us first find out what the static in determinacy of this structure is.
For that let me draw the structure again. We have to find out the static determinacy of truss.
First and foremost, find out the number of members, it is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12.
The number of members is 12. Number of support reactions are 1 2 at the hinge and 3 at the
One at the roller, so total 3. Now we have to find out how many joints; 1, 2, 3, 4, 5, 6, 7, 8.
How many members do you have? 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15; so we have
15 members. I missed the 3 verticals when I was doing the computation; so there are 15
members. Let us look at it. How many unknowns do we have? For each member - one force.
The actual force in the members is unknown and the number of reactions are unknown. The
total number of unknowns are: m plus r which is 15 plus 3 which is equal to 18 and number
of equations are two for each joint, so it is 2 into 8 16. So, the static indeterminacy of this
problem is 2.
You see we are going through all the steps, in that we have first the computation of the static
indeterminacy of this structure. Once we have the static indeterminacy, then we find out the
number of redundants. Redundants are those which if we put equal to 0, then we have a stable
statically determinate structure. Let us look at that. In this particular case, this is my X1 and
this is my X2.
2
These are my members, therefore if I look at my base structure, actually this member and this
member are the same size because the panel size of this and this is the same. So this is my
base structure. Now, I am solving the first part of the problem which is forces in the members
due to loading. So let us put the loading: 150, 200, 150. This is the loading and this is a
statically determinant structure. I need to find out all the forces in all the members due to the
loading alone.
If you look at it, we see sigma fx is equal to 0 gives me this, and sigma fy is equal to 0 and
taking moments about any point you will see the symmetric loading. So therefore without
much ado, I am going to write down the reactions. If you look at the reactions then this is
250, 250 here and 0 is the horizontal truss here. So once we have done that, we can actually
start solving for the structural loading and therefore if you look at it, I can start at this point
(Refer Slide Time: 11:22). Let me just put it down. This is A, B, C, D, E, F, G, H. Taking the
method of joints at this point we will see that this is equal to (because the vertical component
of it has to be equal) this and therefore the horizontal component of this is equal to this (Refer
Slide Time: 11:50)) so this is 250. Now, if we go over here and take moments of joints here,
this turns out to be 250 and this turns out to be 150.
Once we have that, then again I can actually go everywhere, if I do this here this (Refer Slide
Time: 12:25) will also turn out to be 250 over root 2, this is going to be equal to 250.
Similarly this will be 250 and this will be 150. Once we do that we can find out the forces in
these members also; without much ado; I am just going to go ahead and do this. If you look at
this, the vertical component of this member has to be equal to this minus this (Refer Slide
Time: 13:18). If you really look at it this will become 100 root 2 and if we do that then this
becomes 350; if you take moments at this particular point, this member turns out to be a zero
force member, this one is 350 and similarly this is 100 root 2. So I have got all the forces and
all the members.
Now the thing is what is the next step? Remember that these members are cut but they still
exist. Remember that we know that the forces and these members are 0 because we have cut
them. However, members still exist so I should actually put it down, 0 and 0. However, one
3
other point, again I am going back. How do we get the additional equation? We have 16
equations and 18 unknowns. We actually need two more equations so that we can solve for
X1 and X2. How do we do that? By actually finding out the displacement of these two points
relative to each other given the loading, and also given X1 and X2. Once we find those out
then we can find out that the compatibility condition over here will be that this displacement
has to be equal to 0. The compatibility condition here is that the displacement has to be equal
to 0 so that gives us two additional equations. Once we have those two additional equations,
we can solve for them, get X1 and X2. Once we get X1 and X2, all the member forces can be
found out. Now how do we find out the displacement at that point? Now I want to introduce
one concept and that is, that if you note, that all I need to do, just look through this:
I need to actually solve two other problems. Let me go through it and then explain to you
what I am doing. What I have done here, is that I need to look at two additional analyses. If
you look at this, what does this represent? Here, what I have done here is corresponding to
redundant force, I have put a unit force here and corresponding to the second redundant force
this is 0. The second structure has the force corresponding to the first redundant as 0 and the
force corresponding to the second redundant as 1 and I claim that if, I solve these two
additional equilibrium problems I should be able to analyze the structure.
Let me explain how that is. Let us look at this; one of the first things is the compatibility. For
the compatibility I have to find out the displacement at this point due to the loading, plus due
to X1, plus due to X2. Actually, the displacement at this point I need to find out under this
loading, under this loading of course not equal to 1 equal to X1 and under this loading where
X2. If you add this and this and this with this not 1 but X1 and X2, then you see that is the
actual structure that you have, and this point is referred (Refer Slide Time: 18:55). What you
are trying to do is again use superposition. Find out the displacement at both the cuts due to
all the loads and then sum them all up and that will give you the actual displacement.
Now, the loading we have done; if I need to find out this displacement and I am using the
principle of virtual displacement what would I do? I would use a virtual force system where
this was equal to some arbitrary value; I might take it equal to 1. If I wanted to find out the
4
displacement here, what would I have to do? I would have to take a unit force corresponding
to this and then find out the work done. So this would be the virtual force system to find this
displacement. This would be the virtual force system to find this displacement under this
loading.
Furthermore, under this loading I need to find out the displacement here. So under this
loading you see this problem. Again it is the same loading excepting that is not equal to 1,
there it will be multiplied by X1. That will give me the displacement at this point. The real
force system is this for finding out this displacement here. This is the virtual force system for
finding out the displacement here. Then to find out the displacement here, due to this loading
- this is the virtual force system Now let us look at this particular case. In this particular case
under this loading, if I have to find out the displacement here - then this is the virtual force
system. If I have to find out the displacement here - then this itself is the virtual force system.
You see the point that I am trying to make is, that as long as I have these three: this one, and
this one and this one equilibrium; I can actually get all the variety of things. Therefore, I am
going to solve for this and I am going to put is equal to 1 here; because if it is the actual load
then all I have to do is find out the forces here and multiply by X1. That is not just the forces
but even the displacements I get here I multiply by X1 and I get the actual displacements due
to X1.
First, let me solve these structures and then we will go on to see what additional points we
have to note. I am going to make a general kind of development on the theory that we have
already done where the displacement plus displacement plus displacement is equal to 0. I am
going to just put that down in proper format so that you get the complete way of solving all
problems. First let me analyze this problem. Under this loading it is obvious, since there is no
external loading, sigma fx and sigma fy and sigma m is equal to 0 as around any point shows
that these reactions are 0. If these reactions are 0 then these member forces automatically are
equal to 0. If we look at this here also, this is the force right? So this force is equal to 1. If I
take the vertical component of this at this point it will be taken by this, so therefore this is
going to be equal to 1 over root 2. The horizontal component of this will be taken by this, so
this is 1 over root 2. Then again if I look at this point, this is equal to 0, this is equal to 0 and
then since this is equal to 0, this will be equal to 0. So this is equal to 0, because if you look at
this since this is equal to 0. At this particular point we have to be careful to see what are the
other forces we get.
Now let us look at something where we know. Do I know all the forces? I do not know all the
forces; there are three unknowns here. Which place should I go to? I will go to this point
where I have two unknowns. If I look at this particular one, the vertical component of this is
going to be equal to this. Therefore if I put that you will see that this will be equal to 1. Then
if you look at this particular one, the vertical component of this is equal to horizontal
component of this. So this is going to be equal to 1 root 2. Where do we go? Here we know 3
of them.
If you look at this particular one I know 1, 2, 3 so I have only 2 unknowns so I can solve for
them. If I look at the vertical component does it give me anything? No. Can I look at the
horizontal component? You know this horizontal component this one does not contribute;
only this contributes. So if we look at this, the horizontal component of this is 1 over root 2
and this is opposite, so summation equal to 0. Therefore this force has to be equal to 0.
Now if this force is equal to 0, then only the vertical component of this comes in here and if
we look at this, this becomes 1 over root 2. Now let us come here to this point. I know this
force; I know this force; I know this force; I know this force. So I just need to check whether
this will come out. What will this be? The horizontal component. If I take the horizontal
component of this, I get minus 1 upon root 2, so this becomes equal to 0. If you look at the
vertical component this is 1 over root 2; everything checks and if I come over here all forces
in the members are 0, so therefore it checks. You will always get three checks.
Here I have got sigma fx and sigma fy equal to 0 and here I have got sigma fy equal to 0; these
are the three checks that I have in this entire procedure. That means these are the forces due
to this application of X1 equal to 1 and X2 equal to 0. Similarly, if I apply X2 equal to 1 and
X1 equal to 0 then we just get exactly the same kind of thing. Only thing is that on this side
because this X1 these panels are non-0 and these panels are 0. Here you will just have the
opposite. These panels are 0. I am not going to go through these steps; I will request you to
go through these steps yourself. As I said analysis is not something that I am going to be
doing. I am just writing down the values; statically determinant analysis you should already
be totally used to it. Then this is 0, this is 0 and this is 1 over root 2. This is equal to 0 and
this is equal to 1 over root 2. I think that takes care of it; those are the two analyses. Now,
having done these three analyses let me write down all the values that I have in a particular
pattern. I am going to write it down in a table.
Refer Slide Time: 28:15)
First the table will have the member; so members are here, then I am going to write down the
load. This I am writing down is the actual force in members due to the loading. Next I am
going to write down and I will call this P1i , this is the force that I have got in the members
(Refer Slide Time: 28:55) due to X1 equal to 1 and X2 equal to 0; these are what I call as P1i. I
will explain to you why I call it P1i and then this I will call P2i. I am going to write it down as
P2i Let me write them all down. I have member AB, then I have member BC, then I have
member CD, then I have member DE, then I have member AF, BF, BG, FC, FG.
Now I am going to write down all the members here. But you must understand the point that
there are a total number of 18 member forces so I have to actually put it down. FG, then I
6
have CG, then I have CH, then I have GD, then I have GH, DH and HE. How many? It is 15,
i.e. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15. I have 15 members here and I am going to
write down the values of the loads that I get. In member AB I got a compressive force and
compressive force for me is going to be equal to negative.
I am just actually trying to explain it to you. What is member AB? 250 compressive, that is
all I have done 250. Then if you look at this AB, in this 0, in this 0 so 0.
All I am doing here is actually just putting down all the values for you. This is going to be
compressive minus 350. Here this is going to be minus 1 over root 2 and in this case 0. This
is CD minus 350, then 0, minus 1 over root 2 and then DE is minus 250 for the loading, for
this it is 0 and for this it is 0. AF is plus 250 because it is tension and in this case AF is 0, in
7
this case it is 0. BF is plus 150 here, minus 1 over root 2 here and 0 here. BG is going to be
plus100 root 2, here BG is equal to 1, here BG is equal to 0. FC is equal to 0 here because FC
is X1, so here it is going to be 1 and here it is going to be 0.
This is FC, the force here is X1, so obviously it will be 0 here, 1 corresponding to X1 is equal
to 1 and 0 when X1 is equal to 0. FG is plus 250 here then it is minus 1 over root 2 here and 0
here. CG is going to be equal to 0 here, minus 1 over root 2 here and minus 1 over root 2
here. CH is equal to X2 so this is going to be 0 here (Refer Slide Time: 34:06), 0 here and 1
here. We look at this for X1, 0, 1, and 0; for X2 0, 0, and 1. GD is going to be equal to 100
plus 100 root 2 here. GD is going to be 0 here and is going to be equal to 1 here. GH is going
to be plus250 here, it is going to be 0 here and it is going to be minus 1 over root 2 here and
finally I have DH. DH is plus150 here, DH is 0 here, it is minus 1 over root 2 and finally HE
is plus250, 0 and 0. These are all the members that I have.
Now, understand that this is the real loading; so in the real loading case if I want to find out
the displacement corresponding to in FC the separation, then what is my virtual displacement
pattern? This one and if I want to find out the displacement corresponding to X2 in CH, this is
going to be virtual force.
Here, I am going to now put it down on paper what virtual displacement gives me in the first
case 1 into delta10. What is this delta10? It is this separation (Refer Slide Time: 36:25). This is
my virtual displacement pattern where I have put X1 is equal to 1. So for that one the virtual
work principle is going to be 1; that is the virtual force corresponding to delta10 is equal to
summation over all the members i and each member deformation. What is the real
deformation? Real deformation is the real load multiplied by Li upon EA note that EA here is
the same for all the members; this is the real deformation multiplied by the virtual force
which is going to be equal to P1i. Is that clear? Understand that delta10 is nothing but
summation of. Since EA goes outside; because EA is same for all the members this is
going to be equal to P1i Li. So that gives me delta10. How do I get delta1x? delta1 due to X1
delta X1 is equal to what? Let us look at it this again. Here, the real load is going to be P1 X1
because P1 X1 gives me the forces in all the members due to the load X1 the real load. So this
8
going to be this load into Li upon EA multiplied by this one; so the virtual force pattern is
going to be P1i. If you really look at this, this turns out to be equal. X1 and EA are independent
of each other so I can take it outside, so I get X1 upon EA, inside P1i squared Li.
What do I want to find out? I wanted to find out delta21 that is the load corresponding.. The
real load is X1 and I want to find out the displacement corresponding to X2. If I look at this
now, I am just going to write these down; I do not need to talk about them anymore. The real
load is 1. So the real load is P1i into X1 into Li upon EA and the virtual load is P2i because
there is a displacement that you are finding out, if I put this in, I get X1 upon EA summation
P1i P2i Li. I can put them down in this fashion.
Using the virtual work I got delta10 is equal to 1 upon EA Pi P1i Li. delta20 is equal to 1 upon
EA summation Pi P2i Li. delta11 this is due to X1 is going to be equal to X1 into 1 upon E P1
square Li. EA is already outside so the EA is not required. Of course if EA is not the same
you can put EA inside also. Then delta21 is equal to X1 into 1 upon EA P1i P2i Li. delta22 is
equal to X2 1 upon EA P2i squared Li and delta12 is the displacement at 1 due to X2 is going to
be equal to X2 1 upon EA summation P1i P2i Li.
If you do this, let us look at what this term represents (Refer Slide Time: 42:03). This term
represents the displacement at 1 due to a unit X1. Displacement at 1 due to a unit load X1.
That can be defined as a flexibility coefficient 1-1. What is flexibility coefficient? Understand
displacement due to unit load is flexibility. I am putting down f11 because that corresponds to
delta11; this is the displacement at one due to a unit load at one. What is this? This part is
displacement at two due to a unit load X1 so f21. What is this, displacement at two due to a
unit load at one? What is this? delta22 load displacement at two due to load at two, so this part
is equal to f22. Displacement at two due to a unit load at two and this part is equal to f12,
displacement at one due to load at two, so this will be displacement at one due to a unit load
at two. All these are nothing but the flexibility coefficients for this structure.
If I put them together you will see that I can rewrite this equation in this manner: delta10 this
is the displacement at one due to the loading, plus f11 X1 displacement at one due to unit load
multiplied by the load itself, plus f12 X2 displacement at one due to the load at two multiply
unit load at two multiplied by X2 and this. If you look at this, what is this displacement at one
due to loading, displacement at one due to X1, displacement and one due to X2. If you add all
of the three up and that is the original structure. This is the displacement in the original
structure. What is the displacement in the original structure? Those two points cannot move.
Therefore the displacement in the original structure is this. Now let us look at the second
compatibility condition. This is the compatibility condition corresponding to the first load. So
what I am saying is that at this point due to summation of all the effects these two points
cannot go apart because FC is a actually a member.
Similarly, for the other one f21 X1 plus f22 X2 is equal to 0. This is displacement at two due to
X1 because due to unit multiplied by X1 this gives me the displacement at two due to X1.
Displacement at two due to X2, displacement due to the loading all of them together add up to
0. These are my compatibility conditions. And I can solve these to get X1 and X2. I can
rewrite in this fashion f11, f12, f21, f22, X1, X2 is equal to 0. For this particular structure I shall
write down the Li. The only thing that I need to do over here is write down what the Li are for
each and then all I need to do is just multiply all those values. Note that once I find out P1i P1
P2 Li I can find out all these coefficients and am going to do those, but first I will write down
the Li.
10
If you look at the Li, this is going to be equal to AB-AB is going to be equal to 5 root 2. BCBC is going to be 5. CD-CD is going to be 5. DE-DE is going to be 5 root 2. AF-AF is going
to be equal to 5. BF-BF is going to be equal to 5. And if you really look at it: BG-BG is equal
to 5 root 2. FC-FC is equal to 5 root 2. FG-FG is equal to 5. CG-CG is equal to 5. CH-CH is
equal to 5 root 2. GD-GD is equal to 5 root 2. GH- GH is equal to 5. DH-DH is going to be
equal to 5. HE-HE is going to be equal to 5. So I have got all my displace lengths etc and I
can now do all of these computations. I will write down the values for you right now.
In this particular case delta10 is equal to 823.22 upon EA. delta20 is identical because this is a
symmetrical structure 823.22 upon EA and f11- f11 is equal to 24.19 upon EA. F22 is also the
same: 24.19 upon EA and f12 is equal to f21 is equal to 2.5 upon EA. This is not equal to 0 this
is equal to minus delta10 and minus delta20. These terms come in here. This is the equation
11
here. I know f11 f12 f21 f22, I know delta10 and delta22 and I can solve these equations for X1 and
X2. I will actually write down the values of X1 and X2 for you and they turn out to be equal to:
X1 is equal to X2 is equal to 30.9 Kilonewtons.
Now how do I find out the force in all the members? The force in all the members is going to
be equal to Pi plus P1i plus P2i X2. I can find out the actual force in all the members in this
fashion. I want to end it by talking about one very fundamental point over here and that is
you see f12 is equal to f21 in this particular case. Is that always true? You will see that what
does f12 represent? F12 represents the displacement at one due to unit load at two and f21
represents displacement at two due to unit load at one. The loads in both the cases, in one
case it is applied at two and in one case it is applied at one, the loads are the same. So what
can we say about the displacements?
Let us go back to the Maxwells Betti Reciprocal Theorem. If you really look at that, what
does that say? It says due to a load at one, the displacement at two is equal to the
displacement at one, if you applied the same load at two. That is what the reciprocal theorem.
So If you really look at this: The same load is being applied - one unit load. Once it is being
applied at one and we are finding out the displacement at two, and in one case we are
applying the load at two and we are finding out the displacement at one. Obviously they are
going to be the same. Therefore this is a fundamental fact that the coefficients f12 and f21 are
always equal to each other always.
In fact we can make this general; we can say that this is always the case if that is the case and
these two are equal to each other. What can we say about this matrix? This is the symmetric
matrix and this is always true; the flexibility matrix is always symmetric. This is due to the
load alone; we also looked at how to get for the temperature. I am going to leave this for the
next lecture, as to, how to compute X1 and X2 and then of course, all the member forces; due
to the temperature in the members AB BC CD and DE.
Thank you.
12
Prof. P. Banerjee
Lecture 09
Good morning. Last time in the last lecture, we saw how to use the force method to solve a
truss analysis problem. Just to quickly review that process, if you look at what we looked at,
this was the truss example we looked at.
It was a four-panel truss. This was the loading and we calculated the forces in the members
due to the loading. For that, we took x1 and x2 as the two redundants and then we went ahead
and found out the forces in the base structure that is, without these two, both being 0, we
found out all the forces in all the members, then we applied x1 is equal to 1 and x2 is equal to
0, found out the forces, applied x2 is equal to 1, x1 is equal to 0 and found out the forces and
after that, we put all of those values that we computed in one table; then, from that table we
computed all these various parameters.
Ultimately, these were the two equations that we got and we solved for X1 and X2 and once
you solve for X1 and X2, then you know what the values are in those parameters and you can
find out all the forces.
Today, what we are going to be doing is we are going to be looking at the second part of the
problem which we did not solve: forces and members due to the temperature only, where AB,
BC, CD, DE are subjected to an elevated temperature of 25 degrees where alpha is 1.2 into 10
to the power of minus 5. Today, we are going to be looking at how to do this particular
problem.
Note that this remains as our x1 is equal to 1 and x2 is equal to 1. These are still our virtual
force and real force system depending on how you are going to be using them; I have already
talked about them last time, I am not going to repeat that.
The only thing that happens differently is that these equations. There is also another
equation here, which I did not write down that day. That is 1 upon EA sigma Pi P2i Li. What
is this term? This term is nothing but the real deformation in member i because of the
loading. In the case where we have the temperature problem, all that happens is you calculate
the deformation in the member directly; you do not calculate forces, you calculate the
deformation directly and your problem statement becomes this way.
Delta10 is equal to summation over all the members deltai into P1i and delta20 is equal to i
deltai P2i. These come from your virtual force equations; I do not want to go into that all over
again. This is the only difference.
If you look at it, these remain the same. Only this and this for loading is this; when you have
temperature, this is different and this is given by these (Refer Slide Time: 06:14), where these
are the temperature deformations that you get in every member. Let us look at it. We have
only member AB, member BC, member CD and member DE. These are the only members
that are subjected to the elevated temperature, so let us find out deltaAB due to the elevated
temperature. How will you find that out? Alpha is 1.2 into 10 to the power of minus 5 into
deltat (deltat is 25 degrees) into L.
What is L AB? L AB is equal to 5 into root 2. What does this become? This becomes 1.2 into
25 is 30, 30 into 5 is 150, so this becomes 1.5 into root 2 into 10 to the power of minus 3
meters. Similarly, deltaBC is 1.2 into 10 to the power of minus 5 into 25 into its length, which
is 5. So, this is equal to 1.5 into 10 to the power of minus 3 meters. Similarly, deltaCD will
turn out to be 1.5 into 10 to the power of minus 3 meters and deltaDE will turn out to be 1.5
into root 2 into 10 to the power of minus 3 meters. We have computed these; these are the
only ones (Refer Slide Time: 08:23). Here, all that will happen is delta10 will only have
deltaAB into the P1 of AB. P1 of AB.
P1 of AB is 0. Then, we have deltaBC, which is 1.5 into P1 BC, deltaCD is P1 AC and deltaDE is
this into this. If you look at the entire procedure, since P1 AB, P1 CD and P1 DE are all 0, the
only thing that contributes is BC. So, delta10 will turn out to be equal to minus (1.5 into root 2
into 10 to the power of minus 3 multiplied by 1 by root 2), because it is minus 1 over root 2.
Delta1 becomes minus 1.5 into 10 to the power of minus 3 meters, which is minus 1.5
millimeter. Similarly, we can compute. Here again, for P2i, you will see that only CD
comes in and CD has exactly the same concept. Therefore, if you look at it, you will see.
Sorry, I have made a mistake here (Refer Slide Time: 10:03). deltaBC is not 1.5 into root 2, it
is. You basically have minus (1.5 over root 2), so minus (1.5 over root 2), that is
approximately 1 millimeter.
Similarly, delta20 also turns out to be exactly the same. By computing in that case, deltaCD
will be 1.5 into 10 to the power of minus 3 multiplied by minus (1 over root 2), it becomes
minus 1 over 1/2. Essentially, once we have computed delta10 and delta20, we substitute the
same equations that you have which is delta10 plus f11 X1 plus f12 X2 is equal to 0 and delta20
plus f21 X1 plus f22 X2 is equal to 0.
Since we know these, these, these, these, these and these, we can solve for X1 and X2; the
value of X1 and X2 turn out to be equal to plus17.6 Kilonewtons and so is X2. Once you know
this thing, the force in the member is going to be equal to P. In this case, what is P? P is the
load due to the temperature alone, so this is 0. You are going to have P1i, which is the force in
the member, into X1 plus P2i into X2. This becomes the force in all the members due to
temperature where we have already got P1i and P2i, you know X1 and X2 and you can obtain
the forces. In essence, these are the forces that you find out in your truss for temperature.
Now what we want to do is to go ahead and look at a frame problem. We have already looked
at how to tackle a realistic truss problem and today what I am going to be doing is I am going
to be looking at how to treat a frame problem. I will take an actual problem and go through
all the numbers, so that at the end of it all, it helps you to be able to solve a frame problem.
Let us take a frame; I am going to take a simple frame. Essentially, if you understand the
concepts with a simple frame, you should be able to use it to solve any problem. I will go
through all the details so that you know how to tackle the problem. Let us take a situation
where we have this as 5 meters, this as 5 meters and let me call this a, b, c and d. These are
the loads on the structure (Refer Slide Time: 14:16). This is acting at the center span, so it is
2.5 meters; this is acting at b, the lateral load. This is a typical kind of load you would have: a
gravity load and a lateral load due to wind or it may be even earthquake, does not matter. So,
this is the load. We have to draw the bending moment diagram for this particular structure.
Remember I am going to explicitly state that we only consider flexure; in other words, both
axial and shear deformations are neglected. This is the problem.
How many degrees of static indeterminacy? I will leave it up to you to go through your steps;
using my procedure, you will see that the starting indeterminacy is 1. I am going to take this
reaction as my X1; so, this is the problem statement. What is the first step? The first step is to
take the base structure where X1 is equal to 0, find out all the forces in the members and find
out the deformations, then put X1 is equal to 1 and find out the bending moment diagram.
Once you get the bending moment diagram, you can then go through the process. Let us go
through this particular process.
10
The first part is where X1 is equal to 0; let me put it explicitly: X1 is equal to 0. This is the
loading, this is 10, 10. What will be the reactions at this point? The reaction over here will be
10, horizontal reaction will be 10 and you will see that this gives rise. This is 5 meters and
this is 2.5 meters, 2.5 meters. If you look at this 5 into 10, if I take moments about this point 5
into 10 plus 2.5 into 10 that is the net moment that has to be resisted here. So, you have a
resisting moment here equal to 75.
Once you have that, you can draw the bending moment diagram. Let us see how I will draw
the bending moment diagram. The bending moment diagram over here would land up starting
from c. The value over here would be 25 that would be the same value here and then this
becomes 75, the sense of the bending is this way and the sense of the bending over here is
this way (Refer Slide Time: 17:45). This is your bending moment diagram, M diagram. This
is the M diagram for X1 is equal to 0; so, this is the base structure bending moment diagram.
Note that I have also made the assumption that this EI and this EI are the same; so, if I just
divide by EI and EI would give me the M by EI diagram. What is M by EI? This is the
curvature. So, it directly gives the curvature diagram for the real curvatures under the loads.
What is the next step? The next step is for X1 is equal to 1.
11
Let us find out what the reactions at this point are. The reaction at this point is going to be 1,
this is going to be 0 and what is the moment going to be equal to? This is 5 and this is 5. If
you take moments about this point, this gives rise to this, so you are going to have a moment
in this direction, which is equal to 5. I am not going to write Kilonewton because this is just a
unit, there is no force; so, this is the moment 5. How will the bending moment diagram look?
Note: I do not care which direction I draw the diagram as long as I show the sense. I am
drawing it in the same direction as that, but note that here, the bending is hogging; in other
words, tension at the bottom, compression at the top.
Here it is sagging, so it is going to be tension at the bottom and this is going to be hogging;
so, although I have drawn it on the same side, you can draw it on the other side, it does not
really matter, you can draw it. However, you have been asked to draw. So, I think as long
as you are consistent. This is the point that I always make: as long as you are consistent, it
does not matter. This value is equal to 5 and you will see that at this point, it is going to give
rise to a sagging moment. It is actually this way and that is correct; if you look at this, you
will get it over here; here, it is actually the opposite way.
12
If you look at this, this is this way, this way and this way; this is the opposite way. I am so
sorry, I made a mistake here. Please ensure that you understand. This is hogging because this
creates a curvature in this direction and that is what you are doing here and this also is this
way. Over here, you have it as sagging; this is sagging. So, what do we have? Let me just put
it together in the other one.
13
I will draw it small so that I have both of them. The actual M by EI diagram is starting from
c. This is hogging, hogging 25, 75; these are Kilonewton meters because the loading is
Kilonewton and the forces are meter. These are 25 by EI and 75 by EI. Note that if you have
two different EIs, then the moment diagram and the M by EI diagram would not be the same.
But in this case, I have taken them to be same and this is the small m1 diagram. The m1
diagram I have shown it on the same side, but this one is opposite (Refer Slide Time: 22:53).
This is the M by EI diagram, this is the m1 diagram.
Now, what do we do? Note that, if I want to find out the displacement at this point then my
equation becomes delta10 is equal to integral over the whole length m1 into M upon EI dx
over the entire length of the frame. That is delta10. If you look at this particular situation, all
you have to do is take this area and at its centroid find out this value, take this area and the
centroid, find out this value.
I will go through the steps. Let us look at what you will get over here. In this particular one,
what is the area under the curve? 25 upon EI multiplied by 2.5 by 2 that is the area under
this curve. Now let us look at this; which point would this be? This point would be 2.5 by 3
from this end, which is actually 5 by 6. If you look at this, since this is 5, this is 5 by 6 from
this point. If you take that, you will see that this the value at that particular point, the centroid
of this graph, is going to be equal to 25 by 6; this is going to be the value at that particular
point. Is that clear? It is going to be equal to 25 upon EI multiplied by 2.5 by 2, which is the
area of this curve, multiplied by 25.6, which is the value at the centroid of this. This is the
integral of this area, the integration over this length.
Then, we look at it over this length. This can be broken up into two parts, one a rectangle of
25 by EI. So, you are going to have plus25 upon EI into 5 (that is the area under this curve)
multiplied by the value at. that is going to be 5 because it is a constant; it does not matter
where the centroid is, it is going to be the same. The next one is going to be the difference,
which is 50 upon EI. So, 50 upon EI multiplied by 5 divided by 2 (the area of this triangle)
multiplied by 5, which is the centroid at this point. What we get is 1380 by 2EI. This is the
value of delta10. Then, we need to find out the deflection due to X1 is equal to 1. What is the
deflection due to X1 is equal to 1? That is my f11 that is my flexibility.
14
Displacement at one due to unit load at one and that is equal to m1 squared upon EI dx. What
is the flexibility? Flexibility is displacement at one due to X1 is equal to 1. What is going to
be the actual displacement due to X1? It is going to be f11 into X1. I am introducing the
concept of flexibility which I have introduced with the truss; I am doing the same thing over
here. If you look at this, if I draw this, this will then be this integral with itself upon EI (Refer
Slide Time: 27:41). If you look at this, this is going to be equal to 5 into 5 by 2 (that is the
area under the curve) multiplied by 10 by 3 (that is the value at the centroid). Sorry, this is
going to be 5 upon EI because this is the real part. Then, plus 5 upon EI multiplied by 5
multiplied by 5. What are these? This is the integral of m1 square dx over this length and this
part is the integral over this length. This is going to be equal to 500 by 3 EI; this is my f11.
What is my compatibility condition? Delta10 plus f11 X1 is equal to 0. Since this is only one
redundant, this is the only additional equation that I require. From this, by plugging in the
value of delta10 and f11, I get X1 is equal to minus. I am sorry, I have made a mistake. I just
wanted to make a point here: since this and these are opposite, the delta10 turns out to be
minus 1382, so here, X1 is equal to 8.28 Kilonewton. I have got the value of the reaction at
the support d that we have. Once I have got that, what is the next step?
15
The next step here is to actually draw the bending moment diagram. However, before
drawing the bending moment diagram, let me draw the, this is my loading. I know this
value, this reaction; so, this is a, this is b, this is c and d. I have found out this reaction and
this reaction value is 8.28 Kilonewtons. Once I know this, I can actually find out the reactions
at this point; this reaction is going to be 1.72 Kilonewtons, this is going to remain 10
Kilonewtons and what I get is this: if I find out that the moment over here is 10 into 5 plus
2.5 into 10 that is 75 minus 8.28 into 5. This is going to give me a moment in this direction
whose value is going to be equal to 33.60 Kilonewton meter.
Once I have found out the reactions, I can proceed to finding out the forces in the bending
moment diagram. Let me just see how I will draw the bending moment diagram. Let me go
piece by piece in this particular case. I will take bcd, I will separate out bcd and ab separately.
If I separate out, then what are the things that I generate if I take out bcd?
16
This is 8.28, this is 10. Note that I am going to isolate the joint and then I am going to have
the member. This is a, b, c and d. The only difference is that this cut that I have made is
actually infinitesimally small; it is just so that I can isolate the joint b and look at it
separately. When I make a cut here, what are the forces that I generate over here? The forces
that I generate over here are a shear, the shear is going to be in this direction, then a moment;
let us just take positive moment right now, so the positive moment will be in this direction
and what else? There could be an actual force too. Note that you do not have to consider
shear deformations and actual deformations but the forces are still going to be there.
This is going to be M, this is going to be V, this is going to be P and can I find out these
values? Sure, because I have this; here, I know this is roller support, so moment is going to be
equal to 0. What is V going to be equal to in this particular case? V is going to be equal to
1.72 because you take sigma fy, you will see that is equal to 1.72 Kilonewton. What is the
moment over here? Let us take the moment about this point; this is going to give you 10 into
2.5 this way and this going to give you 8.28 into 5 the other way. This is going to give you
41.4 minus 25. What is that going to be equal to? It is going to be equal to 16.4. Which
direction is it going to be? Actually, if you look at it, this is going to be this way, this is going
to be this way, so the net moment is actually going to be if I am showing it this way, it is
going to be minus 16.4 Kilonewton meter.
What can we say about P? Take sigma fx is equal to 0, you will see it is 0 Kilonewton. I have
found out the forces here. Once I find out the forces here, just the opposite ones are going to
be coming here. Let me take this one. This one, I know all the reactions that are coming here,
1.72, this is going to be 10 and you have already found this out as 33.60 Kilonewton meter. If
17
you look at this particular case, how do I find out the ? When I make a cut here, I am going
to get reaction V here, I am going to get a moment here and I am going to get an axial force b
here.
Let me just see, I will make them separate so that the P prime. What will be P prime equal
to? Sigma fy equal to 0 gives me that P prime is equal to 1.72 Kilonewton meter. What is V
prime going to be equal to? If you look at this, sigma fx is equal to 0 will give me 10
Kilonewton meter. What is M prime going to be equal to? Let us see. 10 into 50 is going to
be in this way, 33.6 is in this way, so what is it going to be equal to? If you look at this, this is
going to be giving this way. If you look at your moment, it is going to remain. This
moment is going to be equal to 50 minus 33.60, so that is going to be 16.4 Kilonewton meter.
Let us review this again. Taking moments about this point, what do we generate?
We generate that this moment plus this moment is equal to this moment (Refer Slide Time:
36:43). M is M plus10 into 2.5 minus 8.28 into 5 is equal to 0, that is sigma M is equal to 0.
If you take that, you will see that this is going to be plus16.4. If you take V, V plus 8.25
minus 10 is equal to 0, so V is going to be equal to this and sigma fx is equal to 0 is going to
give me P is equal to 0 here. In this particular case, what do you get? If you take sigma fy is
equal to 0, you will get P minus 1.72 is equal to 0, so you get P; you get V prime minus 10 is
equal to 0, so you get V prime. Similarly, moment plus 33.60 minus 10 into 5 is going to be
equal to 0, so moment is going to be equal to this. I have found out at the ends of the two
members. Similarly, I can actually find out the forces at the joints also. Let us look at that; I
am going to spend some time looking at equilibrium here, because after this I am not going to
be looking at equilibrium.
18
I have already done this, so I am just going to put them down: 8.28 Kilonewton here, 1.72
Kilonewton, load 10 Kilonewton, moment over here is 16.4 Kilonewton meter. What else?
Over here, I have 1.72 Kilonewton meter, this is going to be 10 Kilonewton, the moment over
here is 33.6 Kilonewton meter. At this point, I have 10, I have 1.72 and I have moment equal
to 16.4 Kilonewton meter. Now, let us do it on this side. This side is going to give rise to this
force, which is going to be equal to 1.72. It is going to give rise to moment here. Here, I have
the 10 Kilonewton force, over here, I am going to have 1.72 which is going to be in this
direction, 10 and the bending moment is going to be in this direction. Let us look at the
equilibrium of the joint.
Let me take equilibrium of the joints sigma fx equal to 0. I have minus 10 here, so I will put
minus 10; I have plus 10 here, plus 10 that is it, there is no other force here, so this is equal to
zero, automatically satisfied. Then, we look at sigma fy equal to 0; sigma fy equal to 0 is
going to give me that this is upwards 1.72, there is no load over here and over here, it is
downwards 1.72, so it is zero. Check. Then, let me take a moment equal to zero, let me take
moments about this point. These distances are zero, these do not give rise to any moments.
What is the moment? I have a clockwise moment here and an anticlockwise moment. The
clockwise moment is 16.4, the anticlockwise moment is 16.4. Check. That means all that I
have done is correct and so now once I have this particular loading done, the next thing that I
am going to be doing is I am going to be actually putting together the bending moment
diagram. Since I know all of those, how will the bending moment diagram look?
19
This side remains the same. It is going to be 16.4, 33.6, so ab, there is no mistake, it is 33.6;
and this is 16.4, this is also 16.4, this is this way, and let us see what the moment at this point
is going to be.
If I take a cut here. I am taking a cut exactly to the left of this load so what is going to be
the moment? It is going to be 16.4 and if you take moment about this point, it is going to be
16.4 plus 1.72 into 2.5 so that is going to be equal to. This is going to go this way, this is
going to be equal to 41.6, so it is going to be 20.8 (Refer Slide Time: 42:27). This value is
going to be 20.8, this is going to be 16.4 and all of them are in Kilonewton meter. This is the
bending moment diagram for the actual structure where I have included both the effect of the
actual loads as well as the redundant force. This in essence is my force method solution.
I hope that with this kind of procedure I have gone through a detailed frame analysis problem
so that you know exactly how the force method is used to solve a frame problem. Till now, I
have always been talking about the basics, so this time I have gone through all the details so
that you can actually feel comfortable using the force method for solving a frame problem.
Just to refresh your memory, up till now, all that I have done was going to be a review. You
already hopefully used the force method etc., for solving both truss as well as beam and
frame problems. Let me just go back and say, in essence, what the force method entails.
20
This is irrespective of whether you have a truss or beam frame.

If you notice, the only difference between a beam and a frame that I pointed out that day was
that in a frame, all that happens is this: what is shear over here becomes actual force over
here, and whatever is shear over here may become actual force here; it would have if this had
not been there. So, the only difference between a beam and a frame is that in a beam, if you
21
have multi-span beams, a shear in one beam will always be a shear in another beam and axial
force in one beam would be an axial force in another beam. But in a frame, a beam and a
column, the shear forces in the beam may land up giving axial forces in the column and the
shear forces in the column may land up being axial force in the beam; that is all that is there.
Otherwise, equilibrium and all the forces that you develop are identical whether you have a
beam or a frame.
The equilibrium considerations may be different in a beam and a frame, but as far as the
solution is concerned, there is no difference between a beam and a frame, you only consider
flexure, and in a truss you only consider axial. So, what does the force method entail? The
force method entails the following: one, first find static indeterminacy this is fundamental
to using the force method. If, by now, you do not have confidence in obtaining the static
indeterminacy or structure, there is no way you are going to be able to use the force method.
Given any structure, you should be able to find out the static indeterminacy of the structure. I
have spent the first two lectures looking at how to determine the static indeterminacy for a
truss, as well as the beam frame structure. So, by now, you should be able to get the static
indeterminacy. If you cannot, please go back and look at enough problems, look at any book
on structural analysis, take up some problems, essentially structural analysis of statically
indeterminate structures, take up any structure and find out the static indeterminacy. If you
are confident that you have found out the static indeterminacy properly, you can use the force
method easily. So, find the static indeterminacy
Once you find the static indeterminacy, you identify the redundant forces. What are the
redundant forces? Those are the forces on the structure; they could be a support reaction, they
could be internal forces. In the frame problem that we looked at, we took the support to be a
redundant force.
If you look at the truss problem that we solved, we took internal members forces to be
redundant forces. The only thing is that the redundant forces have to be such that if they were
0, the structure would still remain stable; that is most important. You cannot take out
redundant forces, put something equal to 0 which will make the structure unstable; that
cannot be done. It does not matter if you have static indeterminacy in support reactions; you
can remove some support reactions and take them as redundant.
If you do not have redundants, for example, if you looked at the truss structure that I had, I
had a hinge at one end and a roller at one end. I cannot remove a support reaction without
making the structure unstable. Therefore, in that particular case, I had to take redundant
forces which were internal. In the frame case that I considered, I could remove the roller
support and still have a stable structure; therefore, I removed the stable support. Is that clear?
Overall, the concept is that you identify redundant forces such that you get a stable, statically
determinant base structure by taking redundant forces equal to 0.
22
Second, what you do is, you find the member deformations under loading. How do you find
out member deformations? The way you find out the member deformations if you have a
loading is that you find out the loads if member forces due to by solving the statically
determinant problem and you find out the member forces. Based on that, once you know the
member forces, for example, if it is axial force, if it is a truss member's axial force, the axial
deformation is given by PL upon EA, that is the axial deformation. If you look at flexure,
then the flexural deformation is given by M by EI. So, these are the member deformations
that you find out under the loading. If it is temperature loading or a lack of fit, then you know
the member deformation directly. That is the overall.
Third, find member forces due to unit redundant forces. If you have two redundants, you will
apply one redundant first and put the other redundant equal to zero. You put X1 is equal to1
and X2 is equal to 0; then, you find out another member force putting X1 is equal to 0 and X2
is equal to 1. Find the member forces; this helps you to essentially find deflections
corresponding to redundant forces using virtual work.
Once you find the deflection, the next step, step five is apply the compatibility conditions and
from that, solve for redundant forces. Seven, knowing the redundant forces, find the internal
forces and support reactions. This is your structural analysis. I have essentially read out the
procedure for using the force method. I hope at the end of this lecture, you are confident in
using the force method to solve any truss or beam frame problem. Next time onwards, I am
going to start with looking at the matrix approach to the force method. In this way, I am
going to be introducing you to the matrix methods that are actually used in structural analysis
of real structures. Thank you.
23
Prof. P. Banerjee
Lecture 10
Good morning everybody. We have spent the last few lectures looking at the force method.
Today, I am going to introduce another method. The force method actually lies under a broad
class of what are known as compatibility methods, because the equations were essentially based
on getting a compatibility equation and the number of compatibility equations is equal to the
number of redundant forces that you generate.
Today, we are going to be starting to look at a completely different approach. This is broadly
classified as displacement methods just like we have the force method. Initially, what I am going
to be doing is, I am going to look at what is known as the classical slope deflection equations,
look at beams, and then quickly I will move on to more complex application of the same sort of
methods.
We will see that just as we have used the principle of virtual force in the force method,
somewhere along the line, we will be using the principle of virtual displacement to write down
the displacement equations. Let me just quickly move on to how to use the displacement method
and today, I am going to concentrate essentially on what is known as the slope deflection
equations and how to use the slope deflection equations to solve a statically indeterminate
structure.
Let us look at what the slope deflections equations are. The slope deflection equations are given.
Remember we had said when we had started looking at kinematic indeterminacy or the degrees
of freedom, that a kinematically determinant structure is one such as the one that I have over
1
here. In this case, the number of degrees of freedom that you can release, if you look at the
release degrees of freedom, we can look at it this way, we can say this rotation we can release. I
will call this a, b. I will call this thetaab; in other words, it is the rotation at a in the member ab.
There is the rotation ba and then there is also the deltaba, which is the lateral displacement of b
relative to a, please remember that, it is relative to a, that is deltaba. Another point that you will
slightly notice if you look at standard slope deflection equation books, this theta is defined in this
manner. I always define anticlockwise rotation as my proper orientation; I will always have
anticlockwise rotation as positive in my notation system.
Note that because I have drawn it in this way, the slope deflection equations become this. I am
just writing down the slope deflection equation. If I call this as L, the flexural rigidity at EI is
given as 4 EI by L thetaab plus 2 EI by L thetaba minus (6 EI by L squared deltaba); and Mba is
equal to 2 EI by L thetaab plus 4 EI by L thetaba minus (6 EI upon L squared deltaba).
These are my slope deflection equations. These are the slopes, this is the deflection and
essentially, these equations relate the moment at the two ends of the beam and they give it in
these terms. Now, these are the slope deflection equations and normally, this is where you start
off. However, what I am going to do is this. This goes along with my basis of never writing
down an equation without knowing where it comes from. So, we are going to develop these
equations of motion so that we know how they are developed, such that if there are any different
kinds of situations that we come up with, we can always obtain the slope deflection equations for
those kinds of situations. Let us see these slope deflection equations; please look at the slope
deflection equations. Mab is equal to 4 EI upon L thetaab plus 2 EI upon L thetaba minus (6 EI
upon L squared deltaba); Mba is equal to 2 EI upon L thetaab plus 4 EI upon L thetaba minus (6 EI
upon L squared deltaba). These are the classical slope deflection equations. Now, let us see how
we can develop this.
(Refer Slide Show: 08:15)
For this, let me go back to my old friend, the simply supported beam. In the simply supported
beam, I can apply any kind of load; I am going to apply some specific loads here. What I am
going to do is I am going to apply a moment M1 at this point and I am going to apply a moment
M2. This is my loading on this beam; this is the only loading on this beam, this is a, b; so, I need
to find out thetaab and thetaba. Under this application, this beam will deform and I want to find
out what these rotations are under these loads. This is my question here. How am I going to find
this out? I am given the fact that the length here is L and since only flexural deformations are
considered, a flexural rigidity is given by EI. My brief is, given M1 M2, find thetaab and thetaba.
Remember that since I am given M1 and M2, I can consider this structure to be first subjected to
M1 plus M2.
You see, principle of superposition. If this is my loading, I can say that this loading is equal to
M1 is equal to M1, M2 is equal to 0 plus M1 is equal to 0 and M2 is equal to M2; I can always say
this. What would be the thetaab due to these two applications? I can find out thetaab and thetaba
due to this, I can find out thetaab and thetaba due to this. If I add the thetaabs, then I will get these
thetaabs. Therefore, the question here is very very simple. I want to find out how much.
Note that I define thetaab also as anticlockwise, as positive and also thetaba. Similarly here thetaab,
thetaba. Once I find out. In each case, I add the two up and I get these thetaabs that I am
interested in. Let me solve this problem first and then I will solve this problem.
Under this load, how would I find out the rotations here? First and foremost, find out. This is a
standard displacement problem which have you already solved earlier; so, I am going to use the
principle of virtual force to solve this problem. First, I am going to apply the real load. When I
apply the real load, I get. This is my bending moment M1 here, 0 here, linear and this is the
bending moment; I am going to leave it up to you to get this. Next, I want to find out the rotation
at this point. What do I do? I apply a unit virtual load. What do I get? I will get again linear, 1. If
I want to find out this rotation, what do I do? I apply a moment here corresponding to the
3
rotation and draw the bending moment diagram. This is the bending moment diagram. All I need
to do is. To calculate thetab, I need to do. 1 into thetaab, that is my external virtual work is
equal to.... This is going to be my M1, this is going to be my M2, this is going to be equal to m1
M1 upon EI dx, which is equal to M1 upon EI into. This length is L, so it is L by 2.
This is the area under this curve and the value at this at the centroid of this axis is equal to twothirds. So, this is equal to M1 L upon 3 EI. Similarly, thetaba is equal to m2 M1 upon EI dx. That
is equal to M1 upon EI into L by 2, that is again the area under this curve, and multiplied by
the. At the centroid, what is the value here? Since the centroid is two-thirds or one-third from
this distance, it is going to be equal to one-third and note that it is opposite sign because if this is
hogging, this is sagging. This is going to be minus1 upon 3. This is going to be equal to M1 L by
6 EI; because of this load, the thetaab anticlockwise and anticlockwise thetaba are going to be
equal to this. What does thetaab positive and thetaba negative mean? It implies that it actually is
something like this, because thetaab is positive but thetaba is negative, which means it is
clockwise rotation; clockwise rotation is negative, so this is how it looks. Under this loading, you
would expect it to behave in this manner. So, these are the values. Now, let us look at what if I
apply M2.
I am going to apply M2. I need to find out how much thetaab is (that is positive anticlockwise)
and how much thetaba is, which is positive anticlockwise. Under this loading, what would be the
bending moment diagram? Again, I am not going to be spending time telling you how to do it;
you should be able to do it by now and this is going to be this sense that is my M2. Then, to find
out thetaab, I need to apply a moment here and this is going to be equal to 1 this way, and virtual
force here to find out this rotation is going to be equal to minus 1 this way. Again, taking the
principle of virtual force, external work done to find out thetaab is 1 into thetaab, so 1 into thetaab
is equal to. This is M2, this is m1, this is m2, this is going to be m1 M2 upon EI dx. That is
equal to m1 M2 upon EI. So, this is going to be area under this curve M2 upon EI, multiplied by
area under this curve L by 2, multiplied by the value at this centroid. This centroid is one-third
4
from this; if you take one-third from here, you will get one-third. Note the opposite sign, so it
will be minus 1 upon 3. This is going to be equal to M2 L upon 6 EI.
Similarly, if I do 1 into thetaba, you will see that it is the area under this curve, which is M2 upon
EI into L over 2, multiplied by the value at this point. The value at one-third from this point is
two-thirds and both of them are the same, so two-third. This is going to be equal to M2 L upon 3
EI. What does this mean? Thetaba is positive that means it is this way; thetaab is negative, that
means it goes in this fashion; and indeed, under this load, you will expect it to move in this
fashion; so, these are consistent.
Since I have obtained it for each individually, I am just going to add both of them up. When I
add both of them up, what you get is thetaab is equal to M1 L upon 3 EI minus (M2 L upon 6 EI),
thetaba is equal to minus (M1 upon 6 EI) plus M2 L upon 3 EI. I have got the thetaab due to M1
and application of M1 and M2 together. Let us look at this. These are the rotations due to these
moments; that means these rotations are related to the moments through these two equations. I
can rewrite this in this fashion. I can consider this as a vector, which is equal to L upon 3 EI,
minus (L upon 6 EI), minus (L upon 6 EI), L upon 3 EI, multiplied by M1 and M2. All I have
done is I have written this in terms of a kind of matrix equation.
If you look at this, this implies thetaab is equal to this into this plus this into this, which is what
you get here; this is equal to this into this plus this into this, which is this. In other words, given
M1 and M2 which are external loads, you are able to get thetaab and thetaba. You have to
understand that I have actually taken loads M1 and M2. Now, I can also say, look, those are
actually. What is the bending moment at a in ab due to M1? You will see that the bending
moment at ab due to M1 is M1. What is the bending moment at b? It is going to be the same as
M2. This is a fact; you can actually go through it and do it yourself. In other words, what I am
saying is that I can also replace M1 and M2 by Mab and Mba and that is what I am going to do. I
am going to say that M1 is equal to Mab and M2 is equal to Mba. If I do that, that means thetaab
5
and thetaba are related to Mab and Mba through this. I can do the flip side, I can then write down
Mab and Mba in terms of thetaab and thetaba, that is what I am going to do. How would that be?
If thetaab and thetaba are equal to. I am going to now take L upon 6 EI outside, so that this
becomes 2, minus 1, minus 1, 2 into Mab, Mba. This is equally valid because this is what I have
got from the previous case. Instead of writing thetaab and thetaba in terms of Mab and Mba, I am
going to write down Mab and Mba in terms of thetaab. What do I need to do? Look at this: if I premultiply this by the inverse of this, what would I get? I would get this. The inverse of this
multiplied by thetaab is equal to the inverse of this into this, which is going to be a unit matrix,
into this. What would be the unit matrix? It would be the same. This implies that Mab, Mba is
equal to this inverse into thetaab, thetaba. What is the inverse of this? It is simple.
How do you find out the inverse? You find out the factors and then, if you look at this, the
inverse of this is L upon 6 EI, 2, minus 1, minus 1, 2. Please go back and look at it. It will
become 6 EI upon L, that is just this flipped, inverse is obviously just this upon this. What is the
inverse of a scalar? It is 1 upon the scalar. So, 1 upon this is equal to 6 EI upon L and the inverse
of a matrix is 1 upon the determinant; you have to find out the determinant. The determinant is 2
into 2, 4; minus 1 into minus 1, which is 1; this is 3, so it is equal to 1 upon 3 multiplied by the
cofactor of this, which is this, so this (Refer Slide Time: 25:26); the cofactor of this is minus of
this, so that is that; the cofactor of this is the minus of this, so that is that; and the cofactor of this
is this, so that is this. If you look at it, this becomes equal to 2 EI upon L 2, 1, 1, 2.
If I rewrite, plug this into here, what I get is Mab, Mba is equal to 2 EI upon L 2, 1, 1, 2 into
thetaab, thetaba. If I write this in a proper format, it will become Mab is equal to 4 EI upon L
thetaab plus 2 EI upon L thetaba; and Mba is equal to 2 EI upon L thetaab plus 4 EI upon L thetaba.
Simple. If you look back at the slope deflection equation, what did I write down?
4 EI upon L thetaab plus 2 EI by L thetaba. Forget this term for now. These two terms, I have just
developed. These terms have not been developed because I did not consider this delta. I only
considered thetaab and thetaba to occur due to M1 and M2. Now, how do I include this? Simple.
Bear with me and I will take you through this. These two terms (Refer Slide Time: 27:30), this
7
relationship has already been developed here. Now, we are going to develop the deflection part.
We have already computed the slope part, now we are going to be computing the displacement
part. How do we compute the displacement part? Just bear with me.
Take this situation; this is a pin here and a pin here. Suppose I were to remove this and let this go
up by delta. Note that this delta is small compared to the L and so it is essentially a small
displacement. How would this displace? Note that this is free to rotate. So, if you look at the
displacement pattern for a simply supported beam it would just be this, it will be a straight line
where this angle would be delta upon L. Now, let us look at the system that we were looking at,
the original.
How will I get a displacement here? I will just have to release this just like I released that and
displace this by L. Note that the fixed end remains. How will this displacement look like? Note
that here because they were simply supported and hinged, when I moved it by delta, I got a
displacement pattern which was exactly like this; here, because they are both fixed, the
displacement pattern, the rotation has to be 0 here and the rotation has to be 0 here and that is the
displacement. This and this do not look like each other at all (Refer Slide Time: 29:43). Can I
make this behave like this? Sure. Think about it.
Remember that when I had this under an application of Mab and Mba, what happened? It is going
to be this where I got thetaba and thetaab and I know the relationship between thetaab and thetaba.
Let me take this situation. What I am going to do is. You see, this is a small displacement, so it
does not really matter. I am going to now apply on this a moment Mab and moment Mba; I am
going to apply these two moments.
Of course, I am applying them here and I am just showing it over here because these have
displaced it. Remember that these displacements are always very small. I just exaggerate them to
show them clearly. If I apply moments here, what would happen? Think about it. Could I apply a
8
moment Mab and Mba? Under Mab, this happens (Refer Slide Time: 31:13). So, can I apply a
moment Mab such that this is the displacement pattern about this? Remember that this is also
length L because this is small. This is length L, this happens. In other words, I apply a moment
Mab and Mba on the simply supported beam to ensure that this displacement pattern essentially
displaces in this way, so that it mimics this displacement pattern.
What are my rotations? Let us look at my rotations. What is my rotation? This is the undisplaced
shape. Note displacement is this way. This is my thetaab and this is my thetaba. What are my
thetaab and thetaba? Let us look at this. What is this angle? Delta by L. Is this clockwise or
anticlockwise? From the undisplaced to the tangent, that is equal to . thetaab is going to be
equal to minus (delta upon L). Why minus (delta upon L)? delta upon L is the magnitude of this
rotation, that is for sure; and the fact that it is clockwise makes it minus (delta upon L).
Similarly, what is thetaba equal to? From here to here, this is clockwise; so, thetaba is also minus
(delta by L). Can I find out an Mab and an Mba which would give rise to these? Sure, I can,
because I have already developed this.
I can find out Mab and Mba given thetaab and thetaba just by applying this equation which I have
developed; so, I am going to use that; let me use that. What happens? Mab is equal to 4 EI by L
thetaab plus 2 EI by L thetaba. Similarly, Mba is equal to 4 EI upon L thetaba plus 2 EI by L thetaab.
That we have already got there; just substitute the values of thetaab and thetaba in here. You will
see that Mab is equal to 4 EI upon L into minus (delta by L) (note that thetaab is minus (delta by
L)) plus 2 EI upon L. thetaba is also minus (delta upon L). Plug that in. What do you get? You
will see that this will become minus (4 EI upon L square delta) and minus (2 EI upon L square
delta) and this becomes minus (6 EI upon L square delta). By definition, what is this delta? I
have defined this as deltaba, so I have got Mab in turn. You can similarly get Mba, you will get it
equal to this.
That means if I substitute all these factors which is a thetaab, a thetaba, and a delta, if we put all
those factors in, then what would be my Mab? I just add all of them up. When I add all of them
up, what do I get?
Mab is equal to 4 EI upon L thetaab plus 2 EI upon L thetaba minus (6 EI upon L squared deltaba).
Mba is equal to 2 EI upon L thetaab plus 4 EI upon L thetaba minus (6 EI upon L squared deltaba).
We have just developed the slope deflection equations from first principles. This is what I want
you to do. You have to understand where they come from and then, once you know those, you
can actually develop these equations for any kind of member. Here, we considered that EI was a
10
constant over the entire member. It is not necessary that the member has to be uniform all the
time, you could have a non-uniform member also. How did we evaluate this? We first applied
loads M1 and M2 and found out thetaab and thetaba.
Once we found those out, we knew that M1 and M2 are equal to the bending moments of that
point, which is Mab and Mba. Once we did that, we took the inverse. We got thetaab and thetaba in
terms of Mab and Mba, then we inversed it and we got Mab and Mba in terms of thetaab and thetaba.
In essence, that is your slope deflection equations. Then, we got the deflection part just by being
innovative and seeing how to relate a deltaba in terms of a spurious thetaab and thetaba. So, this is
the slope deflection equation. What can we do with the slope deflection equation? Let us look at
a simple structure that we have. Let me take a very very simple structure.
I am going to take a very simple structure and then, I am going to apply a moment here. I am
going to take a very very simple structure just to illustrate how to use the slope deflection
equations. I have a, b and c. There is only one load and that is the moment. How did this moment
come? Maybe it is because of some applied load over here, I am not bothered about that. Is this a
statically indeterminate structure? Of course, it is. What is the static indeterminacy of this? You
have 1, 2, 3, 4. You will see that static indeterminacy is 4 here, but in this application of slope
deflection equation, you do not worry about what is the static indeterminacy. I found out the
static indeterminacy just to illustrate to you that this is a statically indeterminate beam and there
is no way you can find out the bending moment diagram for a static indeterminate beam. My
point here is to find the bending moment diagram.
How do I apply the slope deflection equations to get this bending moment diagram? Instead of
finding out the static indeterminacy, what you have to find out is the kinematic indeterminacy of
this structure. Remember how we developed that? Kinematic indeterminacy or I also called it as
degrees of freedom how do I find that out? Let us go through this. How many joints? a, b, c, 3;
so, 3 into the 3 degrees of freedom per joint. Then, we have to subtract the restraints. How many
11
restraints? Three here. There is translation here, there is translation here and there is rotation
here; they are 0. This is also restrained; 3 plus 3 and there is 1 degree, so that is 7. Plus, I have
one constraint. What is that constraint? The actual force in this member is 0, so that is 1. How
many degrees of freedom? 9 minus 7 minus 1, that is 1. What is the degree of freedom? You will
see that the degree of freedom is essentially this rotation. What I want to say is that the whole
point in this exercise is to find out theta1. This is unknown displacement. All I need to do is find
out theta1 and hopefully, if I find out theta1, I should be able to find out the bending moments
given theta1. That is the reason why. Since we are trying to find out rotations and
displacements, this method classically is called as the displacement method of analysis. Let us
see how this.
Think about it. In the force method, what did you do? You found the static indeterminacy. Once
you find the static indeterminacy, you identified redundant forces; once you found out redundant
forces, you actually wrote down compatibility equations corresponding to the redundant forces
and that is how you solved it. In this particular case, it is just the opposite. In the force method,
actually, the redundant forces. The only thing about the redundant forces is that when you put
redundant forces equal to 0, you need to have a stable statically determinant structure; that is all
that is there. Therefore, you know there is uncertainty about which redundant force. In other
words, there are so many redundant force systems that can create a stable statically determinant
structure that you and I may not identify the same redundant forces ever and still solve this
problem correctly.
Whereas, if you look at the displacement method, degrees of freedom, once you have the degrees
of freedom, can any of you tell me that this is kinematic indeterminacy or degree of freedom is
one? Can you identify a degree of freedom which is different from this rotation? No. This is
restrained completely, this is restrained completely, this is not allowed to go this way because of
axial rigidity. This is not allowed to go this way (Refer Slide Time: 43:39), so the only thing that
can happen is this. In other words, in the displacement method, the degree of freedom
identification is unique. You and I and everyone else would choose the same degree of freedom;
there is no other degree of freedom to choose in this particular case. There is a slight amount of
uniqueness. In other words, once you identify the degree of freedom, you or I or anybody else
cannot choose any degree of freedom. It has to be a unique degree of freedom.
This, you will see later, makes the displacement method particularly amenable to use in
computers because the computer is not going to be confused. In the redundant force, the
application of force method, how is the computer going to determine the redundant forces? You
have to identify for it, but degrees of freedom, the computer knows which degree of freedom.
Given a structure, it can always identify the degrees of freedom very very easily. You can
develop an algorithmic way which can be implemented on a computer using the displacement
method and this is the reason why the displacement method is the more popular method for
analyzing structures in general. Now, let us look at how to solve this problem.
12
This is my problem. This is my degree of freedom and I have to find out this degree of freedom
given this moment and hopefully if I find out the degree of freedom, then I can get the bending
moment diagram. Ultimately, never forget that the overall factor here is to find out the bending
moment diagram. What are the steps? The first step: give theta1 equal to 1. What will happen?
Think about it. theta1 is equal to 1 implies that at this point, the joint rotates by 1. If it rotates by
1, since this cannot rotate, it is going to take up something like this. This is the displacement
pattern under theta1 is equal to 1. Note that when I say theta1 is equal to 1, understand it is not
one radian that I am saying. All I am saying is that theta1 is equal to 1, it is very small. You can
you can take 1 into 10 to the power of 4, who cares?
The whole point is that it is unit rotation; however, the rotation still remains small. All
displacements that we are going to consider in this case are going to be small. This is my rotation
pattern. What next? I take this to a member level; I take member ab and I take a member bc.
Under this rotation, this thing looks like this and bc looks like this. Note I have a situation where
in a member, I know the rotations at the two ends and the displacement, if any, of b relative to a.
Since I know these, can I find out the bending moments at the two ends, which are Mab, Mba
here? Here, it is going to be Mcb and this is going to be Mbc. Can I find these out in terms of these
rotations? Sure. Apply the slope deflection equations. What do they give me? thetaab is equal to
0, thetaba is equal to 1 and deltaba equal to 0.
13
I just need to plug that into the slope deflection equations and my Mab turns out to be equal to 4
EI upon L thetaab, which is 0, plus 2 EI upon L thetaba, which is 1, minus (6 EI upon L squared)
into 0. This is equal to 2 EI upon L. Similarly substituting, you get Mba is equal to 4 EI upon L or
I can put this in this terms. These are the moments given. Theta1 is equal to 1. If theta1 was
actually theta1, which is an unknown, then what would be Mab and Mba? Just multiply by theta1.
This is linearity. We can just find out Mab in terms of theta. Similarly, for bc, if I apply it, you
will see that Mbc (I am not going to go into the details, you can apply it yourself), you will get it
equal to 4 EI by L theta1 and Mcb is equal to 2 EI by L theta1. I have got the member in moments
now. Once I have got the member in moments, what do I do? Let us look at it; let me draw it.
14
I am now drawing the original structure; the original structure is this way. What do I have? I
know Mab, Mba, Mbc, Mcb; I have got these in terms of theta1. By equilibrium, let us look at a few
things. By equilibrium, what is the moment at this point? The moment at this point is in this
fashion, that is equal to Mab and the moment at this point is equal to Mbc and remember that at
this point, I had applied a moment M1. If I take moment equilibrium of this, I am only taking
moment equilibrium of this, what do I get? I get Mab plus Mbc minus M1 is equal to 0. Therefore,
Mba plus Mbc is equal to M1 this is my equilibrium equation. Now, what is Mba equal to? I you
look back, Mba is equal to 4 EI upon L theta1? What is Mbc equal to? Also 4 EI upon L theta1 is
equal to M1. That means 8 EI upon L theta1 is equal to M1, which implies that theta1 is equal to
M1 L upon 8 EI. I have found out my unknown rotation. Once I have found out my unknown
rotation, can I find out my Mab, Mba and Mbc?
Since I have got theta1, my Mab is equal to 2 EI upon L into theta1 (theta1 is M1 L upon 8 EI),
which will become M1 upon 4; Mba, which is 4 EI upon L into theta1, is going to be M1 upon 2.
Similarly, Mbc will be M1 upon 2 and Mcb will be M1 upon 4. I have Mab, Mba, Mbc and Mcb and
since I know all of these, I can draw the bending moment diagram because this is now a
statically determinate structure where I know the member and moments.
15
If I were to show you the member and bending moment diagram, this is M1 upon 4, this one is
M1 upon 4. If I were to draw the bending moment diagram, it would look like this. I will leave
that as an exercise for you where this one is M1 upon 2, this one is M1 upon 2, this is this way
and this is this way. Why is there a discontinuity at this point? What is the discontinuity? M1.
What is the applied moment over here? M1. So, this is the discontinuity.
How did we go about applying the displacement method? We actually found out the kinematic
indeterminacy then found out the degree of freedom. Then the next step is, given a unit
displacement corresponding to the degree of freedom find out the displaced shape of the
structure. Once you find out the displaced shape of the structure, you can find out the displaced
shape of each member. For each member then, once you know the displaced shape, you know
the member and rotations and the displacement of one end relative to the other. You can apply
that into your slope deflection equations to get the bending moments at the two ends.
Once you get the bending moments at the two ends, you can then apply equilibrium equations to
relate the unknowns and find out the value of theta1 in this particular case. Once you know the
value of the rotations, then you can always go back, substitute and get the member and moments
exactly. Once you know the member and moments, then you have a statically determinate
structure for which you can find out the bending moment diagram. These are the steps in what is
known as the displacement method.
Do not worry if you have not been able to understand the quickness with which I have gone
through. I have just established the procedure for you today. I have taken a very simple question
and gone ahead with it. This I am going to expound on. Since I know that this is a topic that
you have not covered earlier, I am introducing the topic to you today; later on, you are going to
see how this method is developed over the next many lectures.
16
Prof. P. Banerjee
Lecture 11
Good morning. In the last lecture, we looked at how to use a different method what we
called the displacement method to solve or analyze a statically indeterminate structure. I am
not going to use the word statically indeterminate anymore because if you look at the
displacement method, there is no computation of static indeterminacy in the entire procedure.
In fact, I will show you later that even a statically determinate structure can be solved using
the displacement method. The only point that the displacement method concentrates on is
actually the kinematic indeterminacy or the number of degrees of freedom in a structure
because the displacements corresponding to these degrees of freedom are what we first find
out, based on which we complete the analysis of the structure.
Today, I am going to continue looking at how to use the displacement method. Last time, I
introduced you to the slope deflection equations and we saw how you could use the slope
deflection equations to solve a particular problem. Remember that what we did was, we had a
problem in which we had a single degree of freedom, we actually wrote down an equilibrium
equation corresponding to the degree of freedom and that equilibrium equation actually
enabled us to solve for the unknown displacement corresponding to the degree of freedom,
based on which we could find out the bending moment diagram for that. The way I have
written the slope deflection equations are not complete, I will explain what I mean by that.
Let us look at a particular problem, let us look at this problem.
I have a loading here, let us say P, and I want to know what my Mab and Mba are. If you look
at this particular case, you will agree that given this load, Mab and Mba are going to exist; they
cannot be equal to 0. Do you agree to that? Let us look at the slope deflection equations.
What do the slope deflection equations say? Mab is equal to 4 EI by L thetaab plus 2 EI by L
thetaba minus (6 EI by L squared deltaba); Mba is equal to 2 EI by L thetaab plus 4 EI by L
thetaba minus (6 EI upon L squared deltaba); these are the equations. Now if I look at the slope
deflection equations, what is thetaab? What is the rotation at this point? Since it is the fixed
end, this is equal to 0. What is thetaba? Since this is a fixed end, it will be equal to 0. What is
the relative movement of this point to this point? These points cannot move vertically, so this
is equal to 0. Then, what is Mab equal to? According to this equation, Mab is equal to 0.
Similarly, you will see that Mba is equal to 0, but this is wrong. Why? Because we can just
see that there has to be an Mab and Mba; they cannot be equal to 0. Therefore, the way we
wrote the slope deflection equations are okay as long as you do not have a load on the
member.
When you have loads only on the joints, these kinds of equations are okay. Remember last
time, when I solved, where did I apply the load? The load was a moment applied at the center
support, which was at the joint; it was not on a member; but, this kind of load is on a member
and this is not valid when you have loading on a member. What do we do? How do we solve
this problem? Obviously, slope deflection equations still have to be used because they are the
fundamental equations in the displacement method. How do we solve this problem? Let us
look at the problem that I have defined.
Let us try to solve this problem. In other words, the slope deflection equations that we wrote
are incomplete and I will show you how incomplete they are. In this case, there will exist an
Mab and there will exist an Mba. How do I find out this Mab and Mba? What are these
moments? These are actually the moments at the fixed ends; so, I will call them as (FEM)ab
and (FEM)ba; these are moments at the fixed end. If we are somehow able to compute these
and if we add these to the expressions for the slope deflection equations, then maybe we will
have something. Let us try to find out how we can get this (FEM)ab. If you look at normal
books, actually at the back of the book, you will always see that given various kinds of loads,
they will give you expressions for these fixed end moments.
For example, let me just tell you that if you look at any book on structural analysis and you
say that you have a load at the center, let us say L by 2, L by 2, they will tell you (these are
given in handbooks), this fixed end moment (Refer Slide Time: 08:52) is going to be equal to
P L squared by 8; this is also P L squared by 8, but in the opposite direction it will be in
this direction. Note that the fixed end moment is force into displacement, it is not L squared,
it is PL upon 8 (Refer Slide Time: 09:33). The point is that you will always see that I will
never be able to tell you the formulae, because I really do not try to remember formulae.
Again, as I have said all along, the entire focus of this particular course structural analysis
as I teach it, is not to give you a whole set of formulae that you are going to have to
remember to be able to solve. For example, I gave you the slope deflection equations and
then I explained to you how to actually obtain the slope deflection equations from first
principles.
As far as the fixed end moments are concerned, I am going to do exactly the same thing. I am
going to tell you how to compute the fixed end moments given a load, so that you would
never have to remember a whole bunch of formulae. Of course, if you are just doing an
analysis, it always helps, but when you are actually learning a course, learning a whole lot of
formulae does not, in my opinion, help you in understanding how to solve problems. I am
going to again go back and explain to you how to compute the fixed end moments so that
given any loading (you might have a loading which may not be there in any handbook), you
know how to get the fixed end moments. I am going to spend the whole of today explaining
to you how to obtain fixed end moments and then I will show you how to apply this entire
procedure for a given structure. Let us look at this.
Let us look at this situation. I am going to back to the original where I have a uniform beam
whose length is L and EI is the flexural rigidity at any cross section; it is a uniform beam. In
this beam, let us look at a particular example; I am using this only as an example. You will
see that my procedure can be used for anything and this load is applied at the mid-span and
our entire goal over here is to find out the fixed end moment at a and the fixed end moment at
b. How will we solve this procedure? How will we get these? The procedure is very simple. I
always go by the principle of superposition, so I am going to take this structure.
I am going to say that this structure is equal to this plus. What I have done over here is, I
have taken this structure, which is a fixed beam, I have made it into a simply supported beam
and I have said that in a simply supported beam, you know that the bending moment here and
the bending moment over here are 0. What I have taken is. In this structure, I have made
these fixed end moments (Refer Slide Time: 13:27) as external loads and I have said that this
entire thing is equal to this plus this, but that is not good enough because we know this, we do
not know these, we have to find these. What else has to be satisfied? Note that under this
load, what is going to happen? This is going to become something like this this is the
rotation; I am just drawing the rotation for this. The rotation, the deflection pattern will be
something like this (Refer Slide Time: 14:06). What will be the deflection pattern for this?
Something like this. What do the deflection patterns have to satisfy?
Note that one difference between this plus this (Refer Slide Time: 14:22) is the fact that not
only has the loading to be taken but the displacement here and the displacements here are 0;
the displacement here and here are 0 because of axial rigidity; this point is not going to go;
displacements are 0 at this point. What additional thing is 0 over here which is not 0 here?
The rotations. Therefore, what we have to say is that we are going to compute (thetaab)0 under
the loading. Note that the way I have shown it, it is negative, because my positive is always
anticlockwise and this is (thetaba)0. Now, what we are going to do is get thetaab and thetaba
under the moments. One additional factor that this has to satisfy is that (thetaab)0 plus thetaab
due to the fixed end moment is equal to 0 and (thetaba)0 plus thetaba due to the moment is
equal to 0. Therefore, you have this compatibility that you have to satisfy: the moment at this
point is equal to 0. Do you see what I am trying to find out? I am trying to actually find out
the fixed end moment using the force method.
It is very interesting that I am using the force method to solve, because how many redundant
forces do you have in this particular case? These two (Refer Slide Time: 16:15). Since you
have these two as the redundant forces, I am actually writing down compatibility equations
corresponding to this. To find out the fixed end moments which I am going to use in the
displacement method, I am actually using the force method to compute these fixed end
moments. Let me just go through these steps for this particular one so that I can explain to
you how to do this for a general type of loading.
Let us see what are the rotations. Therefore, the goal here first is to find out the rotations at
the two ends under the loading. Let us apply the load. How do I find out the rotations? Find
out the rotation. Under this load, what would be the bending moment diagram look like?
Again, I am not going to spend time in telling you how to compute the bending moment
diagram; by now, you should know this. It is going to be this way (Refer Slide Time: 17:23)
and this bending moment is this way. This is the bending moment; I leave it up to you how to
obtain it. By now, for a statically determinate structure, you should be able to draw a bending
moment diagram, shear force diagram and any other diagram that you have to draw. This is
my bending moment diagram and since EI is a constant, this M upon EI diagram is going to
be equal to PL upon 4 EI. Simple.
Now, I want to find the rotation here. What would I do? Use the principles of virtual force,
apply a unit force here. If I apply a unit force here, what is the bending moment diagram?
Unit. For finding out this bending moment over here (Refer Slide Time: 18:25), what do I
need to do? Let me first do this. This is the bending moment. What would thetaab be equal to?
1 into thetaab is equal to the internal virtual work, which is m M upon EI dx. Area under these
two curves. Note that I have to take two curves actually. Although explicitly you will say
that you can take this entire thing, find out the area under this curve and draw its centroid (I
know where the centroid is), note that these integrals are always valid only where.
In this particular case, actually this integral, since this expression from here to here (Refer
Slide Time: 19:13) is different from the expression from here to here, actually you have to
take two integrals. For each integral, you need to find out its area under the curve. Therefore,
this one is going to be equal to PL upon 4 EI; length is L by 2, so L by 2 by 2, so that is L by
4; this is the area under this curve. Where is its centroid? It will be at one-third from here,
two-third from here. So, two-third of L by 2 is L by 3, that means add L by 3. What is the
corresponding value? You will see that it will be equal to 2 by 3. Note the fact that this is
sagging and this is hogging (Refer Slide Time: 20:07). So, it is actually minus because the
sum total is minus. Similarly, if I take this side, you are going to have. This is one part; let
me add the next part. The next part again will be minus (PL upon 4 EI) multiplied by L by 4
and this is acting at this point, which is 1 by 3. If you look at the 1 by 3, it is going to be 1 by
3. If you add the two of them up, you will see that thetaab due to the loading is equal to minus
(P L squared upon 16 EI).
Let us look at the consistency of units. P is in terms of Newton, Newton L squared, so it is
Newton meter squared. What is the unit of E? E is Newton per meter squared. I is meter
fourth, so Newton per meter squared into meter fourth is Newton meter squared; this is
Newton meter squared divided by Newton meter squared it is dimensionless. What is the
unit of theta? You will see that it is radians, which is dimensionless. So, this is consistent; this
is my (thetaab)0. Let us find out thetaba under the same loading.
Under the same loading, this is acting at L by 2, same M by EI diagram, I am just drawing it
all over again. This is my M by EI diagram and now I want to find out. This is a, this is b,
this is c, a, c, b (Refer Slide Time: 22:14). Now, I want to find out thetaba. What do I do? I
apply a unit force corresponding to the theta, which is a unit moment, and then find out the
bending moment diagram. The bending moment diagram over here is going to be 1. This is
my small m diagram and therefore, 1 into (thetaba)0 that is the external virtual work, is
going to be equal to the area under this curve. I am not going to go through; I have already
gone through the steps last time; I am just going to write down L by 4. If I take this, it is
going to be 1 by 3; note that both of them are plus, so this is going to be plus (because both of
them are of the same sign) plus PL upon 4 EI into L by 4, this part is going to be 2 by 3 and
therefore, (thetaba)0 is going to be plus P L squared upon 16 EI.
What does that mean? It means that under this loading, the displacement pattern is this way
(Refer Slide Time: 23:51). Note that this one is equal to P L squared upon 16 EI, this one is
equal to P L squared upon 16 EI, note that this is minus which means clockwise, this is plus
which means anticlockwise (Refer Slide Time: 24:09); everything works out perfectly and
since the loading is symmetrical, even the displacement pattern will be symmetrical. We have
found out thetaab and (thetaba)0. Now, we need to find out what are going to be thetaab and
thetaba due to the fixed end moments.
In other words, I am going to put M1 and M2. If I plug that in, what do I get? Let me apply
them separately because anyway I can do a superposition; I am going to take this plus this
(Refer Slide Time: 25:09). Is that okay? This is the same thing, right? Two loads acting, I am
just considering it as two separate loads. I can find out the thetaab due to this moment, find
out the thetaab due to this and sum them up. Under this loading, what kind of moment
diagram do I have? I have M1 and this one is going to be this way, it is linear. How do I find
out thetaab? Apply a unit moment here and for finding out thetaba, apply a unit moment here.
For this, the bending moment diagram is this way: 1 into and for this, it is. For this, its
bending moment diagram is M2. I have drawn all the bending moment diagrams. Why?
Because these are the real loads. Finding out this gives me the curvature diagram and these
are the virtual moments. I am going to find out due to this and due to this and add it. If you
look at it, (thetaab)M due to this is the area under this curve and then, the value of this at its
centroid. What is the area under this curve? M1 upon EI multiplied by L by 2 and it is at twothird the distance from this point, so that value is going to be two-third. If you look at both of
them, they are hogging, so this is plus. This is the thetaab at this point due to this load. Now, I
am going to add the thetaab at this point due to this load. For that, this is the real curvature
and this is the virtual. So, what is the area under this curve? It is going to be equal to M2 upon
EI multiplied by L by 2 multiplied by. This is sagging, this is hogging (Refer Slide Time:
28:02), so it is going to be minus and two-third, you will see that at the CG, the M1 value is
equal to minus one-third. This is my thetaab. Similarly, for thetaba, this is the virtual and these
are the two reals. For this one, the area under this curve is M1 upon EI into L upon 2
multiplied by minus 1 over 3 (at this point, this is sagging, this is hogging, so it is minus 1
over 3) plus M2 upon EI that is due to this loading where this is the real curvature and the
area under the curvature diagram is going to be this, and since these are both sagging (Refer
Slide Time: 28:51), this going to be plus 2 by 3.
I am going to just put all of them together and you will see what I get. What is the big deal
about this? I already know this. Remember we computed last time, L upon 3 EI, L upon 6 EI,
L upon 6 EI; we have already done this. I use this to obtain the slope deflection equations. I
have done this already and I have got the same thing. Now, the only point here to note is.
What is (thetaab)0? Let me write that down. I have already computed (thetaab)0: it is equal to P
L squared upon 16 EI; and (thetaba)0 is equal to P L squared upon 16 EI. The only thing is that
if M1 and M2 are the fixed end moments, then (thetaab)0 plus thetaM; now, I need to find out
those fixed end moments. What I am going to do is I am going to put this plus this equal to 0.
Therefore, you will see that ultimately, if I say that . I am not going to explicitly put them
equal to 0 but I am going to show you what we are going to be doing. What we are going to
say is that this implies that due to the fixed end moments, this is equal to minus (thetaab)0. In
fact, this is what I am going to do: I am going to say that thetaab or thetaba, I am just writing
down, is equal to minus of thetaab due to loading.
Once I say that and I substitute that in, what do I get? I get that for fixed end moments. and I
am going to put down (FEM)ab L upon 3 EI minus (FEM)ba L upon 6 EI) is equal to minus
(thetaab)0, because (thetaab)M is equal to minus and so minus (FEM)ab L upon 6 EI) plus
((FEM)ba L upon 3 EI) is equal to minus (thetaba)0. This is by definition, because this plus this
is equal to 0 for fixed end moments. If I rewrite this, look at what comes up. This becomes L
upon 3 EI, minus (L upon 6 EI), minus (L upon 6 EI), L upon 3 EI into (FEM)ab, (FEM)ba and
this is equal to minus of (thetaab)0, minus (thetaba)0, minus of (thetaab)0 and (thetaba)0. Can I
10
find out the fixed end moments? Sure. Take inverse of this but note what the inverse is; you
have already done this.
You will see that (FEM)ab and fixed end moment at ba is going to be equal to 4 EI upon L.
We have already done this. When did we use this? To develop the slope deflection equations.
If you look at this, this becomes equal to. Minus here because it is minus (thetaab)0,
(thetaba)0. If you look at this, (FEM)ab is equal to 4 EI by L into minus (thetaab)0 plus 2 EI
upon L into minus (thetaba)0. Similarly, fixed end moment at ba is equal to 2 EI by L into
minus (thetaab)0 plus 4 EI by L into minus (thetaba)0. Once I have described this, does this not
remind you of something? This is the actually the slope deflection equations that we have
already developed last time. Therefore, the only thing that you do is to find out the fixed end
moments; you just compute the thetaab and thetaba due to the loading and take the negative of
that; substitute that into the slope deflection equation and that gives you the fixed end
moment. Let us see what happens if I do that for this particular case.
11
We have got for this case P at L by2. What is the fixed end moment? What were (thetaab)0
and (thetaba)0? We have already computed this; P L squared upon 16 EI. I am going to plug
those into my slope deflection equations and what do I get? (FEM)ab is equal to. Let us see.
Let us go back; whenever I rush, I get into trouble.
12
thetaab was negative of (P L squared upon 16 EI) and thetaba was plus P L squared upon 16
EI. Let us put that in.
13
If we go back there, we actually see that thetaab is minus of (thetaba)0, which is minus; those
are the values; I am going to plug in the values of the fixed end moment. I will have 4 EI
upon L that is the slope deflection equation, negative of (thetaab)0. Since (thetaab)0 is minus,
it is going to be negative, it is going to be plus, so plus, then plus 2 EI upon L. What is the
negative of (thetaba)0? It is minus, so I am going to put minus (P L squared by 16 EI). The
fixed end moment at ba is equal to 2 EI upon L; again, minus of this is P L squared by 16 EI
plus 4 EI upon L into minus (P L squared upon 16 EI). If you do this, this 4, 4, EI, EI cancel;
EI, EI cancels; EI, EI cancels; 4 goes into this 4 times; this goes into it 8 times; L takes away
this square; this L takes away this; so, you have PL upon 4 minus (PL upon 8), which is equal
to PL upon 8. If you look at (FEM)ba, you will see that EI, EI cancel; L cancels this; 2 cancels
this; this cancels; this cancels this; EI, EI cancel; so, you have PL upon 8 minus (PL upon 4),
which becomes minus (PL upon 8). What does that mean? Fixed end moment positive PL
upon 8, is anticlockwise and minus is clockwise.
So, under this load, PL upon 8, PL upon 8 we have obtained this from first principles. Now,
I am going to quickly go ahead and look at some other kind of loads so that you can convince
yourself that you can compute all the fixed end moments that you have in handbooks, using
this method. What are the steps in this method? I am going to write down the steps.
First, take the simply supported and obtain thetaab and thetaba due to the loading. Two, get the
fixed end moments (this is fixed end moments; this is my notation) by substituting minus
(thetaab)0 and minus (thetaba)0 into the slope deflection equations. Let me again take this.
14
Let me take this case: w Kilonewton per meter, length L, EI. What are the fixed end
moments? Take a simply supported beam and take udl with w. What does my bending
moment diagram look like? My bending moment diagram is a parabola where this is equal to
w L squared upon 8 EI (Refer Slide Time: 41:53). My curvature diagram or M by EI diagram
is going to be w L squared upon 8 EI. How do I find out (thetaab)0? Put a moment at this
point. This is sagging moment, this one is hogging moment, and for this moment, apply a
moment there. 1, this is my m1, this is my m2 and therefore, (thetaab)0 is equal to area under
this curve. What is that equal to? It is equal to two-third w L squared upon 8 EI into L. Where
is the area under this curve? Where is its centroid? The centroid is in the center. For this, this
is sagging, this is hogging, so centroid value will be minus half. Subtract, subtract, it is going
to be minus (w L cubed upon 24 EI); similarly, for (thetaba)0, you will get plus w L cubed
upon 24 EI. Take the negative of these and substitute into the slope deflection equations.
15
What do you get? Fixed end moment at ab is equal to 4 EI by L into negative of minus omega
becomes w L cubed upon 24 EI plus 2 EI upon L and thetaba is plus, so negative of that is
minus w L cubed upon 24 EI. If you look at this, L, L will make it w L squared; EI, EI
cancels, 6, it is going to be minus w L cubed by this. If you put it together, you will see that
this turns out to be w L squared upon 8. Similarly, when you find out the fixed end moment
at ba, you will find it is equal to minus (w L squared upon 8). What does that mean? That
means that under this loading, FEM over here is going to be w L squared by 8 (Refer Slide
Time: 44:46), and this is going to be clockwise w L squared by 8 (Refer Slide Time: 44:48).
Check and you will see that this is indeed what is given in the handbook.
Let me finally do one more problem and convince you that I have not constructed this to
make life complicated for you. It is just that for any loading, you can always put that same
loading on the simply supported beam, find out the rotations of the two ends, put the negative
of the two rotations into the slope deflection equations and you will get the fixed end
moments.
16
I am going to do a final one before I show you how this is going to get used. This is P and
this one, I am going to say, is applied at a and b from the right end, where the total length is L
in EI. In this case again, I am going to just go through the steps very quickly because we have
already spent a lot of time on this. Note that if you draw this, you will see that the bending
moment diagram looks like this; this is why the bending moment diagram and L EI is going
to be equal to the M by EI diagram. Again, for thetaab, this is my virtual moment diagram and
for thetaba, this is my virtual moment diagram. These two are different. Let me find out the
area under this curve, then the area under this curve and add.
17
1 into (thetaab)0 is equal to. The area under the left-hand curve is going to be equal to Pab
upon L EI this is the area under this curve, multiplied by a by 2, multiplied by its centroid.
What will be the centroid? If you note, the centroid is going to be two-third a from this side;
or conversely, I can say that it is equal to L minus (2 by 3a) from this side. What will be the
value at this point (Refer Slide Time: 48:10)? You will see that this is going to be equal to L
minus (2a by 3); I am going to put that down; it is going to be equal to L minus (2a by 3);
then, the entire thing divided by L. Note that since this is hogging (Refer Slide Time: 48:37),
this is going to be minus on the outside; the area under the other curve is going to be equal to:
b by 2 this is the area under this curve and it is at two-third b from this end. The value of
this is going to be equal to again minus outside, two-third b upon L; because it is 1 at L, this
is going to be two-third b upon L; this is thetaab.
Let me put these things down properly. L upon L is 1, so it is going to be 1 minus (2a upon
3L). I am going to make this P a squared b upon 2 EI and inside, it is going to be equal to 1
minus (2a upon 3L) plus P a b squared upon L EI, multiplied by minus (2b upon 3L) this is
thetaab. Similarly, you will see that thetaba is equal to just the same and just the opposite way;
it is going to be Pab squared upon L EI; the only difference you will have over here is that.
over here, all of them are minus, so this is going to be minus and plus; I am taking the minus
inside; it will be minus and plus; so in this case, this is going to be positive because both of
them are sagging; so, it is positive. You will have Pab squared upon L EI; on this side, it is
going to be 1 minus (2b upon 3L); and then, you are going to have P a squared b upon L EI
multiplied by 2a by 3L. These are very complex equations but simple to actually go through.
18
Now, I have to substitute the negative of these into the moment equation. If I plug that in,
what do I get?
I will just do it for Mab and then for Mba, I can just write it. The fixed end moment at ab is
equal to 4 EI upon L into minus of thetaab, it is going to be P a b squared upon L EI and
minus here (Refer Slide Time: 52:24), so you will see that it is going to be equal to 1 minus
(2a upon 3L) plus P a b squared upon L EI into minus of that becomes plus, so, this is
going to be 2b upon 3L plus 2 EI upon L minus of that, this is going to be P a squared b upon
L EI, into minus of (2a upon 3L). Then, I have plus P a b squared upon L EI and that is going
to be equal to 2b upon 3L minus 1. Now, I am going to substitute all of these in and write it
out throughout explicitly and then what I get is this. I am going to take Pab outside just for
the sake of completeness, so I get Pab upon L. Note that EI, EI cancels out, so this is going to
be ab upon L. What do I get inside? Note that I have taken L and L outside, so I get just 4
into a, 4a; then, minus (8 a squared upon 3L). Here, I get 4 into 2b, so it is going to be plus 8;
then, b comes out, it is going to 8 b squared upon 3L; I get that from these two terms and then
I am going to find out from these two terms. I have taken Pab outside, I have a here, I have 2
here; so, this is going to be equal to minus (4 a squared upon 3 L) 2 into 2 is 4a and on
this side, I am going to get plus 2 into 2, which is 4; ab goes out, so it is b squared upon 3L;
then, minus 2b. Here, if I take Pab upon L squared, you will see that this becomes 12 a
squared upon 3L. What is that equal to? I am just going to write it down; 4 a squared upon L
plus b squared; so, what I get is 4 a minus (4 a squared upon L minus (4 b squared upon L))
minus, minus, you get plus over here, that is what I have done minus 2b. Now, note that this
19
becomes a plus b by L, 4 times ab plus L and you can substitute all of those in; ultimately,
you will see that this turns out to be this.
When you substitute, you will get. This is P, put a and b, you will get that this is equal to P
into a b squared by L squared and this turns out to be P a squared b upon L squared; now,
substitute. If you want to check, put a and b as L by 2, L by 2 and you will get PL upon 8, PL
upon 8, which is what you got earlier. I am going to stop over here.
20
Prof. P. Banerjee
Lecture 12
Good morning. In the last lecture, we found out how to get fixed end moments due to load on
the member itself. Today, what I am going to do before I start off looking at problems etc., I
am going to give you a few exercises which I would like you to try out on your own using the
procedure that I developed in the last lecture. What I am going to do is, I am going to give
you a few exercises.
I have already done this for you: this is Pab squared upon L squared, this is Pa squared b
upon L squared, this is L and of course, it is a uniform beam. Then, I will give you a few
others. This one I have already done; so, using exactly the same procedure, I would like
you w over length l, this is equal to w L square upon 12; this also I have done for you, do
some of the more interesting ones. This is a linearly varying intensity of load where w is the
intensity at the end. This is a partial uniform loading, uniform loading over part of the,
where this is a and this is l minus a; this is equal to w a squared upon 12 into (6 minus 8a
upon l plus 3 a squared upon l squared); and this is w a cubed upon 12 l into (4 minus 3a by
l). Satisfy yourself that if you put in a is equal to l, then you are going to get this. This is
important because if a is equal to l, then you have to recover this (Refer Slide Time: 05:08)
from this.
Finally, I will give you one more. Acting at a distance a and b and then, this is this way, this
is this way, this is equal to M upon l squared a into (2b minus a); this is equal to M upon l
squared b into (2a minus b). Satisfy yourself that if a is equal to b, then this has to be equal to
M upon 4; these two would be M upon 4. I would like you to actually look at all these
solutions that I have given you; you should be able to establish the fixed end moments by
using the method that you have already developed; so much for all of this. What does this do
to the slope deflection equations?
1
What is the definition of fixed end moment? Let me first write down the entire equation.
These are the complete equations; look at this. If thetaab is equal to 0, thetaba is equal to 0 and
deltaba is equal to 0 that is the fixed fixed case, what is Mab equal to? The fixed end
moment. That is what it should be. Similarly, this (Refer Slide Time: 08:06). These are the
final slope deflection equations which actually present the member force displacement
relation; essentially, if you look at this, this is the member force displacement relations that
you have. Now, this is the final equation that we are going to be using. Of course, we will see
later that there are several things that come out of this, which we have to use, but now we will
start looking at problems and see how we can solve those problems to be able to get your
solution for equations. Let me give you one thing. What I want to give you is that you should
be able to. Remember the problem that I had given you?
Remember this problem? Try solving this problem and defining the degrees of freedom.
Right now, how many degrees of freedom would this have? One, two, three, three degrees of
freedom and you can set up three equations etc. I leave it up to you; you can solve this. Let
me solve some other problems and I will revisit this problem in a different light, I will talk
about it a little bit later.
Let me now look at some standard problems that we can solve. The loading here (Refer Slide
Time: 10:35), this is 10 meters, 10 meters, this loading is 4 meters, this load is 120
Kilonewton and I have a loading here, this is 50 Kilonewton per meter. You have given the
fact that this is EI is the same; the multi-span uniform beam continues over two supports. Let
us now do this and I will write this as a, call this b and I will call this c. EI is a constant. Let
us look at this particular problem. How many degrees of freedom? One, two, three; so three
joints into three is equal to nine. How many restraints? These are the unrestrained degrees of
freedom, restraints are three here because it is fully fixed, one here and one here, so that is
five. Then, we have constraints. What are the constraints? The constraint is that both the
members are actually rigid; if they are actually rigid, then one into two (one constraint per
member, there are two members), so two, the total degrees of freedom are two. What are
they? They are the rotation here and the rotation here (Refer Slide Time: 12:45). Let me look
at what is the possible kind of situation that you are likely to have.
Look at this. What is this? This by definition is thetab and the way I have shown it, this is
actually minus thetac because it is clockwise. I can define it in this fashion. Let me actually
do it in that way since that will make it easier for us to understand. I will draw this and then
do this (Refer Slide Time: 14:03) so that by definition, this is thetac. That means what we
have are two unknowns. Here, the thetaa is equal to 0, we do not know thetab and thetac; what
we do not know are thetab and thetac, but for the member boundary conditions, we know
theta; thetaab is equal to 0, thetaba is equal to thetab, thetabc is equal to thetab and thetacb is
equal to thetac. What is deltaba? Since there is no deflection, deltacb is equal to 0. Always note
that the delta is of the right end relative to the left end and that is why it is always done in this
fashion. These are our boundary conditions and compatibility conditions where the unknown
displacements are these. Therefore, the overall thing is that we have to use the slope
deflection equations to be able to find these values; once we find these values, if we find
these two (Refer Slide Time: 15:51), remember that we can find these and then we can find
out the member end moments and that becomes a statically determinate structure. Let us look
at how to solve the problem. The whole point here is that once we have written these
compatibility equations, we write the slope deflection equations for each member. What
would be the slope deflection equations?
Slope deflection equations would be Mab is equal to 4EI by L thetaab plus 2EI by L thetaba
minus (6EI by L squared deltaba) plus the fixed end moment due to any load on the member
ab. If I substitute the values of thetaab and thetaba that I have written down here (this is my
thetaab, thetaba and deltaba, if I substitute those in, what do I get? I get 2EI by L thetab plus
fixed end moment at ab. I am not going to write down this equation, I can write down the
same equation. Similarly, when we put it in, we will get 4EI upon L thetab plus fixed end
moment at ba. So, you see Mab and Mba are in terms of thetab. I will find out these fixed end
moments a little bit later. Let me write down the equations for Mbc by writing down the
equations of motion, slope deflection equations and substituting that thetabc is equal to thetab,
thetacb is equal to thetac, and deltacb is equal to 0, what I get is the following: 4EI upon L
thetab plus 2EI upon L thetac plus fixed end moment at bc; similarly, Mcb is equal to 2EI by L
thetab plus 4EI by L thetac plus fixed end moment at cb. We have got these equations, but we
need to find out what these are for the given loadings. What is the loading on ab? Note that if
there is no loading on member ab, then these (Refer Slide Time: 19:41) will be automatically
0, but let us see what are the loads on the member.
Member ab has a load here, which is equal to 120 Kilonewton. This is a (Refer Slide Time:
20:03), this is b, this is 4 meter and this is 10 meter. I need to find out these fixed end
moments by plugging it into the formulae that we have already developed. Note that my
definition of fixed end moment is positive if it is anticlockwise, so this is going to be this way
and this is going to be this way. This (FEM)ab is positive, it is (plus 120 into 4 into 6 squared)
upon L squared. What is that equal to? That is equal to plus 172.8 Kilonewton meter.
What about fixed end moment at ba? If you look at fixed end moment at ba, that is going to
be negative (120 into 4 squared into 6) upon 10 squared, this is going to be equal to minus
115.2 Kilonewton meter. These are the fixed end moments at ab and ba in member ab due to
this 120. Similarly, I can find out what is the loading on member bc. The loading is 50
Kilonewton per meter; this is 10 meter, so these are going to be this way. My fixed end
moment at bc is equal to plus (w L squared) by 12 which is equal to plus 416 Kilonewton
meter, and fixed end moment at cb is equal to minus (50 into 10 squared), minus because I
am assuming anticlockwise to be positive please note that and please note my sign
convention that I assume anticlockwise to be positive, others may choose clockwise as
positive. There are various ways of solving this problem but understand that as long as you
use one particular method, then you have to be consistent and you cannot keep changing your
values. This is my (FEM)ab, this is my fixed end moment at ba, this is fixed end moment at bc
and this is my fixed end moment at cb. Let me plug those into my equations. What do I get?
I get that Mba is equal to 4EI and now I am going to put in the value of 10, so it is 4EI by 10
thetab minus 115.2; Mab is equal to 2EI upon 10 thetab plus 172.8; Mbc is equal to 4EI upon
10 thetab plus 2EI by 10 thetac plus 416.7; Mcb is equal to 2EI by 10 thetab plus 4EI by 10
thetac minus 416.7. These are my member end moments. Let me look at what I have to do
now. I have to actually write down two equilibrium equations so that I can solve for thetab
and thetac. Let us see what those equilibrium equations are.
Let us look at the equilibrium conditions here. At joint a, you do not have any equilibrium.
The equilibrium says that Mab is equal to the moment at a; that does not give you anything.
Let us look at joint b. At joint b, what do we have? From ab, we have Mba, then we have the
joint and then from cb, we have Mbc. If I look at the joint itself, what do I have? Mab and Mbc.
Then, if I take the moment equilibrium at the joint, what do I get? I get Mba plus Mbc is equal
to 0.
7
Similarly, if I look at joint c, what does it tell me? This is Mcb and by definition, since the
moment is 0, Mcb is equal to 0 these are my two equilibrium equations. Now, I substitute
Mba and Mbc that I have written down earlier and if I put those in by substituting these that I
evaluated into these equations, what I get is 8 EI upon 10 thetab plus 2EI by 10 thetac is equal
to 301.5 and the other equation is 2EI 10 thetab plus 4EI upon 10 thetac is equal to 416.6. So,
I have two equations in thetab and thetac and I can actually solve this; the easiest way to solve
this is by writing it in matrix form and solving it.
I am going to put 2EI by 10 outside, so inside what I have is 4, 1, 1, 2 into thetab, thetac and
this is equal to 301.5 and 416.7. Then, what do I get? If I solve for this, I get thetab is equal to
(728 upon EI) radians, thetac is equal to (minus (1406 by EI)) radians. I have solved for thetab
and thetac and I can actually draw the deflected shape now, because if you look at the
deflected shape that we had drawn earlier, now I know what thetab is and I know what thetac
is.
Essentially what this means is that the actual deflected shape goes in this fashion, this is
negative and this is positive. Anyway, that does not really matter; what was my interest? My
interest was in the bending moment diagram. Now, when I have these and I put that into the
equation that I have here and since I know thetab and thetac, I can put those in and get my
Mab.
My Mab is equal to 27.2 Kilonewton meter, Mba is equal to minus 406 Kilonewton meter, Mbc
is equal to plus 406 Kilonewton meter and Mcb is going to be equal to 0 Kilonewton meter. If
I look at this, what am I getting? I am getting the following.
Let me now draw this. I am going to break whatever I have broken these up, this is my
moment at the fixed end. This is equal to . To balance this (Refer Slide Time: 33:12), I
need something like this and to balance this, I need Mab. So, this Mab is equal to 27.2
Kilonewton meter. Then, I have this load 120 Kilonewton then I have minus 406? What does
minus 406 mean? Minus 406 means it is clockwise, so clockwise and this way. Then, I have
this, this will be going this way and what is this value? This is 406. Note at bc also, I have
406, anticlockwise, so this is correct, it goes this way and then on this side, what do we have?
0. We have this and now, we have this. These are my member end moments that I have
obtained over there; I have put them in here and now, I need to find out the shear forces also.
How do you evaluate that?
Take member ab. Member ab at this end has 27.2. I am just writing down the values, I am not
writing down the dimensions because we already know what the dimensions are. In addition
to that, at 4 meters, you have 120. How do I obtain the shear force at a? To find out the shear
force, I can take moments about this end because I have shear force over here too. What
would this shear force be equal to? Take moments about this end. What you will have is Va
into 10, that is clockwise, then this is anticlockwise, so minus 27.2, this is clockwise, so it is
plus 406, and this is anticlockwise (Refer Slide Time: 36:23), so this is minus 120 into 6 is
equal to 0. This gives me that the shear force is equal to 720, so it is going to be 72 and
then minus, this is going to give you 379.3, I will just put that in, divided by 10, so it is going
to give minus 37.9 (I am dropping the 3, of course) and so Va is equal to 34.1 Kilonewton.
Let me look at this one here. Obviously, the shear at b, Vb is going to be 120 minus 34.1 and
that is equal to 85.9 Kilonewton meter. This is the shear at a and similarly for this (Refer
Slide Time: 37:41), with the loading of 50 Kilonewton meter, you will see that V at c turns
out to be equal to 290.7 Kilonewton meter that essentially comes from 406.5 divided by 10
plus 250, and V at d turns out to be equal to 209.3 Kilonewton meter. In one place, you have
additional and in one case, you have less. These are our shears. Now, let us draw the bending
moment diagram. How does it look? The bending moment diagram would look in this
fashion.
10
Let us draw the shear force diagram first. You know that the shear force diagram is a simple
one; I am going to draw it in one direction and be consistent, I am drawing it in this direction.
What is the value? 34.1. What is the direction? It is this way up, this way down. That is a
constant till we hit 4 meters; from 4 meters, it goes to minus 85.9; this minus does really
matter, plus, minus is really irrelevant for us. This is going to be this way. In reality, I always
consider this as positive but it does not matter, I am just drawing it. Then, what do we have?
It has to go to 290.7 on the other side; so, here, it is going to go up, it is going to be 290.7;
from there, because it is linear, this is going to be 209.3 and because it is UDL, it is going to
go like this (Refer Slide Time: 40:30), this is going to be this way, this is going to be this way
and this is my shear.
Now, what is the reaction at b? The reaction at b is going to be 290 plus 85 is equal to 376.6
that is the upward reaction at b. What is the reaction at c? The reaction at c is going to be
209.3. That is my shear force diagram and now let us look at our bending moment diagram.
This is going to be equal to 27.2, then it is going to go down (because this is constant) to this
value, which is going to be equal to 109.5 Kilonewton meter and from there, it is going to go
up at this point to 406.5 that is the bending moment at b. Note that this goes this way, this
goes this way, ultimately, this goes this way and then 0; at this point, it just turns out to be
438 Kilonewton meter. This is our M diagram and this is again this way (Refer Slide Time:
43:10). This is your bending moment diagram and shear force diagram and that is the total
analysis of this particular problem.
This, in essence, gives you what the displacement method is. In the displacement method,
what did we first do? We first identified the degrees of freedom. Once we had the degrees of
freedom, what was our next step? The next step was to evaluate the fixed end moments in
each member; the fixed end moments will only exist if there is a load on the member; if there
is no load on the member, then it does not exist. You find out the fixed end moments; now, I
have already told you that you should be able to find out the fixed end moments given any
loading from first principles that is what I did in my last lecture; you should be finding out
the fixed end moments.
11
Once you know your fixed end moments, you write down the member force displacement
relationship which is given by the slope deflection equations. Substitute everything in. Next,
write down the equilibrium conditions; the number of equilibrium conditions will always be
equal to the number of degrees of freedom that you have. Essentially, the equations are in
terms of the unknown displacements or in this particular case for the beam, they are the
rotations. You find the rotations by solving those equations and once you find out the
rotations or the displacements, what do you do next? You then substitute back into the
member force displacement relationships, which are the slope deflection equations. Now, you
can find out the member end moments; once you know you the member end moments, then
you can take each member and find out its reactions, its shear force, bending moment,
everything; then, you put it all together and this is what you have. I am just going back to the
original problem this is the original problem.
Given this loading on this multi-span beam, we use the displacement method to generate the
shear force and bending moment diagram. To take it forward, let us look at another aspect of
this particular problem and that is, what do we do when we have support settlement?
Remember that I talked about support settlement when we looking at the force method. Let
me take this particular problem itself.
12
This is 10 meters, 10 meters; no loading on the structure; remember I said that there is no
loading on the structure. Why? Because we can always consider the loading to be done
separately and we can then consider other conditions. Remember that the principle of
superposition is always valid in all the problems that we do because we are only considering
linear elastic structures; when you have linear elastic structures, you always can superpose.
Let us just say that due to the loading that we considered in the previous case, this support
settles. We have already found out what effect the load has, now we want to find out what
effect this settlement has; let us just assume a particular value and it goes down by 0.03
meters. Then, what do we have? This is going to become something like this (Refer Slide
Time: 47:51). Again, I have thetab unknown and thetac unknown. Only thing is, since there is
no loading, fixed end moments are all 0; they are all 0 because there is no loading.
Just quickly, you can go back to my equations, I am just writing down one of them. Substitute
these values in and we have the same compatibility conditions. Mab is equal to 2EI by 10
thetab. Do we have a deltaba? Note that we do have a deltaba. This is gone down relative to
this. I had always said that deltaba is positive going up, so when it is going down, it is
negative. Here, you have deltaba as minus 0.03 meters. What about deltacb? If you look at
deltacb, ba has gone down relative to c, which means c has gone up and so, it is positive, plus
0.03 meters. If I substitute this in, this becomes plus 6EI upon L squared multiplied by 0.03
meters; this is the way that you have to solve it.
What I will do is, I am going to spend the next lecture looking at how you can solve this
particular problem in more detail. Now, let us just put down some values, because I will
require the values of E and I to be able to get estimates on everything that we have here. I am
going to put down some values of El: E is 200 gigapascals and I is equal to 2000 into 10 to
the power of minus 6 meter fourth. These are some of the values that you have been given
and so, once you know E and I, you will be able to solve for this. More about this when I go
into my next lecture. Just to review, what is the procedure for the displacement method?
13
One: determine the number of degrees of freedom and identify them; two: determine the
fixed end moment if loads on members exist; three: write the slope deflection equations for
each member; four: write the equilibrium conditions for the structure; five: solve the
equations for unknown displacements; and six: substitute the known displacements into the
slope deflection equations so that you can get your member end moments and once you get
your member end moments, you should be able to solve it.
I am going to stop here and in the next lecture, I am going to take up this particular problem
and show you how to solve this problem for support settlement using the displacement
method. Thank you very much.
14
Prof. P. Banerjee
Lecture - 13
Good morning. In the last lecture, we looked at how to solve a beam problem using the
displacement method and in the end, I gave you a problem which included a support
settlement; we are going to look at that problem in this lecture. Let me just reiterate that
particular problem.
What you have is this, the support has settled and the support settlement is 0.03 meters. So,
what will the deflected pattern look like? This is thetac (Refer Slide Time: 02:15) and this is
thetab the two unknown rotations, the two displacements corresponding to the degrees of
freedom; so, thetab is unknown, thetac is unknown.
Going back, what do we have? By compatibility condition, thetaab is equal to 0 and thetaba is
equal to thetab; deltaba, now, this is interesting. What is deltaba? deltaba in the previous case
was 0, but here, you see b, a, c; b has gone down relative to a, so that means deltaba is not 0.
Remember deltaba was considered to be positive if b went up relative to a; since it is going
down, it is negative and the value is 0.03 meters. Let us look at member bc; thetabc is equal to
thetab, thetacb is equal to thetac and what is deltacb equal to? b has gone down relative to c, so
we can say that actually c has gone up relative to b, so it is plus 0.03 meters. Note that we are
given that E is equal to 200 GPa and I is equal to 2 into 10 to the power of minus 3 meter
fourth these are what are given; we can substitute them into equations. By the way, what
are the fixed end moments?
Since there is no loading, note that (FEM)ab, (FEM)ba, (FEM)bc and (FEM)cb are all 0. Now,
substituting into the equations, what do we get? Let us see what we get. We get the equation
Mab minus 2EI by 10 thetab plus note that it is actually minus 6EI L squared into delta but
since delta is negative, this is going to be 6EI upon 100 into 0.03. Then, Mba is equal to 4EI
upon 10 thetab plus 6EI upon 100 into 0.03 and Mbc is equal to 4EI upon 10 thetab plus 2EI
upon 10 thetac minus 6EI upon 10 squared into 0.03. Note that in bc, deltab is positive, so that
is why this is this way and Mcb is equal to 2EI upon 10 thetab plus 4EI upon 10 thetac minus
6EI upon 10 squared into 0.03. If I plug in the values of EI, see what I get.
I get Mab is equal to 2EI by 10 thetab plus . You plug in the value of EI, note I am just
going to go through a little bit; let me show it to you, since you might be a little bit confused.
What I am doing is 6EI upon 10 squared into 0.03; EI is 200 Gpa, which if you write in
Kilonewton (note that everything is in Kilonewton) becomes equal to 200 into10 to the power
of 9 Newton per meter squared; this is going to be equal to 200 into 10 to the power of 6
Kilonewton per meter squared divided by 100 meter squared multiplied by I. What is I? 2
into 10 to the power of minus 3, that is 6EI upon L squared meter fourth and then we have
0.03 meters. Let us see what units I land up getting.
This is going to be equal to Kilonewton (Refer Slide Time: 09:20), this is going to be meter
fourth at the bottom, meter fourth cancels here, so this is going to be Kilonewton meter, this
goes into this, this goes into this three times and this 3, if I bring it here, this will become 1;
this will become 6 into 6, 36, 36 into 2 is 72, into 10 is equal to 720; that is what I have over
here, this is 720 Kilonewton meter. These are the units and then, 4EI upon 10 thetab plus 720
Kilonewton meter; Mbc is equal to 4EI upon 10 thetab plus 2EI upon 10 thetac minus 720, the
same; and then, Mcb is equal to 2EI upon 10 thetab plus 4EI upon 10 thetac minus 720. Now,
the equilibrium equations are the same.
The equilibrium equations are Mba plus Mbc is equal to 0 and Mcb is equal to 0. This gives us
that Mba plus Mbc is equal to 8 EI upon 10 thetab plus 2EI upon 10 thetac. If you have plus
720, minus 720, that cancels out and this becomes 8 EI upon 10 thetab plus 2EI upon L thetac
is equal to 0. The second equation Mcb is equal to 0 gives us 2EI upon 10 thetab plus 4EI upon
10 thetac is equal to 720. Solving for this gives us that thetab turns out to be equal to (minus
514 by EI) radians and thetac is equal to (plus 2057 by EI) radians. What does that mean? It
means that if I were to draw the deflected shape over here due to this settlement (Refer Slide
Time: 12:58), the system becomes this way; this is positive and it is clockwise because this is
negative and this is anticlockwise (Refer Slide Time: 13:27) because it is positive. This is the
deflected shape and now, once we get thetab and thetac, we can substitute it back into the
equations for Mab.
In other words, once I have found out thetab and thetac I can substitute thetab and thetac into
these equations, plug in the values of thetab and my final Mab is equal to plus 617.2
Kilonewton meter, Mba is equal to plus 514.4 Kilonewton meter, Mbc is equal to minus 514.4
Kilonewton meter and Mcb is equal to 0. Once you get these moments, then you can draw the
bending moment diagram just we as we had done last time; it is not very different from what
we had done, excepting that there is no loading.
The interesting point to note here is how much did the moments come out to be? Think about
the loading; you had 120 Kilonewton loading and then we had 500 Kilonewton on 10 meters
and what kind of forces did we get? We got much less moments than we got due to a
settlement of only 3 centimeters. In other words, settlement problems can actually give rise to
significantly higher stresses in structures than loads can and that is the reason why support
beams are always very very support settlement can have very differential settlement, can
have extremely problematic consequences for your structures unless you design them for
these kinds of loads. So much for this particular analysis; I think I have done enough number
of problems for beams.
What I would like to now do is, I would like to go on to solving a particular problem for a
frame, so that it gives you an idea of how the whole concept of displacement method can be
used for frames. You will see another very interesting thing and that is, in a beam, we only
had thetaab and thetaba; in frames, you will see that deltaba will also come in and it will be an
unknown. In a beam, the only problem that we did where deltaba came in was when we had
support settlement; we knew the value of delta. In frames, you will see that that will also be
an unknown and therefore, you have some interesting equations of equilibrium that you have
to develop when you go through it. Let me take a very simple problem and go through it for
you.
This is a frame that we have and if you look at this particular frame, it has fixed supports at a
and d and it is a single-storey, single bay frame with an overhang. At the end of the overhang,
you have a load, you have one load in member bc; member ab has a load, bc has a load, ca
has a load and cd has a load. In this particular case, you will see that I am going to eliminate
ce because I am not quite interested. If you look at ce, can I draw the bending moment
diagram for ce? Sure, I can.
At this point, the bending moment is going to be 0 (Refer Slide Time: 20:09) and at this
point, it is going to be 50 into 5, 250 Kilonewton meter; I know the bending moment in here,
so I do not really need to consider this. So, what I am going to do is, I am going to consider
that this load (Refer Slide Time: 20:31) is going to get transferred to c with the appropriate
equivalent forces. With this problem, I can solve it as far I need to only solve abcd with the
given fact that the loading from ce comes out to be this.
I can solve this problem. 15 meters, 20 meters, load here (Refer Slide Time: 21:12), load here
and this load, I can transfer here as 50 Kilonewton, but just transferring the load is not good
enough. What happens? Equivalent load system, I also have to transfer a moment and the
moment is equal to 250 Kilonewton meter (Refer Slide Time: 21:29). Now, I am solving
abcd. This is the problem that we have to solve using the displacement method.
What is the first step? The first step is to find out the degrees of freedom. How do you find
out the degrees of freedom? Let me go through all the steps of the degrees of freedom. How
many nodes? 1, 2, 3, 4 so four joints; three into four is the unrestrained degrees of freedom,
twelve. How many restraints? Restraints are three here and three here (Refer Slide Time:
22:22) because you are restraining all three degrees of freedom at this point and you are
restraining all three degrees of freedom, so six restraints. Constraints? I have three, axial
deformation is neglected, three members, three constraints. So, how many degrees of
freedom? Three. What are those three degrees of freedom? Theta at b, theta at c (Refer Slide
Time: 22:57) and delta (Refer Slide Time: 23:03); this delta is the same as this delta because
bc cannot actually deform.
Let me do that again; I have thetab, thetac and delta. Now, I am going to introduce some
concepts and that is, what do those? Before I write down the slope deflection equations, I
have to be able to relate these (Refer Slide Time: 23:57) with respect to the member end
deformations. For that, I am going to now start a new procedure by which I am going to find
out, given each degree of freedom, how the displacement occurs, so that I can always write
down in terms of their displacement what the member end displacements are; then, when I
take all three of them together, that becomes a sum total.
Here, I am going to be drawing. What happens when you have? thetab I am going to put.
Since I need to evaluate it together, I am going to draw it piece by piece; so, thetaba is equal
to 1. How will the displacement look? thetaba is 1, so, it is going to go like this, this is going
to go like this, this point cannot go because delta is equal to 0 and thetac is equal to 0, so this
has to be fixed and then, this remains exactly this way; this is my 1, 1.
Let me write down all the member end displacements relative to that, but first I will draw all
of them and then I will write them down. Let me put thetac is equal to 1. What is thetac is
equal to 1? You will see that if thetac is equal to1, this will go like this, this will go like this
(Refer Slide Time: 25:59), this will come here and finally, delta is equal to 1 and is going to
be this way; note that when delta is equal to 1, thetab and thetac is equal to 0, so this is going
to go this way (Refer Slide Time: 26:25). The reason why I am doing this is that you know,
if I show all of them together, it is going to look very very complicated; so, what I do is I take
each one separately and then I know that all three act together. Once I know all the shapes
for this is equal to 1 and here, this is 1 and this is 1. Once I have this, I can write down
each and every for member ab, what are the deformations in terms of thetab, in terms of
thetac, in terms of delta; and now, with this, look at how their deformations go.
I am now going to write down the kinematic relationship and that is, for member ab, what is
thetaab equal to? Look at this. thetaab is always going to be 0 because at that particular point,
you have the theta will be 0. What is thetaba equal to? Let us see thetaba. When thetab is equal
to 1, thetaba is equal to 1 and so thetaba is equal to 1 into thetab; when thetac is equal to 1,
thetab is equal to 0; when thetadelta is equal to 1, thetab is equal to 0 and therefore, we are
going to write down thetaba is equal to thetab only. Similarly, write down deltaba. When thetab
is equal to 1, what is deltaba? 0. When thetac is equal to 1, what is deltaba? 0. When delta is
equal to 1, what is thetaba? Let us look at this, deltaba (Refer Slide Time: 28:45), so ab has to
go up this way (Refer Slide Time: 28:47) for it to be positive; it is going down, so it is
actually minus 1, so it is minus delta.
Similarly, let us look at member bc. What about member bc? thetabc is equal to what? What is
bc equal to? Let us look at it. bc is equal to 1 into thetab (Refer Slide Time: 29:24) plus 0 into
thetac plus 0 into delta; that means thetabc is equal to thetab. What about thetacb? thetacb is
equal to 0 into thetab plus 1 into thetac plus 0 into delta, so it is equal to thetac. Similarly,
deltacb; c relative to b, 0 into thetab plus 0 into thetac plus 0 into delta; note that this has to
move this way for it to be delta, so deltacb is equal to 0.
Now, I am looking at member cd, this is the way I am looking at it (Refer Slide Time: 30:28).
I am looking here to here. What is member cd? deltab, cd is equal to 0 into thetab plus 1 into
thetac plus 0 into delta. So, thetacd is equal to thetac, thetadc at fixed end is 0; what is deltadc
equal to?
Let us look at it. deltacd is equal to 0 into thetab (Refer Slide Time: 31:05) plus 0 into thetac
plus now, let us look at this, what does this mean? cd, d has to move in this direction for it
to be positive delta; however, since this has gone up (Refer Slide Time: 31:22), we can say
that relative to c, this has actually gone this way (Refer Slide Time: 31:26), so it is negative
delta, let us make it negative delta. We have now got all the relationships between the
member displacements, the member end displacements and the degrees of freedom thetab
thetac and delta. Now, we substitute these into the equations, but before we substitute, we
have to find out the fixed end moments; we have to find out the fixed end moments in ab, bc
and cd due to the loads.
Member ab has a load. What is the load? This is a, b, 10 meters, 5 meters; at this point, we
have a 20 Kilonewton force. What is this going to give me? The fixed end moment at ab is
going to be equal to b and note that this is going to be this way (Refer Slide Time: 32:51) and
this is going to be this way; this is positive, so this is going to be plus ((20 into 5 squared into
10) by 15 squared); this is going to be equal to 22.2 Kilonewton meter.
The (FEM)ba is going to be minus it is negative (20 into 10 squared into 5 by 15 squared)
is equal to minus 44.4 Kilonewton meter. Let us look at what happens in bc. bc has 10
Kilonewton force acting; this is 8 meters, this is 12 meters (Refer Slide Time: 34:09), this is
going to be this way, this way, so fixed end moment at bc is equal to plus ((100 into 12
squared into 8) upon 20 squared) this is equal to plus 288 Kilonewton meter. Then, the
fixed end moment at cb is equal to minus ((100 into 12 into 8 squared) upon 20 squared)
this is going to be equal to minus 192 Kilonewton meter. Now, look at cd; cd does not have
any load on it, so the (FEM)cd and dc are going to be equal to 0.
Let us look at the equilibrium equations. If you look at the equilibrium equations, what we
get is that Mab is going to be equal to 2EI upon 15 multiplied by thetab plus 6EI upon 15
squared into delta plus 22.2 Kilonewton meter. Mba is equal to 4EI upon 15 thetab plus 6EI
upon 15 squared delta minus 44.4. Then, we have Mbc is equal to 4 into 4EI (note that it is 4
into 4, the EI itself is 4EI; remember that) upon 20 into thetab plus 2 into 4EI upon 20 thetac
plus 288 (note that delta is equal to 0, so it does not contribute). Mcb is equal to 2 into 4EI
upon 20 thetab plus 4 into 4EI upon 20 thetac minus 192.
Finally, Mcd is equal to 4EI upon 15 thetac plus 6EI upon 15 squared into delta; note that it is
minus again, so it is minus into minus, plus, and that is it, there is nothing else, there is no
fixed end moment. Mdc is equal to 2EI by 15 plus 6EI upon 15 squared delta. These are all the
moments that we have. We have been able to write down the member end moments in terms
of the unknown displacements that we have. Once we have done that, then we have another
aspect: equilibrium. Here, equilibrium is going to get really really complicated. Let us look at
what equilibrium actually entails.
Again, I am going to separate it out so that you can understand what I am trying to say; I am
separating out all the members from the joints and representing them separately. I am first
putting down all the moments that we have computed and all the forces that we have. These
are the moments that I have; I am putting them all in the positive; this is Mab, this is Mba, this
is Mbc, Mcb, Mcd, Mdc; so, this happens to be Mab, Mba, Mbc (Refer Slide Time: 41:22), Mcb,
Mcd, Mdc (Refer Slide Time: 41:32). These are the values that I have already put down just
now; these are all in terms of thetab, thetac and delta the unknown displacements
corresponding to the degrees of freedom; I have put them down.
Now, what I need to do is, I need to actually put down all the other forces that I have on the
system. This one (Refer Slide Time: 42:02) has 20 Kilonewton here. Then, what else do I
have? I have let me just look back at the problem itself; I have 20, then I have 100; in
addition to that, I have 50 (Refer Slide Time: 42:33) and 250 Kilonewton meter here.
Anything else? No, that is it, these are all the loads.
Now, I need to be able to do is write down the equilibrium equations. What are the
equilibrium equations? One equilibrium equation obviously is moment equilibrium, so
summation of all moments at b is equal to 0 this is going to give you that Mba plus Mbc is
equal to 0. Similarly, one is moment. What is this moment (Refer Slide Time: 43:27)?
Corresponding to the rotation; this is corresponding to this rotation. Sigma Mc is equal to 0
and that is going to give you that Mcb plus Mcd is equal to 0. Is it? No, it is not. Mcb plus Mcd
plus 250 is equal to 0. Note that when we are taking equilibrium, the external loads also have
to come in, so it is plus 250 is equal to 0. These are the two equations corresponding to the
two rotations (Refer Slide Time: 44:09).
Now, I have a displacement; corresponding to that, what do I have? I have to write down an
equation which says reactions have to be equal to 0 and what reactions have to be equal to 0?
Let us see. If I take the reaction that I get here, let us see what I get. This is going to be like
this, so the loading will be like this (Refer Slide Time: 44:37); let me just put it; so, this is
going to be this way, this is going to be this way.
Similarly, you will get over here, this goes this way, so this has to go this way, this will go
this way, this has to go this way and this has to go this way (Refer Slide Time: 45:06). The
other equation that we get is going to be that this force the reaction at a plus this force the
reaction at d. The other equation is corresponding to this delta, so it is going to be sigma fx
is equal to 0 of the entire thing and that is going to be give me Ra plus Rd minus 20 (this is the
force (Refer Slide Time: 45:54)) is equal to 0; this is my third equation, which essentially is
this.
Look at the other reaction. This is Va (Refer Slide Time: 46:04), Vd. The other equation Va
plus Vd is equal to 100 plus 150 is not going to give me an additional equation; you will see
why. You will always see that since you have a rotational degree of freedom, you have a
moment equilibrium at b; since you have a rotational degree at c, you will have a moment
equilibrium at c; rotations go with moments and displacements go with forces. So, since the
displacement is in this direction, I need a sigma fx is equal to 0 to be able to generate that and
that is the reason why I have that. If I had any degrees of freedom in this direction (Refer
Slide Time: 46:58), then I would have to take Va plus Vd. That is the overall scope of things.
Once you have these, I know I can substitute them in; this one, I have to derive, I can find
out? How do I find this out? I can find out that this is going to be equal to Vab; Ra is equal to
Vab and Rd is going to be equal to Vdc. I can find out these from this equation (Refer Slide
Time: 47:32).
What is Va equal to? Let me evaluate that. Va is going to be equal to. Vab is going to be
equal to I can easily evaluate it (Mab plus Mba) divided by 15 (these are the moments,
reactions due to this) plus 20 into 5 upon 15; that is going to be Vab; Vdc is just going to be
equal to (Mdc plus Mcd) upon 15; and you know Mab, Mbc. If you put all of these together,
what you get is Mba plus Mbc gives you let me write it down.
Mba plus Mbc gives me 4EI upon 15 thetab plus 16 upon 20 EI thetab plus 8 EI over 20 thetac
plus 6EI upon 225 into delta plus 243.6 is equal to 0 that is my Mba plus Mbc. If I look at
Mcb plus Mcd plus 250 is equal to 0, Mcb is going to be equal to 8 EI upon 20 thetab plus 16EI
upon 20 thetac plus 4EI by 15 thetac plus 6EI upon 225 delta plus 250 minus 192, 58 is equal
to 0 that is my second equation. My third equation is going to give me that this plus this
(Refer Slide Time: 51:24) minus 20 is equal to 0.
Now, you can evaluate that, I am not going to go into that in detail, I will leave it up to you.
In fact, what I can do is, if you look at this (Refer Slide Time: 51:28), since these are all
divided by 15, I can just multiply by 15 over here and I will get M ab. If you look at this, from
this equation, I get Mab plus Mba plus Mdc plus Mcd minus 300 is equal to 0 and all I need to
do is substitute all these terms in, let me do that. I am going to put that equal to 2EI plus 15
thetab plus 4EI upon 15 thetab, so this is 6EI upon 15 thetab plus Mdc and Mcd, I am going to
just put it in, 6EI upon 15 thetac plus M; let me just go back.
Mab as 6 plus 6, then I have Mcd has 6 plus 6, so this is going to become equal to 24EI by 225
delta and then, I have 22 minus 44, that is minus 22, minus 22 minus 300, so this is going to
be equal to minus 322.2 is equal to 0. Now, I have three equations. I will just put it down
finally in an equation format.
We look at it in an equation format. This is going to be thetab, thetac, delta and on the other
side, I am going to have the values. It is going to be minus 243.6, minus 58, plus 322.2 and in
here, you will have all the expressions; for example, in this plus this, (Refer Slide Time:
54:53) I can take EI outside and so, it will be 4 by 15, plus (16 by 20), 8 by 20, 6 by 225. So,
0.8 plus this thing gives me 380 and this is going to be 16 by 15. So, this is by 60, this is 16
plus 48 (Refer Slide Time: 55:25), so 64 by 60; this is going to be 8 by 20, this is going to be
6 by 225, this is going to be 8 by 20, this is going to be again the same 16 plus 48 upon 60,
this is going to be 6 by 225, this is going to be equal to 6 by 15, this is going to be 6 by 15

and this is going to be 24 upon 225.
If you solve for these equations, you can get thetab, thetac and delta. I am not going to go
through these details, I will leave it up to you to solve it. Once you get thetab, thetac and delta,
you can find out these values (Refer Slide Time: 56:41) and once you find out these values,
you can always draw the bending moment diagram for the given frame. The only difference
that we had in a frame was that in a beam, you only have rotations as degrees of freedom but
in a frame, in addition to rotational degrees of freedom, you also have displacement degrees
of freedom and therefore, corresponding to displacement, when you take equilibrium
corresponding to rotations, you take a moment equilibrium equation and corresponding to a
displacement, you take a force equilibrium equation that is how you go about solving it. I
am going to spend some more time over the next couple of lectures looking at a few more
problems and seeing how we can solve those.
I hope I have been able to explain a little bit over the last two lectures on how to use the
displacement method to solve beam and frame problems. Thank you.
Prof. P. Banerjee
Lecture 14
Good morning. Last time, we looked at a couple of problems introducing the concept of the
displacement method for solution of a beam with settlement and then, we solved a frame
problem where we saw that the displacement was also an unknown and therefore, we had to
write an additional equation to be able to solve for the displacement also.
Today, I am going to take the same problems and I am going to introduce another concept to
you. Remember we wrote the slope deflection equations as 4EI upon L thetaab plus 2EI upon
L thetaba minus 6EI upon (l square) deltaba plus (FEM)ab, that is the moment at ab. Today, I
am going to introduce you to a concept where I am going to eliminate the delta. In other
words, I am going to redefine my thetaab and thetaba so that you do not need to consider delta
at all in the equation.
If you look at the slope deflection equations, they look this way. Mab is equal to 4EI upon L
thetaba plus 2EI upon L thetaba plus (FEM)ab; in other words, I am eliminating the only
thing is that these (Refer Slide Time: 03:18) have to be defined differently. Of course,
similarly, Mba is going to be equal to 2EI upon L thetaab plus 4EI upon L thetaba plus
(FEM)ba; the only thing is that thetaab and thetaba are not the same as those defined in the
earlier equations.
Let us look at how we defined thetaab in the earlier equations. Here, thetaab and thetaba are the
rotations from the undisplaced position to the tangent to the elastic curve; so, thetaab and
thetaba as defined in the earlier equation were rotations from the undisplaced position to the
tangent to the elastic curve from the original to the tangent, that is how rotation is defined;
and I took counterclockwise as positive, so that is this. Let us look at what happens when we
have just a displacement. When we have a displacement, what is theta from the original to the
tangent? In this case, when you have deltaba, what is thetaab and thetaba? Both are equal to 0,
because the tangent to the elastic curve at both positions from the original displaced position
(Refer Slide Time: 06:25), so this is the original displaced theta, this the original, this is the
original tangent, final tangent, original tangent, final tangent, original tangent, final tangent,
original tangent and final tangent they are equal to 0. This is the previous definition that we
had.
Today, I am going to introduce you to a new definition. The only thing now I say that this is
defined differently, this is the chord (Refer Slide Time: 07:24) that goes from a to b. From the
chord that joints a to b to the tangent is my thetaab and thetaba. What are thetaab and thetaba? In
this particular case, you may ask how it is different from the original, but the whole definition
is different; this is the rotation from chord joining a to b to the tangent to elastic curve.
Therefore, it is now rotation from the chord, it is not rotation from the tangent in the
undisplaced position to the tangent of the displaced position, it is rotation from chord joining
a to b to the tangent to the elastic curve. When there is no displacement of b with respect to a,
they are the same as what I had originally defined, but let us now look at this case where I get
deltaba; in this case, only deltaba.
Now, b has moved here (Refer Slide Time: 09:19), chord joining a to b is this line, chord
joining from a to b to the tangent of the elastic curve, so in this particular case since this is
l, what is thetaab equal to? From the chord to the tangent; thetaab is equal to minus deltaba
upon L. From the chord to the tangent; since these are clockwise (Refer Slide Time: 10:05),
this is my new definition of thetaab from the chord joining a to b to the tangent, so now you
see that this becomes my definition of thetaab and thetaba; now, the proof of this is to show
that both of these give me the same moments.
In this particular case, what is Mab equal to? Mab is equal to 4EI upon L thetaab plus 2EI by L
thetaba and Mba is equal to 2EI by L thetaab plus 4EI by L thetaba. Here, what would Mab be
equal to? Mab would be equal to minus (6EI upon L square deltaba) and Mab would be equal to
minus (6EI upon L square deltaba) these would be the moments at the end.
Let us see whether we get the same Mab and Mba here. There is no difference, so Mab would
be equal to 4EI upon L thetaab plus 2EI by L thetaab and Mba would be equal 2EI by L thetaab
plus 4EI upon L thetaba. Now, let us look at this see there is no difference because thetaab
and thetaba are also the same.
Let us now look at this one. Here, Mab was equal to this. Now, Mab is going to be equal to 4EI
by L thetaab plus 2EI by L thetaba; now, let me plug this in (Refer Slide Time: 12:33), 4EI by
L thetaab into minus (deltaba upon L) plus 2EI by L thetaba into also minus (deltaba upon L).
Substituting this, you will get minus (6EI upon L square deltaba) and you will get a similar
Mba also. The point to note is that we have got exactly the same Mab and Mba as we had got
using the earlier definition and therefore, there is no difference in the entire method.
The only point here is that we have negated the effect of trying to take deltaba in the equation;
we have simplified our slope deflection equation to just containing three terms one
containing thetaab, one containing thetaba and one containing fixed end moment. You may ask
what is the reason behind this; we will slowly come to what the reason behind this is.
The point that I am trying to make is that this definition of thetaab and thetaba from the chord
to the tangent of the elastic curve is completely equivalent to the previous definition. The
only thing is that in the previous definition, you had thetaab, thetaba, deltaba all these as
terms, whereas here, you only have thetaab and thetaba and the advantage of this will become
obvious a little bit later on. Let us now see whether we can use this concept to solve the
problem that we had solved last time. We will look at the problems and see if we get the same
answers, I think that is the key thing.
Here, we are given that this is a (Refer Slide Time: 14:53), this is b, this is c and this goes
down by 0.03 meters, you were given that E is equal to 200 Gpa, I is equal to 2 into 10 to the
power of minus 3 meter fourth. This is what you were given and essentially, what you were
asked to find out is the bending moment diagram due to the settlement alone, only b settles by
0.03 meters.
Let us try to solve this problem using the new approach that we have. If I were to take just
this displacement (Refer Slide Time: 15:42), what would it look like? This is fixed, this is
displacement, this is displacement that is due to 0.03. Then, let us see what happens when I
have thetaba. We know that the degrees of freedom are this and this, so I have this and then I
have this, this is thetab is equal to 1, this is delta is equal to 0.03, this is thetaba is equal to 1
and finally, we have thetac is equal to 1.
So, this is delta (Refer Slide Time: 17:07), this is thetaba and we can find out all these things
that come out of it. Let us see. Due to all of these, what is thetaba? Note that thetaab is now
defined from the chord to the tangent; from the chord, so here, the chord would be this a
with b; this would be this way; here, the chord is this, here, the chord is this. Let us look at
what thetaab is; thetaab here is equal to 0, so thetaab in terms of thetab is 0, thetaab in terms of
5
thetac is 0, and in terms of delta, it is going to be equal to from the chord to the tangent
(Refer Slide Time: 18:17), it is positive, so thetaab in this case is equal to 0.03 by L, where L
is 10; remember this was 10 and this was 10 (Refer Slide Time: 18:39), thetaab is 0.03 by 10.
What about thetaba here? We will see thetaba is also 0.03 by 10.
Let us see what thetabc is from the chord to the tangent, from the chord to the tangent, from
the chord to the tangent (Refer Slide Time: 19:09), clockwise; so, thetabc is equal to minus
(0.03 by 10) and thetacb is the same; for thetab is equal to 1, thetaab is equal to 0, thetaba from
the chord to the tangent is 1, from the chord to the tangent is thetabc is equal to 1 and thetacb is
equal to 0. Here, thetaab is equal to 0, thetaba is equal to 0, thetabc is equal to 0 and thetacb is
equal to 1.
Note one thing: when you only have a rotation, the new definition of theta and the old
definition of theta remain the same, the only difference is when you have a displacement, the
new definitions of thetaab and thetaba are different from the old one; also note that here (Refer
Slide Time: 20:23), I do not have any delta anywhere.
Once we have that, I can write down thetaab as equal to 0 into thetab, so, 0 times thetab plus 0
times thetac plus 0.03 upon 10; thetaba is equal to 1 times thetab plus 0 times thetac plus 0.03
upon 10; thetabc is equal to 1 into thetab plus 0 into thetac minus 0.03 by 10; and thetacb is
equal to 0 times thetab plus 1 times thetac minus 0.03 upon 10 these are my equations and
now, I substitute these into my slope deflection equations.
What do my slope deflection equations look like? Mab is equal to 4EI upon L thetaab plus 2EI
upon L thetaba plus (FEM)ab.
Note that in this particular case, since there are no loads, (FEM)ab is equal to (FEM)ba is equal
to (FEM)bc is equal to (FEM)cb and all of them are equal to 0.
I am going to substitute these in. I am going to get 4EI upon 10, thetaab is 0.03 by 10 plus 2EI
upon L, thetab is thetab plus 0.03 by 10 plus 0, which is going to be equal to 0.12 plus 2, so
0.16, it is going to be 0.18 EI upon 100 plus 2EI upon 10 thetab. Similarly, if you substitute
everything in, you will get Mba is equal to 0.18 EI plus 100 plus 4EI upon 10 thetab; Mbc is
going to be equal to 4EI upon 10 thetab plus 2EI upon 10 thetac minus 0.18 EI upon 100; and
Mcb is going to be equal to 2EI upon 0 thetab plus 4EI upon 10 thetac minus 0.18 EI upon 100.
These are going to be the equations.
Again, going into the fact that equilibrium equations give us I am not going in, I have
already done this problem earlier, so I am just substituting them in. This one gives me 8 EI by
10 thetab plus 2EI upon 10 thetac is equal to 0; the other equation was Mcb is equal to 0 and
that gives me 2EI by 10 thetab plus 4EI by 10 thetac is equal to 0.18, EI is 200 into (I am
8
doing it in Kilonewton), so it going to be 200 into 6 into 2 into 10 to the power of minus 3
upon 100.
After that, you are going to see that once you substitute these equations, this is going to land
up being equal to when I substitute these equations, you will see that these equations look
identical to what I had obtained earlier and you will see that thetab is equal to minus (514
upon EI) radians, thetac is equal to 2057 EI radians and when you substitute, you will get
exactly the same equations. The most important thing is that once I have utilized this
procedure, the solution process remains identical; I have just proved that to you with one
problem. What is the advantage of this definition? I will shortly come to that by actually
taking up the other problem that I had discussed last time that was the frame problem.
Let us just revisit that problem and I will introduce you to the concept of. Remember I said
that when you were using. What virtual work method did we use for the force method? We
used the method of virtual force. I had said that I was going to use the method of virtual
displacement when I came to using the displacement method that is what I am going to
illustrate to you and this is where it will become obvious as to why I was using this new
definition that I had today. This is 12 (Refer Slide Time: 27:53) and this is 5. I am just
restating the problem that we had done last time. Today, what I am going to introduce to you
is the concept of the principle of virtual displacement and we will see how .
Last time, if you remember, to be able to generate the third equation, we had to go through
quite a bit of involved computation, where we computed shears etc. Once I give you slightly
more complicated problems, especially with inclined members, you will see that this is going
to become a very very messy affair because not only will you have to compute shears, you
will have to compute axial forces and then resolve them, find out the vertical reaction,
horizontal reaction all kinds of things.
Just to be able to get the third equation, you will probably have to solve ten different
equations to ultimately get to the third equation; it gets quite messy. I will illustrate this
concept to you as we go along; there are several lectures that I am going to spend on actually
9
solving example problems for you. But, I will introduce the concept in today's lecture so that
the concept remains clear in our mind while we are solving all the particular problems. I am
going to introduce the concept using this example problem which we had already solved in
the last lecture.
Let us just put one or two things down and then go back to the concept that we are going to
have. Remember that I am not going to go through the steps of trying to define what
everything is; I am just going to put down my new definition. These are my degrees of
freedom: thetab (Refer Slide Time: 29:57), thetac, and the translation of this is delta. Then, I
have this load here, here I have a load of 50 Kilonewton and I have a moment of 250
Kilonewton meter, I have a 100 Kilonewton load here and a 20 Kilonewton load here, this is
20 meters and this is 15 meters.
These are the degrees of freedom: thetac (Refer Slide Time: 30:37), thetab and delta. One
thing you must note is that always, my structure degrees of freedom are implicitly assuming
this as my structural Z coming out (Refer Slide Time: 30:51); so you will see that if I have a
displacement in this direction, it is taken positive in this direction, a rotation is taken positive
in this direction (Refer Slide Time: 31:07) so that it comes out of the paper, which is in the
positive Z direction. If I ever have any displacement in the Y direction, I will have it positive
in that. My definition is always completely related to the idea that positive displacements are
along the positive axis of the structure.
10
That is the reason why I always define anticlockwise moment as positive (Refer Slide Time:
31:41) because that then gives me the concept of positive; it can be generated very easily
because in positive degree of displacement, the corresponding degree of freedom is aligned
along the direction. Now, let us look at this. First and foremost, we have to compute the fixed
end moments. There is no problem in computing the fixed end moments; the fixed end
moments are identical to what we had the last time.
Let me just go through and quickly get you the fixed end moments. Here, this is a (Refer
Slide Time: 32:25), this is b, this is c, this is d. The (FEM)ab is equal to plus 22.2, (FEM)ba
was equal to minus 44.4, (FEM)bc was equal to plus 288 Kilonewton meter and (FEM)cb is
equal to minus 192 Kilonewton meter. In the computation of the fixed end moments, you do
not have any specific problem; the only thing is that we need to actually go and define the
displacements corresponding to the degrees of freedom and find out how the structure moves.
11
Now I have thetaba is equal to 1, so thetaba is equal to1 is going to give me this (Refer Slide
Time: 33:41), this and that is it; this is 1, this is 1, this is thetab is equal to 1 and thetab is
equal to 1 automatically assumes that thetac is equal to 0 and delta is equal to 0; then, I am
going to put thetac is equal to 1. Remember that for rotations, the definitions do not change.
Let us put delta, delta give me this (Refer Slide Time: 34:31), this is delta, delta, so this is
delta is equal to 1.
Now, our definition is from the chord delta to the tangent; so if you look at this particular
point, what is thetaab equal to? 0. What is thetaba equal to? 1. thetabc is equal to 1, thetacb is
equal to 0, thetacd is equal to 0, thetadc is equal to 0. Here, thetaab is equal to 0, thetaba is equal
to 0, thetabc is equal to 0, thetacb is equal to 1, thetacd is equal to 1, thetadc is equal to 0; no
difference; remember that whenever you have a rotation, there is absolutely no difference in
the computation, I mean with respect to last time and this time.
Here, once you have a displacement (Refer Slide Time: 35:42), it all changes because it is
from the chord (the chord is the one that connects the displaced positions of) to the tangent,
from the chord to the tangent (Refer Slide Time: 36:05), so, what is thetaab equal to? Delta by
15; if you look at this angle, it is delta by 15 so, thetaab, look at it from the chord to the
tangent it is anticlockwise, so it is positive delta by 15. What about thetaba? Similarly, from
the chord to the tangent, it is anticlockwise (Refer Slide Time: 36:32), it is delta by 15. What
about thetabc? From the chord to the tangent, it is 0, from the chord to the tangent, thetacb is
equal to 0 (Refer Slide Time: 36:45) and from the chord to the tangent, anticlockwise,
positive; thetacd is equal to delta by 15 and thetadc is delta by 15. We have figured out
everything in terms of the displacements.
12
Therefore, I can write down thetaab is equal to 0 into thetab plus 0 into thetac plus delta EI by
15, so this is just delta by 15; thetaba is equal to 1 times thetab plus 0 times thetac plus delta by
15; thetabc is equal to 1 into thetab plus 0 into thetac plus 0 into delta; thetacb is equal to 0 into
thetab plus 1 into thetac plus 0 into delta; thetacd is equal to 0 into thetab plus 1 into thetac plus
delta by 15; thetadc is equal to 0 into thetab plus 0 into thetac plus delta upon 5; these are my
rotations.
Now, all I need to do is find out the moments and so, I am just putting them down, I am not
going to actually calculate. By now, just by substituting into the equations, you will get Mab is
equal to 2EI by 15 thetab plus (6EI by 15 squared) into delta (just plug this in (Refer Slide
Time: 39:26) and you will get 6EI delta) and Mba is equal to 4EI upon 15 thetab plus (6EI
upon 15 square) into delta (when you plug this in (Refer Slide Time: 39:46), you will get 4EI
into theta plus 2EI etc). You will get all this by substituting these into the equations. Then,
13
you have the fixed end moments, so this going to be equal to plus 22.2 (Refer Slide Time:
40:06) and this is going to be minus 44.4. This way, by substituting bc and cb, you can find
out Mbc and Mcb and you can find out Mcd and Mdc by substituting them in; I am not going to
write those down.
Let me put forward the thought process to you. Essentially, I know that Mab is a function of
thetab, thetac, delta all of them. Of course, it is not necessary that all of them will be a
function; for example, Mab does not have thetac in it, but I am just putting down in general
that these will be functions of this. Mcb is function some other functions f1, f2, f3, f4 and
Mcd is going to be equal to f5 of (thetab, thetac, and delta) and Mdc is going to be f6 (thetab,
thetac, delta). All are different functions, but essentially they are functions and then, there are
some fixed end moments. These are my real moments and these happen due to the loading
that we have on the structure. Now, I am going to use the principle of virtual displacement.
What am I going to do? I am going to say that I am going to apply three different independent
virtual displacement patterns.
14
What are my three independent virtual displacement patterns? One is this (Refer Slide Time:
42:11), one is this, one is this and note that these are now virtual displacement patterns; I am
going to find out the work done by all the forces subjected to these virtual displacements. The
work done by the real forces subjected to these virtual displacements is going to give me the
virtual work done in the structure.
Once I have the virtual work done in the structure, what can I do? I can write virtual work
done by the external loads equal to the virtual work done by the internal loads and once I put
that, all the total virtual work is equal to 0 and from that, I get an equation. What will be the
equation in terms of? Equation will be in terms of thetab, thetac because these are the internal
forces and I am going to find out for each displacement pattern what are the thetas are; once I
have those thetas, I can just say what the work done by the internal forces is. If this is my
virtual displacement (Refer Slide Time: 43:29) times, it is going to be thetaab into Mab plus
thetaba into Mba etc... and in that way, we can continue doing. So, that is my virtual work
equation and that virtual work equation is what is going to give me the three independent
equations and I can solve for the displacements. I have three equations, three unknowns, I can
solve for them. That is in essence the method of virtual displacement.
15
Now, what I am going to do is, I am actually going to put it into practice and actually take
this particular problem. Let me just put them down and you will see that this (Refer Slide
Time: 44:20) is equal to 2EI by 15 thetab plus (6EI upon 15 square into delta) plus 22.2; this
is equal to 4EI by 15 thetab plus (6EI upon 15 square into delta) minus 44.4; this is equal to
16EI upon 20 thetab plus 8 EI upon 20 thetac plus 288; this is equal to 8 EI upon 20 thetab
plus 16EI upon 20 thetac minus 192; this is equal to 4EI upon 15 thetac plus (6EI by 15
square into delta); and this is equal to 2EI by 15 thetac plus (6EI upon 15 square into delta).
These are the moments which you can get. Then, what is the next step? The next step is to
find out the work done by all the forces.
Let us look at what happens. I am breaking it up because I need to know what is the work
done by all the forces. Here, you will have Mab (Refer Slide Time: 46:37), here you will have
Mba, here you will have Mbc (Refer Slide Time: 46:49), here you will have Mcb, here you have
16
the load 50 and a moment 250; then, you have the moment Mcd and then you have the
moment Mdc. These are all the moments and note that I am just putting down all the loads;
this is 100 and here, you have 20 (Refer Slide Time: 46:21).
One aspect that is very important is that when you have loads acting on the member, what
you need to do is you need to find out the reactions at this (Refer Slide Time: 47:45). Think
of this as a simply supported beam with Mab and Mba. What I need to compute for getting the
virtual work properly is, I need to find out the reactions at the end of the member due to this.
Since I am considering it to be a simply supported beam, what are the reactions going to be
equal to? This is 5, this is 10, so if I find out the reaction, I am only finding out the reaction
due to the load only, remember that, not due to the unbalanced moments etc.; only the load.
What is the reaction at this point due to this load (Refer Slide Time: 48:23)? If I take
moments about that point, it is going to have 5 into 20, so this is going to be anticlockwise
and if I put this in this direction, this is going to be clockwise, so this into 15 is equal to 20
into 5; this is going to be equal to 100 upon 15 and this is going to be 200 upon 15.
Similarly, here, we are going to have 8 (Refer Slide Time: 49:01) and 12, so just due to the
load alone, you will see that this is going to be equal to 1200 (Refer Slide Time: 49:08), this
is going to be 60 and this going to be 40; if I substitute, then this is going to 20 upon 3 and
this is going to be 40 3. What have I done? On this member, there is nothing. Note that these
are applied at the nodes, so do not I need to find out, it is only for the members I need to find
out the reactions.
I have just found out the reactions, so what is my virtual work equation? I am going to put
thetab is equal to 1. This is my virtual displacement pattern, so thetab into the moment at b
(which is 0) this is the external work done and I am calculating all the external. Note that
these I have translated (Refer Slide Time: 50:16), so these are no longer. Member loads are
never external, it is only joint loads which are external. How much is this joint (Refer Slide
Time: 50:25) displaced by under thetab? Let me look back at the equation that I have drawn.
17
This is my thetab, I am giving this virtual displacement, so I am going to have that this into 1,
nothing, 0; this is not going up (Refer Slide Time: 50:47), neither is it rotating, so these two
do not do any work.
The external virtual work is equal to this. What is the internal virtual work? The Mba does
work, so it is going to be Mba into thetab because it is 1 into thetab (note that the moment is
this way (Refer Slide Time: 51:17), the rotation is this way, so it is positive work) plus Mbc
into thetab. Then, let us look at all the work done by this and this; these do not undergo any
displacements, neither do those and neither does that, so this is my equation.
This is the external virtual work (Refer Slide Time: 51:46), this is internal virtual work. This
is going to give me Mba plus Mbc into thetab is equal to 0 this is true for any arbitrary thetab;
this implies that Mba plus Mbc is equal to 0. Note that is exactly the equation that we get when
we put equilibrium conditions in the previous case; the only thing that we did was we
computed the virtual work done. Remember I had said that the method of virtual
displacement essentially replaces the equilibrium conditions and I am showing this here I
am getting the same equilibrium conditions. Let me put thetac and I have the same equation;
let me put the arbitrary equation thetac is equal to 1.
18
Then, the work done will be thetac; thetac is in this direction (Refer Slide Time: 53:02), so
what is the moment at that point? You will see that is clockwise moment, so you are going to
get 250 into thetac this is the external virtual work. All the other external forces, including
the support reaction and the joints, do not undergo any displacement; all of them are going to
be 0 and so the internal virtual work is going to be Mab into thetaab (which is 0) this theta is 0,
all the Mba, thetab, 0 etc.
You will see that the only term that exists will be Mcb into thetac plus Mcd into thetac and none
of the others do any work; so this one is going to be (Mcb plus Mcd plus 250) into thetac is
equal to 0. Now, this has to be true for all arbitrary thetacb, so I get Mcb plus Mcd plus 250 is
equal to 0. These look like equilibrium equations but they are not, they are actually weak
solutions, they are actually the work done by this (Refer Slide Time: 54:31).
I have shown to you that when I put thetab is equal to 1 and thetac is equal to 1 as the virtual
displacement patterns, I get back my Mba plus Mbc is equal to 0, which was the simple
equation that we had got earlier. I also got that Mcb plus Mcd plus 250 is equal to 0. In other
words, when I put the thetab and thetac virtual displacements, I essentially get back the simple
equations that I had developed by taking equilibrium of the joints, joint b and joint c, in the
earlier case.
I am going to stop over here because I have come to the end of my lecture. In the next lecture,
I am going to show to you that by taking the virtual displacement delta is equal to 1, how I
can generate a third equilibrium equation easily without having to go through a whole host of
equilibrium conditions for each individual member.
So, on to the next lecture I will show you the actual power of applying the method of virtual
displacement. Thank you.
19
Prof. P. Banerjee
Lecture 15
Good morning. I am continuing my earlier lecture. Just to review quickly, I was using the
method of virtual displacement to generate the equations.
These are all the internal forces and these are the external loads for which I am going to apply
it. Note that the support reactions do not matter because the supports never move, so the work
done by all the support reactions are always 0. It is only the internal forces and the external
joint loads that do any work. First, I put thetab is equal to arbitrary thetab.
Then, I found out another equation applying an arbitrary thetac. Now, I am going to apply an
arbitrary delta, arbitrary delta virtual displacement and let us see what the virtual shape of the
structure is.
This is the shape of the structure and let us look at what thetaab is equal to, it is delta by 15.
thetaba is equal to delta by 15, thetabc is equal to 0, thetacb is equal to 0, thetacd is equal to
delta by 15, then thetacd and thetadc are each equal to delta by 15 and these are for a virtual
displacement delta. This will be the work done by all the forces here. Let us see what the
work done by the forces is; let us see what the forces are; we have already seen what the
forces are; all the forces are noted here. Note that these cannot do any work, the work done
by these are represented equivalently by the work done by the reactions to these loads, so it
becomes on the internal side because it is an internal force. If I want to write down that
equation, I am going to put it equal to. You have to see the work done by the external
forces.
The work done by the external forces is going to be delta prime into 0 plus 50 (that is the
force) into 0 because there is no displacement plus 250 into 0 these are all the external
forces; the reactions are 0 anyway, so the right-hand side the work done by the external
forces is equal to 0. Let us see what is the work done by the internal forces; the virtual work
done by internal forces is going to be equal to first, Mab into delta prime by 15 (note that both
of them are anticlockwise, so it is positive), then work done by Mba into delta bar by 15 that
is the real force undergoing the virtual rotation. What can we say about Mbc? thetabc is equal
to 0, so that is going to be equal to 0; the work done by Mcb into thetacb is equal to 0, so that is
going to be 0; then we have Mcd undergoing thetacd virtual, so that is Mcd; then, we have Mdc
multiplied by its rotation, so that is the work done; these are all the work done by all the
moments. Now, let us find out the work done by all these forces.
What is the work done by these? Let us see. Let us take 20 by 3. What is the work done by
this 20 by 3? The displacement of this point is 0, so the 20 by 3 does 0 work. What about the
40 by 3? You will see that that point is displaced by 1 in this direction. The work done by the
40 by 3 is negative, because 40 by 3 and the displacement delta bar are opposed to each other.
Now, let us find out the work done by this 60 there is a reaction also, that is why it is on the
internal sides, for 60, the vertical displacement is 0 and work done by the 40, the vertical
displacement is equal to 0, so there is no work done by those.
Essentially, what am I left with? I am left with internal virtual work is equal to (Mab by 15
plus Mba by 15 plus Mcd by 15 plus Mdc by 15 minus 40 by 3) into delta is equal to 0 this is
what we get as our equation. Note one particular thing that we have. This is the equation that
we get and now, if I substitute this and put this in. One of the points that I would like to
make over here is that. This is the equation. Let me just put this into a standard format.
(Mab plus Mba plus Mcd plus Mdc minus 15 into 40 is 600 minus 200) into delta bar is equal to
0 because I have taken the 15 on the other side.
Since this delta prime cannot be 0, we get Mab plus Mba plus Mcd plus Mdc is equal to 200
this is my other equation that I generate from the third. Note the advantage of this procedure.
Remember how we obtained it the last time? The last time, the way we obtained it was that
3
we actually found out the reactions and then put the reactions equal to the loading that was
applied on the structure and based on those, we did the whole procedure and that is how we
obtained these equations. Here, just by applying the virtual work equation without actually
looking at any kind of equations of equilibrium, we generated this apparent equation
equilibrium. Note that this equation is actually a virtual work equation, it is a work equation;
the only thing is that it gives you something that looks like a moment equilibrium equation
and if you solve these three simultaneously, you will get thetaab, thetac and delta.
So much for all of this and you will see that these equations are identical to what we got last
time. This is the concept of the application of the principle of virtual displacement. I just want
to go back and review what I have done before I proceed. What I have done is I have
introduced a new concept of using a theta. I have used a simpler form of slope deflection
equations where all you have are the rotations and fixed end moments. The only point here is
that the rotations are defined from the chord rather than from the original position, original
tangent, that is from the chord to the displaced tangent of the elastic curve. This is how we
define the new thetas and once we define those new thetas, we saw that we got exactly the
same equations that we got earlier without considering this new definition the original
definition of theta as well as delta. Finally, I showed you the application of the method of
virtual displacement in obtaining the equations.
This method of virtual displacement essentially becomes very important when we have to
write down the equations corresponding to a displacement quantity because the only thing
that we get directly from the slope deflection equations is actually the moments. All other
shear forces etc. are derived quantities and to write down a force equilibrium equation, we
have to actually go through a significantly large number of equations to get that one equation
corresponding to a displacement. By using the method of virtual displacement and putting
external virtual work equal to internal virtual work and noting that the fact that works is a
scalar, you can add up the work done by all the forces acting on the individual members and
sum them up. We got an equation very easily without having to go through several equations.
This is the advantage of the method of virtual displacement.
Now, what I am going to introduce you to is a couple of other problems and I am going to
again introduce the concept. One of the very important aspects of using the displacement
method is that you saw we have to draw the displaced shapes corresponding to each degree of
freedom, so that we can find out the member end rotations are related to the structural
degrees of displacements corresponding to the structure degrees of freedom.
Now, what I am going to do is I am going to introduce you to a couple of problems where all
that we do is actually just look at how the displacement pattern comes and see if we can relate
the member end rotations to the structure degrees of freedom. Then, you can write down the
member end moments in terms of structure degrees of freedom and then you can use virtual
displacement to write down the equations. Essentially, this becomes a very simple procedure
once you get the kinematics. This is called kinematics because when you want to find out
how does the structure displace given a degree of freedom, given a displacement
corresponding to a degree of freedom that is what we are going to concentrate on because
once you know that, the displacement method becomes very very simple to apply. Let us now
look at a few problems. Let me take you through to some complicated issues, not as simple as
they looked last time.
Note that here also, if you assume axial rigidity, there are only three degrees of freedom and
the three degrees of freedom are this (Refer Slide Time: 14:28), this and this ; identical. All I
have done from the previous problem is I have taken this straight member and I have inclined
it. Let us put down some values just to satisfy ourselves. I will make this this is 3, 4, so I
will make this 9 , I am going to make this 12 , make this 15 ; this is a , this is b, this is c and
this is d; this is 9 meters, 15 meters, 12 meters. Right now, all I am interested in is what
happens when I give this equal to 1, this equal to 1 and this equal to 1 independently; that is
all I need to draw; once I know that, given a loading, I can always find out everything else.
Right now, we are going to concentrate only on kinematics because you will see that
kinematics is important to relate, to get the member end moments due to the displacements
corresponding to structure degrees of freedom and it is also important because those become
my independent virtual displacement patterns. I am using the same displacement patterns that
I generate for each individual degree of freedom because I know that if I choose one degree
of freedom equal to 1 and the other degrees of freedom equal to 0, they are each going to be
independent because that is the whole definition of independent degrees of freedom. In other
words, I could have actually chosen completely different independent deformation patterns
for the virtual displacement and been absolutely correct. The only problem is that to figure
out two independent displacement patterns is not as easy as it looks.
However, since my degrees of freedom are independent of each other, I know that if I take
one displacement equal to 1 and the other displacements equal to 0, I am going to generate a
pattern which will be completely independent when I take another degree of freedom equal to
1 and the others equal to 0; those are going to be independent and that is the reason why I
choose those patterns as my virtual displacement patterns. For this, there are three degrees of
freedom and so, if you could figure out three independent virtual displacement patterns, you
could actually write down three independent equations.
I am not that this thing in terms of trying to figure out what would be an independent pattern
and therefore, what I tend to do is, I tend to use the simplest independent patterns that I can
generate, which are the displacement patterns corresponding to the degrees of freedom
5
themselves. Therefore, you see how important the displacement pattern becomes every time
and that is why the kinematics. Here, I am going to first put thetab is equal to 1 and note that
automatically I am assuming thetac is equal to 0 and delta is equal to 0, I do not need to
explicitly state it. This one is actually not a problem; you know the inclined member does
nothing to the displaced shape, so, thetaab is equal to 0, thetaba is equal to 1, thetabc is equal to
1, thetacb is equal to 0 and of course thetacd and thetadc are equal to 0; so, I could relate it. I
am going to put them down. If this was real, these would be real and if this is virtual, these
would be virtual; I mean identical. Then, let me put thetac is equal to 0. Once I get thetac is
equal to 1, we have to be consistent, so let me put down what. This is thetabc, thetacb is
equal to 1, thetacd is equal to 1. Note that these are all thetas being taken from the chord to the
tangent all of them from the chord to the tangent, but remember I told you that when you
put thetas, the chords are the undisplaced directions, so there is no change.
Now, for the interesting one and let me draw this; for this one, I will use a different colored
pen to highlight to you what actually happens. Now, note that I am going to put displacement
delta is equal to 1. What happens? This member, since this is moving perpendicular, there is
no change of length, so there is no problem; this is going to remain like this. Let us look at
this member. For this member, for it not to change its length, along which direction can it
move? It can only move perpendicular. Please note that these are small displacements, so the
perpendicular movement does not increase because ideally you will see an arc but when your
displacement is small, the arc and the tangent are the same; so, this can move along this
direction only.
What about this member? Since this end has gone here, the only way. This cannot change
length, this has already gone by 1 from here to here, so this was the member. How will this
member move without increasing its length? Only if was perpendicular. Look at where this
point goes. If this member was not there, this point would move here (Refer Slide Time:
23:43), but since this member is there and it is fixed at this point, it can only move along this
line, it can only move along this line. This point now has to move because it cannot be here;
this point is not on this line, so if I move it only up to here, this would increase this length.
Therefore, this now has to move; when it has to move, how does it move? It moves
perpendicular so that this member does not change length. The whole point here is that since
axial rigidity is there, nothing can change its length. Where would the point go? The only
way is where these two lines intersect; this is the point where it would go. How would it look
now? The thing is thetab is still equal to 0. This point cannot go anywhere, so this would still
be perpendicular to this. What about this (Refer Slide Time: 24:47)? Since thetab is equal to
0, this would still be parallel to this, this would be parallel to this, this would be parallel to
this and here, this would be parallel to this. That ensures that thetab is equal to 0 and thetac is
equal to 0, which is what you need. If this was 1, because delta is equal to 1, how much is
this? Let us see what this is.
If you look at this particular one, this is the angle theta, so that is the angle theta, this is
perpendicular and this is perpendicular. Which angle is represented over here (Refer Slide
Time: 25:46)? If you look at it, by similar triangles, this would be the same as this (Refer
Slide Time: 25:53); satisfy yourself. So, if you look at this, this is 12, this is 9, therefore 12
by 9; since this is 90 degrees and this is 90 degrees , tan theta is equal to 12 by 9; here also,
tan theta is equal to this by this, so what is this equal to ? This upon this is equal to 12 by 9,
so you will see that this is equal to 9 by 12 and 9 by 12 is 3 by 4. If this is 1 and this is 3 by 4,
you will see that this is equal to 5 by 4. Once I have drawn the shape, the point is to find out
thetaab and thetaba and here, I need to connect.
Whenever I have a displacement, I need to connect the chords and take displacement, rotation
from the chord to the tangent. If I do that, what do I get? If you look at it, from the chord to
the tangent, what is that? You will see that this displacement is 15 because 12 square by 9
squared, square root is a hypotenuse and that is going to be equal to 15; this is 12, this is 9,
this is 15 and 5 by 4 is the displacement here, so 5 by 4 divided by 15 is this rotation, this
rotation is equal to 5 by 4 divided by 15 that is equal to 1 by 12. Similarly, here also, this is
1 by 12. Let us look at this one. This is 3 by 4 and what is the dimension? The length is 15, so
3 by 4 by 15 is equal to this angle is equal to 1 by 20. Similarly, this angle (Refer Slide Time:
28:17) is 1 by 20.
Similarly, over here also, we have to join the chord, the two members, two points of the two
ends and join it by a straight line; so, what is this? This is delta and what is this length? 12.
This is 1 by 12 and this is 1 by 12. Having written those down. This is from the chord, so
what is thetaab? Thetaab from the chord to is anticlockwise, positive, this is 1 by 12. What is
thetaba? Thetaba is equal to from the chord, anticlockwise, so 1 by 12. What is thetabc? Let
us see. From the chord to the tangent, clockwise, so it is negative 1 by 20. Similarly, thetacb is
negative 1 by 20; from the chord clockwise (Refer Slide Time: 29:28), so thetacd is equal to 1
by 12; similarly, thetadc is equal to 1 by 12.
This one is fairly complicated and that is the reason why I have spent so much time actually
drawing. Now, you can see that delta is equal to 1. Up till now, what I did was, these were
simple problems and so now, if I look inclined, you actually have to spend a lot of time
figuring out where each point goes. Once you know where each point goes, you can then
draw the displaced shape because delta is equal to 1, thetab is equal to 0, thetac is equal to 0
you know those. From that, you can draw the displaced shape given delta is equal to 1; this is
the displaced shape if delta is equal to 1.
Once you have delta is equal to 1 that means this is equal to one-twelfth of delta, one-twelfth
of delta, minus delta by 20, minus delta by 20 that is the overall concept. Then, you can
write down thetaab is equal to 0 into thetab plus 0 into thetac plus 1 by 12 into delta; that
means thetaab is equal to delta by 12 and then, you can substitute in the equation and you can
get your expressions for Mab, Mba, Mbc and Mcb. Once we have this, you can see that this
kinematics gives you all the equations for Mab and then, these are also independent virtual
displacements, so these can then be used to actually write down your virtual work equations.
So much for the equilibrium.
Now, I am going to introduce you to the kinematics of various kinds of members because
once you know the kinematics, you can actually write down the equations for any kind of
structure. Let me now look at another equation. I am going to take the same thing and I am
going to say that this member bc is flexurally rigid also.
If member bc is flexurally rigid, let us draw this. The same equation and a, b, c, d; now, this
member is flexurally rigid. What happens when the member is flexurally rigid? Let us see.
We will first find out the degrees of freedom; one, two, three, four, so unconstrained degrees
of freedom are equal to 3 into 4, 12. How many restraints? Restraints are equal to 2 into 3
because both of them are fixed, so three constrains per joint, that is minus. Then, how many
constraints do we have? One constraint is axial rigidity of all the three members, so that is 1
into 3 members. In addition to that, bc is also constrained to be flexurally rigid. Now
remember, I had said how many for flexurally rigid, for each member? One member and two
constraints. So how many constraints do I have? 3 plus 2 is equal to 5. How many degrees of
freedom? 1. What is the degree of freedom? I am going to take it as this. Note that what has
actually happened is that because of this being flexurally rigid, this member cannot rotate
because this cannot rotate and here also, it cannot rotate; so, thetab and thetac both have
disappeared as independent degrees of freedom, because no way can this joint displace. If this
joint does displace, then this has to go this way and there is no way this can go, because then,
for it to be rigid, this point has to go and violate. You can see that.
Similarly, for thetac, if this rotates and this rotates (Refer Slide Time: 35:24), this has to come
down here, and it has to, it will violate this. There is only one degree of freedom and for that
one degree of freedom; I have to draw the displaced shape. It is the same, so 9, 9, 12, 15 and
12; rigid, rigid and this is rigid, so it can only deform linearly. Let us see what happens here.
It is very interesting to note that this point is going to go there, this point is going to go there
because delta is equal to 1. I have already shown where is point is going to go; this point is
going to go here, so this is gone by 1, this is going to come down here, this point is going to
be here, this is 1, this is 3 by 4. The points go exactly where they went earlier; the only
difference that happens is how are these points connected.
Note, since this member is rigid, this member is rigid, the only way these two points can be
connected to each other is in a straight line because this is the only way that member bc can
deform, it can only go straight. Now, if it goes straight, what happens? If it goes straight,
what happens? Let us see what is the angle here? 3 by 4 by 15, that is 1 by 20. Now, note that
that means this also has rotated by 1 by 20. The interesting point here is that since this
8
member has gone this way, there is no way that this tangent can remain in this way. Why?
Because that will violate the continuity of the joint. That means that if it rotates by 1 by 20,
this also has to rotate by 1 by 20, the tangent at this point has to rotate. So, how will my
displacement of this look? This will look like this way and here, again, this was this way; so
this also has to rotate in this fashion by 1 by 20. You see, what has happened is if I were to
plot this, this is how the thing has rotated due to delta.
Now, let us see what my member end rotations are. For that, I have to draw this again, this
one I do not need to draw. These are the chords, so once member bc is rigid, if a member bc
is rigid, its slope deflection equations are no longer valid because Mbc will be given by 4 EI
upon 1 into thetabc. But what is thetabc? 0, because it cannot rotate. Theta is from the chord to
the tangent, that is 0 and when it is rigid, EI is infinity, therefore, essentially what you have is
a 0 by 0 situation; therefore, Mbc and Mcb can only be obtained by equilibrium at the joints,
that is all, there is no way. Therefore, bc does not come into the picture at all. But what
about ab and cd? These are flexible members, so these slope deflection equations have to be
written. You need to find out from the chord to . (Refer Slide Time: 40:10). What is this?
This is going to be the same as 1 by 20. This is 5 by 4 divided by length, which is 15, so this
is going to be 1 by 12.
What about this? What is this angle? From the original to the chord, it is going to be same as
1 by 12. From the chord to the tangent is how much? It is 1 by 12 plus 1 by 20. Here, from
the chord to the tangent, what is this? 1 by 12. Here, from the chord to the original
undisplaced tangent, it is equal to 1 by 12. What is it from the original tangent to the new
tangent because of this rotation? That is 1 by 20. What is the total from the chord to the
tangent? It is equal to 1 by 12 plus 1 by 20. Let us see what thetaab is equal to; thetaab is equal
to 1 by 12. What is thetaba equal to? 1 by 12 plus 1 by 20 which is 2 by 15. This one (Refer
Slide Time: 41:54), I do not need to find out because I cannot write down slope deflection
equation for this. thetadc is going to be equal to 1 by 12 and thetacd is equal to 1 by 12 plus 1
by 20, which is 2 by 15. These are the member end rotations in terms of delta.
Here, the point that I was trying to make is that the kinematics of the structure is very very
important and once you can draw the displaced shape corresponding to each degree of
freedom, you are well on your way to using the displacement method for solving any
structure. The advantages in the displacement method as you see it is that it actually makes
life much easier for you, in general.
Now, I am going to actually take this particular problem, this problem and I am going to
actually put some loads on it and see what happens in reality. Let me take a very simple load
initially; let us not complicate the issue too much; let me say that this structure is only
subjected to 20 Kilonewton and due to this 20 Kilonewton, I need to draw the bending
moment diagram for this particular structure. I have got these. From these, what do I get?
This is a single degree of freedom, so what do I get from this? I will write down the equation.
Thetaab is equal to delta by 12, thetaba is equal to 2 delta by 15, thetacd is equal to 2 delta by
15. Wait a minute. I made one mistake and that is that we did not find out what the values
were. Let us go back here. From the chord to the tangent, anticlockwise; from the chord to the
tangent, anticlockwise; from the chord to anticlockwise all of them are positive and I can
actually write them down. Thetadc is equal to delta by 12. I am going to substitute this into my
equation. What are my fixed end moments? Since there are no loads in the members, fixed
end moment is equal to 0. What is my Mab? All my fixed end moments are 0, so what is Mab
equal to? Mab is equal to 4 EI and length of ab is 15 times delta by 12 plus 2 EI by 15 into (2
delta by 15). What does this become? This becomes one-third, this is EI upon 45; this is
going to be equal to 5, this is going to be equal to 45, so 5, 5 plus 9, so this going to be 9 EI
delta by 225. Mba is equal to 2 EI by15 into (delta by 12) plus 4 EI upon 15 into (2 delta by
15). Here, I have 6 and it will go to 450, this is going to be equal to 5 and this going to be
equal to 160, so this is going to be (5 plus 160) into EI by delta. If I divide throughout by 5, I
get 35 by 90 EI delta.
In the same way, I can find out Mcd. You will see that Mcd will be equal to 35 by 90 EI delta
and Mdc is going to be equal to 9 EI delta upon 225. I have obtained these equations in terms
of the delta and then, I need to write down the virtual work equation. How will I write down
the virtual work equation? Find out all the work done by all the internal forces undergoing the
displacement. What is the external virtual work? External virtual work is delta prime into 20.
What is the internal virtual work? Internal virtual work is going to be equal to Mab into 1 by
12 plus Mba into 2 by15, of course, delta remains, this is going to be delta prime, this is going
to be delta prime plus Mcd into 2 by15 delta prime plus Mdc into 1 by12 delta prime. This is
equal to this. From this, if I put it in, I am going to get that this implies that 20 is equal to Mab
by 12 plus 2 Mba by15 plus 2 Mcd by15 plus Mdc by12 see how easily we got the equation.
If you had to actually do it in the other way, you would have to find out all the moments, you
would have to actually solve everything, find out all the reactions and from the shear and the
actual force, you would then get the horizontal and then equate the horizontal reaction that
these two equal to 20 it will take you a few hours to do that. Satisfy yourself; try it in that
way and you will get this equation.
10
Here, you will see that all these are in terms of delta, so I will substitute these values in. I will
get only one equation in delta and I can solve for delta. Once I solve for delta, I can
immediately plug it back in here and I get Mab, Mba, Mcd and Mdc. In other words, once I
solve for it, I can get Mab. Now, I know Mab. I am not solving them but I know Mab, Mba, Mcd
and Mdc.
These can be evaluated by plugging it into that equation, solving for delta and then plugging
them back in here. I have only found out Mab. How do I draw the bending moment diagram?
This is where equilibrium comes in. You cannot forget equilibrium. So, what you have to do
is, you have to do this. You cannot forget equilibrium because it cannot be violated.
Therefore, this is Mab, this is Mba, this is Mbc ; we did not evaluate Mbc because bc was rigid,
but nonetheless even in a rigid member, you are going to have bending moments, remember
that; this is Mcd and this is Mdc . These are my bending moments where I know this, I know
this (Refer Slide Time: 51:58), I know this and I know this. Can I find out Mbc? Sure, I can.
How? Equilibrium says that Mba plus Mbc is equal to 0, since there is no moment applied.
Similarly, over here (Refer Slide Time: 52:21), you will have Mcb plus Mcd is equal to 0. You
can actually evaluate this and this.
Once you know these, since there are no bending moments, there are no other loads on the
members, just the bending moments themselves, you can find out the reactions and draw the
bending moment for the structure. The whole point here is just to go back. You can draw the
bending moment; I am going to leave that to you. I have already found out Mbc. There are no
member end moments, so you can always find out the reactions and then, the bending
moment diagram will look like this (Refer Slide Time: 53:06), straight line here, continue
straight line here, come here, continue straight line here (Refer Slide Time: 53:11) this is
your line and you have got the bending moment diagram for the load of 20 Kilonewton force.
The point that I am trying to make here is this: what are the important aspects to the entire
thing? The important aspects to this are a) finding out the number of degrees of freedom,
identifying. For the degree of freedom, let us come down here. You will see that for the
degree of freedom, you get the displaced shape of the structure; the displaced shape of the
11
structure can be obtained through kinematics. Please keep practicing this; you have to get
kinematics.
Once you get the kinematics, you have got all the thetas and member end deformations. Once
you have got the member end deformations, you can get the member and then apply the same
delta as my virtual displacement pattern and so, I can write down my virtual work equation;
from the virtual work equation, I can actually substitute these and get an equation for delta
prime, from where I can solve delta prime. Once I solve delta prime, I plug those back, get
Mab, Mbc and then use equilibrium to obtain your bending moment diagram.
I hope over the last couple of lectures I have been able to impress upon you how the method
of virtual displacement becomes a very powerful tool in simply analyzing a frame by the
displacement method. I shall continue by actually solving a few more problems, so that you
can gain confidence of how to actually use the displacement method and the method of
virtual displacement to. In fact, the method of virtual displacement is used to generate the
equation, it is not a separate method; the method is the displacement method in which you
generate the equations to solve for the displacement using the method of virtual work, virtual
displacements. Once you solve them, you can get the member end moments; once you know
the member end moments in any member, it becomes a statically determinate structure and
you can draw the bending moment diagram.
Thank you very much. I look forward to continuing with you on using the displacement
method.
12
Prof. P. Banerjee
Lecture 16
Today, we are going to continue looking at the displacement method, which we have been
spending the last few lectures looking at. Today, I am going to introduce a new concept that
will lead to a simplification of applying the displacement method, essentially because it
eliminates a few degrees of freedom which you may need to consider otherwise. Remember
that everything is essentially member force deformation relationships and that is what I am
going to be looking at right now.
So if you look at the displacement method again, just to lay the foundation and in the
displacement method for the member force deformation relationship, this is our structure and
we say that this is Mab, thetaab and then, we have Mba, thetaba and there is also a displacement
deltaba. Remember I am just writing down the original slope deflection equations from last
time and they come in this fashion. Mab is equal to 4EI by L thetaab plus 2EI by L thetaba
minus (6EI by L squared deltaba) plus (FEM)ab, where EI is the constant flexural rigidity and
L is the length; Mba is equal to 2EI by L thetaab plus 4EI by L thetaba minus (6EI by L squared
deltaba) plus (FEM)ba. thetaab and thetaba are the rotation of the tangent from the original
tangent.
Remember that from the original undisplaced tangent to the displaced tangent are the
definitions and deltaba is the deflection of b. This is a, this is b, deflection of b is relative to a,
upwards of b relative to a; thetaab and thetaba are positive anticlockwise and the fixed end
moments and moments are positive anticlockwise. This is the original definition and in the
last two lectures, I introduced you to the concept of Mab is equal to 4EI upon L thetaab plus
2EI upon L thetaba plus (FEM)ab; Mba is equal to 2EI upon L thetaab plus 4EI by L thetaba plus
fixed end moment at ba.
Here, what are the definitions of thetaab and thetaba? These here (Refer Slide Time: 06:03) are
defined from the chord to the displaced tangent; this chord is the straight line joining a and b
(that is the chord) from the chord to the displaced tangent that is thetaba. Here, maybe I will
call it theta this way just to distinguish from these (Refer Slide Time: 06:26), where these are
from the chord. The fixed end moment here and here remain the same. These are the two
definitions that I have looked at where it is essentially the fixed, where you have both Mab
and Mba. Today, I am going to introduce to you a different member. This is based on fixed
fixed member and this is the force deformation force or member rotation relationship for a
fixed fixed beam. Today, I am going to introduce you to a different, I mean, a beam with
different end conditions.
This is the beam that I am going to introduce you to. This is EI upon L (Refer Slide Time:
07:32) and note over here that the difference between the original fixed fixed beam and this
fixed hinge beam (this is a hinge here) is the fact that you do not have. You know that Mba
is equal to 0. Note that thetaba is not equal to 0. Rotation can occur, but Mba cannot be nonzero, because at a hinge, this is the n hinge and at the n hinge, you know that Mba is equal to
0. Today, I am going to introduce a different concept. This is Mab, thetaab. Note that over
here, what I have to understand is what happens, what is the additional Mba has to be equal to
0?
In other words, what does that mean? It means that Mba is equal to 2EI by L thetaab (I am still
using the definition of from the chord) plus 4EI by L thetaba plus fixed end moment at ba is
equal to 0. I will come to the fixed end moment later on but for now, let us just say that this is
0 for now, let us say it is 0 (it is normally not 0, but for now, let us call it as 0). Then, what do
you have? You have 2EI upon L thetaab plus 4EI upon L thetaba is equal to 0. What does that
mean? It means that thetaba prime is equal to minus (2EI upon L divided by 4EI upon L, so
that is half thetaab prime). Do you acknowledge that? This is just by putting it equal to zero; I
am assuming this fixed end moment equal to zero. thetaba is equal to minus of and now, I
substitute. Therefore, if Mba has to be equal to 0, thetaba has to be equal to minus half of
thetaab.
Let us see what happens to Mab. Mab which is equal to 4EI upon L thetaab prime plus 2EI
upon L thetaba prime plus (FEM)ab, which for now I am considering. In other words, I am
considering the situation where there is no member load. What is this equal to? Here, I
substitute this here (Refer Slide Time: 11:12), so this becomes 4EI upon L thetaab prime plus
2EI upon L and this is equal to this. If you look at this, this becomes equal to 3EI upon L
thetaab prime.
In other words, what happens is, if you look at it, Mba is equal to zero and Mab is equal to 3EI
upon L thetaab prime. In other words, the only relationship that we are left with in this
particular situation is the force deformation relationship; it is Mab is equal to 3EI upon L
thetaab prime because Mba is equal to 0. What is thetaba equal to? It does not really matter in
this particular case, because remember the only reason why we are finding out rotations and
displacements is essentially to find out the moments; and what happens here is, since you
know that the moment over here (Refer Slide Time: 12:41) is 0, you really do not need to find
out this at all; you do not need to find it out. However, if you are really interested in finding it
out, since this is equal to 0, you can always find out what thetaab is equal to from this
formulation. For a fixed hinged member, this is the force deformation relationship; this angle
(Refer Slide Time: 13:23) is from the chord to the.
If theta is taken from the chord to the displaced tangent and if you want to define it in the
original sense, then you can show. I am not going into the details here but exactly in the
same way, you can show that this is the original definition. This is from the original tangent
to the displaced tangent (the rotation is the angle from the original tangent to the displaced
tangent) and deltaba is the displacement upwards of b relative to a. This is it. Now, what about
the fixed end moment? Up till now, we have not considered the fixed end moment. The
advantage of this is that I can consider the fixed end moments separately and add it because it
is after all superposition, so let us look at what happens. To consider the fixed end moment, it
is very easy. This is equal to this (Refer Slide Time: 15:16), where. Note that for this, we
know the fixed end moments, so it is (FEM)ab, given any loading.
We know the fixed end moment (Refer Slide Time: 15:32) and we know ba. You know how
to compute these also, I have shown it to you. This is here; this is the modified fixed end
moment, so we can call it modified fixed end moment; this is equal to this plus this. What is
the reason? Look at this. I am saying that this is the original one which we know how to
compute and this plus this (Refer Slide Time: 16:52). What I am doing is, I am actually
applying the negative of the fixed end moment. Why am I applying the negative of the fixed
end moment? Because this plus this will ensure that this becomes equal to 0, there is no fixed
end moment because there is no fixed end; and all I have to do is find out this, because then I
know that the modified fixed end moment at ab is equal to (FEM)ab plus (FEM)ab prime,
because the principle of superposition is valid. So, all I have to do is find out. I know this,
this I know from before (this, all I have to do is find out what this is, given this load. Let us
see. If I apply a load.
Let us go back to the original equation. Let me say that I am applying a moment here. We
need to find out what is the fixed end moment at this. If I can do that, I have solved my
problem. Let us see what happens. If you look at this particular situation, what do you have?
You have thetaab is equal to 0 and thetaba is not equal to 0 this is the condition. Here, what I
am doing is, I am finding out Mab. This is the only loading on the system, so Mab is equal to
4EI into thetaab, which is 0, so it going to be just 2EI upon L thetaba. Do you agree to that?
We do not know what thetaab is, but do you agree to this? Since there is no load, there cannot
be any fixed end moment. What is Mba equal to? Mba is equal to 2EI by L into thetaab, which
is 0 into plus 4EI upon L thetaba and since there are no moments applied, there is no fixed
end moment. What is Mba equal to? By definition, we know Mba is equal to M, which implies
that thetaba is equal to M L upon 4EI.
All I need to do is substitute that into this and what do I get? 2EI upon L into M by 4EI,
which if you notice is equal to M by 2. In other words, what is this Mab? Mab is the fixed end
moment since thetaab is equal to 0; that means if I apply a load M here (Refer Slide Time:
20:43), the fixed end moment is M by2. I have derived it using the equations and since fixed
end moment is equal to a, since I am applying minus fixed end moment at ba, what is this
fixed end moment equal to? It is going to be equal to minus fixed end moment at ba upon 2.
4
What does that become? Ultimately, the fixed end moment modified at ab is equal to fixed
end moment at ab due to the fixed fixed condition minus ((FEM)ba upon 2), where these are
obtained using the using the fixed fixed. We know how to obtain this; let us apply it to a
particular equation and see whether we get it from first principles also.
I am taking an example of load P being applied at L by 2 (Refer Slide Time: 22:24) and what
are the fixed end moments? We can derive them but we already know what the fixed end
moments are. These are equal to PL upon 8, PL upon 8 this way. If I were to find out for this
situation, load applied at the center, what is this fixed end moment equal to for the modified
beam? The fixed end moment is equal to PL by 8 (the fixed end moment at ab) minus PL
upon 8 divided by 2 (note that the fixed end moment here is negative because it is clockwise).
This becomes PL upon 8 plus PL by 16 and this becomes 3 PL upon 16; so, the fixed end
moment here is equal to 3 PL upon 16 (Refer Slide Time: 23:55) when this end is hinged.
Now I can derive this from first principles. How will I derive it from first principles? Let us
think back. I want to satisfy you that whatever we have done, we should be able to obtain it
for this also from first principles. Let us see what happens when I apply first principles.
How do I go from first principles? I consider this, apply the load here and under this load,
what is going to happen is that it is going to go like this. What I am saying is that this plus
a moment applied here such that if this is (thetaab)0, I will get minus (thetaab)0. It is going to
be equal to this. If this is the case, then this (Refer Slide Time: 25:46) is equal to (FEM)ab.
We have already looked at this; the only difference between the original and this one is that
since this is the only one fixed, we do not need to satisfy this (thetaba)0, and this can be thetaba
due to M.
Nowhere does (thetaba)0 plus thetaba due to the moment have to be equal to 0, no. Why?
Because this is a hinge, the rotation can go anything. The only thing is we cannot apply a
moment here and I am not applying a moment anywhere. Note that I am only applying the
moment here (Refer Slide Time: 26:34); since this is a hinge, its moment is 0 and this is a
hinge, its moment is 0. Since I am not applying a moment, this plus this is not going to give
me a moment here and this plus this is not equal to 0, because there can be rotation; if
rotation is allowed, it cannot take a moment; so you see the difference between the original
one and this one. Let us see what happens here, let us go through the steps.
Let me apply the load; due to the load, the bending moment is PL by 4 and it is sagging.
Now, I need to find out thetaab due to the loading. How do I do that? L is this. Virtual force I
need to find out this, so I apply a unit virtual force and I find out the moment diagram. The
moment diagram looks like this (Refer Slide Time: 28:05), this is my small m diagram, this is
my M diagram. Since EI is a constant, this gives me directly the M by EI diagram and this
gives me the small m diagram. What is (thetaab)0 equal to? (thetaab)0 is equal to the area under
this curve half; because there are two areas, I have to consider PL upon 4EI into L by 2 into
1 by 2, which is half length into base (that is the area of the triangle). Since this is this way, I
am taking that as positive, this is going to be negative.
The value at this point (the centroid) is equal to minus 2 by 3; the centroid of this part is at
two-thirds of L by 2, which is one-third L, so we look at one-third L from this side. Thus, you
will see the value is equal to 2 by 3 and negative because if I consider this positive, then this
is negative plus. I have to consider this part, so it is going to be PL upon 4EI into L by 2
into 1 by 2 multiplied by the centroid value at this point. Look at the centroid; again, twothirds L by 2, that is at one-third, so the value at this point is equal to one-third and again
negative, because if this is positive, this is negative. This is equal to P L squared upon 16EI.
This is identical. Remember we had computed this earlier also. This P L squared by 16EI
still remains the same. Now, what I need to do is I need to find out a moment that will give
me minus thetaab.
What I need to find out is due to a moment here (Refer Slide Time: 31:02), what is the
rotation here? How do I find that out? You will see that it comes out from the original
equation that we have the force deformation relationship, but I will do it from first
principles. Due to this moment, what is the moment diagram? The moment diagram looks
like this and this is equal to M and since EI is a constant, I am drawing the M by EI diagram.
I want to find out this thetaab, so I apply a moment here and this is unit moment here and this
is going to be the small m diagram. I am not writing them big because you know them
already.
Here, what is the thetaab due to the moment? You will put this and what you get is the area
under this curve, which is ML upon 2EI, M upon 2EI into half of length. Where is this? This
is at two-thirds the distance at that value and note since this is hogging, this has the same
direction , so it is positive and this becomes equal to ML upon 3EI. You could have found
this out easily because Mab is equal to 3EI upon L thetaab you know this and you see that
you will get exactly the same relationship from here too. Since we know the thetaab and this
has to be equal to. If M has to be the fixed end moment, then this (Refer Slide Time:
33:21) plus this has to be equal to 0. Therefore, what we have is fixed end moment at ab
becomes equal to 3EI upon L. Note that this was minus, so you see what we are getting is that
thetaab is equal to minus (P L squared upon 6EI); minus is obvious because this is clockwise,
minus means clockwise. This plus this (Refer Slide Time: 34:01) is equal to 0 because that
is what you have here.
It has to be minus of (thetaab)0 (Refer Slide Time: 34:15). Therefore, if we put that in, what do
we get? We get it equal to minus (3EI upon L) because this goes on the other side
multiplied by minus of (P L squared by 16EI); this becomes plus 3 PL upon 16. What is the
fixed end moment? For a load applied at L by 2, the fixed end moment over here is positive 3
PL by 16, which means anticlockwise you have already derived that earlier. Therefore,
essentially what happens is that this I have derived from first principles and we have got
exactly the same thing from here too and since we had this from the same, therefore I can say
for certainty that I can write down the fixed end moment for this modified beam by just
taking the fixed fixed and just subtracting ba by 2; I will always do this. This is the uniform
equation that I am going to use because I know the fixed end moment for the fixed fixed case.
Now, what I am going to do is to see how this changes.
Ultimately, if you look at the equation for this, it turns out to be Mab is equal to 3EI upon L
into theta (this is where you take it from the chord plus and here, I am going to write down
the expression for the fixed fixed case these are the fixed end moments from the fixed fixed
case; and this is algebraic if the sign of ba is different from ab, it becomes positive. This
becomes my new equation; I am going to use this to solve a problem that we had solved
earlier.
What was the problem? I will make the problem statement for you; the problem statement is a
problem that you have seen earlier and I am going to just put it down. I am going to use this
procedure to solve this. This was the problem; this is a, this is b, this is c; this was the
problem that we looked at earlier. In this particular case, if you look at member ab, at b, you
have continuous, so there is going to be a moment here and so we cannot consider this in this,
but if you look at bc, c is an end support and at an end support, we know that moment is
equal to 0.
Remember we had applied that moment at c is equal to 0 as one of the equilibrium equations.
In this particular case since I know that the moment is equal to 0 and I am not interested in
knowing what the rotation over here is, what is the only degree of freedom that I have?
Thetab because note that I am not interested in thetac because I know that the moment over
here is equal to 0. So, I am going to use the modified for the ab, I am going to use the
original element and for bc, I am going to use a modified element; let us see what happens. I
am going to define this problem in a different way. Earlier, what we had taken ab as fixed
fixed, bc as fixed fixed, and we had defined thetab and thetac, both of which are unknown
displacements. Here, I am only going to consider one unknown displacement and let us see
how we can solve this problem. You see the advantage of this problem automatically with
only one degree of freedom, we only need one equation and things will become much simpler
if you put that in. Let us go back and revisit that problem.
What we are going to do is, we going to be considering ab as a fixed fixed. As far as ab is

considered, we are going to say Mab is equal to 4EI upon L thetaab plus 2EI by L into thetaba
(note that since there are no displacements, I am not even writing down the delta because the
10
chord or you can say take it from the chord, it does not matter) plus fixed end moment at
ab; and Mba is equal to 2EI upon L thetaab plus 4EI upon L thetaba plus (FEM)ba. So, for ab,
this is what I do and the fixed end moment comes from this 120 Kilonewton force acting at 4
meters. For bc, I am going to consider it this way: b is fixed but c is the end hinge. Whenever
you have an end hinge, you know that the bending moment at that point is equal to 0 and so, I
am going to use this and I have this loading, this is 50 Kilonewton per meter. I am going to
put Mab is equal to 3EI upon L thetaab plus (here, I am going to put down that these fixed end
moments are the ones I had considered; they were fixed fixed) [(FEM)bc minus ((FEM)cb by
2)]. For ab, I consider this (Refer Slide Time: 42:30) and for bc, I consider this. Now, for
thetab is equal to 1, let us go through the steps. For thetab is equal to 1, what does the thing
look like?
Note that that is the only degree of freedom so this is going to become 1. I am going to have a
displacement pattern that looks like this and note that over here, it does not have to go to 0
because this is a hinge and it can take any value.
11
What is the value over there? Actually, if you notice, this one is Mbc (Refer Slide Time:
43:20) and it is equal to thetabc (Refer Slide Time: 43:22) and since in this equation there is
no thetacb, I do not really care what the rotation over here is. If I put this, I get that for thetab
is equal to 1, thetaab is equal to 0, thetaba is equal to 1 and thetabc is equal to 1 and thetacb I
do not know and I do not care because thetacb is not going to get into my force deformation
relationships anyway. Once I have that, it means now I can put down the expressions. I get
Mab; thetaab is equal to 0, so I get 2EI upon L and since thetaba is equal to thetab, this becomes
thetab plus (FEM)ab. Mba is equal to 4EI upon L thetab plus (FEM)ba. These do not look any
different from what we had originally; they look identical. The only difference here is in this
equation Mbc is equal to 3EI upon L into thetab (since thetabc is equal to thetab) plus
((FEM)bc minus ((FEM)cb by 2)). I need to find out ab and bc and note that I have already
done those earlier, so I am just going to show you that we have already evaluated these.
12
These were the fixed fixed, so ab is equal to plus 172.8, (FEM)ba is equal to minus 115.2 and
(FEM)bc is equal to 416 and minus 416. Having done that, I can plug in those values directly.
Therefore, what do I get?
My (FEM)ab is equal to plus 172.8 Kilonewton meter, (FEM)ba is equal to minus 115.2 and I
need to evaluate (FEM)bc minus ((FEM)cb by 2), which is going to be equal to 416.7 minus of
(minus 416.7 by 2); this becomes 416.7 plus 208.3, which is equal to 625 Kilonewton meter.
These are the fixed end moments that I have; I have put them in. Once I have all of these
relationships, I need to write down the equilibrium condition. Since there is only one, the
equilibrium condition, satisfy yourself, the equilibrium condition becomes Mba plus Mbc is
equal to 0. If I put that in, what do I get?
13
I get 4EI upon L into thetab minus 115.2 this is Mba and Mbc is equal to 3EI upon L thetab
plus 625 (this is the fixed end moment) is equal to 0. This becomes 7EI upon L thetab is equal
to 499.8 and thetab becomes equal to minus 71.4 L upon EI. Once I have my thetab, I
substitute it into my equations and I get Mab is equal to 27.2 Kilonewton meter, Mba will be
equal to minus 406.5 Kilonewton meter and Mbc will be equal to plus406.5 Kilonewton
meter. This is what you got the last time, so you will get those values just by substituting this.
Now, since L is 10, this is multiplied by 10 (Refer Slide Time: 50:20).
The beauty of this essentially becomes this: that there is just one degree of freedom, so the
equations become much easier. The major advantage is that I have already made two degrees
of freedom into a single degree of freedom. You may say that there was not too much of a
difference between two degrees of freedom and a single degree of freedom, but when you go
into multiple degrees of freedom and you can eliminate a few of these degrees of freedom,
you will see that it becomes a big issue.
This simplification does nothing to the solution process; it still remains exactly the same; the
only thing that happens is that you land up getting a much better solutions base. That is all I
have to say in this particular case. We will see that as we go along, we are going to get many
many problems. By the way, let us go back to this; I have made one small mistake and I just
want to correct that mistake because you may find yourself. This becomes actually 509.8
and if you look at this, this becomes 72.8 and these give you this; this basically becomes
thetaba is equal to minus 728 upon EI; you will see that 728 gives you this; I have just made a
mistake in putting it down over here; satisfy yourself that this is going to happen. Now, I am
going to just state the basic concept that you have to work with.
Right in the beginning, you have to decide whether you want to use this (Refer Slide Time:
52:49) or whether you want to use this. If either joint of the member is continuous, in other
words, some other member is connected over here, then you always have to use this. You can
use this only when this end happens to be the end support of a structure and so, this can only
be used if there is an end support; you cannot use this if there is another member connected
over here; you cannot. If there is another member, then the moment at this point is not equal
14
to 0. Is that clear? You can use this only if you have an end support; otherwise, you have to
always use this and use the member force deformation relationship for this; and for this, I
have just shown you today how the force deformation relationship is different for this one.
I hope today's lecture has made it clear to you that even in the displacement method, you can
actually bring in simplifications to enable you to reduce some of the degrees of freedom and
thereby reduce the equilibrium conditions that you need to specify; ultimately, you reduce the
number of algebraic equations that you need to solve for finding out the displacements and
hence the member end forces. Thank you very much.
15
Prof. P. Banerjee
Lecture 17
Good morning. In the last lecture, we saw how we got a modified member force deformation
relationship. In this lecture, I am actually going to take you through one problem completely so
that you can understand what it is and I am going to take a frame problem; let us take the
problem. Last time, time before last, we looked at some kinematic relationship, so I am going to
take that same problem.
Under normal circumstances, you will see that this has two degrees of freedom one degree of
freedom is this lateral displacement and one degree of freedom is the rotation at this hinge (Refer
Slide Time: 03:08). However, this is an end hinge and I am going to consider this member ab as
a modified member so that I will write down the modified member force deformation
relationships. Here bc is rigid, this is EI, this is also EI, this is totally rigid flexurally as well as
axially, these are axially rigid but flexurally flexible and so is this one; this is being subjected to
this load and the question is to find out the bending moment diagram. Note that last time, I had
said that since member bc is a rigid member, you cannot write down a member force
deformation relationship because deformations are 0; so you cannot write a member force
relationship. However, you can write it for ab and cd.
First, let me write down the member force deformation relationship. Let me go through it as an
algorithmic state, just like I had done for the force method. What is the first step? Identify the
number of degrees of freedom and identify the degrees of freedom. If the degrees of freedom are
related to the end hinge, then we can forget those degrees of freedom (those are non-essential
degrees of freedom). So, although this has two degrees of freedom (you can evaluate it to be
1
having two degrees of freedom), this is a non-essential degree of freedom because the moment at
this point is equal to 0. We have only one essential degree of freedom, so this is a single degree
of freedom structure. Having done that, the next step is to write down the member force
relationships for each member.
For member ab, what do we have? We have this situation: this is a, this is b. What is the member
force deformation relationship? You are going to see that it is only going to be Mba and thetaba.
Note that none of the members have any member loads, so Mba is going to be equal to 3EI upon
L thetaba that is the only member end moment; since bc is rigid, you cannot write down the
force deformation relationship because deformations are 0 and so, you cannot evaluate bc; there
is bc, there is cb but we cannot write down any force deformation relationship for this; and for
cd, since it is fixed fixed, the member end deformation becomes this. Here, note that I am using
the modified definition of thetab, so it is going to become equal to 4EI upon L thetacd plus 2EI
upon L thetadc; I am not putting the additional terms because this definition of thetacd is from the
chord joining c to d to the displaced tangent, same as this; and Mdc is equal to 2EI upon L thetacd
plus 4EI upon L thetadc. There are three members ab, bc and cd; and these are my member
force deformation relationships. The next step is to relate the member deformations with the
structure displacement. How do I do that? Kinematics.
What do I do? I write down; this is my structure and what is my displacement? I am going to put
delta is equal to 1; all that delta is equal to 1 means is that this point has to move horizontally by
1. Can this point move vertically? No, because since this point cannot move anywhere, if this
point were to move vertically, it would involve axial deformation; this cannot move vertically
but would only move horizontally.
Therefore, this point, we know where it is, this point (Refer Slide Time: 8:54) is at a distance 1
from where it is. Let us look at this. Left to itself, this would go 1 this way. However, this point
cannot go anywhere displaced, so this point can only move along its tangent and if it moves
along its tangent, this member also can move along its tangent and this is the point and we have
already looked at this last time; this is where your point b has gone.
Now, since point b has moved here and c has moved here, this member has to remain straight
because it is a rigid member; a rigid member can only go straight; this angle then becomes 3
upon 4 divided by 15, which is 1 by 20. Therefore, the tangent over here automatically has to
follow this and it becomes like this; this tangent also has to follow that, so this also becomes like
this; this is 1 over 20, this is also 1 over 20 and this goes like this. Note that since here it is a
hinge, it is free to rotate and it will rotate in this fashion.
Now, to find out the from the chord, I need to actually connect these joints; this is a, this is the
displaced position of b, so I will call it b prime, this is c prime; a prime and d prime are exactly
where it is because these are supports. This is going to be in this fashion, this angle is going to be
equal to 1 upon 12, is this angle is going to be 1 upon 12.
Similarly, this angle is 5 upon 4 upon 15 becomes 1 upon 12 and this is also 1 upon 12. Having
drawn this, we need to put down what these are. We are not interested in thetaab because
remember that in member ab, the only member deformation is Mba is equal to thetaba; keeping
that in mind, we only need to know what thetaba is. Thetaba is 1 by 12 plus 1 upon 20 is equal to
3
(5 plus 3 equal to 8) upon 60, which is equal to 2 upon 15. Is it positive or negative? From the
chord to the tangent, it is positive. What is thetabc? Note again that I do not need to find out
thetabc and thetacb; I know that they are 0 because from the chord, these are 0; I know those, they
are rigid, there can be no deformation. I am interested in finding out thetacd. Why do you find
these out? It is because we can relate all these in terms of the delta that is the reason why we
find this, this and this; these are the only three that are there, so you only need to find out these
three. When you find out those three, thetacd is equal to 1 upon 12 plus 1 upon 20, that makes it
plus 2 by 15, again anticlockwise; and thetadc is equal to 1 upon 12. Having these inputs, what do
I get?
I plug this into the original equations and I get Mba. Mba is equal to 3EI upon 15 (L for ab is 15)
into thetaba, which is 2 by 15 delta. Mcd is equal to 4EI upon 12 (length of cd) into thetacd, which
is equal to 2 by 15 delta plus 2EI upon 12 into thetadc, which is 1 upon 12 delta. This becomes
6EI delta upon 225; this becomes 1 upon 12 and inside, you will get EI delta, you can take EI
delta outside and inside, I get 8 upon 15 plus 1 upon 6; this is going to be 30 is the nearest, 30
would make it 16 and this would make it 5, so 21 upon 30, this would become 21 EI delta upon
360, which becomes equal to 7 EI delta upon 120 that is Mcd; and Mdc is equal to 2EI upon 12
into (2 by 15 delta) plus 4EI upon 12 into thetadc, which is 1 by 12 delta and if you plug that in,
you get (EI delta upon 12) outside, inside you get 4 upon 15 plus 1 by 3, so this becomes 4 plus 5
equal to 9 upon 15, which becomes 3 upon 5, which in turn becomes EI delta by 20. Note all of
these are positive for delta. Having obtained these, what is my next step? My next step is to
actually write down the equation by taking this as my virtual displacement; I am going to use the
principle of virtual displacement where this is my virtual displacement; I can put any arbitrary
virtual displacement, so I opt to put virtual displacement equal to 1. What is the work done by all
the forces?
Let us see what the work done by all the forces is equal to. What is the work done by the external
force? It going to be equal to 1 (unit virtual displacement) into real force, which is 20
Kilonewton. What is the internal virtual work done? Internal virtual work done is going to be
equal to thetaba (that is the virtual thetaba), it is going to be 2 by 15 into Mba (the real moment)
plus 2 by 15 into Mcd plus 1 by 12 Mdc; this then becomes equal to 2 by 15 multiplied by 6EI
delta upon 225 plus 2 upon 15 multiplied by 7 EI delta upon 120 plus 1 by 12 into EI upon 20.
If I take it through, you will see this is equal to this, so all I am going to do is, I am going to
multiply by 15 both sides. This will become 300 is equal to 12 and I am going to take EI delta
outside; inside becomes 12 upon 225 plus 14 upon 120 plus this is 240 and if I take out 15,
this goes into 15 16 times, so this is 1 upon 16; please make a note that this will actually go 4
times, this when divided by 3 becomes 75, so 4 upon 75 plus this becomes 7 upon 60, this is 15
into 4 and this is 15 into 5, so the LCM is going to be equal to 15 into 5 into 4 into 4; this is
going to be equal to 4 cubed, so that is equal to 64, this is going to be equal to 20, this is 140; this
is going to be equal to 75; this is equal to 279 upon 1200; this divided by 3 becomes equal to 93
upon 400 EI delta. Delta is going to be equal to 120000 upon 93EI; I have found out delta. Once
I have found out delta, I can substitute back into Mba, Mcd and Mdc to be able to evaluate those
values. How would my equation and then how would my boundary look?
Is this positive? This is positive. If these were positive, it means Mba would also be positive; Mba
is positive, Mcd is positive and Mdc is positive. Mba is equal to 2 upon 75 EI delta, which is 2
upon 75 into 120000 upon 93 (note that this is also EI and so EI, EI cancel) this is the value of
Mba. Mcd is equal to 7EI delta upon 120, so this is going to be equal to 7 upon 120 into 120000
divided by 93 and Mdc is equal to EI delta upon 20, which is going to be equal to 6000 by 93.
These are the values.
You should be able to obtain these values. I can put down some values over here; 25 will go here
3 times and this will go 4800, which also when divided by 3 becomes 1600; ultimately, Mba is
equal to 3200 by 93, Mcd is equal to 7000 by 93 and Mdc is equal to 6000 by 93 these are in
Kilonewton meter.
What is Mbc equal to? I know that Mab is 0, I have found out Mba, I have found out Mcd and I
have found out Mdc. What do I need to find out? I need to find out what Mbc and Mcb are. Note
that from equilibrium, Mbc has to be equal to minus Mba; similarly, Mcb from equilibrium has to
be equal to minus Mcd. Therefore, if you look at this particular situation, this becomes
approximately 35; let us just put it down; this would be approximately about 78 and this would
be about 66.
How would the bending moment diagram look? At this point, this is 0 and this is anticlockwise,
so the bending moment diagram will look like this. This is anticlockwise and since it is
anticlockwise, I am going to be drawing it on the compression side; this is going to be this way
and this going to be this way. Over here, this is this way; over here, this is this way; over here,
this is this way and over here this is this way (Refer Slide Time: 30:51). The values are: this is
3200 upon 93 Kilonewton meter both of them; here, both of them, the value is 7000 upon 93
Kilonewton meter; and over here, it is going to be 6000 upon 93 Kilonewton meter. This is my
bending moment diagram for the 20 Kilonewton meter force.
You saw the steps: the first step is to identify the degree of freedom; the next step is to write
down all the force deformation relationships for all the members that are possible; third, get the
kinematic relationship between the member end deformations and the displacement
corresponding to the structural degree of freedom; then, use the virtual work equation and write
down the equation corresponding to each structural degree of freedom; then, solve for the
displacement corresponding to the structural degree of freedom and from that, derive the member
end forces; once you get the member end forces, you have got a structurally determinate member
and you can draw the bending moment diagram for each member; put it all together and you
have the bending moment diagram for the structure this in essence is your displacement
method.
Now, what I am going to do is, I am actually going to sit and look at another problem. The
reason behind that is very simple: the overall scope of what we want to do is that you have to
understand the various ways of tackling this problem. How we apply the virtual work principle
and how we get the member end kinematics these are the two fundamental points in the whole
displacement method. How do we choose the members? Normally, in all members where the two
joints at the two ends are continuous joints, you will always have a fixed fixed member. The
member force relationship is very simple: if you have a member load, then you need to compute
the fixed end moments; if you do not have a member load, you do not need to compute the fixed
end moments.
Next is if you have one end of a member which is a support and that support is a hinge support,
we can eliminate that degree of freedom and modified beam, so that the modified member force
deformation relationships. Once you have done the member force relationship, the next step is
the kinematics the relationship between the members; you essentially apply each unit
displacement corresponding to each degree of freedom and see how the structure deforms and
from that evaluate what the member end deformations are this aspect is called kinematics.
Once you have the kinematics, then applying the virtual work principle becomes reasonably
simple. The only point that I would like to make is that every case becomes different. For
example, if you do not have member loads, then you do not need to consider any additional
forces; but if you have member loads, then in the virtual work equations, you actually need to put
in how much deflection the entire reactions are undergoing and that will give you the overall
scope of the virtual work equations.
Once you have got your virtual work equations, then it is a simple algebraic matrix solver to get
all the displacements corresponding to the degrees of freedom and then substituting back, you
get the member end moments. Once you get the member end moments, you should be able to
draw the bending moment diagrams, shear force and everything you should be able to find it
out. Every member becomes a statically determinate structure. Now I am going to actually take
you through one or two problems to show you how to tackle this problem.
For this, let us now look at this situation. This is the structure (I keep using different structures).
Under these loads, I have to draw the bending moment diagram that is my question. I also wish
to know another additional point: I wish to know how much this support (Refer Slide Time:
38:07) displaces by under these loads that is also another point I would like to know my
delta.
From this point onwards, I am going to stop using thetab, thetac, delta the way I have been using
until now; I am going to start defining the structure degrees of freedom using a notation r.
Suppose I have three degrees of freedom in a structure, I am going to call them r1, r2, r3; r1 may
be a rotation, r2 may be a displacement, all that does not matter to me, I am going to define them
as displacements. If the displacement happens to be rotational, a rotation is also a rotational
displacement, it is a general displacement, just like when we say force, I mean not only a linear
force but I also mean moment. This is the problem but I am not going to be able to solve this
problem today, so what I am going to take you through is some of the basic parameters so that
next time when we meet, we can actually try to go through the solution process. Let us see what
the first step was and let me write this down, so that it gets into your head that this is the
displacement method.
One: determine kinematic indeterminacy, which is equal to the number of degrees of freedom
that is the first step. Two: identify the degrees of freedom; in other words, identify the
displacement quantities that you want to be your degrees of freedom; normally, this is selfexplanatory but in peculiar situations, you may have to choose this. Three: write the force
deformation relationship for each member in the structure; the assumption over here that we are
making (we will release that much later) right now, just for simplicity, is that each member is
considered to be axially rigid; I am going to write that down, implicit assumption: axial rigidity
(it is not required but we are making) this is implicit; therefore, the force deformation
relationship essentially becomes moment rotation relationships. You may not be able to write
this if a member is flexurally rigid also, just like we saw that previously, you had member bc
which was flexurally rigid; we could not write the member force deformation relationship for
that. One other thing is write member force deformation relationship before that, actually, this
is 3b; 3a is identify for each member whether it is standard or modified this is important. Why?
9
Because just like we showed that if you have a situation where you have a member whose one
end is a hinge support, then we can use the modified and then we write appropriate member force
deformation.
Now, first important step over here is this (Refer Slide Time: 43:15): determine kinematic
indeterminacy. After that, identifying the degrees of freedom is not that important. Once you
have identified the force deformation relationship, standard or modified is also easy. Writing
down the force for each is also easy. The fourth step is: obtain expressions for fixed end
moments if member loads are present; if you have member loads, then you need to do this but if
the loads are adjoints, then you do not need to do this.
Next is kinematics this is another very very important step because the entire success of your
displacement method actually depends on the kinematics; if your kinematics is wrong, your
solution process is going to be wrong. The next step is the virtual work equations for each degree
of freedom. How do we get them? We take kinematics; for each degree of freedom, we find out
the displaced shape and that is the kinematics find displaced shape for displacement
corresponding to each degree of freedom.
The next important step is the virtual work equations for each degree of freedom. There are some
considerations that we will show you over and over again, which go into this. Seven: solve
equations to get displacements. Eight: obtain member end moments. Once you know the member
end moments, you can get. This in essence is the displacement method. Once you have this
displacement method, let us go through some of the steps for this particular problem that we
have identified.
First, the degrees of freedom. Determine the kinematic indeterminacy. How do we go about it? 1,
2,3, so how many joints? The unrestrained degrees of freedom are equal to 3 into 3, 9. How
many restraints? This is the fixed (Refer Slide Time: 47:19), so in a fixed you have two; and this
10
is the fixed roller; in other words, this cannot move in this direction; it cannot rotate but it can
displace this freely. How many are we restraining? The rotation as well as this motion, so there
are two; so restraints is minus 4; constraints are two, so minus 2. How many does that leave us
with? Three degrees of freedom. What are those three degrees of freedom? Let us go back. Let
us understand what these are. One is the rotation thetab I will classify this as r1 (remember I
was saying I was going to use); I am going to call thetab as r1. What else? The delta at c this I
am going to classify as r2. What else? Let us see what happens. Suppose I stop this from going
up.
Can this go up and down?
I have made a mistake here, 3, there are three restraints here (Refer Slide Time: 49:38), two
restraints there, so that is 5; so 9 minus 5 minus 2 is 2 degrees of freedom. Here, the point to be
noted is that there is . Can this rotate? This can rotate. If this displaces upwards, what happens
over here? This joint can rotate in any direction that it wishes, because ab is constrained to move,
so you have two degrees of freedom. We have identified two degrees of freedom for this
structure and then, what is the next step? We have also identified the degrees of freedom, so we
have completed the first two steps of the displacement method.
The next is to identify for each member the force deformation relationship, whether it is standard
or modified. Is any member a hinge? No, it is not hinged, so both ab and bc are standard
members; so 3a has been completed. What is 3b? 3b is write the member force deformation
relationships for each member. Is that going to be for ab and bc? Both of them are going to be
4EI upon L into theta that point plus 2EI upon L, we already know that. Obtain expressions for
fixed end moments. Does ab have any load on it? No, so fixed end moments at ab and ba are 0.
Does member bc have a force on it? Of course, we have to determine the fixed end moment at b
and fixed end moment at c. We will do all of this in my next lecture.
I hope at the end of this lecture that you are reasonably clear as to what are the steps that are
involved in a displacement method. Over the next few lectures I am going to solve different
problems and show you how the different considerations come into the aspect. One, determining
the kinematic degrees of freedom; two, the kinematics; and three, how to write down the virtual
work equations these are the three important points and each problem that I am going to be
looking at will have a different aspect to all these three points. Thank you very much. Look
forward to seeing you next time.
11
Prof. P. Banerjee
Lecture 18
Good morning. Today, we are going to be solving some more problems on the applications of
the displaced method and I am going to take up a specific problem.
If you look at this particular problem, the various questions are one: find this displacement
(Refer Slide Time: 03:19), two: find this rotation and three: find the bending moment. I will
call this as r1 and this as r2. The question is one: find r1 and r2 and two: draw the bending
moment diagram this is the problem statement. Let us see how we can go about solving this
problem. First and foremost, . Let me go back to that problem itself. First, we need to find
out the degrees of freedom; this is a, b, c. The unconstrained degrees of freedom are equal to
3 into 3 equal to 9; restraints are 3 here and 2 there, which is equal to minus 5; constraints are
2 members (1 into 2), axial rigidity is minus 2, so this is equal to two degrees of freedom and
the two degrees of freedom are r1 and r2 these are the displacements corresponding to the
degrees of freedom. What is the next step?
Step 2: for ab, what is the member forced deformation? Mab is equal to 4EI by L thetaab plus
2EI by L thetaba . We are doing the member deformation relationships and for ab, Mab is
equal to 4EI upon L thetaab plus 2EI upon L thetaba plus (FEM)ab. Then, Mba is equal to 2EI
upon L thetaab plus 4EI upon L thetaba plus (FEM)ba. Now, what are the fixed end moments
here? (FEM)ab is equal to 0, which is equal to (FEM)ba? Why is that? There is no member
load on ab; since there is no member load on ab, the fixed end moments are automatically
equal to 0. Now, we look at bc. Mbc is equal to 4EI by L thetabc plus 2EI by L thetacb plus
(FEM)bc and Mcb is equal to 2EI by L thetabc plus 4EI by L thetacb plus fixed end moment at
cb.
If you look at the end, this is fixed, therefore, ab is normal. In bc, c is a fixed roller and when
it is fixed, there is a moment; since there is a moment, we have to use the original; if this had
been a hinge, then of course we would have used the modified, but in this particular case, it is
not and therefore, we are going to continue in this way. What about the fixed end moment at
bc? We have to find that out; let us find that out. The next step is The fixed end moment at
bc and cb are going to be there because there is a load in member bc, so we have to find out
what that is. How do we find that out?
Note that there is a slight difference here: the member is inclined and the loading is this way,
so I can actually write this load of 100 Kilonewton. This is a 3, 4, 5 triangle and therefore, if I
write down the vertical component and then the horizontal component, the vertical
component of this is going to be 80 Kilonewton and the horizontal component is going to be
60 Kilonewton.
Now, the horizontal component, since the member is axially rigid, there is absolutely nothing;
therefore, this problem essentially becomes one where you have a 10 meters with 80
Kilonewton and we have to find out what the fixed end moments are for b and c. We know
that it is equal to PL by 8; this becomes 100 Kilonewton meter, 100 Kilonewton meter; this
implies that (FEM)bc is equal to 100 Kilonewton meter and the fixed end moment at cb is
equal to minus 100 Kilonewton meter this is something that we know as of now. What is
the next step? We have written the member end deformations and we have found out the
fixed end moments. The next step is to find out the kinematic relationship for each degree of
freedom the relationship between the member end deformations and the structural degrees
of freedom. How do we do it? Well, kinematics.
The first step is this: give r1 is equal to 1 and r2 is equal to 0. If you give that, what happens?
This point is to move up by 1; this point can only move up because that is the only direction
that it can go, so this goes up. How do I find out where this point goes? This is important; this
is what I mean by kinematics. Therefore, what we do is, we let that go up by 1 that is the
only way it can go. We know where this point is (Refer Slide Time: 11:52). This point is
fixed and therefore, it is not going to go anywhere.
How can this point go? This point can actually go on this. Now, let us look at this carefully.
Since I have taken this up by 1, if you look at the components of this one, this is vertical
(Refer Slide Time: 12:27), along the member and perpendicular along the member; these are
the two things that we have to find out; you can find that out with the 3, 4, 5 triangle; please
do the; you will see that this is 3 by 5 and this is 4 by 5. In other words, what has actually
happened is. Suppose this point was not fixed, this could have gone up also with it by the
same amount.
Here, the point to be noted is that in actual fact, what has happened is that it has gone along
this way and this way, but you know that this point cannot go there; that means this member
has become something like this if we displace it, but this cannot go there because this point
can only move along this line. So, what has to happen? Since this member is this way, the
only way this member can move is along this and when we take it along this, this is the point
at which it is going to get the cut and that is the point where this member is going to be. What
is that point going to be? Let us look at that. This is 1 and this is a 3, 4, 5 triangle, so if 4 is 1,
that means this is going to be equal to 3 by 4 and this total distance is going to be 5 by 4.
Once I know this point, I can now draw my displaced shape. Note that this cannot rotate, so
this remains straight and this also has to remain parallel to the line; this will go in and here
also, it cannot rotate, so this is our rotation.
If we look at this, from the chord, this angle gives thetaab and what is that angle equal to?
Thetaab from the chord to the tangent is anticlockwise, so plus 3 by 4 (the displacement) upon
10, which is going to be 3 upon 40. What about thetaba? It is the same thing 3 by 40 and
positive (from the chord to the tangent), so plus (3 upon 40).
4
What about this one? The total displacement of this is 5 by 4, this is moved by 5 by 4 and the
length is 10. Therefore, looking from the chord to the tangent, it is minus because it is
clockwise from the chord. Thetabc is going to be minus (5 by 4 divided by 10), which is going
to be minus (5 by 40) and thetacb is the same. We have found out thetaab, thetaba, thetabc and
thetacb in terms of r1; when r1 is equal to 1; these are the values. Now, r2 is much easier. All
these computations are for r1. For r2 is equal to 1 and r1 is equal to 0, the displacement pattern
is going to be very very simple; it is going to be this. If we put this, what do we get as thetaab?
Thetaab is equal to 0. Thetaba is equal to plus 1 (from the chord to the tangent that is
anticlockwise); similarly, from the chord to the tangent, so thetabc is equal to 1; and thetacb is
equal to 0. We have done the kinematics and this kinematics is going to be important even
when we do virtual displacement; so remember this kinematics.
By substituting the kinematics into the equations, what do we get for ab? Let us put down
values. Mab is equal to 4EI upon L (which is 10) into thetaab, which is (3 by 40 into r1 plus 0
into r2) plus 2EI upon 10 into (3 by 40 into r1 plus 1 into r2) and FEM is 0. Mba is equal to 2EI
upon 10 into (3 upon 40 into r1 plus 0 into r2) plus 4EI upon 10 into (3 by 40 into r1 plus 1
into r2). This implies that Mab is equal to 3EI by 100 r1 plus 3EI by 200 r1 plus 2EI by 10 r2,
which becomes equal to 9EI upon 200 into r1 plus 2EI upon 10 into r2 that is Mab. Similarly,
Mba is equal to 3 by 200 r1 (Refer Slide Time: 20:38) plus 3 by 100 (Refer Slide Time:
20:39), so we get exactly the same thing: 9 EI upon 200 into r1 plus 4EI by 10 into r2. These
are Mab and Mba in terms of r1 and r2. Similarly, we can write down for bc.
We get Mbc is equal to 4EI by L (which is 10) into thetabc, which is minus (5 by 40 into r1
plus 1 into r2) plus 2EI upon 10 into thetacb, which is minus (5 upon 40 into r1) plus fixed end
moment at bc, which is equal to plus100. Similarly, Mcb is equal to 2EI upon 10 into minus (5
upon 40 into r1 plus 1 into r2) plus 4EI by 10 into minus (5 by 40 r1) minus 100. This implies
that Mbc is equal to. This is going to be minus 5 by 100 and then minus 5 by 200; 5 by 100
is 1 upon 20; 1 upon 20 and this is 1 upon 40; so this is going to be equal to minus 3EI upon
40 into r1) plus (4EI upon 10 into r2) plus 100. Mcb is equal to minus (3EI by 40 into r1) plus
(2EI upon 10 r2) minus 100 this is Mcb. I have got my Mab, Mba, Mbc and Mcb. Next, what I
have to do is, I have to find out the work done by the loads, so I am going to write down the
virtual work equation. The first virtual work equation is going to be written down for this as
virtual displacement; if I put this as virtual displacement, then what do I get?
I am going to put the first virtual work equation as r1 is equal to 1; since it is in an arbitrary
displacement pattern, the first displacement pattern is going to be this pattern. My external
work will be the load into 1. What is the external load? There is only one external load, which
is 50 tons because that is applied at the joint. How much displacement does that undergo?
Since the load is applied here and it is undergoing three-fourth in the same direction, the
external work done is 50 into 3 by 4. If you look at everything else, what is it going to be?
Think about it. There is no other external load that does work; this is the only one that does
work and so this is going to be equal to 75 by 2 this is the external work in Kilonewton. The
internal virtual work is done by all the moments undergoing the rotations: one moment is Mab
into the rotation that it undergoes, which is (theta bar)ab plus Mba multiplied by (theta bar)ba
plus Mbc multiplied by (theta bar)bc plus Mcb undergoing (theta bar)cb. These are all the
moments and they also take care of the work done by the shears. The only thing left behind is
the load 100, which gives rise to reactions at the supports.
The reaction that I get is actually equal to 50 and 50 vertical in both the directions. I just
have to find out what is the work done by each vertical reaction. Is this point moving
vertically? If you look at it, it is not, it is only moving horizontally, so this one does not do
any work; however, this vertical reaction does work. How much work does it do? It does into
1. This is going to be equal to plus 50 into 1; this is my internal virtual work. Having written
that, I can put internal virtual work equal to external virtual work and if I substitute all the
terms that I have, you will see that this essentially becomes the following: (9EI upon 200 into
r1 plus 2EI upon 10 into r2) multiplied by (theta bar)ab, which in terms of r1 is equal to 3 by 40
plus Mba, which is equal to (9EI upon 200 into r1 plus 4EI upon 10) into 3 upon 40 plus Mbc,
which is equal to (minus (3EI upon 40) into r1 plus 4EI by 10 into r2 plus 100) multiplied by
minus (5 by 40) (note that this work done is multiplied by the rotation; rotation being
clockwise, it is negative work) plus Mcb, which is equal to (minus (3EI upon 40 into r1) plus
2EI by 10 into r2 minus 100) multiplied by minus (5 upon 40) that is all the moments plus
50 into 1 is equal to 75 by 2. That is my first equation; it is fairly complex, but I can always
simplify it and write it in terms of r1 and r2 and a number.
What about the second one? In the second virtual work equation, r1 is equal to 0 and r2 is
equal to 1 this will give me independent. What is the external virtual work done? You will
see that since that point does not displace, it is going to be equal to 50 into 0, which is 0.
Internal virtual work is going to be again Mab into thetab etc., Mab into thetab is 0 plus Mba
into thetaba, which is 1, plus Mbc into thetacb, which is 1, plus Mcb into thetacb, which is 0, plus
the work done by the reactions, which is 0. Therefore, the second equation becomes our most
common Mba plus Mbc is equal to 0, which we know already; we got it from virtual work.
Therefore, if you notice, when you have a rotation, the virtual work boils down to the fact of
it being the equation of joint b and that is what we are interested in. We have written the two
equations and note that once I substitute for Mba and Mbc (I will just put them down), you will
get that 9EI by 200 into r1 plus 4EI upon 10 into r2 minus 3EI by 40 into r1 plus 4EI by 10
into r2 plus 100 is equal to 0. This is easy to do, so what you get is 5, 15 minus 6 by 200;
minus 6 by 200 becomes minus 3 by 100EI into r1 plus 8EI by 10 into r2is equal to minus 100
that becomes my second equation. Let us go through the steps of the first equation.
I am going to actually go through this. Let me put down all the things that we are going to
get. This equation is simple and if I expand the first equation, it is going to be equal to 27EI
by 8000 into r1 plus 6EI upon 400 into r2 plus 27EI upon 8000 into r1 plus 12EI upon 400 into
r2; then you have minus and minus, so it is plus 15EI upon 1600 into r1 minus 20EI upon 400
into r2 minus 500 upon 40 plus 15EI upon 1600 into r1 minus 10EI upon 400 into r2 plus 500
by 40 plus 50 is equal to 75 by 2. If I put in all the r1s together, you will see that if I put 8000,
8000, this will become 75, 75, 150; 150 plus 54, so that is 204EI upon 8000 into r1; all of
them are 400 in r2, so it is 6 plus 12, 18 minus 20 is minus 2, minus 12 upon 400, so minus
3EI upon 100 into r2; this cancels this, plus, minus; minus goes on the other side; so this
becomes minus (25 by 2). We can simplify this, we can divide by 4, so we get 51 upon 2000
EI into r1 minus 3EI upon 100 into r2 is equal to minus 25 that is one equation. Let me write
down those two equations and you will see that those two equations land up being this way.
If I take out the EI and write it in matrix form, this will become 51 upon 2000, this is going to
be minus (3 upon 100), here I am going to have minus (3 upon 100), plus 8 upon 10
multiplied by r1, r2 is equal to minus (25 by 2), minus 100. Note something interesting; you
will see that this is equal to this. In fact, that is also always true that if you write it down
properly, this and this are equal to each other and it is symmetric. It turns out that we can
solve for r1 and r2 by taking the inverse of this; the inverse of this is going to be equal to
20000, this is going to be 100000, this is going to be minus 18 upon 20000 and this is going
to be 408; so 408 minus 18 is going to be 390, so this is going to be equal to 20000 upon
(390EI) multiplied by the inverse; this is 100; so I can find out r1 and r2. Once I know r1 and
r2, then I substitute these back into the original equations.
The original equations being Mab is equal to this into r1 and Mba is equal to this into r1 and r2.
For bc, Mbc is given in r1, r2 and Mcb is given in r1, r2, so you can find out Mbc and Mcb. Now,
I have found out r1 and r2. There are several interesting points to note over here. One is that
this can be written in terms of stiffness matrix into r is equal to some force; we will see later
when we use them. Right now, this is not matrix method; I am just writing down two
equations, which I am solving using matrix algebra; I am just solving two simultaneous
equations using matrix algebra. Right now, I have not yet introduced you to the concept of
the stiffness method; understand I am still using the displacement method excepting that I am
slowly beginning to introduce the concept of K r is equal to f; this term is K, this is the r
vector and this is like the equivalent load vector which represents all the effects of all the
loads that come in. Let us just assume that I have been able to find out Mab, Mba and Mbc.
Now, where do I go from there?
I have found out Mab this is known; Mba this is known; Mbc this is known; Mcb, this is
known. The point that I would like to make is that I have not written down the values of r1, r2,
Mab, Mba, Mbc and Mcb just because I want you to actually solve this problem yourself; when
you solve this problem yourself, you are going to get the values. Just before I start my next
lecture, I am going to give you these values so that you can check whether you have done the
problem correctly. Now, I am just giving you the steps that you have to go through; I will
give you the solution at the beginning of the next lecture so that you have an opportunity to
actually try and solve this problem. Let me just go back to what I was trying to state.
Ultimately, how you get your bending moment diagram is this way.
You know Mab, you know Mba, you know Mbc and you know Mcb. Over here, what do you
have? You have this load and over here, what are the things that you have? Note that at this
point, I know that there is not going to be any vertical reaction. This is because it is a roller.
So, what is the only force that is going to come? This is what I am trying to say: be
innovative, incorporate, do not try to solve problems just by themselves. Let me just put it.
What I have over here is Ma, Va, Ha and then I have this over here; here, I am going to have
Mc and the reaction is going to be just Hc because vertical cannot be there; since this is going
to be this way, you might as well apply this, apply this; the fact that this is going to be like
this; this is going to be Hc and it is only going to be vertical; since this is going to be
vertical, I know that over here, I am going to get both the horizontal and vertical. Note that
these are not the shear and this not the axial force. Once I evaluate these, I can actually
evaluate all the support reactions. What is this going to be equal to? This is going to be equal
to 100 Kilonewton, then this is going to be 100, this is going to be 100, this is going to be 100
and over here, this is going to be 100.
Here, I have this, this is going to be here, this is going to be here, this is going to be this way,
this is going to be here and here, I have 50. One of the other equations that I get is Ha plus 50
is equal to Hc. Can I find out Ha? Sure, I can find out Ha. This is this way and over here also,
it is this way. Ha is equal to minus (Mba plus Mab) upon 10, minus because both of them give
the same; that is my Ha.
10
Now, these two I know, so I can find out Ha; once I know Ha, I can find out Hc. I have got my
horizontal I know this; I know this, I know this, I know this, I have found out this. Once I
have all of those, I know all of these, I can draw the bending moment diagram; I leave it up to
you; I will draw the bending moment diagram next time.
Now, what I want to do is, I want to bring to you a concept and I would like you to solve this
problem yourself; I will come back next time and solve the problem. I have done enough
problems using the displacement method and now, I am going to propose problems for you to
do; once you have done the problem, I am going to go back, look at my next lecture and you
will get an opportunity to find out how to do the problem correctly; but I recommend before
you look at the next lecture, I would suggest that you solve this completely. I am going to
give you the answers next time and I would like you to look at one other problem; let us look
at the problem that I would like you to look at.
This looks suspiciously like the problem that we solved today, does it not? Excepting that I
have made a small change. This is a, this is b, this is c. What I have done is, I have made this
fixed roller into a hinge roller support and I am still going to say that you have to solve this
problem using these two displacements as your degrees of freedom. Actually in this particular
case, you have a third degree of freedom but remember that this is an end hinge, so I can
actually treat bc as a modified member and keep ab as an original member; bc is a modified
member with a fixed at this end and a roller at this end. What I would like you to do is, given
this problem, one: find r1 and r2 and two: draw the bending moment diagram. This is exactly
the same problem as I have done.
I am leaving you at the end of this lecture with two things to do. This is the first assignment
that I would like you to do in the displacement method; please go ahead and solve the
problems; first problem: having done everything, find out r1 and r2 and then draw the bending
moment diagram using the procedure that I have outlined; second: take this problem, solve
this problem through and find out r1 and r2 and draw the bending moment diagram. I am
going to give you the answer to this and solve this next time. Another interesting point that I
am going to bring out and this is of course what I tend to do always; and that is not just the
11
ability to solve problems, but you see, what I have do over here is I have changed a fixed into
a hinge.
Next time, I am going to actually look at what effect does changing support conditions have
on the displacements and the bending moment diagram in the structure at all times.
Therefore, one aspect that I would like you to look at when you are doing this assignment is
not just get the numbers, but once you get the numbers, you also try to see what are the
effects of changing this fixed to hinge, because I have got exactly the same member with
exactly the same load and since I have the same member and the same load, the only
difference here is this support condition; this difference between the two answers is going to
give you the effect of changing this fixed to hinge and that is something that I am going to
concentrate on the next time. Thank you very much. Do please solve the problems and if you
want to know how the problems and their answers are, wait for my next lecture.
Thank you.
12
Prof. P. Banerjee
Lecture 19
In the last lecture, we looked at a particular problem on a frame, which was loaded we saw
how to develop the equations. Then, I spent some time looking at how to solve that particular
problem. I promise that today I will start off the lecture by actually giving you the solutions. I
hope you have had an opportunity to look at those solutions.
This is the final solution the equations that I looked at. Remember that this was the abc
frame with a 100 Kilonewton load here and 50 Kilonewton load here. r1 and r2 were the
displacements corresponding to the degrees of freedom and when we solve for r1 and r2, we
got r1 is equal to minus (2000 by 3EI), which essentially means that r1 is 2000 upon 3EI
downwards; r2 was minus (150 upon EI), which essentially means that r2 is equal to 150 upon
EI clockwise I hope you got these answers too; if you have, pat yourself on the back
because you have understood the displacement method reasonably well. Now that you know
r1 and r2, if you substitute them back into the equations (the member end moment
relationships), then you get Mab is equal to minus 60 Kilonewton per meter, which essentially
means that Mab is clockwise 60 Kilonewton per meter; Mba is equal to minus 90 Kilonewton
per meter, which basically means Mba is clockwise 90; Mbc is plus 90, which essentially
means it is counterclockwise; and Mcb is equal to minus 80, which means it is clockwise
these can be obtained directly by substituting these values into the equations for Mab, Mba,
Mbc and Mcb. What does that mean? Once we have found those out, what do we do? Last time
I derived the expressions, I had listed out the expressions for the support reactions in terms of
50 plus Ha is equal to Hc and Ha is equal to minus (Mba plus Mba) upon 10.
Having substituted Mba and Mab for Ha, we got that Ha is equal to 15. Therefore, Hc is equal to
65 these are the support reactions. This is a clockwise moment of 60 Kilonewton per meter
here, a vertical reaction of 100 Kilonewton, a horizontal reaction of 15 Kilonewton and a
horizontal reaction of 65 Kilonewton in this direction, which is clockwise these are the
support reactions. Once you know those, you can always draw the bending moment diagram
and this is what the bending moment diagram looks like. It is linear here, continuous here;
this is superposed on top of this, so at the center of the beam, this is 115 Kilonewton per
meter; and at the end of the beam it is 90 and 80 here and 90 and 60 here.
This is the sense of the bending moment, this is the tension on the top; on the left hand side
here, tension is on this side; here, tension is at the top; here, the tension is at the bottom ; here
again, the tension is at the top. I hope this is exactly what you have got from your solution. I
just gave you the solution to this particular problem. Remember this solution. It is interesting
to note here that almost everywhere. What is the maximum bending moment here? The
maximum bending moment is on the right hand side is 60, on the left hand side is 90, on the
top it is 90 and on the bottom it is 115. Fairly uniform. What are the displacements? The
displacements are 667 upon EI and 150 clockwise, downwards, this is clockwise. Now, let us
look at what happens to that particular problem where all that we have done is taken this and
made it into a we have removed the fixity here so that this becomes a hinge roller. By the
way, before I end this particular problem, I want to tell you that this actually is the part of a
particular problem that we solve in general. This is actually the solution to this problem.
This portal frame problem with vertical loads of 100 uniformly distributed and an equal and
opposite lateral load of 50, 50, essentially using the fact that this structure. Here again, this
would be 8, 8, this would be 10 and this would be 6. This is a portal frame and this portal
frame by using symmetry can simplify to this particular problem. Although this particular
problem that I am solving does not look like a real problem, it is actually a very real problem.
This is the solution to this problem using symmetry. I will take an opportunity to discuss this
particular topic of how to use symmetry to make structures simpler so that they have lesser
number of degrees of freedom a little bit later in this particular course. Remember that this
problem is actually not a problem that is constructed just like that it is actually a real
problem.
Now, we come to the next problem that we have. If you look at the next problem, there are
identical loads but the only thing that we have done is, we have removed the fixity over here
3
so that the moment at this point is 0, so that it becomes essentially a hinge roller support;
fixed support, hinge roller support. In this particular case, what is the difference in the
solution process? Member ab continues to be a fixed-fixed member because it is fixed at one
end and continuous at another end. However, we made bc into a modified fixed-hinged
member because it is continuous here, so it is fixed; this is an end support here we know
that the moment at this particular point is equal to 0, so that makes bc a modified fixedhinged member. Let us see how we proceed with that.
We already know that this has two degrees of freedom but there is a third degree of freedom
which is the rotation at this point but since we are considering the fact that the moment at this
point is 0, this is not an essential degree of freedom and so we are left with two essential
degrees of freedom in this particular structure. Having identified the two degrees of freedom,
the first step is to write down the member force deformation relationships.
Here ab is a fixed-fixed member, so Mab is equal to 4EI by 10 (since L is equal to 10) into
thetaab plus 2EI upon L into thetaba plus fixed end moment at ab. Since there is no load in
member ab, obviously the fixed end moment is 0. Then, Mba is equal to 2EI by 10 thetaab plus
4EI by 10 thetaba plus fixed end moment at ba, which is also equal to 0 that is the force
member relationship for Mba; note that the way we have defined it in this particular equation
actually turns out to be. What are thetaab and thetaba? Remember that these are the rotations
of the tangent from the chord joining ab remember that; thetaab and thetaba are rotations of
the tangent at a and b, respectively, considering the rotation from the chord joining a and b;
remember that; I just wanted to tell you that this is the solution.
Now for bc, since bc is a modified member, Mbc is equal to 3EI upon L and L is 10 (because
8 and 6, 10) into thetabc plus (fixed end moment at bc minus (fixed end moment at cb upon
2)). Remember this particular thing? Here, what are these fixed end moments? These are the
fixed end moments in a fixed-fixed beam such that when the load is. Remember that we
had evaluated these two quantities last time. This turned out to be plus 100 Kilonewton per
meter and this had turned out to be minus 100 Kilonewton per meter. When you substitute
that into this equation, you get 150 Kilonewton meter. Therefore, Mbc becomes equal to 3EI
4
upon 10 thetabc plus 150. We know that Mcb is equal to 0 and we do not need to know thetacb
because we have condensed out thetacb. Remember that thetacb is not equal to 0; thetacb is not
equal to 0 remember that. In the modified fixed-hinged member, thetacb is not equal to 0.
However, we consider the fact that Mcb is equal to 0 and so thetacb can be statically
condensed out and we are only left with thetabc; thetacb is not an essential deformation
quantity required to define the force member relationship in bc. We have got the relationship
for Mab, Mba and Mbc in terms of its member end rotations. Remember that this thetabc is also
from the chord to the tangent. What is the next step? The next step is to find out the kinematic
relationship.
First, we put r1 is equal to 1 and r2 is equal to 0 and second, we put r2 is equal to 1 and r1 is
equal to 0. First and foremost, for this, this is simple and it turns out to be this way and this is
this way, this is equal to 1 and this is equal to 1. Therefore, what we have here is that for r2 is
equal to 1, thetaab is equal to 0, thetaba is equal to 1 and thetabc is equal to 1 these are the
only essential degrees of freedom that we are interested in and we have evaluated them. Note
that thetacb does not have to be equal to 0 because this is a hinge and we acknowledge the
fact that the moment at this point has to be 0 and we do not need to find this out. If you look
at this one, note that the displacements of the points remain exactly the same as we had
computed last time, so this is 3 by 4 and the displaced shape looks this way. Remember that
here, r2 has to be equal to 0, so the tangent at this point has to remain straight. Similarly, the
tangent over here has to remain straight so this goes, but over here, it does not have to remain
straight because it is a hinge. This is 1, this is 3 upon 4 and you can compute it remember
last time we had done it if this is 3 by 4, then this is going to be 9 upon 20 and this is going
to be 4 upon 5, so that is 16 by 20; 9 by 20 plus 16 upon 20 we had evaluated that this was
equal to 5 upon 4.
If I draw the chords, because ultimately it is the tangent from the chord, my chord goes this
way. This is my angle this angle is equal to 5 upon 4 divided by 10, so it is 5 upon 40; this
angle is equal to 3 by 40 and this angle is equal to 3 upon 40 these are our rotations.
Therefore, thetaab is equal to plus 3 over 40 (from the chord to the tangent, it is
anticlockwise), thetaba is equal to plus 3 upon 40 (from the chord to the tangent
anticlockwise, so positive) and thetabc is equal to minus (5 upon 40) because from the chord
to the tangent, it is anticlockwise, so it is minus.
Note something very interesting: if you look at these values and if you look at these values,
the fact that this point has become a hinged roller and not a fixed roller as earlier has done
5
nothing to the shapes, excepting for the fact that at this point, you do not have 0 rotation but
instead, you have 0 moment. In other words, in the earlier case, it had gone like this and gone
like this because this cannot rotate. Here also, this would go like this and then come like this
because it cannot rotate, but as far as these values are concerned, they do not change at all
because the tangent and the chord remain exactly the same. Therefore, these things are not
determined closely by the fact that this fixed roller has become a hinged roller.
The point that I would like to make is that the kinematics definitely depends on the support
conditions. However, if we change a fixed support condition to a hinge support condition and
keep the geometry the same, it has no effect on the kinematics, on the overall kinematics. Of
course, as I said, the displaced shape may look different, because if it was fixed, it looks like
this and would look like this. However, as far as the kinematics is concerned (that is the
relationship of the member and rotations with the degrees of freedom of the structure), they
are no different from what we had evaluated last time. Thus, this is an important point to note
that kinematics is essentially driven by geometry rather than by changes in support
conditions.
Of course, one thing is there: in this particular case, it has not changed. Why? Because the
support condition change has actually introduced a degree of freedom; however, because we
have used the modified method, we have the same number of degrees of freedom and for
those degrees of freedom, the kinematic relationships remain the same. However, if we had
not used the modified beam, then we would have had another additional degree of freedom
and then of course, we would have to do the kinematics of that degree of freedom; remember
that. That is the point that I am trying to make to you. Having done that, let us now write
down the relationship.
For ab, since the kinematic relations are the same and the member force deformation
relationships are the same, Mab and Mba are identical to what we had computed when the c
support was fixed. It has no bearing on the member end force deformation relationship in ab
and therefore, I have just written down the value of Mab, which of course incorporates that
thetaab is equal to 3 by 40 into r1 plus 0 into r2 and that thetaba is equal to 3 by 40 into r1 plus 1
theta into r2. If you substitute those into the forced deformation relationships, you get Mab in
6
terms of r1 and r2 in this manner. Similarly, if you substitute again thetaab and thetaba in terms
of r1 and r2 into the member end deformations, you get this Mba. This is identical to the
previous problem; note it.
Let us look at bc now. Because of the fact that c is a hinge, we had to write a modified
member end force relationship which only relates the moment at bc with the theta at bc,
because we had to incorporate the fact that Mcb is equal to 0. I have substituted thetabc; if you
look at the kinematics, thetabc is equal to minus (5 by 40) into r1 plus 1 into r2; if I substitute
that into this and rewrite it, I get Mbc is equal to minus (3EI by 80 into r1) plus 3EI by 10 into
r2 plus 150 these are our relationships. Note that this relationship (Refer Slide Time: 24:35)
is different from the previous problem. Why? Because bc in the previous problem was a
fixed-fixed member whereas in this particular problem, it has become a modified fixed hinge.
We have written Mab, Mba, Mbc in terms of r1 and r2. What is our next step? Our next step is to
write down the equilibrium equations and for that, we will take help of our virtual work
principle the principle of virtual displacement and get the first equations.
We have two unknown displacements and therefore we essentially need to write down two
equations. The first equation, the virtual displacement pattern is r1 is equal to 1 and r2 is equal
to 0; the bar actually gives the fact that they are virtual; this is a virtual displacement pattern.
If I substitute those in, you will see that the external virtual work is 50 Kilonewton is
undergoing a 3 by 4 virtual displacement, so real force into virtual displacement is the virtual
work that is the external virtual work. The internal virtual work is going to be equal to Mab
into (theta bar)ab plus Mba into (theta bar)ba plus Mbc into (theta bar)bc plus now Mcb is 0, so 0
into (theta bar)cb plus 50 into 1; that is the reason why we have no interest in finding out what
(theta bar)cb is; it is because we are going to multiply it with 0.
We only need to know thetaab, thetaba and thetabc that we know; in terms of r1, that is 3 by
40, 3 by 40 and minus (5 upon 40). I substitute Mab from the previous equation and multiply
it by this plus this plus Mbc, which is this multiplied by minus (5 by 40) plus 50 into 1. This is
the support reaction due to the member load of 100 Kilonewtons that, if you remember I
had told you, is something that you have to consider; that is the support reaction that goes up
7
by 1, so therefore the virtual work is 1. Obviously, the virtual work equation says the external
virtual work is equal to internal virtual work, so all we do is we take this and equate it to this.
If we do that, what happens? I have actually gone and multiplied all of those.
This is the work done by Mab, this is the work done by Mba, this is the work done by Mbc and
so this is the external virtual work and this is my equation. If I plug in these, this is 8000,
8000 and 3200, so LCM is 16000; if I put that, this becomes 54 plus 54 plus 75 that is r1;
then, 6 plus 12 minus 15 upon 400 EI into r2 is equal to this is 75 by 2, so by 4, it will
become 150; this minus comes here, plus; and this 50 goes here, becomes minus 200.
Substituting all of those, what I get is 183EI upon 16000 into r1 plus 3EI upon 400 into r2 is
equal to 25 by 4.
Please note that these are simple numbers and that is the reason why I am actually going into
this problem. Otherwise, as far as you are concerned, you can actually take out these and
make them into fractions and do the solution; you will get the same thing, only thing is that
these will be in decimal points; it does not matter. I have done it this way because I could do
it; in some other particular problem, since it gets too complicated, I might actually just
calculate on the calculator itself and put down in decimal engineering digits. I have kept it
this way but I leave it up to you to do it any which way you want to. That is the first equation.
Now, for the second equation, we need a completely independent virtual displacement pattern
so that we can get an independent virtual work equation. Note that I cannot highlight enough
that to get two independent equations, you require two independent virtual displacement
patterns. Since getting independent virtual displacement patterns is hard, what we tend to do
is we know two independent virtual displacement patterns automatically. They are first, the
one where r1 is equal to 1 and r2 is equal to 0 and the second one where r1 is equal to 0 and r2
is equal to 1 we know that these are independent patterns. As I had told you earlier, you
could take anything else and get, as long as you use two independent patterns, because if you
use two independent patterns, then and only then will you get two independent equations,
which you can use to solve for r1 and r2. I use obvious independent patterns and that is why I
am using r1 is equal to 1, r2 is equal to 0 and the other one r1 is equal to 0 and r2 is equal to 1.
8
For r2 is equal to 1, the 50 Kilonewton does not undergo any load, so the work external
virtual work is equal to 0. Internal virtual work is Mab into thetaab, which is 0, plus Mba into
thetaba, which is 1, plus Mbc into thetabc, which is 1, plus 0 into this value we do not know
what it is but since 0 into anything is 0, we do not care plus this is the work done by the 50
Kilonewton, because it does not undergo any load. If you put this in, you get this equation.
Remember I told you that if you use the rotational kinematics, you essentially get back the
equation that you would get by taking the equilibrium of joint b. Substituting for ba and bc
into this and then putting the terms together, what we get is 3EI upon 400 into r1 plus 7 EI
upon 10 into r2 is equal to minus 150. This is my second equation and I know that this is
independent because my virtual displacement pattern is independent of the other. Now I write
down the two equations.
This is what you get; look at this. This is my first equation, which is 183EI by 16000 into r1
plus 3EI by 400 into r2 is equal to 25 by 4; the second one is 3EI upon 400 into r1 plus 7EI
upon 10 into r2 is equal to minus 150. I have written these equations in matrix form just to be
able to solve it. Therefore, r1 and r2 will be the inverse of this into this. The inverse of this is
written in terms of 1 upon the determinant; 1 upon the determinant is this; the determinant of
this is 1272EI upon 160000, so inverse is this; the cofactor of this is this (Refer Slide Time:
33:11), the cofactor of this is this, the cofactor of this is the minus of this, the cofactor of this
is minus of this.
Again, like last time, this matrix is symmetric. The reason for it to be symmetric is the force
at a particular point that gives the displacement at another point. By the Maxwell-Betti
reciprocity theorem, you know that the force load at one point that gives a displacement at
another point is the same as the load at the second point giving a same displacement at the
first point; and this is essentially because of that reason. Remember when we are using the
force method, we got that the flexibility coefficients were the same these are known as
stiffness coefficients. What are stiffness coefficients? Stiffness coefficients are the force to
produce a unit displacement that is stiffness. What is flexibility? The displacement due to a
unit load. This is the load required to get unit displacement. You can see that flexibility and
stiffness are actually the opposite of each other and if you take a single spring, the flexibility
of the spring is equal to 1 upon the stiffness of the spring the stiffness constant, so the
flexibility constant is equal to 1 upon the stiffness constant. This goes that they have to be the
same; if you do not get them to be the same, you have done something wrong somewhere.
Having got this and putting this in, I get r1 is equal to 691.824 upon EI and r2 is equal to
minus (221.698 upon EI). Let us compare that to the situation where you have the fixed case.
In the fixed case, r1 was equal to minus (666.67 upon EI) and r2 was equal to minus (150
upon EI) that is for fixed roller; this is for hinge roller. Let us look at the effects of
removing the fixity at this point to make it into a hinge, in other words, eliminating the
rotational restraint what that has done to the actual displacement of the structure under the
same loads; the loads for both the structures are identical, the geometry is identical, the only
difference between the two structures is that in the previous case at c, you had a fixed roller
10
and in the second case that we are dealing with right now, it is a hinge roller. All we have
removed is the restraint of rotation and see what that has done to the displacement shape. It
has had a significant effect. Look at the rotation. The rotation has gone up significantly. More
interestingly, you see that r1 was downwards when it was fixed and now that you have
removed the restraint, it is actually upwards. Is it not very interesting? You would think that
you have actually just taken a support at one end and all I have done is removed the rotational
restraint but the whole displacement pattern now changes. The support at c instead of
moving down, it is actually moving up and the value is much larger.
This brings us to a particular point: when I removed the restraint, what did I do to the
structure? Think about it. When I put more restraints on the structure, what do I do to the
structure? When I add restraints to a structure, I actually make the structure stiffer. When I
make a structure stiffer for the same load, what would you expect the displacements to be?
Less, is it not? Look at that, it is less. Now, The point is when I remove the restraint, I made
the structure more flexible and therefore, the displacements have to go up and they have gone
up. Therefore, the behavior of the structure is actually reflected in the results that you get
this is very very important. The displacements have gone up and now I can put these
displacements into the equations for the moments. These are the equations where Mab, Mba
and Mbc are in terms of r1 and r2 and since I know r1 and r2, I can substitute into that equation.
When I substitute into that equation, what I get is this: Mab is minus 13.21, Mbc is minus 57
and Mbc is plus 57. What were they when it was fixed? This was minus 60, this was minus 90
and this was plus90. What has happened to the member end moments? They have gone
down. Why? When I made the fixed roller into a hinge roller, what did I do? I increased the
flexibility of the structure. Here is where I make the points that I have tried to make: when
you make a statically indeterminate structure more flexible, a, you increase the displacements
we have already seen that the displacements are significantly larger than in the fixed roller
case; and b, you make the member end moments smaller. Why? Because the structure is more
flexible. When a structure is more flexible, it can actually displace to relieve the stresses in
the structure; when it can deform more to release the stresses, the stresses (or in this
particular case, the member end moments) reduce. Therefore, for a stiffer structure, what do
11
you do? When you make a statically indeterminate structure stiffer, you reduce the
displacements but you increase the member end moments. When you make it more
flexible. Why the member end? Because when you make it stiffer, it is less able to deflect
and therefore, the deflection is small and the loads are essentially transferred through forces.
When the structure is more flexible, it can deform and relieve the stresses. Therefore, the
member end stresses are less. Does that mean that if you do not have a serviceability
criteria. This is where I come into the design aspects. You have two design aspects: one is
to design for strength and two is to design for serviceability. Serviceability always puts a
restriction on your displacements and the strength is so that it can resist the forces that it is
subjected to. When you make a structure more flexible, what happens? Displacements go up.
If you do not have a serviceability problem, then you might think that you get the
displacements go up and the moments to go down and that you can actually make the
structure sleeker because it needs to resist less forces, but you know this is incomplete
information. Let us look at what happens to the bending moment diagram due to these that
is the next step that we are interested in.
We have got the member end moments. Therefore in this particular case, here, the member
end moment is minus 13.21 since it is clockwise, minus 57.5 since it is clockwise, plus57.35
is anticlockwise these are the member end moments. This is the joint at b, this is the joint at
a and this is the joint at c. From this if I take Mab, since it does not have any loading, the
shears at this point are going to be equal to this into this (Refer Slide Time: 43:47) is equal to
this plus this. The couple generated by the shear is going to negate this moment and from
that, we get Ha is equal to this plus this upon 10, that is equal to 7.076; that means Ha is in
this direction and the value is 7.076. In the previous case, what was this? 15. Hc becomes
57.076. What was it before? 65. We seem to be going in the same direction, meaning that the
member end moments and the support reactions seem to be all less. Still we think that things
are going to improve, but now since I have got this, I can actually find out the support
reactions and this is ultimately what the structure looks like.
12
Once I have the support reactions, this becomes a statically determinate structure and I can
draw the shear force bending moment diagram for this. If I do that, this is clockwise 13.21,
this is 7.076, 100, since this is a roller, the entire 100 comes here and here, since it is a hinge,
there is no moment, there is only 57. Till now, the structure is more flexible, the support
reactions are less, we are seemingly. In other words, if your serviceability is not a criterion,
making the structure more flexible by removing a restraint of support seems to make the
structure much better. However, let us look at the bending moment diagram. I leave it up to
you to do it.
Remember I had told you that I am not going to be doing equilibrium for you. Once I have
got the support reactions, you should be able to generate the bending moment diagram, shear
force diagram for any member. I have drawn the bending moment diagram; this is the
bending moment diagram. The bending moment over here has to be 13.21 because that is the
13
support reaction; the bending moment over here has to be 0 because at this point, you have a
hinge, so you cannot have a moment; and if you look at this, what happens here? What was
this? 60 Kilonewton meter, 90 Kilonewton meter, this was 115 Kilonewton meter, and this
one, which is 0 now, was 80 Kilonewton meter. Let us look at it from a design point of view:
is flexibility good? As far as this member ab is concerned, it is good; it has actually reduced
the bending moments significantly from the previous case. However, in this case, what has
happened? In the previous case, this was 115 but now this is 171.23 significantly more,
significantly more. In other words, the member bc.
Until now, we were saying make it flexible, it is going to get much much better as long as
you do not have a serviceability limit. In other words, the rotation and displacement can be
anything that it wishes to be. Then, we thought that introducing flexibility was being
wonderful because it was reducing all kinds of bending moments in the structure, but you
have a problem: it has not. This particular one is more than a one and a half times and this
member bc has to be designed for this. In other words, what has happened here is that making
the structure more flexible does indeed increase the displacements and reduce the member
end moments.
However, if any structure has a load on a member, then this is not a good idea because the
maximum bending moment in the member goes up significantly if you have a member
loading. Now, would you say that increasing the flexibility of the structure makes it better?
No. In fact, by restraining the rotation here and making the structure stiffer, you may have
increased the moments, however, you have made the moments much more even. The
difference between this, this and this and this, this and this the lowest moment here is 60
Kilonewton meter and the highest is 115; here, the lowest is 13.21, highest 171. There is a
tremendous amount of bending moment variability and it implies that curvature goes up and
you might actually have cracking problems, etc., if you have a reinforced concrete.
Therefore, by and large, increasing restraints in the structure is actually a good idea, of course
as long as you do not keep increasing it to a point at which what happens is, the displacement
goes down and almost the entire forces are transmitted. At that particular point, if you take
restraints beyond a certain point, you reduce the effect of the loads, but the member end
moments go up significantly higher and you no longer reap any benefits of making the
structure stiffer. The point that I was trying to make is that there is an optimum stiffness for a
particular type of loading that good designers always look for. Ultimately, that was my take
on the behavior of structures but as far as getting the solution is concerned, you can see that if
you have a structure, then the displacement method for a statically indeterminate structure.
I am now ending the entire discourse on using displacement method for a statically
indeterminate structure. I want to end it by going back through the steps.
The first step: determine the number of degrees of freedom of a structure, given all the
restraints and constraints that you have that is number one. Number two: once you have
determined the number of degrees of freedom, you have to define the displacements
corresponding to the degrees of freedom. Three: determine what each member is and write
down the member end force deformation relationships. Four: determine the kinematic
relationship between the displacements corresponding to the degrees of freedom and the
member end deformations that is where the kinematics comes in. Five: once you have got
the kinematics, you can write down the member end moments in terms of the displacements
corresponding to the degrees of freedom. Six: use the virtual work principle, specifically the
virtual work displacements, to write down independent equations corresponding to each
14
degree of freedom. Seven: solve those equations for finding out the displacements
corresponding to a particular loading system. Eight: substitute those displacements into the
member end displacement relationships to get the member end moments. Nine: once you
have got the member end moments, do equilibrium and find out the support reactions and the
bending moment diagram, shear force diagram, whatever you have to do that is your
analysis completed.
The next lecture is going to be my last lecture on the displacement method. Till now, I have
only looked at loading in the member. Now I am going to look at other situations other
kinds of member loads. Up till now, I have only looked at flexural deformations and flexural
member end force deformation relationships. These are not valid in trusses, so I shall look at
trusses briefly so that you can apply the displacement method for all types of planar trusses
and planar frames. Thank you very much.
15
Prof. P. Banerjee
Lecture 20
Good morning. In the last few lectures, we have been looking at the displacement method and
in the last lecture I told you that I have done as much as I can do to explain the different
nuances in the application of the displacement method. In this particular lecture, I am going
to end my discourse on the displacement method by actually taking you through the last few
points that I would like to make in terms of the lecture. That essentially boils down to how to
treat. Support settlement and support flexibility are not issues in the displacement method
because when you have support displacement and support flexibility, all it does is that it
introduces another degree of freedom. Remember what we had talked was the various things:
one was member loading, other was support settlement, support flexibility and then, we had
lack of fit lack of fit was specifically in terms of trusses and since I have not yet dealt with
truss, I shall do that in this particular lecture; the only other thing, if you remember, was
temperature-related effects.
Today, I am going to start off by talking about temperature effects in the displacement
method. If you are looking at a beam frame, we are only considering flexure; then we only
consider the differential temperature. If you have a cross section in which this is the top and
this is the bottom, this is your depth and this is your neutral axis. Then, essentially, what you
are interested in is this variation. At the neutral axis, temperature is 0, this is the temperature
difference delta T; it is not as if at the center you have freezing, it is just that there is no rise
of temperature at the neutral axis. Here, you have delta T at the top, you have delta T at the
bottom, obviously for it to be 0; this is only the differential part; in reality, you might have
two parts to it; I am only looking at the differential part because that is the one that causes
flexure.
If you remember, I had actually computed earlier that the curvature in the cross section due to
delta T is given by (alpha into (delta Tbottom minus delta Ttop)) upon 2 d, where d is the depth
of the cross section at that particular point this is the curvature. Now, remember that
temperature is a member effect brought in through the fixed end moments. The point to note
here is that if you look at the member force end, member deformation, Mab is equal to 4EI
upon L into thetaab (this thetaab is due to displacements) plus 2EI upon L into thetaba (again
thetaba is due to displacements) plus fixed end moment at ab. The member load effect is only
incorporated in the displacement method by introducing the concept of the fixed end
moment; the fixed end moment is one in which thetaab and thetaba are put equal to 0 and you
find out the member end moments due to the load only. Here, there is no load, there is a
temperature. However, it is again a member effect.
Remember I talked about that? What are member effects? Member effect is lack of fit,
temperature, member load all these are effects at the member level; in the displacement
method, member-level effects are only incorporated in finding out the fixed end forces. In
this particular case, since we are looking at flexure, the fixed end forces are the fixed end
moments. Ultimately, if we can find out the fixed end moment due to a particular temperature
profile, we have solved the problem. I am going to solve it for the simple case. This is the
curvature induced due to the fixed end moments (Refer Slide Time: 07:46). How do I find out
the fixed end moments? The way we found out the fixed end moments was by going back to
first principles.
We found out the rotations, we found out the member end rotations due to the member
effect? What is the member effect? Member effect is temperature. How do I compute these?
Using the virtual force, thetaab is equal to 0 to L M1 into curvature dx; thetaba is equal to 0 to
L M2 into curvature dx. What is M1? M1 is the bending moment in the member due to a unit
virtual force corresponding to thetaab that gives me the virtual displacement; this is my M1
diagram. This is M2 corresponding to thetaba (Refer Slide Time: 09:48) and this turns out to
be my M2 diagram. For the temperature effect, we can find out thetaab and thetaba. I have
already solved this earlier, but I am just going through the steps again just to refresh it.
Now, I am going to assume the situation that everywhere along the length of the member, the
top fiber sees delta Ttop and the bottom everywhere sees delta Tbottom. In other words, the
temperature gradient is uniform along the entire length. I am actually solving for a specific
case where I have uniform temperature gradient across the cross section over the entire
length; in this particular case, beta turns out to a constant given by this; this is a constant, so I
just need to integrate M1 dx. Therefore, I get thetaab is equal to area under the curve L by 2,
so it is going to be (alpha into L into (delta Tbottom minus delta Ttop)) upon 4d; M1 is just the
opposite of this because this is the curvature to get positive curvature, so this is negative, so I
am going to get thetaab is negative. Thetaba is plus alpha L into (delta Tbottom minus delta Ttop)
upon 4d, because this is the same as that. Therefore, these are my thetaab and thetaba. How do
I get fixed end moments from these particular equations? We had to apply the moment which
would give me just the opposite of these thetaab so that the sum total of them gave me 0, so
that I could get my fixed end moments. Without much ado, I am going to go into computation
of the fixed end moments.
What are they? (Fixed end moment)ab is going to be equal to 4EI by L into ((minus alpha L
into (delta Tbottom minus delta Ttop)) upon 4d) plus 2EI by L into ((alpha L into (delta Tbottom
minus delta Ttop)) upon 4d) this is equal to minus 4EI alpha plus 2EI alpha (Refer Slide
Time: 14:18), this is going to be equal to minus (2EI alpha into (delta Tbottom minus delta
Ttop)) upon 4d. This will be equal to minus EI alpha delta Tb minus delta Tt upon 2d that is
my fixed end moment at ab.
Let me check if I am dimensionally consistent. What are the units of alpha? alpha into delta T
is going to be meter by meter, which is essentially dimensionless. What are the units of E?
Newton per meter squared, this is meter fourth, so Newton per meter squared into meter
fourth is going to be equal to Newton meter squared; Newton meter squared divided by d,
which is meter, is going to become Newton meter and that is the fixed end moments; the units
are consistent. This is the fixed end moment at ab (FEM)ab.
Similarly, you can actually get that the fixed end moment at ba is going to be equal to (EI
alpha into (delta Tbottom minus delta Ttop)) upon 2d. Therefore, if I have a uniform temperature
gradient with the bottom being hotter than the top, it expands in this way so that you get it to
be equal to the opposite. I would just like to make a point here: remember that thetaab is equal
to minus and thetaba is equal to plus.
Therefore, when we substitute in here, this has to be plus and this has to be minus, therefore
this becomes plus and this becomes plus. Remember that the rotation that we get is this plus
this rotation has to be equal to 0. Therefore, when this is minus, this (Refer Slide Time:
18:24) plus this is going to give you 0; then this plus this is going to give you 0. Therefore,
the fixed end moment here has to be positive.
Similarly, the fixed end moment at this has to be negative. If we show it for uniform
temperature gradient where delta Tb (the temperature differential at the bottom) is greater
than delta Tt (the temperature differential at the top) and it is uniform across the entire length,
then we get (FEM)ab and fixed end moment at ba equal to. I will just write it down, I have
got (FEM)ba over there. (FEM)ab is equal to this (Refer Slide Time: 19:35). Note that these
two fixed end moments that I have written down over here are essentially for the case where
the increase in temperature at the bottom is more than the increase of temperature at the top.
Mostly, it would be increase of temperature in the bottom and a decrease of temperature at
the top because you have to have neutral axis 0 for pure flexure.
Note also the fact that it is uniform over the entire length; in other words, the top fiber across
the length sees delta Ttop that is a negative temperature; delta Tbottom is uniform across the
length that is a positive temperature to get the directions that you have. Of course, if delta
Tb and delta Tt turn out to be different, you will get them to be minus. This gives you the
expression. Now, if there is a different variations of temperature across the lengths, then of
course you will not get these two, you will get some other expression, but the fact is that you
should, by now, be totally certain about how to evaluate this effect.
Once we have the fixed end moments, there is no difference between the member loads or
member temperature, excepting that once you have got the member end moments in
temperature, you do not have additional bending moment in the member due to the
temperature effect as you do for the load. In other words, the simply supported beam subject
to temperature has 0 moments, whereas a simply supported beam subjected to a load will
always have a bending moment diagram associated with it, which has to be superposed on; in
the temperature, you do not have the superposition. Otherwise, once you have calculated the
fixed end moments, you use the same principle that we have already; there is no difference
when you have to consider the temperature effect.
We are now done with considering the temperature effect and how to incorporate it into the
displacement method. I am now going to move on to look at the displacement method for a
truss. How do we apply the displacement method? The reason why I am looking at this is that
5
the displacement method is exactly the same excepting for the fact that in a flexural member,
the member end force deformation relationship essentially related the moments with the
rotations. In this particular case, the force deformation relationship in a truss, it only has axial
forces and axial deformations, so the member end force deformation relationship should
essentially relate the axial force and the deformation in the member. Let us look at how to
tackle that.
This is my truss, I am going to consider it to be fixed over here (Refer Slide Time: 23:43).
Just like I considered a simply supported which was the simplest statically determinate
flexural member, I am going to consider the simplest statically determinate axial member,
truss member. Here, I have load F and due to this, I have a displacement which is u. Since
this is member i, I will call this force i and member i. How do I determine this relationship? I
know that ui is equal to Fi Li upon (EA)i this is assuming that EA is a constant and only the
load is applied at the member end; I know this, I have already evaluated this earlier.
That means my force deformation relationship will be Fi is equal to (EA)i upon Li into ui
this is my member end force, this is my member end displacement and this is (Refer Slide
Time: 25:13). Look at the difference between the member end for flexure: in flexure, you
have two rotations and two moments to define the member end forces; for the truss, you only
have a single force and a single displacement; and this is the force deformation relationship
for a member. By and large, in trusses, you do not have forces acting in the center; you
normally always have member end moments and that is the reason why the force deformation
relationship is essentially this for a member; so I can write it as F is equal to EA by L into u.
What was it for a flexure? It was Mab is equal to 4EI upon L thetaab plus 2EI by L thetaba. Of
course, in a flexural member, you may have member loads and that is why you have the fixed
end moments etc. In this case, you normally do not have that and therefore you do not have
this aspect. Of course, you might have a temperature problem and therefore, this can be
written generally as F is equal to EA by L u plus fixed end force. Which is the fixed end
force? This is the force developed here due to a member effect we will see that later.
Let us look at this moment (Refer Slide Time: 27:20). EI by L is Newton meter, so the unit
here is Newton meter it is moment per unit rotation, so this unit is Newton meter per radian
and it is like a torsional coefficient, torsional stiffness constant. Here, EA by L is Newton per
meter something like a stiffness constant or spring constant, so this is similar to that, it is
force per unit displacement. Once we have this, how do we tackle a particular problem? Let
me take a specific problem here and see how to solve it. I will take a simple problem here
because you can actually take it for any kind of thing. I just want to reduce the number of
degrees of freedom and that is why I am considering this effect.
The reason behind why I have chosen this is in the displacement method, what do you start
off with? You start off with defining the degrees of freedom, so this is essentially all truss
members; although this does not look like a truss that you would have normally, the reason
why I am taking this is because it is simple; I will take up another thing also in which we will
see how it works.
This is a statically indeterminate truss. You cannot find out the forces in each of the members
given this load. How do you use the displacement method? Let us go through the steps. First,
determine the degrees of freedom. I have a, b, c and d. Remember that when you have a truss
member, you do not have the rotation, so you only have two displacements and so the
unconstrained degrees of freedom are 2 into 4, so 8. There are three constraints, each one has
two constraints, so constraints are equal to six, so there are two degrees of freedom (Refer
Slide Time: 30:03). What are those two degrees of freedom? I will define them as r1 and r2.
Note that I always define my displacements along the positive. This is my coordinate system
for the structure, so my degrees of freedom are positive in the positive direction; so r1 and r2
these are my two degrees of freedom. Two: Define the degrees of freedom r1 and r2; r1 is
the horizontal deflection of d and r2 is the vertical deflection of d; we have defined that. The
third step is defining the member end force deformation relationship.
How many members do I have? I have ad, I have bd and I have cd. For ad, what is the
member end deformation relationship? Fad is equal to EA upon L; L is equal to 5, so EA upon
5 into uad; then Fbd is equal to EA by 3 into ubd; Fcd is equal to EA by 5 ucd simple
relationships but we have written the member end force deformation relationships. I am
actually going through the displacement method as I have defined it. The next step is the
kinematics. What do I need to do? In terms of r1 and r2, I need to find out uab, ubd, ucd. What I
need to do is put.
Next is the kinematics. How did we do it? Put r1 is equal to 1 and r2 is equal to 0 and look at
the displaced shape. When I do this, what happens? This becomes 1. Note that since they are
hinged, they go like this (Refer Slide Time: 33:25). Now, I need to find out elongation. How
do I find out elongation? I drop a vertical along this direction. I will show this a little bit
better because it gets a little bit complicated, so let me draw it properly; otherwise, we will
8
have a problem. Since this point has come here and these points remain where they are, your
new lengths are these; I need to find out what the increase in length is. Let me take ad. It is
given in this form where this is 3 and this is 4; this point has gone here (Refer Slide Time:
34:40) and I am going to expand it by 1. This is my new shape and I need to find out the
length. To find out the length, what do I do? I drop a 90 degree here because in small
displacements if it goes perpendicular, then there is no change in length; so all I need to do is
now find out, if I have dropped a perpendicular here, how much this is and this is going to be
my uad (Refer Slide Time: 35:18). How would I find that out? Let us draw it.
If you look at this and look at this (Refer Slide Time: 35:50), I say this makes an angle theta,
where this is 90 degrees; since this is 90 degrees, since this is theta and this angle is 90 minus
theta, what happens is that since this is 90 degrees, this angle becomes theta; and since this
angle is theta, this angle is theta and this angle is 90 minus theta.
If you look at the change in length, the change in length is given by uad, which is given as r1
into cosine of theta; if r1 is equal to 1, this is 1 and this is cosine theta uad is cosine theta. In
this way, I can find out that whatever angle that it makes with the horizontal, the ui of that is
given by cosine theta. In this particular case, it is going to be that uad is equal to cosine of
theta, which is 4 by 5 positive 4 by 5. What about ubd? What is this theta? It is 90 degrees.
What is cosine of theta? 0. It makes sense, right? If it moves perpendicular to itself, the
change of length is 0. What about ucd? It is going to be equal to the angle made with the
angle. It is going to be shortening and that is minus 4 by 5 these are the displacements given
in terms of cosine theta. Similarly, you will see that corresponding to r2 in other words, the
kinematics for trusses is actually very simple; it does not get very complicated.
For trusses, you can actually write it down. In this particular case, this is my structure and I
give r2 is equal to 1 and r1is equal to 0. If you put that, what happens? This will go this way
and the displacement pattern will be this (Refer Slide Time: 39:53) and this; and all we need
to do is find out how much of shortening you have. The shortening or lengthening will
depend on again, perpendicular from this. I leave that up to you: you will see that uad is
going to be equal to minus r2 sine theta, where theta is this angle and this happens to be minus
9
because it is shortening. Same thing for all of them; in this particular case, you will see uad
will be equal to minus 3 by 5 and here, when you do sine theta for the vertical member, sine
90 is 1, you will see that ubd is equal to minus 1 (that is true because here, whatever is the
displacement is directly the shortening of the vertical member); ucd is going to be equal to
minus 3 by 5. We are done with the kinematics now. If you plug this into the kinematics and
into the equations, what do you get?
For ad, my Fad becomes equal to EA by 5 and we have 4 by 5 r1 minus 3 by 5 r2 all I have
done is I have substituted uad in terms of r1 and r2; I have got the influence coefficients for r1
and r2 from the kinematics and that is all I do. Similarly, for bd, you have Fbd equal to EA by
3 into (0 r1 minus 1 r2) and for cd, Fcd is equal to EA by 5 into minus (4 by 5 r1 minus 3 by 5
r2). These three expressions give me the forces in the members ad, bd and cd in terms of the
displacements corresponding to the degrees of freedom. Once I have those, what is the next
step? I am following exactly the same steps as I did for the flexure; only thing that I have in
this particular case is that I have instead of flexural equations, the truss equations.
10
The next step is virtual work. Since I have two degrees of freedom, I have to write two
independent equations. My first equation is my virtual displacement pattern: r1 is equal to 1,
r2 is equal to 0 and I find out the external work done. What is the external work done? Let us
look at what the force was. What were the forces? There was a horizontal force of 10 and a
vertical downward force of 10, so external virtual work is going to be 10 into 1 plus 10 into 0,
because r2 is equal to 0; so this is going to be equal to 10. What about the internal virtual
work? The internal virtual work is going to be Fad into uad plus Fbd into ubd plus Fcd into ucd
how are these are related to r1? I already have the influence coefficients, so this is going to
give me Fad into plus 4 by 5 plus Fbd into 0 plus Fcd into minus 4 by 5. What I get is the
following. I am going to substitute Fad, Fbd and Fcd in.
I am going to get (EA by 5 into (4 by 5 r1 minus 3 by 5 r2) multiplied by 4 by 5) plus (EA by

5 into (minus 4 by 5 r1 minus 3 by 5 r2) into minus 4 by 5). Taking 4 into 4, 5 into 5 into 5,
11
this is going to be 16 125 upon EA into r1 minus 12EA upon 125 r2 plus 16EA upon 125 r1
and this one (Refer Slide Time: 47:33), minus, minus gets plus, it becomes plus 12EA by 125
r2, which is equal to this cancels this (Refer Slide Time: 47:50), this becomes 32 upon
125EA r1 is equal to 10. In this particular case, we actually get an uncoupling; normally, you
would not get an uncoupling. r1 is equal to this.
The other equation is going to give me. The virtual displacement pattern is r2 is equal to 1
and r1 is equal to 0 independent. In this particular case, the external work done is minus 10
into 1, so this is going to be minus 10. Internal virtual work is going to be equal to Fad
multiplied by r2, which is minus 3 by 5, plus Fbd multiplied by minus 1 plus Fcd multiplied by
3. We will see that the positive and the negative will cancel out; what we will be left with is 9
EA by 125 r2 plus EA by 3 plus EA by 3 r2 plus 9EA by 125 r2 is equal to minus 10 and so
this is going to give me r2.
What is r2 going to be equal to? Let us take stock of that. This is going to be equal to minus
10 multiplied by 375, this is going to be 125 into 3, 375, so this is going to become 27 plus
27, 54; 54 plus 125 is equal to 179EA, so 179EQ. This is my r2 (Refer Slide Time: 50:58).
This r2 is negative. Why? Because r2 is taken to be positive upwards and so all that means is
r1 is positive, it is to the right and r2 is negative, which essentially means that r2 is downwards
this is how it should be because the way we have defined r1 and r2 is positive to the right: r1
is positive and r2 is upwards positive, so obviously under the loading, you will see that r2 will
go down and r1 will go up. Now, how do I find out these values?
12
Once you know r1 and r2, you can substitute and you can get Fad is equal to EA by 5 into (4
by 5 (which is r1) into 1250 upon 32EA) plus (r2 is minus, so this becomes plus) (3 by 5 into
10 into 375 upon 179EA). EA, EA cancels, EA cancels here; you get 5, 250, 50, 25 by 4, this
becomes 25 by 4 plus goes 2, so this is 6 into 375, this is 1125 into 2 is 2250, 2250 upon 179.
This is my Fad and in exactly the same way, I can find out Fba and Fcd. Once I know these, I
know my member forces and I have analyzed the structure I have found out the
displacements and I have found out the member forces.
In trusses, it is relatively easier because the member force deformation relationship is simple,
the kinematics is very simple the kinematics essentially depends on sine theta, cosine theta,
so kinematics is simple. Therefore, actually, the displacement method is very simple for
trusses but the only thing is that for regular trusses, the number of degrees of freedom is so
large that you cannot do a hand computation that is the reason why I have taken a simple
example with axially loaded members to illustrate the concept. You will have to use
computers to be able to solve for larger trusses.
However, the procedure still remains the same: determine the number of degrees of freedom,
define the degrees of freedom, define the member force deformation relationship, do the
kinematics, do the virtual work and then solve for displacements, incorporate the
displacements and get the member end forces, member forces that is all there is to the
displacement method. Thank you very much. I hope at the end of this lecture, series of
lectures, you are now comfortable with now applying the force method and the displacement
method for obtaining forces and displacements in both plane trusses and plane frames.
13
Prof. P. Banerjee
Lecture 21
Good morning. In the last few lectures, we have been looking at the displacement method for
analysing structures. Today, what we will start off with is actually looking at a method that is
much older than the displacement method that I talked to you about this is called the moment
distribution method. The problem with the displacement method is that for it to be usable by
hand, you cannot really go beyond two or three degrees of freedom because to solve by hand
very large simultaneous equations is reasonably difficult.
Nowadays, of course, what do we do? We use the computer to solve the problem. In fact, after I
have gone through this part, I am going to go into the matrix methods, which are essentially
formulations that are very computer-friendly. I would also like you to understand the moment
distribution method because this is a method by which you can actually analyse very large
structures with lots of degrees of freedom without recourse to a computer.
In fact, till the advent of computers in the '90s, moment distribution was the method that was
used extensively by designers to actually analyse structures, to find out the forces for which they
have to design. Of course, the relevance of the moment distribution method today has gone
down. However, it is still is a very very quick method by which you can check by hand as to
whether your computer simulations are giving you reasonably accurate results. That is the reason
why in the next few lectures, I am going to talk about the moment distribution method and solve
examples that illustrate how the moment distribution method is used for analysing beams and
frames; the moment distribution method is essentially a method used for beams and frames. Let
us have a look at the moment distribution method.
It is a class of iterative methods that do not actually develop all the equilibrium equations. In
fact, that is the advantage of this approach: since you do not develop all the equations, you do not
have to work with a very large number of equations at one time; let us see how that works. For
that, I am going to actually take a problem that I have already discussed earlier. This is the beam
problem. Do you remember?
There are two 10-meter spans fixed at a, this is b (Refer Slide Time: 04:52), this is c each 10
meters long. This one has a 120 Kilonewton at 4 meters and over bc, you have a 50 Kilonewton
meter uniformly distributed load. There are two equations: one is summation moments at b equal
to 0 and summation moments c equal to 0 these two equations. We have already got these two
equations and I am not going into the details of how we have obtained that; we have already
obtained it; please look back and you will see that these are the equations.
Since both are 10 meters, I am going to define I upon l, which is I upon 10 to be K. There is a
reason why I am doing this it will soon become obvious to you. If you do this, then these
equations can be written in this fashion (Refer Slide Time: 05:45). The first equation I am going
to write is thetab in terms of the moment and thetac and the second equation I am going to write
down is thetac in terms of thetab.
When you do this, this becomes EK thetab is equal to 37.7 minus 0.25 EK thetac and EK thetac is
minus 104.2 minus 0.5 EK thetab these two equations I am going to solve iteratively. How am I
going to solve iteratively? Initially, I will assume that thetac is equal to 0, I will assume that EK
thetac is 0; if EK thetac is 0, that means EK thetab will be 37.7; this value EK thetab I am going to
substitute here (Refer Slide Time: 06:53) and find out the thetac value that is the first cycle; the
next cycle I have found this; plug that in here, get this, so you see how iteratively I am looking at
it.
This is how the iterative solution looks. Step one: assume value of thetac is 0 and from that, you
can find out it is 37.7. In the second step, you take 37.7, find out EK. In the third step, use this
(Refer Slide Time: 07:25), put it into the first equation, get this. In the next step, use this, get
this. Do you understand the way this is being solved? The way this is being solved is that in my
first step, I am assuming thetac to be 0; then, I can find out thetab; once I find out thetab, the
second step is that I plug in that value of EK thetab here to get EK thetac; once I get that, the third
step is that I plug this in here (Refer Slide Time: 07:53) and keep doing this. The problem with
this is that you have to understand that this iteration has to converge; if it does not converge, then
it is not useful. It can be shown mathematically that this converges and I will show you the
details of this method here.
If you look at this, EK thetac, EK thetab, put the value of thetab, you get thetac keep putting that
in; first step: 0, then you evaluate thetab, then you put that into thetac, get thetac; then, you put
that EK thetac assumed value and solve for EK thetab; then you use that value of EK thetab and
solve for EK thetac; continue that way and see what happens; assume a value of EK thetac, get
EK thetab; then you assume that value of EK thetab, you get the value of EK thetac; you assume
that value of EK thetac, you get back the original value. What does that mean? That means that
the two equations have given you the same solution they have converged. Therefore,
ultimately, the value of EK thetac is this and EK thetab is this.
This is the standard iteration procedure known as GaussSeidel iterative procedure. In the
GaussSeidel iterative procedure, it can be shown that this will always converge. Therefore, let
us go back and look at what this implies by looking at what effect this has. Of course, once you
have got this, you can plug in these values into Mab, Mba, etc., and get it. However, moment
distribution is not the GaussSeidel iteration procedure.
Let us look at what this implies in terms of the moments themselves. Note EK thetab and EK
thetac we evaluate not because we want to find out what the values of thetab and thetac are;
understand that very rarely are we interested in what the rotations are we are interested in
calculating the rotations, so that we can calculate the moments. Now, let us see what these
moments actually look like.
The equations come out this way this is for the same equation. Mab is equal to 2 into EK thetab
.... I am just rewriting what we had already written, excepting that this K is I upon 10. Mba is
equal to 4 (EK thetab) plus 115.2; Mbc is equal to 4... I am sorry, this is this (Refer Slide Time:
11:41) and this is negative; this is 4 (EK thetab) plus 2 (EK thetac) plus 416.7; Mcb is equal to 4
EK thetac plus 2 (EK thetab) minus 416.7. These are of course positive; I have taken
anticlockwise as positive. These were the expressions for Mab. Then, you had the two
equilibrium equations: one which said that Mba plus Mbc is equal to 0 and the other one which
said Mcb is equal to 0 these were the equations we had written down earlier and I am just going
through that. Now let us see what actually happens. You know Mba plus Mbc is what gives me
this equation and Mcb is equal to 0 is what gives me this equation (Refer Slide Time: 13:03) you
can look at it.
(Refer Slide Time: 13: 09)
This one here is actually plus and minus. This is minus and this is plus, plus, and minus. Now, let
us see what these mean. I am going to write down this iterative solution that I have written now,
excepting that I am going to write down the values of Mab and Mba.
If we look at the moments, I am going to draw again. First one is step number, then I am going to
write down EK thetab, EK thetac; for example in the initial step, EK thetac is taken to be 0 and
EK thetab is also taken to be 0 right in the beginning; that is the first step, then, we plug in EK
thetac. Let us see what happens to Mab, Mba, Mbc and Mcb. When it is 0, 0, Mab is going to land up
being 172.8, this is going to be minus115.3, this is going to be plus 416.7 and this is going to be
minus 416.7 these are the moments when I take it. Then, then we have the first step. In the first
step, what did we say? We took this equal to 0 and therefore, this became minus 37.7; if this
became minus 37.7 and if we plug into all of them... I am putting thetab equal to minus 37.7 and
thetac is equal to 0, so I can compute; when I compute these, what I get is this becomes 97.4, this
becomes minus 266, this becomes plus 265.9 and this becomes minus 492.1.
I am going to now look at this (Refer Slide Time: 16:17). Look at this: when I substitute and I
get the value of thetab, what I am actually doing actually is releasing. This is both the joint at b
and the joint at c clamped; when they are clamped, these are the fixed end moments; then what
am I doing? I am actually keeping thetac clamped and releasing the clamp at b, so that thetab can
occur, so that when I release the clamp at b, what happens? Mba plus Mbc has to be equal to 0, for
all practical purposes, this is equal to 0 that is what I have done; I have actually released it and
made it equal to 0. However, Mcb is not equal to 0; so next, what do I do?
In the second step, I clamp b (Refer Slide Time: 17:30) at the rotated position and release thetac.
When I release thetac , what do I get thetac to be equal to? It becomes plus123.1. If I plug this in
and compute, this remains the same because this is not affected by thetac, this remains the same
because this is not affected by thetac, but this changes because it is affected and this becomes
plus 512.1; this becomes plus0.0. What have I done? I have actually kept thetab and released
thetac; as soon as I release thetac, what happens to Mcb? It has to be equal to 0 that is what you
get; so you see how....
Now, what is the third step? I keep thetac clamped at that position and further release thetab.
When I release thetab I get that equal to minus 68.5. Note that this is exactly the number that I
had over here, excepting that I just made one mistake: I took clockwise as positive, so this is...
bear with me, this is anticlockwise, my anticlockwise is positive. In the first step, this (Refer
Slide Time: 19:27); in the second step was this; third step is 123.1 minus 68.5 that is what I
have over here. I have clamped it here, released it here; when I released it here, what does my
value become? It becomes 35.8 and this now becomes minus 389.0 and this is plus 389.0;
however this becomes plus 61.3.
What we have done is as soon as we release thetab.... Again, this balances out. Now, I am going
to just write down the all the subsequent ones which we have computed take this as this and
compute this; so this becomes plus 138.5. I am going to write down the ones that we have
already got: plusplus 138.5 and you get minus 72.3.
Then, the sixth step is take minus 72.3 and you get plus 140.4. The seventh step is, take 140.4
and you get minus 72.8. The eighth step is, take 72.8 and you get 140.6. The ninth step is
identical because you get the same value. This is the converged value and for each one of those,
you substitute thetab and thetac. I will just write down the values that you get: 35.8, 28.2, 28.2,
27.2, 27.2, this one you get as minus 389.0; then you get plus 404.4, then you get plus404.4, then
you get plus 406.4, plus 406.4 ; here you get minus 419.7, here you get plus404.5, here you get
plus 408.3, here you get plus 406.3, here you get plus 406.7; here minus 61.3, plus 0.1 you are
releasing this; then the next one you are releasing this, so you get minus 7.3; here, you get plus
0.1 this is the one that you are taking to 0; then, you get minus 0.7; this is where you release
thetab; so, alternately you are releasing thetab, thetac; finally, you have plus0.1, which is the
release (this should actually go to 0 but we have a least square).
Now, this is the solution that you have. We can say what the procedure is: you start with both b
and c clamped, so you clamp b and c. What are your moments at ab, ba, bc, and cb? They are the
fixed end moments and you know what those are. The first step is that you keep thetac clamped
that means you get thetac is 0 and release the clamp at b. Immediately, as soon as you release the
clamp at b, what happens? The net unbalanced moment here gets distributed on both sides. Here
(Refer Slide Time: 25:14), what you have is balance of moments but that leads to an unbalanced
moment here. When you release this, you get EK thetac as this, then immediately you have this
going to 0 because you have released the clamp at c the moment at c has to go to 0 but that
creates an unbalanced here. Now, you have to distribute those moments; when you distribute
those moments, it creates an unbalanced again here.
Understand the point: what you are doing essentially is clamping, releasing, clamping, releasing,
clamping, releasing; the whole procedure that you have over here is essentially a whole set of
clamping and releasing physically clamped, release thetab, clamp thetac, clamp thetab at the
rotated position, release thetac, clamp thetac at the released position rotated position, release
thetab; keep doing that till you have a situation where even when you release, you essentially get
the same thing, which basically means that you have balanced forces here (Refer Slide Time:
26:34) and you have balance here. Therefore, these are no longer clamped because you have
released; and these are the final member end bending moments.
What are you doing? You are distributing moments, so this in essence is the background behind
the moment distribution method. Although we looked at it, it was a GaussSeidel.... This is the
GaussSeidel iterative procedure. When we look at it in terms of Ms, what do we get? We see
that it is essentially distributing unbalanced moments at all times and ultimately getting to a
situation where there is no subsequent distribution, because all moments are balanced where they
are supposed to be balanced. Note that here you have an unbalanced because it is a fixed end
(Refer Slide Time: 27:35) a fixed end can take a moment, whereas you cannot have an
unbalanced moment at b and you cannot have an unbalanced moment at c. With this background,
let us try to see what the method actually is; let us go through these steps. I will just finish off
putting this together and we will put down what the moment distribution method actually looks
like. It actually is a moment distribution. We are not going to do the GaussSeidel; we have just
introduced the GaussSeidel iterative procedure to show you physically what the moment
distribution method does. Let us look at this situation.
You have a situation where when you clamp it, you have Mba this is the not the final moment.
At a particular step, you look at Mba plus Mbc and you get a net moment this is the unbalanced
moment. When you add, this should be 0, so this is the net unbalanced moment. When we
release this, what do we get? We get EK thetab is equal to minus Mb because when we add this
(Refer Slide Time: 29:24) plus this, this is the procedure; I am writing down what the procedure
is. What does balancing this moment do then? What it does is the new Mba turns out to be equal
to EK thetab. I am sorry I am making a mistake here, this will be some n time EK thetab because
that is the net unbalanced moment that we have. This one is n1 EK thetab plus Mba; Mbc is equal
to n2 EK thetab plus Mbc these are the new.
What are n1 and n2? They are the numbers and you will see here that since we are having Mba
plus Mbc, n1 plus n2 is equal to n because it is this plus this (Refer Slide Time: 31:23). If you now
look at this, what do we get? This plus this is equal to this plus this. If you look at this, n1 plus n2
is equal to n but this is equal to minus and this plus this is equal to Mb, so what you get is Mba
plus Mbc is equal to 0, balanced. We have released thetab to get a balance.
Therefore, what is the distribution factor? What is this n1? If you look back, you will see that n1
is equal to 4. What is n2? You will see that n2 also is equal to 4 and so n1 plus n2 is equal to 8. In
fact, we know n is equal to 8. We get to a kind of a situation; although this is the procedure, what
we are really doing is we are not doing this procedure; what we are actually doing is taking the
unbalanced moment; so we have a situation where we have the unbalanced moment, so we have
an unbalanced moment.
We take minus Mb and distribute this moment to ba and bc. How do we distribute this? This will
get distributed exactly. Let us see what we get. EK thetab is equal to minus Mb upon n. How
much are we adding? We are adding n1 EK thetab to ba. What is that equal to? That is equal to
minus n1 upon n into Mb. To b, we actually apply n1 upon n minus Mb and here we apply n2 upon
n minus Mb. Minus comes here actually (Refer Slide Time: 34:26). If we look at these, what are
these? These are the independent factors that you have. So we can now say that Mba .... This is
the distribution factor. The distribution factor is this way. We will call it the distribution factor to
i is equal to ni upon summation ni of all the members coming at the joints. What is ni? If you look
at it, you will see that since both n1 and n2 are 4, this essentially turns out to be equal to Ki upon
summation Ki, where Ki is equal to I upon L of the ith member.
Here I will call this (Refer Slide Time: 36:22) as my Kba and this I will call as my Kbc. Therefore,
(distribution)ba is equal to Kba (remember K is equal to I upon L remember that) upon (Kba plus
Kbc). The distribution factor for bc is equal to Kbc upon (Kba plus Kbc). The first step is finding
out the distribution factors required to distribute the unbalanced moment to the individuals. Now,
there is another step. This step is essentially releasing thetab but understand one thing: as soon as
I release thetab, let us see what happens to Mba.
Mba is equal to 4EK thetab plus (FEM)ba. What is Mab? When we release thetab, what do we get?
This term comes in. As soon as we release thetab, what happens? This is the unbalanced moment
we have distributed to ba, but what is the additional moment at Mab? Mab is equal to 2EK upon
thetab. When we do the balancing, what do we get? If we look at this, this additional moment that
we apply to Mba due to the unbalanced is equal to 4EK thetab, but delta Mab is equal to 2EK
thetab. Therefore, delta thetab is the net this thing. If you look at it, delta Mab upon delta Mba is
equal to 2EK delta thetab upon 4EK delta thetab which is equal to half. So in addition, as soon as
you have the distribution of the moments you also have to carry over this distributed moment to
get the... and what is the carryover factor? It is half.
Now, this carryover (Refer Slide Time: 39:25) is half whether you apply at ab or ba, because you
will always see that if you have Mab is equal to 4EK thetaab plus 2EK thetaba, Mba is equal to 2
EK thetaab plus 4EK thetaba. So, if you look at this, this way (Refer Slide Time: 40:00) if you
release thetaab, you need to carry over to Mba, which is half, and when you release thetaba, you
need to distribute it, so half is the carryover from any clamped end when we release the clamped
end. That is known as the carryover factor and for a uniform beam, this happens to be equal to
half.
Note that if you transfer a moment to a fixed end, remember that you do not release a fixed end,
so there is no carryover from the fixed end to the other end. Only when you release do you
generate this carryover. Now having put this in, let us see how we solve this particular problem,
the problem that I had defined how we are going to solve this particular problem using the
moment distribution method.
Let us now go back and I will show you that exactly the same steps are being followed,
excepting that explicitly nowhere do we actually compute. This is the structure and now I know
what the fixed end moments at ab is. The (FEM)ab is equal to plus 172.8 Kilonewton meter,
which basically means... this is positive (Refer Slide Time: 42:20), so this is this way; then, the
(FEM)ba is equal to minus 115.2 Kilonewton meter. The first thing that we do is compute the
fixed end moments. The (FEM)bc is equal to plus 416 Kilonewton meters and (FEM)cb is equal to
minus 416.7 Kilonewton meter. I calculate first off the fixed end moments for each member ab
and bc. This is for member ab and this is for member bc, so I have calculated the fixed end
moments for the members. What is the next step? The next step is to compute the distribution
factors and the carryover factors.
We have member b and member c, so the distribution factor for ba at b (this is at b) is going to
be equal to (I by 10) upon (I by 10 plus I by 10), which is for bc, so it becomes equal to half. Dbc
is equal to (I by 10) upon (I by 10 plus I by 10), which is equal to half. Now, the carryover factor
from b to a is equal to half carryover factor from a to b is 0. Why? Because a is a fixed end. The
carryover factor from bc is half, the carryover factor from c to b is half. At c, (distribution
factor)cb is 1 because there is no other member at that point. We have calculated the distribution
factors and now, I am going to put this all in a table.
I have ab, ba this is member ab; and now for member bc, I have bc and cb. Here, there is no
distribution, so there is no distribution. Here, we will just calculate the distribution factor: half,
half; here, the distribution factor is 1.0; this way, the carryover factor is half; this way, the
carryover factor is 0; this way, the carryover factor is half; this way, the carryover factor is half.
Now let us start the procedure. The first step is writing down the fixed end moments. This is plus
172.8, then we have minus 115.2, plus 416.7, minus 416.7. The first step is calculating the
unbalanced moment here. What was the unbalanced moment over here? 301.5. That has to be
distributed.
How do we distribute it? We distribute it half, half. Note it is plus, unbalanced moment is
plus301.5, so the distributed has to be minus 301.5, so it is minus 301.5, which has been
distributed. If you look at it, what is the distribution? Half, half will become... 301.5 will become
150.8, so I will put 150.8 here and I will put minus 150.8 here (Refer Slide Time: 47:33). What
does this mean? This means that I have completed the distribution. If you add these up, what do
you get? Both sides negative. What do you get? Add them up and you will get 266.0 and this
side, you will get 265.9. Do you see something? I will have to carry over these because as soon
as I have release this (Refer Slide Time: 48:16), I have to carry over here; the carryover is half,
so what do I get? It is minus 75.4. Here, I have to carry over half, so I get minus 75.4. Add these
up. You will get.... Let us go back to the old one that I had shown. I am not adding them up, but I
am just showing you: if you look at this 97.4, add these two up, you get 97.4; add these up, you
will get minus 266; add these two up, you will get 265.9; add these two up, you will get 492.1.
Note that here, we are not talking about releasing all that we are doing is distributing moments.
Now what is the next step? The next step is like this has to be equal to 0, so I have to add 492.1.
Now if you add up, you will see that it adds up to 0 but for that, as soon as they release this
(Refer Slide Time: 49:44), this has to carry over here and so this will become plus 246.0. Now,
this is an unbalanced. Here, there is no question of unbalanced because it is a fixed end when
you have a fixed end, you do not require any release. You cannot have an unbalanced moment
here and an unbalanced moment here. As soon as you have released this, you have got this.
Now, you need to distribute this, so distribute it half, half. What do you get? You get minus
123.0, minus 123.0. That balances it out but this causes this to go here and this to go here (Refer
Slide Time: 50:48); this becomes minus 61.5, this becomes minus 61.5; now because of this, you
need to do plus 61.5 here; this becomes plus30.8; now, you need to balance this, you get minus
15.4, minus 15.4; this is balanced but this is going to lead to again carryover here, minus 7.7,
minus 7.7; this gives an unbalanced here, plus 7.7; this now carries over here, plus 3.8; when you
get unbalanced here, you require minus 1.9, minus 1.9; again balance, minus 0.9 and here, you
get minus 0.9.
This way, you continue till you get 0, 0 and then what do you get as a final value? The final
value you get as which we got here; and at every intermediate step if you add them up, you will
see that you will get this, so actually the moment distribution method mimics this; the only thing
is that in the moment distribution method, all we are doing is computing unbalanced moments,
distributing it, carrying it over, distributing it, carrying it over, distributing it, carrying it over.
The physics behind the moment distribution method actually is... you are calculating... what you
are doing is initially clamped-clamped. For this particular problem, there are two joints which are
clamped for which you calculate the fixed end moments; then you release thetab so that you get
no net moment at b, and so Mba plus Mbc is equal to 0. Next, what you do? Release thetac. When
you release thetac and you hold thetab here, what do you get? You land up getting an unbalanced
moment at this point.
Now you release it further and then you release this, release this, release this till you get a
particular point where now you do not have any unbalanced moments and that is the deflected
shape. Note that nowhere in the moment distribution are we looking at the deflected shape; we
are not interested in thetab and thetac we are only interested in what the moments are at the
ends, that is what we are interested in; the whole moment distribution process it does this. Note
that this will always converge; physically you can think about it; what you are doing is you are
starting here this (Refer Slide Time: 54:19), then you do this this is going to cause less, so it
is going to need this, you are going do this.
Ultimately, you are going to get a situation where you converge, so we can mathematically show
that the GaussSeidel iterative procedure does converge and because the GaussSeidel iterative
procedure converges, the moment procedure also converges. Now, never in moment distribution
do you go till you get 0.0 and exact. What you do typically is, you stop after three or four cycles
or till your unbalanced moments get less than one percent of the moments at the end and then
you stop over there; once you know the member end moments, you solve the problem.
The advantage of this method is that it is a completely algorithmic procedure based on
distribution factors and carryover factors being computed and then you keep computing
unbalanced moments and constantly computing the net unbalanced moments distribute it, carry
it over, again look at net unbalanced moments, distribute it, carry it over. In this particular case,
we are doing it; we are distributing b (Refer Slide Time: 55:41), then we are distributing c, then
we are going to b again and distributing b, c, b, c, b, c this is the physical procedure.
We will see in the next lecture that this becomes a very combustion procedure. Ultimately, what
we have to see is the unbalanced moment. We look at all the row joints, look at all the
unbalanced moments, distribute them, then carry over, then continue. In other words, what you
have is you have cycles: one cycle where you distribute, then you carry over; then in the next
cycle again, you distribute and then you carry over; then you distribute, carry over. Which is the
last one? The last one is where you distribute so that you get balanced you do not need to carry
over because moments are small enough so that carrying over will not have tremendous effect.
This is the procedure that we will look at. We will continue looking at this particular problem
and we will also see that the modifications.... Remember in the displacement method, we had the
fixed-fixed beam and then, we had the fixed-hinged modified member. Similarly, even in
moment distribution, you can actually have moment distribution for a modified member and we
will see all of these in the next lecture.
Prof. P. Banerjee
Lecture 22
Good morning. In the last lecture, I introduced you to the concept of moment distribution and the
physical reasoning behind how moment distribution works in terms of relatively removing the
clamp at the joints that is how the moments get distributed. Ultimately, we see that it will
always converge because as you release the rotations at the joints at which your moment
equilibrium has to be maintained, progressively you are going to keep releasing it till you get
equilibrium. The entire moment distribution procedure is an iterative procedure that ultimately
gives you equilibrium at all the joints satisfies all the equilibrium equations. Let us look at this.
Note that what we are normally considering here is the moment equilibrium in fact, we are
currently only looking at moment equilibrium. Let us now look at certain other aspects today.
Remember I talked in the last lecture about how we account for the modified member that we
had. Let us look at this.
Let us look at this particular problem. It is the same problem, excepting that this time I am going
to say that I will consider ab and my clamped clamp, but I am going consider bc as the modified
member. If I consider it as the modified member, what happens? For ab, bc, here we have this
(Refer Slide Time: 04:32) and here we have this. First and foremost, I can calculate the fixed end
moments. The (FEM)ab is equal to plus 172.8, (FEM)ba is equal to minus 115.2; now I am going
to calculate the fixed end moments at bc and cb as if they did not exist as if this is clamped I
am going to compute it that way (it is not, but I am going to compute it). What was the modified
fixed end moment here. The modified fixed end moment remember that we had computed this?
Now my point over here is this: is this not the same as my balancing the force here and then
carrying over half, which is basically plus 208.3?
1
This is almost exactly like and once I release this, that is it this has gone to 0 and it will never
come back again because this is a roller. We can do this even in fixed end moments in which we
compute the fixed end moment as if this was fixed, then do the balance over here, put it to 0,
transfer it over here and then solve the problem, but let us look at what the other things are.
These are the fixed end moments. What is Mab? It is equal to 4 EK (I by L I call it K) thetaab
and since this is the fixed end, this goes to 0 plus 2 EK thetaba (which is of course there) plus
the fixed end moment; this one has . Mba is 2 EK thetaab plus 2 EK thetaba that exists. These
are the moment deformation relationships for ab. For bc, what is it? It is given as Mbc is equal to
3 EK (I by L is K). This is Kab, Kab, Kab, Kab; this is EI by L for bc 3 EKbc into thetabc; that is it,
there is nothing else. Here, Mcb is equal to 0. Can I write this in this fashion? I will write this as
Mbc is equal to three-fourth of 4 EKbc thetabc. What is Mba plus Mbc?
Mba plus Mbc is going to be equal to the moment distribution. Mba is equal to 4 EKab upon (4
EKab plus (3 by 4 into 4 EKbc)). Remember that the 4, 4 cancelled and E, E cancelled and that is
how we got Kab upon Kab plus Kbc in the other case, but here this becomes plus 3 by 4 Kbc.
Therefore, for a modified member, we can define a Kb modified, which is equal to 3 by 4 Kbc. If
we do that for the modified member, then Mba is equal to Kab upon (Kab plus Kbc modified) and
Mbc is equal to Kbc modified by (Kab plus Kbc modified). The only thing that happens is when you
have a fixed-hinged, the modified K becomes 3 by 4 into Kbc. Is that clear? What about the
carryover factor? The carryover factor Mab is still equal to half of Mba, the carryover factor from
Mab to Mba is still half; this is fixed, therefore this does not go.
However, what is Mcb in terms of Mbc? It is 0 because Mcb is 0. When you have a modified
member, the carryover to the other end is 0 there is no carryover because moment cannot be
carried over. With these modifications, let us see how I can solve this particular problem that we
have. In other words, for a modified member, as far as computation of fixed end moments will
come, we will see. Let us not try to make a differential because we will see that this is actually
incorporated in the modified, into this thing. As far the fixed end moments are concerned, there
2
is no difference from the original (fixed-fixed); however, for the modified member, the stiffness
is modified to 3 by 4; the distribution factors this is into Mb, Mb, the distribution factors are
equal to this (Refer Slide Time: 12:44). If you modify it, then we can continue using the same;
the only other thing is that in a modified member, there is no carryover. With this new concept in
mind, let us see how we can solve this particular problem, the same problem that we have.
Let us go through the steps. We have all ready calculated the fixed end moments. Again, ab, ba
this is member ab, this is bc, cb. Let us see what the distribution factors turn out to be equal to.
Let me just do the distribution factors at this point. What is Kab? That is equal to I by 10. What is
Kba? It is equal to I by 10. What is Kbc? Three-fourths into I by 10. What is Kcb equal to? 0.
Therefore, the (distribution factor)ba is equal to 1 over (1 plus 3 by 4), which is equal to 4 by 7
and Dbc is equal to 3 by 4 upon (1 plus 3 by 4), which is equal to 3 by 7. The distribution
factors since I am considering bc as modified, this is 4 over 7, this is 3 over 7, and this one
no distribution because it is a fixed end and here, distribution is 1. There is no carryover in this
direction; however, I am going to say that only for the release of the fixed end moment is there a
half carryover, here there is a half carryover, there is no carryover in this direction because the
fixed end is not to be released. Let us put down the values: minus 115.2, plus 416.7, minus
416.7.
The first thing we do is put plus 416.7, put this equal to 0, carry over here (Refer Slide Time:
16:56), this becomes plus 208.4 and this one is a first-off case so this has gone to 0; this is not to
be considered again because this is a modified beam the fixed end moment alone is released.
Understand that this is identical to doing (FEM)bc minus (FEM)cb by 2 we could have done
that, we would have got the same thing. Once we have done that, this is the unbalanced. What is
the unbalanced? Plus 625.1 minus 115.2 this is the unbalanced, so this needs to be balanced
now.
Balancing will be done as 4 upon 7 and 3 over 7, this is positive, so this is negative, so negative
291.2 minus 218.7 that is a net unbalanced; once I have done that, that is balanced, then I have
a carryover here (Refer Slide Time: 18:21) the carryover becomes minus 145.6. Now, let us
look at it. This is the first level of carryover. Any unbalanced left here? No, this is 0. Here, we
balanced it and we have carried it over there is nothing else; we have ended it, we have
finished it off; and so this one turns out to be equal to plus 27.2, this turns out to be equal to
minus 406.4, this turns out to be plus 406.4, this turns out to be 0 this is what we get; this is
only one iteration. Why is there only one iteration? Since this is released, this is the only one
clamped; as soon as we release it, we have got balance here and that is it, there is nothing else,
thetacb can go any way it decides to that is the reason why this is spectacular. We had nine
steps to get to convergence and by using the modified beam we just did it in one iteration and
reached the solution. It is never going to be this spectacular, but this illustrates how using the
modified element actually helps in speeding up the moment distribution convergence process.
Therefore, to review, I just want to say that for a modified member, you take. For a regular
member, K is equal to I upon L; for a modified member, K is three-fourths I by L only at the
continuous end. Is that clear? You can only apply the modified if the other end is a support at
the support, you never have another member, so that does not come into the picture; even if you
have, note that here you have a hinge support at b also but since there is a continuity, you cannot
take the moment. You know that, you have already seen that and I do not want to belabor that
point again and again. All you have to do is for the modified member at the continuous end, you
just make the stiffness 3 by 4 and you use the modified stiffness in the distribution factor
computation. Secondly, there is no carryover in a modified member ever. This in essence is the
application of the modification or the modified element in the moment distribution method. Let
us look at a kind of a situation. Again, I am going to take the same problem in the earlier case
where we had this situation.
This support went down by 0.03 meters 3 centimeters and we had to find out the moments at
this point. Here, what are the fixed end moments? Let us see. Due to this 0.03, we are going to
have the (FEM)ab is going to be equal to 6 EI upon L square. I am solving the same problem. We
were given E is equal to 200 Newton per meter squared and I was given as 2000 into 10 to the
power of minus 6 meter fourth. Substituting these in, we got that the fixed end moment here was
plus 720 Kilonewton meter. Similarly, (FEM)ba was 720 Kilonewton meter. At the other end,
what did we have? At the other end, I am going to consider bc the same, so it is going to be just
the opposite, so you are going to have minus720 Kilonewton meter and this is equal to the
(FEM)cb. Now, I am going to use the modified members.
All that happens due to support settlement is that I use the same one that I have just used; I am
not going to go through the same points, I am just going to write them down. This is 4 by 7, 3 by
7, this is 1, this way 0, only to release the fixed end moment we have half this way, this way we
have half, this is the fixed end, this way 0 because it is a fixed end (we do not carry over from
fixed end) and this is plus 720, this is plus 720, this is minus 720, minus 720 these are the fixed
moments. First, balance out at cb, put it equal to 0, bring it over here, plus 360. The next step
is. What is the unbalance here? You will see that these two balances each other, the only left
unbalanced is this this has to be divided as 4 upon 360.
What you get is 200 and plus 360, so 4 upon 7 of 360 this goes 5 and 51.4, so this becomes
minus 205.6 and this becomes minus 154.4 this is the balance; then you have half here, so this
becomes minus 102.8 . That is it that is done, so this is it. This alone is going to be plus 617.2,
this is going to be equal to plus 514.4 and this is going to be equal to minus 514.4, this is going
to be 0. Now, this is due to the settlement alone. If I superpose, all I need to do is add the ones I
have got from before, which was plus 27.2, minus 406.4, plus 406.4, 0. If both the force and the
rotation occur together, what we get is plus 644.4, here we get plus 108.0, here we get minus
108.0 and 0. If both the settlement and the loads are together, you do it separately or you can do
them together I do not care, you are going to get the same thing. Understand that all that a
known support settlement does is that it introduces fixed moments and we do the distribution
exactly in the same manner as before. Now let us look at another problem, just to go through the
steps. Up till now, we have only looked at two members; let us look at what happens when you
have multiple members coming into the picture. I am just going to now take up a realistic
problem.
Do not go by what I have drawn; take down the numbers that I have: 30 meter, 40 meter, EI is a
constant, so this is EI, EI, EI, then on this, we have 2.4 Kilonewton per meter, at the center point,
we have a load 40 Kilonewton meter and here at the center point, we have 40 Kilonewton meter.
For this one, we need to find out the end moments and then the bending moment. Once we have
6
the end moments, the bending moment diagrams are obvious. This is a, b, c, d. First and
foremost, compute. I am just algorithmically going through the steps that you do in applying
the moment distribution method to a beam problem.
Today, we are only going to be looking at beam problems; from the next time onwards, we will
see how to consider a frame. Let us look at a beam. The first step is calculate the fixed end
moments and when you take fixed end moments, you do not worry about whether it is a modified
member or not you calculate fixed end moments as if everything was fixed.
For this one, it is going to be equal to 40 into 30 divided by 8 (Refer Slide Time: 31:56). The
(FEM)ab is going to be equal to plus 150 Kilonewton meter; the (FEM)ba was equal to minus (40
into 30 upon 8), so minus150 Kilonewton meter; the (FEM)bc is equal to 2.4 into 40 squared
upon 12 this is equal to 1600 into 0.2, that is plus 320; the (fixed end moment)cb is going to be
equal to minus 620. This is equal to (FEM)cd and this is equal to (FEM)dc because they are the
same thing. Let me then compute the distribution.
Now, ab is a modified member, so Kab is equal to (this is for the member); it is going to be
three-fourths I upon 30 this is equal to I by 40. Since bc is continuous, Kbc is going to be I by
40; Kcd is going to be 3 by 4 because d is a roller, so it is going to be modified (c is going to be
the continuous end, fixed end and d is going to be the roller); in ab, a is going to be the hinged
end and b is going to be the roller. If you look at it, distribution of ba is going to be K b plus Kbc
upon the summation, which is going to be half; Dbc is going to be half; at c, it is going to be I
upon 40, I upon 40, so Dcb is going to be half; and Dcd is going to be half. Having done that, let
us see what the whole thing comes out to be equal to; let us put it.
Note that carryover ba is equal to 0, carryover bc is half, carryover cb is half, carryover cd is 0

and everything else is automatically 0. With these kinds of things, putting them in ab, ba, bc,
cb, cd, dc. What happens over here is ab is a hinged end there is nothing that happens; there is
only one half, that is only once; this way it is 0 carryover. Here, we have half, half; here, we have
half, half; here this is 0, the only thing that we have is half for one, this way it is 0 and here, this
is half, half. What are the fixed end moments? This ab is plus 150, minus 150; this is plus 320,
(Refer Slide Time: 37:39) this is minus 320 and if you look at cd, this is plus 150, minus 150.
The first step is to release these two, so I am going to put minus 150, plus 150. These are
finished and all that we do is take this here (Refer Slide Time: 38:18) and put minus 75 here and
plus 75. This is the one that we need to distribute, so if you look at it 225 and this is 320, so that
net unbalanced is plus 95, so minus 95 has to be distributed, so we do minus 48, minus 48.
Similarly, over here, we do the same. Now here, this is the point that I was trying to make and
this is the problem: I have taken up this problem to show you how we do not do it one after the
other we look at every joint and distribute whatever is the unbalanced moment at that point in
one shot. Again, here at the same point, I am not going to be doing the carryover; first, I am
going to look at the unbalanced.
Look at what we have here: this is 47. I will make this 47 and this 48. Here also, I will make this
48 and this 47. You will see that the same unbalanced moments come up and what we have done
is; we have done both of them together. Now we do the carryover: there is no carryover here, no
carry over here. How was this carryover here and here? We have minus 24, minus 24. Now we
unbalance these, so this is going to be plus 12, plus 12, plus 12, plus 12; now balance: no
carryover here, only carryover here and here, this is going to be plus 6, plus 6. This is going to be
minus 3, minus 3, minus 3, minus 3 that is what it is; then we do again carryover: carryover is
going to be minus 1, minus 1.
For all practical purposes, when I distribute that, I am going distribute plus 1, 0, 0, plus 1 and
that is it it is over, no other carryover, nothing else, so we have reached the end. Add all of
them up, 0; minus 225 and this is going to be equal to 225 plus 47 is 272, minus 272 plus 12 is
minus 260, minus 263; this is 225 and 47 is 272, and this thing is equal to here minus 260 and
minus 3 is minus 263 so it is minus 262; this is going to be plus 262, this is going to be minus
262, this is going to be plus 262 (Refer Slide Time: 42:31) and this is going to be equal to 0.
Once you have this, you have got the member end moments and you can do the distribution. The
point that I am trying to make over here is that note that I did this problem just to show you that
in reality, you do not do the procedure as release one, clamp the other one and so on. Once we
have established it, now you cannot tell what is happening here. I am going to be simultaneously
releasing both and then looking at the effects, then releasing, taking effect, so physically
understanding this is very difficult.
Physically, this does not make sense, but you have to understand that in the moment distribution
the steps are the following. First, compute the fixed end moments. Even if you have a modified
member, you compute the fixed end moments as if it was a fixed-fixed member; the fixed end
moments are computed as if every member was fixed. Then the next step is actually computing
the stiffness of each member.
In this particular case, when you compute the stiffnesses the first step that you do is you actually
go ahead and compute the stiffnesses based on whether a member is the original member or the
modified member. Therefore in this particular case (Refer Slide Time: 44:29); when we
computed the fixed end moments, we computed the fixed end moments as if ab, bc, cd were all
original members; so fixed-fixed, fixed-fixed, fixed-fixed and we computed the fixed end
moments. Actually, accept the fact that ab is a modified member, bc is the original member and
cd is a modified member.
While computing the stiffnesses, that is where we take into effect that ab is a modified member;
so Kba is 3 by 4 of I upon L; bc is a normal member, so it is I upon L; cd is the modified member;
then the distribution factors are computed based on the modified members; the carryover factors
of course depend on whether it is a modified member or a regular member. Then, the first thing
that we do is we have computed the fixed end moments based on fixed-fixed, so the first thing
that we do is actually release the two end releases and carry that over once only and that is it;
then these two are finished because now nothing goes back to them.
Now what we have to do is, we look at both the members. In other words, how many ever joints,
continuous joints we have, we look at their distribution to the unbalanced forces at one shot we
do all the distributions at one shot; so you will see this, this, if there are more members, all would
be distributed. Then, the next phase is the carryover phase: you carry over everything that you
have to; in this particular case, the carryover is only here because there is no carryover on the
other side. Once we have finished the carryover, we look at all the unbalanced and do the
distribution, so you have fixed.
The first it is the modified member carryovers, then distribution, carryover, distribution,
carryover, distribution, carryover, distribution till we hit a situation where the carryover is 0.
9
Ultimately, the distribution is ended and that is your final member force. This in essence is your
overall moment distribution method as we see it for beams. If I want to review what I have done
today, I have looked at the moment distribution method as applied to beams, we have looked at
how modified members can be incorporated into the moment distribution, we have seen the
physical understanding behind the moment distribution method and then we have applied it,
where we actually do it over in a distribution phase, carryover phase, distribution phase,
carryover phase. We also saw how to consider support settlements into the entire consideration.
The other point I would like to make is that even if you have temperature effects, the temperature
effects are again local effects, which only give rise to fixed end forces. Everything else, the
moment distribution procedure it is not different whether you have a load in a member or
whether you have temperature in a member. In other words, at the end of this lecture, you should
be actually able to solve any problem that you have for a beam you should be able to do the
moment distribution method. What I want to do is, I want to give you a problem so that you can
solve it yourself. Please take a note of a problem. Let me give that problem and you have to do
the problem yourself.
This is the structure and I want you to solve this particular problem and get the member end
moments. Once you get the member end moments, you can draw the bending moment diagram
I ultimately want you to get the member end moments and automatically the bending moments
diagram, using all modified stiffness and distribution factors that you can use in this particular
case.
After today, I hope you will be able to apply the moment distribution method to beams for any
kind of situations. I have given you this problem as an exercise you have to do it and satisfy
yourself that you understand the moment distribution method. From tomorrow, I am going to
spend a few lectures looking at how to apply the moment distribution to frames.
10
When we looked at the displacement method, we saw that whether it was a frame or a beam, the
method was really the same; the only thing is that in a frame, you might have to write down
certain equilibrium equations related to displacements but otherwise, the procedure was
identical. In the moment distribution method, you will see that in frames that have displacement
degrees of freedom the entire procedure is very very different. I will first start off by looking at a
frame and see how we can do the moment distribution where it does not have a displacement
degree of freedom.
We will look at that and then we will go on to see how to modify the moment distribution
method so that you can consider displacements or what is known as sway in a frame. Thank you
very much. I hope you have understood the moment distribution. Please do solve this problem
that I have given you and come back later and with confidence.
11
Prof. P. Banerjee
Lecture 23
Good morning. Today, we are going to continue looking at the moment distribution method. In
the last two lectures, we looked at the moment distribution method, its basis and how you use it
to solve beam problems we solved quite a few. I hope you have been able to solve the beam
problems that I had given at the end of the last lecture it is a fairly simple problem and I hope
you have been able to do it. Today, we are going to be looking at moment distribution for
frames.
In moment distribution for frames, you essentially have two kinds of cases: frames without sway
and frames with sway. You will see that frames without sway are identical to the method for
beams and for this, we will develop procedures. Essentially, we understand that in frames, you
have two kinds of situations: one, that of frames where sway is allowed the sway implies
essentially where displacement degrees of freedom exist (since we neglect axial rigidity, sway is
almost always associated with displacement degrees of freedom) and frames without sway are
frames that only have rotational degrees of freedom available in them. Before I start working on
it, I just want to show a couple of things that are essential. One is that there are some
bookkeeping issues when we are doing moment distribution for frames. In other words, you
know in beams, they were all aligned along one direction and therefore, when they were aligned
along one direction, you could just do the moment distribution in a table form and get away with
it; when you have a frame, you know you have a situation let us take a typical frame.
Let me just draw a typical frame. The point is that in a beam, you at the most have two members
meeting at a joint and therefore, putting it together, we did it in a tabular form. In the frames, you
have these two joints where you have four members meeting. When you have four members
meeting, it becomes very difficult to essentially state how to put it in a tabular form. That is the
reason why I am just going to go ahead and look at a situation where how to essentially do the
moment distribution, not so much the mechanics of the moment distribution but more the
bookkeeping aspects of doing moment distribution. Therefore, let us look at some of the things.
The way we put it is, if you have a beam (a horizontal member), the way it is done is that say
this is i and this is j. We write the distribution factors as Dij and Dji and we drew all the processes
in this manner. They start from here, so here you have (FEM)ji and you do the distribution
carryover in this manner and over here. If you have a vertical member (a column), the way you
do it is this way. Let us say this is i and this is j and you have Dij here and j here, Dji here (Refer
Slide Time: 07:05); (FEM)ij and (FEM)ji; and you proceed upward in this direction and here, you
proceed downward in this direction. This is the way.
Therefore, if you look at a situation where we had all four members meeting, let us say this is
joint i, here I have joint j, here I have joint K, l, m. Then, Dij would be here, Dik would be here,
Dim would be here and Dil would be here. If you follow this notation, you will see that, here this
is the bottom node for ik so it comes on this side for the column; it is a right node for ij therefore
it comes over here; it is the left node for il therefore it comes over here (Refer Slide Time:
08:37); it is the top node for im and it comes over here.
You have (FEM)ij going up this way, you have (FEM)ik going up this way (Refer Slide Time:
08:53), you have (FEM)il going down this way and you have (FEM)im going down this way
this is the bookkeeping notation. At this joint, you will have Mij, then over here you will have
Mik, here you will have Mil and over here, you will have Mim. If you look at this, the equation at
this point will be Mij plus Mik plus Mil plus Mim is equal to 0 and the distribution happens also in
this manner; of course, it goes without saying that Dij plus Dik plus Dil plus Dim is equal to one,
by definition. Therefore, it is very important to understand that if we follow this procedure in a
frame, you actually do the moment distribution in an exploded this thing of the frame. Now, let
us quickly look at a couple of problems so that we appreciate what it is that goes into this thing.
Let me first write down, make the problem statement.
The problem statement is this. This is 20 meters, 30 meters, 30 meters, 30 meters. Over here, I
have 10 Kilonewton force acting at the center point, so that is 10 meters from the bottom; on this,
3
I have 10 Kilonewton acting at 10 meters and on this, I have a loading 1.2 Kilonewton per meter.
This is a, this is b, this is c, d, e, f this is the problem. It is very important to understand as to
how many degrees of freedom this structure has. To figure out whether you have a frame with
sway or without sway, you really have to go back and look at the number of degrees of freedom
that you have, so let us go back. How many do we have? We have six joints. I have 3 into 6, 18
this is the number of unconstrained degrees of freedom. Let us look at the number of restraints. I
have two here (Refer Slide Time: 13:03), three here, three here, so 2 plus 3 plus 3, 8, 8 plus 1, 9
so restraints is 9. Constraints is one, two, three, four, and five; we have essentially 1, 2, 3, 4
and 5 axial rigidity. If you look at it, it is a four degree of freedom structure.
What are the four degrees of freedom? One rotation here (Refer Slide Time: 13:37), second
rotation here, third rotation here, fourth rotation here. Therefore, since it only has rotational
degrees of freedom, this is a frame without sway. This is because it only has rotational degrees of
freedom associated with it so it is a frame without sway; we are going to be solving this
problem as a frame without sway. First and foremost, before I start solving the problem, I need to
first write down what the fixed end moments are; let us look at the fixed end moments.
If we look at the fixed end moments, we are going to have (FEM)ab and (FEM)ba are going to be
equal to 0 because there is no loading between a and b. Now, (FEM)bc is going to be equal to 10
(that is the load) into a b squared so it is going to be 10 into 20 squared upon 30 squared. If you
look at this, this turns out to be 44.4 Kilonewton meter. The (FEM)cb is equal to minus ((10 into
20 into 10 squared) upon 30 squared) and that is equal to minus 22.2 Kilonewton meter. Next, let
us look at (FEM)cd. This is equal to 1.2 into 30 squared upon 12 and this turns out to be equal to
90.0 Kilonewton meter; the fixed end moment at dc is negative 1.2 into 30 squared upon 12, so it
is equal to minus 90 Kilonewton meter. Finally, be: (FEM)be is going to be equal to plus 10 into
20 by 8, which is equal to 200 upon 8 is 25 Kilonewton meter and the (FEM)eb is equal to minus
25 Kilonewton per meter. Finally, (FEM)cf is equal to (FEM)fc, which is equal to 0. We have
computed all the fixed end moments. What is the next step? The next step is to compute all
distribution factors. We are going to be using all modified members that we can use.
Note that since is a hinge, ab is a modified member and therefore we have Kab modified is equal
to 3 by the way, all the Is are the same, so you have three-fourths I upon 30, which is equal to
I upon 40 (Refer Slide Time: 18:35); now, Kbc is not a modified member, so it is equal to I upon
30; then, we have Kcd is modified since d is, so it is going to be three-fourths into I upon 30,
which is equal to I upon 40; finally, we have Kbe, which is a normal member I upon 20; Kef is
equal to I upon 20 these are the stiffnesses for each member. Note that the modified ab
essentially happens that this essentially comes at b; on a, it is 0. Therefore, if we look at the
joint b, you have ba which has I upon 40, bc which is I upon 30 and you have be which is I
upon 20. What is the summation of the stiffnesses? The summation of stiffness is equal to I upon
40 plus I upon 30 plus I upon 20. If we look at this, this becomes over 120, it becomes 3I plus 4I
plus 6I, which becomes 13I by 120.
Therefore, distribution factor for ba is equal to I upon 40 divided by 13I upon 120 and this
becomes 3 upon 13, Dbc is equal to bc is equal to I upon 30, so this becomes 4 over 13 and Dbe
is equal to 6 upon 13 these are my distribution factors at joint b. Now, I have another joint
which is c where I have multiple members. Every other joint. Look at joint a joint a is a
hinge, joint d is a hinge, therefore there are no computations at that particular point, excepting
for that one thing that you have which is just to put it together; joint e and f are fixed ends at
fixed ends, you only get moments coming in, you do not have any moments going out; therefore,
only b and c are the joints at which you have to consider the effects.
If you look at joint c, at joint c, the summation is identical and it turns out to be 13I upon 120.
Therefore, Dcb is equal to 4 upon 13, Dcd is equal to 3 upon 13 and Dcf is equal to 6 upon 13
these are the distribution factors. Let us write down what the carryover factors are: the carryover
factor of a to b is half only for the first situation; carryover factor from b to a is 0; carryover
factor from b to c is equal to carryover factor from c to b half; carryover factor from c to d is 0;
carryover factor from d to c is half only for the first, when the fixed end moment is distributed;
then we have carryover factor from b to e is half; carryover factor from e to b is 0; carryover
factor from c to f is half; carryover factor from f to c is 0. These are the carryover factors that we
have, we have the distribution factors, so now let us look at the problem. If we look at the
problem, what do we get?
If you look at it, 3 over 13 is 0.231, 4 over 13 is 0.307 and 6 over 13 is 0.462. Similarly, for these
(Refer Slide Time: 25:28). Let us now finally look at the structure.
let me draw it in a slightly bigger sense. If we look at the structure, note that here we do not have
anything. What is this part (Refer Slide Time: 26:20)? This is actually Dba and Dba is equal to
0.231. This is a, this is b, this is c and this is d and d we have nothing here. For be, bf, what do
we have here? be is going to be here, so be is going to be equal to 0.462; for bc, this is here, so
bc is 0.307; here, this is cb, so this is going to be 0.307; this is cd remember the book-keeping
so this is 0.231; and this is for cf, so cf is 0.462; over here, I need to have it on this side this is
going to be 1.0; this is on this side, this is 1.0.
Let us put down all the moments that we have; if you look at all the moments that we have, on ab
we had 0, (FEM)ba was 0. What was b? b was equal to plus 25.0. For bc, it was plus 44.4, for cd
it was minus 22.2, for this, it was 0, so here also 0, here it was negative 25.0, and for cd, it was
plus 90.0 and minus 90.0. I have just written down all the distribution factors. Remember the
carryover: there is 0 carryover here, here there is carryover in both directions of half, here there
is 0 carryover, here there is half carryover, here this is half carryover in this direction carryover
is 0, in this direction carryover is 0 because this is a fixed end. Ultimately, so ab we have
written down everything all the carryovers.
Let us start the process of looking at the moment distribution. We have written down all the
distribution factors. Remember this is ab, this is ba, this is be, this is bc (Refer Slide Time:
29:58) just remember the way that we have done. Now, what we need to do is we need to
consider one of the ends where we need to. The first thing that we need to do is eliminate d;
there is nothing at a so nothing to eliminate, we need to eliminate d. The first thing that we do
is plus 90.0, eliminate and put it equal to 0; this plus 90.0 will come on this side as plus 45.0
that is the first thing that is done.
Now, what we start looking at is once we have done that, this has gone to 0, this has gone to 0,
modified everything is there. Now, we need to look at the two joints where we need to release
the moments and remember we had said that we were going to do it simultaneously; so all the
joints that have undistributed or net moments we are going to do the distribution in one shot
and do the carryover; so what we do is we do the distribution at all joints and then do the
carryover in the next phase. The distribution: first, let us look at joint b. There is an unbalanced
moment of 69.4 so the 69.4 needs to be balanced out. In what way? We need to balance it out by
taking 69.4 and putting it around. If you look at putting it around, what do we have? We are not
going to do it simultaneously. You can do it simultaneously, but since I have solved the problem
by first releasing joint c and then joint b, let me go with that.
Here, I have plus 90, plus 45, minus 22.2 this needs to be distributed and if I distribute this, I
get. The net unbalanced moment is 135 minus 22.2 is minus 112.7; 112.7 is to be distributed
it becomes minus 26.1 balance; then here when you balance, your balance together here, you
get minus 52.1 and here you get minus 34.6. As soon as you do the balancing, we take care of
this is the thing. Next, we need to do the carryover. There is no carryover here (Refer Slide
Time: 33:02), here there is a carryover the carry over is minus 26.1 and so this is the carryover
from here to here; this is the carryover from here to here it will be minus 17.3 and there is no
carryover, so we have done the carryover phase also. After carryover, you do not write down
anything, so it is still unbalanced but here, there is no question of unbalanced; next thing is the
unbalanced here (Refer Slide Time: 33:38). If you look at it, it is going to be 69. 4 minus 17.3 is
going to be 52.2 that we need to do; this turns out to be minus 16.0, minus 24.1 and minus
12.0.
Since distribution is done, we need to take care. Now, the carry over phase at joint b: this is not
going to carry over here, this is going to carry over here as minus 12.1, this is going to carry over
8
here as minus 8 that is it, we have done the carryover. Now, the next step is essentially you
have an unbalanced moment of minus 8.0 over here, so we need to balance that it becomes plus
8.0, which needs to be distributed and if you distribute it, you get plus 2.5 distributed; here we
get plus 3.7 distributed, then we have plus 1.8 distributed. As soon as we have done the
distribution, we come here; now the next is the carryover phase, so this 3.7 comes over here and
this becomes plus 1.9; here, there is no carryover; here, there is a carryover, which is plus 1.3;
here, there is no carryover.
Now, all we are left with is this plus 1.3, which needs to be redistributed and we redistribute it
it becomes minus 0.4, block here. Here, this will become minus 0.6 and this will become minus
0.3; this minus 0.6 needs to be taken here but 0.3. Remember I had said that. When do you
stop? You cannot keep going forever. I had said that as soon as you get distributed values which
are less than 1 percent of your values, you stop. Over here, this comes out as minus 0.3 carryover
here, but if you carry over this minus 0.4 here (Refer Slide Time: 36:49), it becomes 0.2, which
is much less than 1 percent, so I am not going to do the carryover. The carryover for all practical
purpose is 0 and here there is no carryover.
Since there is 0 carryover, this is balanced and we are done; as soon as we are done, we draw a
double line and we draw a double line here and what you do is you add up all the values that you
have over here and once you have that, you add them; this one if you add up, it turns out to be
0.3 that is be; if you look at this, add this up, you get minus 37.4 this is the moment here; add;
all of this up, you get plus 12.0 this is the moment at b; if we add this up, we get minus 12.3
this is the moment here (Refer Slide Time: 38:07).
Note that if you if you add them up, you will get it equal to 0 and that is how it should be; here
you will get minus 48.4 this is the moment here; this is plus 110.7 this is the (moment)cd; if
you add this up, you get minus 62.3; again if you do the addition, you will see minus 62.3 minus
48.4 plus 110 is equal to 0 so this is balanced; here this turns out to be equal to minus 24.2 and
this is 0. Ultimately at the end, let me put down the values that I get.
This is what I get: Mab minus 0, Mba is equal to minus 12 by 3; Mbc is equal to plus 12; Mbc is
equal to minus 62.3; Mcd is equal to plus 110.7; Mdc is obviously equal to 0; then, you have Mbe
which is equal to plus 0.3; Meb is equal to minus 37.4; Mcf is equal to minus 48.4; and Mfc is
equal to minus 24.2 this is the moment distribution and since this is a frame without sway, this
is where we end we have got the member end moments. What is the next step for drawing the
bending moment diagram?
For ab, 12.3; for bc, it is 12.0, minus 62.3 and a 10 Kilonewton force acting at 10 meters, this is
30 meters, this is 30 meters, then we have cd we have 110.7 with a udl of 1.2 Kilonewton per
10
meter and then we have be over here is equal to 0.3 and here, the applied load is 10 in this
direction; then eb is equal to 37.4; then you have ef cf is equal to clockwise 48.4 and this is
clockwise 24.2. These are your members and for each member now, you should be able to draw
the bending moment diagram.
How will the bending moment diagram for this look? This will look in this way (Refer Slide
Time: 43:10). For this one, it is going to be 12.3. What about this one? This one is going to be
12.0 and 62.3 and at this particular point, this one is going to have 10 into two-thirds, so 20 by 3,
so that is going to be 200, that is going to be 66; at this point, its moment is going to be
something like this; in this particular case, you are going to have a moment this way which is
going to go in this manner; bending moment this way and over here this way; I am always
drawing the bending moment on the tension side, so this is going to be 48.3, this is going to be
this way, so this is in this fashion and over here, this is point, so this is going to be point; this is
going to be 37, and you can compute what those are very easily.
I have just drawn the bending moment diagram and of course, first you need to put it together
and be able to do it.
In essence, what we are saying is that if you have a frame which only has rotational degrees of
freedom, then you can classify that as a frame without sway; and for a frame without sway, the
entire moment distribution procedure is identical to that of beams, the only difference being
since a frame has both vertical and horizontal members, you need to do a little bit of proper
book-keeping. Remember that for a horizontal member, you do it in the left end bottom and then,
you do right end top; then, if you have a vertical member, the top end you do on the left-hand
side and the bottom end you do on the right-hand side; if you proceed in this manner, you will
find that you will not have any problem in solving the moment distribution of frames without
sway. I think you should be able to now solve all problems of frames without sway.
11
Now, how do we tackle frames with sway?

In frames with sway: Let us take this particular same problem. Let me just change it. I am not
drawing it properly but let me just change it one bit with instead of a hinge at this point, let me
put a roller here. It is exactly the problem that we have solved, excepting for the fact that this one
has a roller here rather than a hinge. How many degrees of freedom do you have? You have one
less restraint, so you have five degrees of freedom. What are the five degrees of freedom? One
rotation here (Refer Slide Time: 47:46), one rotation here, one rotation here, one rotation here
and this displacement. Now, you have a frame which not only has rotations but also has a sway.
How do I solve this particular problem? Very simple we solve it as two problems. The first
problem: one, we restrict sway. How do we restrict sway? We do this. When we restrict the
sway, the first thing that we do is find out what is the reaction at this point the horizontal
reaction. Why? Because in the real structure, you will note that since this is a roller, the
horizontal reaction at this point has to be equal to 0. So the first problem: restrict sway, put
loading, solve the problem we have already done it; then compute this (Refer Slide Time:
49:22).
Once you compute this, the second problem is with sway. What you do is you consider the
structure like this (Refer Slide Time: 49:48) you apply a load here and the load that you apply
over here is just the opposite of this. When you put the two of them together, you get the original
structure but in this case, you do not have any loading you only have this. The whole point
comes around to the fact that. What happens when we apply this load? This is where even in
moment distribution, the kinematics of the structure under just the translational sway we need
to find out how the structure actually displaces.
Once we find out how the structure actually displaces, we can actually find out all the fixed end
moments. We can consider it. Remember we solved the support settlement problem? We solved
the support settlement problem. In this particular case, we are going to be solving it as if it was a
settlement that the support had moved, find out how the structure moved, find out all the fixed
end moments due to that movement and then do the moment distribution and we will see how we
are going to go on to solve this problem. In other words, the only thing that happens when you
have a frame with sway is that you have to do two sets of moment distribution one of the frame
with sway; in other words, you are restricting the sway, so you introduce a restraint
corresponding to the degree of freedom that is the sway degree of freedom, then compute the
reaction in the restraint direction at that support; then we consider another problem where we
consider that load to be applied on the structure and consider the frame only with sway and
without loading ultimately, that is the procedure that we are going to be using. I am going to
stop over here. In the next lecture, I will look at the details of how to solve a problem of a frame
with sway.
12
Prof. P. Banerjee
Lecture 24
We have been discussing the moment distribution method over the last few lectures and today,
we are going to be considering the moment distribution of a frame with sway included in it. I
will explain this. I have given you the background behind this in an earlier lecture and let me
illustrate what I meant by actually solving a particular problem. I will start off with a simple
problem and then, we will go on to a more difficult problem in the next lecture, so that you
understand the basis for the entire procedure.
Here the question is to determine the end moments and support reactions this is the problem
statement for you. Let us start the procedure. The first procedure is. Again, to determine. I
just say this is a frame with a sway, but you have to convince yourself, so you actually need to
find out all the degrees of freedom. How many degrees of freedom? One, two, three, four, so 3
into 4, 12 unconstrained degrees of freedom; 3 plus 2, 5 restraints; one, two, three, 3 constraints;
degrees of freedom 12 minus 5 minus 4 is 4 degrees of freedom, so degrees of freedom is equal
to 4. What are the four degrees of freedom? It is 1, 2, 3, and 4. So note that in this particular case
there are three rotational degrees of freedom and one translation, therefore this is a frame with
sway. Let us start. What was the first procedure? The first procedure was to actually restrain the
sway.
The first problem is going to be. How do I restrain the sway? By actually doing this (Refer
Slide Time: 05:44). Therefore, I am going to get a reaction over here. I am going to get a reaction
over here, so this is the frame without sway. Here, what do I do? Well, this is the structure. I find
out. I solve the entire problem I solve it by finding out first the (FEM)ab, (FEM)ba there is
nothing, so this is equal to 0. The (FEM)cb is equal to 0. This is a, b, c, d. The (FEM)cd and
(FEM)dc are equal to 0.
All of them are equal to 0 and then fixed end moments are equal to 0. What would the moment
distribution give me? Note that if fixed end moments are 0, at all joints you have the moment
equilibrium. Therefore, the moment distribution for the frame without sway in this particular
case becomes that we know what the moments are going to be; Mab is equal to Mbc is equal to
Mbc is equal to Mcb is equal to Mcd is equal to Mdc is equal to 0. There are no member end
moments. Therefore, if I take equilibrium of this particular case, since moments are equal to 0,
you will see directly that this r is equal to 20 Kilonewton meter (Refer Slide Time: 08:03). In this
particular case, because of the special load that I have considered here, the computation of this R,
the restraint, is very simple and the frame without sway is actually a trivial solution you do not
need to do the moment distribution at all.
Understand that this is a special case because I am introducing you to the concept of frames with
sway that is why I have chosen a situation where a frame without sway does not actually require
any moment distribution this is not generally the case. Next time, we are going to be looking at
a particular problem, where you will see that you are going to have a situation where the moment
distribution, even for the frame without sway is going to give you a non-trivial solution and then
you will see that to compute this R, you will have to actually solve a lot of equilibrium equations
and get to this particular value of R. Here, it is trivial and therefore I had no problems in
computing the value of R. Now, what do I do? What is the next step?
The next step is to release this R and see what happens to the structure when you allow it to
move. Let us just say that this is some unknown quantity delta. What is going to happen? Note
that when I take this delta, I am assuming that fixed end everything is b and c are clamped. If
b and c are clamped, how will the rotation look like? You will see that it will look like this
(Refer Slide Time: 10:26). Since this point goes here, this point cannot go up it has to go only
this way and the amount it goes by is delta to ensure that bc remains. What are the fixed end
moments?
Calculating the fixed end moments is very easy. All I need to do is find out what the rotation
from the chord is and plug it in to my equation. If I put that, what is my thetaab equal to? Delta,
this is 7.5, so thetaab is equal to delta by 7.5. Is it positive or negative? From the chord to the
tangent, anticlockwise so positive, so thetaab is equal to thetaba is equal to delta by 7.5. What can
we say about thetabc and thetacb? Both are equal to 0. What about thetacd? It is equal to delta by 5
and this is also anticlockwise, so it is delta by 5. Can I compute the fixed end moments? All I
need to do is I need to plug it into my equation and if you plug it into my equation.
What do I get?
I get that the (FEM)ab is equal to 4EI by L L is 7.5 multiplied by thetaab which is delta by 7.5
plus 2EI upon L multiplied by delta upon 7.5 this is equal to 6 EI upon 7.5 square and this is
equal to (FEM)ba. This is the fixed end moment at ab and ba. What are the fixed end moments at
bc and cb? They are equal to 0. What is the (FEM)cd? It is equal to 3EI upon delta multiplied by
delta by 5 so this is equal to 3EI delta upon 25. These are my fixed end moments. The only
problem over here is that I do not know what my values are. If you look at it, we had the
situation that Iab and Ibc were 1.5, so I have to multiply by 1.5 actually, so this becomes 4EI into
1.5 and 1.5 here; this becomes 1.5 here; this one turns out to be equal to 0.0267 delta. (Refer
Slide Time: 15:13) let me just compute this directly this will be equal to 6, so this is going to
be 1.2EI delta upon 7.5; this is going to be equal to 0.1, so this is going to be 0.12EI delta; this
becomes 3, 40, 30 by 4, so 30 by 4 becomes 4.8 upon 30, so this becomes 0.1, 4.8 upon 30 is
equal to 0.16EI delta.
Now if you look at it, what you get essentially is that the fixed end moments turn out to be in this
fashion; 0.16 EI delta, 0, (FEM)cd is equal to 0.12EI delta. Note that the fixed end moments
depend on delta. Here, what I am going to do is I am going to actually assume a certain value of
a delta. If I assume a particular value of delta, then what happens? I am just going to say that let
us assume that EI delta is equal to 100 I am going to assume it and then we will see what
happens. In that particular case, you will see that this will be equal to 160, this becomes equal to
120. Now I have numbers with which I can do a moment distribution. We have done the fixed
end moments, now let us look at the distribution factors.
For the distribution factors I need to know that Kab and Kba are equal to 1.5 I upon 7.5, this is I
by 5. Kbc is equal to Kcb is equal to 1.5 I upon 10. This is equal to 0.2 I, this is equal to 0.15 I and
finally we have Kcd which is equal to three-fourths I upon 5, so this turns out to be 0.15 I. Once
we have this particular thing now we can find out the distribution factors. How many joints?
There are two joints, so (distribution factor)ba is equal to 0.2 I upon the summation of this (Refer
Slide Time: 19:50) which is 0.35 I, so you essentially have 0.571; and Dbc is equal to 0.15 I upon
0.35 I which is equal to 0.429. Then finally at c we have Dcb 0.15, 0.15, so it is 0.5 and Dcd is
equal to 0.5. What can we say about carryover factors? Let me put down the carryover factors
here itself. Cba is going to be half, which is going to be equal to Cbc. However, C(c to d) is going to
be equal to 0. Having put all of that in, let us now do the moment distribution.
Remember left side for this (Refer Slide Time: 21:21), right side for this, bottom for this, top for
this, left side for this and right side for this. Here, I have 1.0 whatever comes in, distributes
here; here, I have 0.571, here I have 0.429, here I have 0.500 and 0.500; and this here, since there
is no moment. What are the values of the moments that I need to put in? (FEM)ab was equal to
plus 160, this is also plus 160. What was the fixed end moment here? 0. What is the fixed end
moment here? 0. What is the fixed end moment here? Plus 120. What is the fixed end moment
here? 0. These are my values. I am going to now start doing the moment distribution and let us
just go through the process.
In this particular case, I am going to do the distribution together, so over here, I get minus 60,
minus 60 distribution over here and this is going to be 4 over 7, so this is going to be 640 upon 7,
which is equal to minus 91.4 and this is going to be 480, so this is going to be minus 68.6. Now
the next step is the carryover, so from here (Refer Slide Time: 24:12) I have carryover to here, so
this becomes minus 45.7; I have carryover from here to here so this is minus 34.3; here I get
minus 30 and no carryover here.
The next step is distribution. This needs to be distributed, so this becomes plus 17.2
distribution done, plus 17.1 distribution done, here the 30 has to be distributed, so what we
have is here, it becomes 17.1, this becomes plus 12.9, so this distribution done. Now, we do
the carryover process: 17.1 over here will become plus 8.6, here, the distribution will go as plus
6.5 and here, there is no distribution.
Now I need the distribution from here (Refer Slide Time: 26:18) to here, so that 17.2 distribution
becomes plus 8.6. Now, I need to distribute this and if I continue with my distribution process, I
am going to come over here, so that I can keep adding and here, I am going to just go in this
direction, so this is going to be distributed over here, it is going to be minus 3.2; this is going to
be minus 3.3, so this is distribution done, distribution done; here, plus 8.6, when we go 4 by 7, so
that becomes 34.4, 34.4 becomes minus 4.9 and this becomes minus 3.7, block here, block here.
Now, let us put this together over here. I am going to have This comes to this point, so this
will go in this direction and this will come here, so this becomes minus 2.5, there is a distribution
here, the minus 3.7 goes over here, it becomes minus 1.9 and the distribution from this turns out
to be minus 1.6. I am going to do my final distribution. because I want to go down. If I do this
distribution, I will get it equal to plus 1.0 that is the distribution, this one turns out to be plus
0.9. When I distribute this, I get plus 0.9 here and plus 0.7 here. Note that I am not going to carry
over any of these because these have gone down in to the less than 1 percent level.
Once I have done that, I can add these up. Note the plus 9 needs to come over here as plus 0.5, so
here also, this goes (Refer Slide Time: 29:42); this also goes this way, this also goes this way.
All I need to do now is add these up and when I add these up, what do I get? I get 178.0 minus
96.3 so I get plus 81.7. If I go through this, I get minus 81.7 here, here I get plus 169.1, this
becomes minus 2 minus 2 minus 43.7 becomes 116.3, 124.9. Let us see what I get over here:
plus 60 plus 77.1 plus 77.9 plus 78 plus 78, plus 78 minus 3.3 is plus 74.7, this also turns out to
be minus 74.7 and we have our member end moments. Once we have the member end moments,
let us see what we get in terms of the values themselves.
This plus 124.9 is going to be in this direction, then I have plus 81.7, then I have this, I have this
member as equal to 81.7 and this is going to be 74.7, this is 0, this is 74.7. Now because of these,
we can find out what the shear forces are, so due to this, I am going to get a shear force in this
direction, this is going to go in this way, so ultimately 124, this is going to be 206.6, so this is
going to be 206.6 divided by 7.5, this is going to be equal to this way, this is also going to be in
this direction and this is going to be 74.7 by 5. If you really look at it, this force R that is going to
require doing this is going to be equal to 206.6 upon 7.5 plus 74.7 upon 5. This is equal to 3 by 4,
30 by 4, so this is going to be equal to 826.4 upon 30; this is equal to 14.94 and this part is going
to be equal to 27.55 plus 14.94 this is going to be equal to 42.49, let us say 42.5. That means to
generate these moments, I require 42.5 but what is the actual R? It is 20. What are the final
moments? They are going to be equal to 20 upon 42.5 multiplied by all the moments.
If you take those, you will see that this turns out to be equal to 57.4, this is going to be equal to
39, this is of course going to be 39, this is going to be 35.7, this is going to be this way and this is
going to be in this way, 35.7, here, this is 0. So if you look at the support reactions, the support
reactions become something like this. At this point, I have a moment of 57.4 Kilonewton meter,
a 7.47 reaction here (Refer Slide Time: 37:31) and 12.86 Kilonewton moment here; over here, I
have this one is downwards, this one is upwards 7.47 and there is going to be a force of 7.14
and this is the load 20 Kilonewton.
Therefore, the point that I want to go back and point out to you is that we did not know what the
delta was. We know that if delta was there, this would be 0.16EI delta and this would be 0.12EI
delta we knew this. You cannot do this, so I am just saying assume; we do not know what the
delta value is, so I am just saying assume EI delta; all of them have EI delta in them, so I assume
that EI delta equal to some value I have assumed EI equal to 100 but you could have actually
taken it to be some other value.
For example, one way that a lot of people work with is take the largest value to be given by 100;
I would have said that 0.16 EI delta is equal to 100; I could say that is equal to 100 but note that
as soon as I say that it is equal to 100, then I know what the EI value is and I can substitute that.
If you look at it, that if you have taken this to be 100 (Refer Slide Time: 39:26), this would be
75. The only point that I am trying to make is it does not matter what you take you need to be
consistent that these values need to be taken together and therefore in this particular case, I have
taken 160, 120 but you might equally have taken it to be 100, 75 it is all relative.
Once you have done that, the fixed end moments you solve through the fixed end moments;
and once you get you do the moment distribution. The moment distribution procedure, if you
note is identical to that of beam again. Once you have fixed end moments, the moment
distribution procedure is unique. You have to compute the stiffnesses and once you compute the
stiffness, you can compute the distribution factors, you can calculate the carryover factors and
then once you put those in, you go through the moment distribution process. This time, I actually
did it so that I am doing it simultaneously I am releasing this and this distribution; then, do
all the carryovers.
Once you do the carryovers, you look at what is the net moment unbalanced balance that and
you keep doing it and as I said, this procedure is going to be convergent and we can get the final
values. Note these values have no meaning if you had taken this to be 100 and this to be 75,
these values will be different; these values per se do not have any meaning, but once you have
these values, these would be the values if you had a certain value of delta; since you do not know
the value of delta, these values have no meaning.
However once you have these values, these are the member end moments; once you have the
member end moments, you actually have to go and put them and compute. You know what force
is required to get these kinds of moments. I computed this and this (Refer Slide Time: 41:54), I
know that this plus this is equal to this. So all I found out R by adding this and this and I got it to
be 42.5. Now I know my R value is 20 Kilonewtons. Therefore, what do I have to do? All I have
to do is, I have got the member end moments; for a force of 42.5, these are the moments. What
would it be for 20? Linear system you can actually scale. What does the scaling mode imply?
For 42.5, you have this (Refer Slide Time: 42:33), so for 20 the value would be 124.9 multiplied
by 20 upon 42.5, multiplied by 20 upon 42.5, multiplied by 20 upon 42.5, multiplied by 20 upon
42.5.
You will see that R is equal to 20, which means all I need to do is scale all these values by 20
into 42.5. Once I scale the values, this is what I get. Once I get this, this is the ultimate solution. I
have got the reactions at the supports, I have got the member end moments so my problem is
solved. In essence, I just want to go back and review that a frame which has sway how do you
tackle that problem? First, you take frame without sway. Now in this particular case, when you
took frame without sway, because there were no member loads, the fixed end moments were 0
and I could compute the restraint reaction directly. Once I calculate the restraint reaction, I can
actually apply an opposite force to get delta. Now I take as if the frame was swayed and so let
me just write down the procedure for you, so that you can get.
This is moment distribution for a frame with sway. First: restrain sway. Two: solve the problem
without sway, because once you restrain sway, you can solve problem without sway; you do the
moment distribution solve the problem without sway is moment distribution; once you do
moment distribution and you get the member end moments, you can. Three: compute restraint
reaction. Four: if reaction R is equal to 0, then frame does not have sway, so end the problem
whatever member end moments you get at the end of the problem without sway, that is your
member end moments; however if R is not equal to 0, then what do you do? Six: assume sway
equal to delta and find fixed end moments. How do you find out fixed end moments? Through
kinematics. You find out the fixed end moments through kinematics.
10
Once you have found out the fixed end moments through kinematics, the next step is moment
distribution of fixed end moments; moment distribution will give you member end moments.
Eight: once you have a member end, you can compute the load required at original restraint point
to get these member end moments. Let this be equal to R1. Then, the actual sway member end
moments are equal to R upon R1 into these member end moments. Ten: total member end
moments are equal to non-sway moments plus sway moments. I just wanted to list out the overall
details. Now, I think you should be able to solve a problem with sway. I am going to take up
some more problems with single sway in the next lecture. I just want to briefly tell you how to do
it suppose we have two sways let me take that problem.
In this particular case, you can find out that you have two sways: delta2, delta1. How would you
solve this problem? This problem is solved in this way. The no-sway case becomes this (Refer
Slide Time: 50:17). Compute R1 and R2 the reactions at the two supports; you can do that
because once you get the member end moments, you can always compute what these reactions
are. Then, plus compute this load which is going to give you a delta compute this; and you
can get the reaction at this point plus now you note that this is the no-sway case, this is the
first sway case, this is the second sway case. You can find out the fixed end moments, do the
moment distribution, compute the member end moments and compute these.
Ultimately, you will see that this (Refer Slide Time: 52:13) plus this, plus this so you will see
that R2 double prime minus R2 prime minus R2 is equal to 0; we also have R1 prime minus R1
minus R1 double prime equal to 0. Note that this is a function of delta2 this is a function of delta2,
this is known, this is known, this is a function of delta1, this is a function of delta1. You have two
equations which are a function of delta1 and delta2 you can actually find out what those delta1
and delta2 are and then add the revised values of this and the revised values of this to this to get
the final moment. When you have double sway, you have to solve equations in the two sways
together. I just wanted to introduce the concept to you you do not have to solve a particular
problem; you can look at it yourself. Finally, just to state that when you have frames, there is a
11
likelihood of sway displacements; then the moment distributions procedure requires you to solve
multiple moment distribution problems to be able to get the final member end moments. I am
going to take up another problem in the next lecture, which will illustrate to you all these
procedures over and over again. I hope you have understood how to do the moment distribution
method for both beams as well as frames and hopefully by the next lecture, which will be my last
lecture on moments distribution, you should be able to understand the moment distribution
procedure completely.
12
Prof. P. Banerjee
Lecture 25
Good morning. We have been spending the last four lectures looking at moment distribution and
I had promised that today, I would take up a problem that would include sway. Last time, we
looked at a problem where the only load that was applied was the sway load and therefore,
computation of the no-sway reaction was relatively simple. This time, we will take up a problem
that will deal with the entire spectrum of issues that you have in using moment distribution for a
frame problem that has sway degrees of freedom.
Let us take one of the old problems that we have all ready looked at using the displacement
method. This was the problem that we had looked at long time ago and I am just going to redo
this problem. The advantage of this is that if we use the moment distribution method. We
already know the answer to this problem when we have used the displacement method.
Therefore, if the moment distribution method gives you the same result, it gives you confidence
that the moment distribution method actually does solve the problem in a similar manner. This is
your essential problem e is the same, material is the same. This problem can actually be done
in this fashion. If you remember, we had transferred that c load as a 50 Kilonewton load here
(Refer Slide Time: 04:23). By the way, this was 5 meters. This is 250 Kilonewton meter and then
you have all these other things. This is the problem that I am going to be actually looking at. In
this case, what is the first step? Well, you know that there are three degrees of freedom, these are
the two rotations and there is a sway degree of freedom.
The first step is essentially to solve the problem the no-sway problem. How does this problem
look with no sway? You essentially have. To stop the sway, I put a support here (Refer Slide
Time: 05:40) so that this does not allow this and essentially, we solve the entire problem to find
out this R. Now, first things first: we need to write down the fixed end moments this we have
already done earlier, so I am not going through the steps. We know that this is plus 22.2
Kilonewton meter, (FEM)ba is equal to minus 44.4 Kilonewton meter, then (FEM)bc is equal to
plus 288 Kilonewton meter, (FEM)cb is equal to minus 192 Kilonewton meter, (FEM)cd is equal
to (FEM)dc is equal to 0.
Another point here is you have to include the fact that (FEM)ce is equal to plus 250 the reason
for this is that although I have put the load here at this particular point, it is very important to
understand that the total joint reaction includes when you take the equilibrium of the joint, you
will see that this 250 Kilonewton meter note that I am putting this as plus 250, but this is
actually a load applied here (Refer Slide Time: 08:00) I can consider it as a moment applied
here and so that would be the moment that I need to consider in the joint equilibrium; that is the
reason why it is different from the other concept; the only thing is that ce will not take any this
thing it is not a member that takes any part in the stiffness.
When we compute the relative stiffnesses, ab is a normal member, so Kab is going to be I1 upon
15; Kba is equal to I1 upon 15; if you look at bc, we saw that the moment of inertia of bc was 4
times I, so we have Kbc as 4 times I1 upon the total length which is 20; Kcb is similarly 4 I1 upon
20; Kcd is I1 upon 15; Kdc is I1 upon 15; note that Kc is equal to 0 because it is a cantilever.
Therefore, this is included in the computation at c. There are two distributions that we have to
do: distribution at b and distribution at c. Therefore, you can see Dba is going to be 0.25, Dba is I1
upon 15 upon (I1 upon 15 plus I1 upon 5), so Dba becomes 0.250 and Dbc becomes I1 upon 5 upon
(I1 upon 15 plus I1 upon 5), which is 0.750; at joint c, we have Dcb is equal to 0.750, Dcd is equal
to 0.250 and Dce is equal to 0 these are the distribution factors.
Carryover factors this is a normal problem, so carryover factor has half from each side
excepting that when you have a carryover from a to b, it is 0 because a is the fixed end; carryover
from d to c is 0 because d is a fixed end; from b to a, carryover is half; from b to c as well as
from c to b, it is half, half and carryover from c to d is half these are standard things, so once
we have done that, we can now put together the no-sway problem. If you look at the no-sway
problem. Let me put it down in this fashion.
Again, being consistent over here, you will see that this is 1.0; this one over here is 1.0, then we
have over here (this is for ba) the distribution factor is 0.25; for bc, this is 0.75; again this one
is for cb, so it is 0.75; this is for cd and ultimately, I am going to put this for ce, but this is going
to be 0.0; the only thing is that here we have plus 250. What do we have here? We have plus
22.2, here we have minus 44.4, here we have plus 288, here we have minus 192, here we have 0
and here also, we have 0. I have written down all the fixed end moments and I have written down
all the distributed factors. Now, what I am going to do is I am going to do this in this fashion: I
am going to do the distribution at both joints b and c.
If you look at this, there is a plus 243.62, 243.6 of, there is a plus 243.6, so we have to do a
negative 243.6 and distribute it that is going to become equal to 182.7 and this is going to be
minus 60.9; this distribution is over, so done here (Refer Slide Time: 14:25). Do the distribution
over here: we get 250 minus 192 that is going to be 58 and 58 has to be distributed. If we
distribute 58 in 3 by 4, we get minus 43.5 (the this thing is plus 58, so minus 58 has to be
distributed) and this is going to be minus 14.5 and this is 0 distribution done, put it here. Now,
the carryover from here to here is going to be minus 91.4, carryover from here to here is going to
be minus 30.5, carryover from here to here is going to be minus 21.8, minus minus 21.8,
carryover from here to here is going to be minus 7.3 you have done all the carryovers.
Now again, we do the distribution and so I am not going to spend too much time on this I am
just going to put down the values and you know what the values are going to be. This is done, so
I can put a; then I do this over here; this is going to be minus 91 distributed, it is going to be
plus 68.6 and plus 22.8; here, we will have plus 11.4; this is distributed (Refer Slide Time:
16:26); here, we will have plus 34.3; there is a carryover here, which is going to be plus 8.2; we
have all ready done the carryover here; the final carryover of this to here is going to be plus 2.8.
Next, we again do this, it is going to be minus 8.6, then minus 25.7 close this; this one needs to
be done, so this is going to be minus 6.1 closed; this is also closed; this is 0; this is going to be
minus 2.1, close, 0.
4
You can keep doing this; I am not going to go any further you know the steps. I am going to
write down the ultimate value that you are going to have at this point. At this point, you are
going to have minus 108.8 at this end (Refer Slide Time: 18:04), here you are going to have plus
108.8, here you are going to have minus 9.4, here you are going to have minus 259.1, here you
are going to have plus 9.1, here you are going to have plus 250 and here, you are going to have
plus 4.6 these are the final moments that you get. When do you stop? You stop when your
distribution gets to be less than 1 percent of the smallest value. Is the smallest value over here?
The highest value is 288, so 1 percent would be about 2.8 and as soon as your carryover becomes
less than 2.8, you stop the entire process; the last process in a moment distribution should always
be the distribution it is just that you do not do carry over, which is less than 2.8 here in this
particular case. Once we have these moments, what is the next step? The next step is you put it
down on a figure.
What you have is. For each one, you know the member end moment; the member end moment
minus means it is going to be this way, so it is going to be 9.4; on this side it is going to be
clockwise so it is going to be 108.8 these are Kilonewton meter, I am not writing them down
and here, you have 20 Kilonewton meter. Understand that for this, I can find out the force over
here. If I look at this particular force, what do I get over here? If you take this moment and you
subtract this aspect of it, you are going to get that this force... this is this way, so this will be this
way, so it is going to be in this direction and this is my horizontal at a.
Similarly, I need to find out at d. You put it all together this is going to be 100, this is going to
have 108.8 here, here this is going to be 259.1, here this is going to be plus 9.1, this is going to
be plus 4.6 and understand that here, you have this and this and here you have a net moment of
250 so that they balance each other out here, these balance each other out automatically. All we
are interested in is actually finding out how much. This is going to be H at d. Let us calculate
what H upon a Ha is going to be.
Ha is going to have two components: 1 component coming from the moment, that is going to be
equal to that is this way, it will be this way, so it is positive, so it will be (108.8 plus 9.4)
divided by 15 and then you have the other load in the other direction and that is equal to 20 into
5 by 15. If you look at this, this turns out to be equal to 1.21 Kilonewton. Let us look at Hd: this
and we are going to have this, so Hd is going to be equal to (9.1 plus 4.6) upon 15, so this is
going to be equal to 0.91. Understand that if I look at this entire structure, Ha, Hd; Ha is equal to
1.21 and Hd is equal to 0.91, we have a reaction R here (Refer Slide Time: 24:00) and you have a
load of 20 Kilonewton here. If you look at that R, this is going to be R plus 20 plus 1.21 minus
0.91 is equal to 0. This implies R is equal to minus 20.3 Kilonewton. In other words, the loading,
the reaction is in this way.
If you look at that, what does that essentially mean? That means that actually the structure wants
to sway in this direction but you are restraining it and therefore you generate a 20.3 Kilonewton
force so R is equal to 20.3 Kilonewton force in this direction that is something that we have
done. Remember I had said that if R was 0, you would know that there is no sway, but here you
have R in this direction, which essentially means that the structure is trying to move in this
direction in the sway case, you have to consider the structure to be moving in this direction, so
let us see what happens.
This is swaying in this direction, so let us make it sway in that direction. You will see that this
joint will go here (Refer Slide Time: 25:49), this joint will go here and all you have is this; the
other part remains straight because it is a cantilever. If you look at this, if we say this is delta and
this is delta, what do we get? We get that the fixed end moment. This is a, b, c, d. (FEM)ab is
equal to plus 6 EI delta upon 15 squared, which is equal to (FEM)ba and you will see that is equal
to (FEM)cd also, which is equal to (FEM)dc. All of them have this. (FEM)bc is equal to (FEM)cd is
equal to 0.
Having done that, we can now say that that the fixed end moment this (Refer Slide Time:
27:06) let us say that this is equal to 100; if we say this is equal to 100, then we can say that 6
EI delta upon 15 squared is equal to 100 Kilonewton meter we are assuming it equal to 100.
We will find out what it actually is a little bit later, because we do not know what the delta is;
therefore we will find out the actual value of this a little bit later.
That is why I am saying that when you look at the sway problem, what do you actually do? You
put it equal to 100 so any one of them to be 100 and you can compute all other fixed end
moments automatically. In this particular case, you will see that since all the moments are equal
to 100, we will get that all fixed end moments are equal to 100 that is incidental in this
particular case. The next step is to find out your. The distribution factors remain the same the
distribution factors are what they are originally, so I am going to just put them down.
Here, 1.0, 1.0, 0.25, 0.75, 0.75, 0.25, 0. Here, what is it? (Refer Slide Time: 29:24) (FEM)ab is
plus 100 plus 100, 0, 0, plus 100, 0, plus 100 this our starting. Ultimately, if you look at it, the
first step, this is minus, so you will have minus 25, minus 75; here you will have minus 75,
minus 25 that is distributed. Next is 0 of course; next, minus 38, minus 38 this is the carryover
from here to here; from here to here, minus 13; from here to here, minus 13; so we are done with
the carryover.
Again, we just have this, so this is going to be one-fourth, so this is going to be plus 9.5 and this
is going to be plus 28.5; this is going to be again plus 28.5 and plus 9.5, 0; carryover is plus 4.8,
plus 4.8; and here carryover is going to be plus 14.3 plus 14.3. Keep doing this what is going to
happen? Ultimately, we are going to be doing the same process. This one needs to be distributed.
I am just assuming that you will be doing all the steps, so you continue with the steps it has not
ended here; you are going to have continuous steps here and these are going to be 0.
Ultimately, I am putting down the final value. The final value over here is going to be plus 81.8,
this is going to be minus 81.8; this is going to be equal to plus 89.9, this is going to be minus
81.8. Note that it is anti-symmetric, so you are going to have exactly the same values coming
through. This is going to be plus 89.9 and this is going to be 0. We have done the moment
distribution I have not gone through all the steps. Again, where do you stop? You stop when
the carryover is less than 1 percent of the largest original fixed end moment. The largest original
fixed end moment was 100 and therefore, as soon as the carryover becomes less than 1, you stop
the process. These are the final, I am just putting down the final ones that we have. If you look at
it, what does this mean? Having gotten these, we need to find out what is the moment that we get
over here.
We are essentially saying what kind of R1; here, this is like a load R1 which would do give us
this. This R1 we need to find out. How do we find out that R1? We actually put down the values
that we get here. This is going to be equal to plus 89.9, 81.8, here this is going to be 81.8, this is
81.8, 81.8, 89.9. If we put this in, this one turns out to be equal to 89.9 plus 81.8 divided by 15,
which turns out be 11.45.
Similarly, here we get 11.45. If you look at the entire structure which is this, we know the
reaction here and here and this is R1, you will see that R1 is equal to 22.9. The main thing is if R1
was equal to 22.9, then the moments over here that you would have was equal to 0, but
understand that we got our reaction at that particular point to be equal to 20.3; so essentially,
really, R1 is equal to 20.3, because the sum of the two together is the actual load that you have at
that particular point, which is equal to 0.
R1 should really be 22 that means all the sway moments have to be factored by 20 by 3 upon
22.9. Therefore, if you look at the sway moments, they turn out to be equal to. Mab is actually
given as 89.9, but in reality, it is this (Refer Slide Time: 36:43); Mba is equal to 81.8 into 20.3
upon 22.9; Mbc is equal to minus 81.8 into 20.3 upon 22.9, which is also equal to Mcb; you will
see Mab is equal to Mdc and Mba is equal to Mcd. When you add these with the non-sway
moments, what do the final moments turn out to be equal to?
You will see that you will get Mab. When you include both the sway and non-sway, the nonsway was equal to minus 9.4 plus 89.9 into 20.3 upon 22.9, which is equal to plus 70.3
9
Kilonewton meter (Refer Slide Time: 38:04). Mba is going to be equal to minus 108.8 plus 81.8
into 20.3 upon 22.9 this is equal to minus 36.3 Kilonewton meter. Similarly, Mbc is equal to
plus 36.3 Kilonewton meter. If you add them up, you get Mcb equal to minus 331.6 Kilonewton
meter (259 and plus). Mcd is equal to 9.1 plus 81.8 into 20.3 upon 22.9 that is equal to plus 81.6
Kilonewton meter. Mdc is equal to plus 84.3. Once we have these, then my final member end
moments turn out to be.
This is the final actual structure that I am going to put down. This is 50 Kilonewton, here we
have 250 Kilonewton meters and 50 Kilonewton shear, so you will get just the opposite on that
side. This one is plus 70.3 Kilonewton meter; this is going to be minus so it is 36.3 Kilonewton
meter; I have this load here which is 20 Kilonewton meter; if I look at this and add the two up I
get 8.94 Kilonewton; on this one, I get plus 84.3 and I also get this to be 81.6 Kilonewton meter;
if you really look at it 160 divided by 15, you get 165.9 divided by 16 you will get it equal to
11.06 Kilonewton meter and over here, you will get this equal to 11.06 here, you will get 11.06.
Let us just go through the steps. This will be like this (Refer Slide Time: 42:26) and the moment
will be like this. What do we have on this one? You will see that I have a 100 Kilonewton force
acting over here. What are the moments? Mbc is plus 36.3. Here, this is going to be this way
minus 331.6. I have got the moments down and let us go through the procedure. Here, this one is
going to be acting in this direction, this one is going to be acting in this direction; over here, you
have this and you will have this; over here, you will have this; over here, you have this and you
will have this. Over here, you will have this, so over here (Refer Slide Time: 43:42); this is the
opposite. If you look at it, this plus this 11.6 will give you net Fx equal to 0 which stands to
reason.
If you look at this one, this is going to be this way, this is going to this way just check it out;
this is 331, this is 250 and 81.6; 250 plus 81.6 is 331.6, which is balanced so there is moment
balance, there is Fx balance, now all we have to do is do Fy balance. In this particular one, you
have a net moment in this direction, so this is going to give rise to a force in this direction, so
10
that is going to be equal to (331.6 minus 36.3) divided by 20 this is going to be a force in this
direction.
If I am taking force in this direction, this is this way plus I have minus (100 into 12) upon 20
this is my net moment. If you look at this, the net force over here is going to be in this direction
and it is going to be equal to compute it. This is going to be this (Refer Slide Time: 45:27),
now you can get this one, this one is going to be equal to your this is your Vbc and your Vcb is
going to be equal to 100 minus Vbc; you can compute this, you can compute this and you will see
that they all balance each other out.
Ultimately, if I were to look at the support reactions, this is going to be equal to 60 and this one
if you look at, it is going to be equal to. let me do this over here. This is going to be equal to
4.7, so this is going to be 295.3 upon 20, this is 60, so this is going to be equal to 60 upon
here, you have 14.8, so 14.8, 60 is going to be equal to upwards 45.6; this is going to be upwards
54.4. Over here, what do you have? The support reaction is 45.6 Kilonewtons, reaction: here, it is
going to be 8.94 Kilonewtons, moment is going to be equal to 70.3 Kilonewtons; over here, the
moment is going to be 84.3 Kilonewtons, this horizontal force is going to be 11.06 Kilonewtons
and the vertical force is going to be equal to 104.4 Kilonewton.
This balances everything out. Ultimately, if you look at the total sigma Fx equal to 0, you will see
80.94 plus 11.06 minus 20 it is checked; sigma Fy is equal to 0 45.6 plus 104.4 is going to be
150 minus 100 minus 50 is 0 check; take moment about A and you will see that you will get it
equal to 0 check. This in essence finally tells you that you have a situation in which you do
have to solve a no-sway moment distribution; by solving the moment no-sway moment
distribution, you get the no-sway moments; from the no-sway moments you compute the
horizontal reactions and you compute the total reaction required to ensure that there is no sway.
Then, the next step is to assume that there is sway and compute the load required to give you the
sway that you get; now that sway, that reaction, the total sum of the two reactions has to be equal
11
to 0 and because of that when you put it all together, you finally get both; you add the sway and
non-sway moments.
These are the non-sway moments (Refer Slide Time: 49:39), these are the sway moments
corrected for the actual values. Then, once you get that, you have got the actual moments due to
both the sway and actual structure and ultimately if you look at this solution that I have given to
you, it is identical to the one that we computed using the displacement method; so essentially, the
moment distribution method, if applied using sway and non-sway distributions together, you will
get the final solution. Now, let me give a particular problem; I will give you a specific problem
and also give you the answers so that you can actually compute what you are going to have.
12
This is an old problem that we have looked at. It is just a little bit of a different loading that I am
putting on it. Here, I have 10 Kilonewton and I have a UDL; this UDL is 4 Kilonewton per meter
length of bar. Here, this is 2EI, this is EI, this is 3 meters up to here (Refer Slide Time: 51:25).
Here, let me tell you that what you have to do is restrain this so that there is no sway and this is
the R that you consider for the no-sway. In other words, for the no-sway you make this into a
hinge.
I will give you the final answers to this particular problem.
When you solve it using the no-sway case, you get R equal to 0.0854 Kilonewtons and it is
minus. In other words, it is actually R equal to 0.0854 Kilonewton going up. Therefore, when
you do the no-sway case, I am going to give you the final answers that I have and find out if you
can solve it. If you can solve it, you know that you have done it correctly. You get 2.408
Kilonewton meter here (Refer Slide Time: 52:59), you get 1.77 Kilonewton here, 20 Kilonewton
here, 11.77 Kilonewton here, the load is 10 Kilonewtons here and that is it that is the solution
that I want you to get. Is that clear? I hope you will be able to solve this particular problem. That
brings me to the end of the moment distribution method as applied to beams and frames beams
and planar frames.
Thank you very much. From next time onwards, I am going to be starting off on mid-fix
methods.
13
Prof. P. Banerjee
Lecture 26
Good morning. Just to review what we have done till now we have looked at the force method
of analyzing statically indeterminate structures, then we have looked at the displacement method
for analyzing structures and finally, we looked at the moment distribution approach, which was a
iterative approach to solving structural analysis. Now, I am going to change gears and move into
what we call as the matrix methods of structural analysis. You will see that the matrix methods
of structural analysis are nothing but the force method and the displacement method written in a
slightly different format. Today, I am going to be discussing how to use flexibility approach.
There are two approaches in matrix methods: one is known as the flexibility approach and one is
known as the stiffness approach. Today, I am going to be discussing the flexibility approach to
structural analysis this is the matrix method.
Flexibility-based matrix methods what exactly is the flexibility approach? Let us now look at a
statically determinate structure. This is a structure and you can see that there are six equations of
motion, four unknown reactions and two forces so it is a statically determinate structure. This
is subjected to loads R1 and R2. Note that in a statically determinate structure, you do not need
any method to compute the forces in the members, because the forces in the members can always
be computed using equilibrium.
However, I am going to state this problem a little bit differently I am going to say find out the
forces in all the members and find out the displacements corresponding to the degrees of
freedom. Here also, you will see. How many degrees of freedom does this structure have? You
1
will see that there are three joints and you have two degrees of freedom per joint in a truss,
remember? Two degrees of freedom: one translation in the horizontal direction and one in the
vertical direction, so two degrees of freedom per joint. You have four restraints, two hinges, so
the number of degrees of freedom are two; and these are the displacements: horizontal
displacement and the vertical displacement.
Let us say that I have this as (EA)1 and this as (EA)2; this is length1, this is length2 Anything
else? You have got the complete information. The problem statement here is that given loads R1
and R2, find a) this is member 1 and this is member 2 find forces in members and b)
displacements R1 and R2 these are the problem statements. How would you do this in the
normal case? You would find out the forces directly in terms of R1 and R2 you can always find
out the forces because all you need to do actually in this particular case is just take equilibrium of
joint b. These are the two unknown forces and take sigma Fx is equal to 0, sigma Fy is equal to 0
and you can find out the two forces member forces are very simple; so finding out the forces
and members is just equilibrium.
How would you find out displacement R1 and R2? Corresponding to R1, if I am applying the
principle of virtual work, I would apply a virtual force equal to 1 corresponding to R1, then find
out the virtual forces and then take internal virtual work and external virtual work you equate it
and you can find out R1. These are the steps that you would follow in a normal kind of a
situation. Let me just write that down in a particular manner.
Using equilibrium, you can actually find out the force in member 1, it is a function of R1 and R2.
Similarly, the force in member 2 is also a function of R1 and R2. These two you can find out
using basic equilibrium. Let me say I am going to use start using notations I am going to stick
through these notations all the way through. This is slightly different: I am going to say this is
equal to S1 and this is equal to S2. Up till now, I have been using force in member 1, force in
2
member 2. I am just defining a notation and I am saying that let me call the force in member 1 as
S1 and member 2 as S2. Then, I can say that Si can be given in terms of bi into R. What does this
mean?
This essentially means that S1 is equal to b11 R1 plus b12 R2 and S2 is equal to b21 R1 plus b22 R2.
What are these coefficients? b11 or I will put it as bij is member force in member i due to a load
Rj is equal to 1. In other words, the member bij is defined as the force in member i due to a load
Rj is equal to 1 this is the definition of bij. In other words, I can write this in this format: I am
using a matrix. If I write S as a row vector containing S1 and S2, I can actually write it as think
about it b11, b12, b21, b22 into R1, R2. This is the vector of member forces, this is the vector of
loads on the structure and this is the matrix which actually is a force relationship that can be
evaluated using equilibrium. Think about how will I evaluate this vector? Think about it. How
will I find out this vector? I could find out this vector by this structure applied to a unit load and
the force so let me put it this way: this is equal to b11 and this is equal to b21.
Therefore, in this matrix, each vector, each column vector can actually be found out by solving
through equilibrium: putting a load equal to one corresponding to the load R1 and unit load and
finding out the member forces by taking equilibrium of joint b. You can find out the member
forces these member forces are denoted as b11 and b21, so you can find out this. Then, how
would you find this out? You could find this out by taking the same structure, applying a unit
load and finding out the member forces. This would be b12, b22. What we are doing here is really
saying that I do not care what load there is on the structure. In this particular structure, what are
the possible applied loads? You will see that the number of applied loads that you will have is
always equal to the degrees of freedom and each load is applied corresponding to a degree of
freedom.
Look at it how many of the loads can you apply on this structure other than these two,
corresponding to the degrees of freedom? Whether you have these loads or not is not relevant.
You might have a situation where only R1 is not equal to 0 or R2 is not equal to 0 and R1 equal to
0, or you may have both R1 and R2. You see, the procedure that I am developing does not depend
on whether a load exists on the structure. These values you can find out, but this is independent
of what these values take up because these are. What this is? These are the member forces due
to R1 is equal to 1, these are member forces due to R2 is equal to 1 and therefore, this is very
simple you can find these out; normally, you know these displacements, so you can always find
out the member forces. Therefore, all I am saying is that I am using equilibrium; however, even
though I am using equilibrium, I am saying that I am writing down the member forces in terms
of the loads and therefore I can always write it in this fashion.
I am going to continue in this fashion; I am now going to put it together; I have just put it
together. This is obtained for all I, so for every member, you can actually compute this; in this
particular matrix, since this is a 1 by 1 in a truss and this is n by 1 depending on how many
degrees of freedom there are and you will see that this is equal to 1 by n; if this is a column
vector, this is a row vector, row vector into column vector gives you a scalar.
We can evaluate this. This can be found out bi can be found through equilibrium. This is the
actual load, these are the actual member forces (Refer Slide Time: 17:48). Now let me find out
what the member deformations are. If I look at deformation in member 1, what is member
deformation? If Si is the load, then you will see that it is Si into Li upon (EA)i. What is Si upon
Ai? Si upon Ai is the uniform stress in the member, then Si upon (EA)i is the uniform strain in the
member and uniform strain integrated over the whole length essentially since you have a
uniform frame, the strain integrated over the whole length is going to be just strain into Li and
this is the deformation in the member.
I have been using till now delta; however, I am going to use consistent formulation, so I am
calling this v1 v1 is the axial deformation in the member 1 due to the load S1. This is the
deformation in the member due to the loads. What is v2? Deformation in member 2 is equal to S2
L2 upon (EA)2. Note that we are just saying the lengths could be different and EAs could be
different; whether they are different or not actually depends on a particular structure. Here, note
that I am not making any assumption excepting for the fact that the structure is a statically
determinate one, so that I can relate the member forces to the only assumption till now, which is
that it is statically determinate and of course that the members are uniform. These are the only
two assumptions: statically determinate structure and uniform members those are the only two
assumptions that I have made. If I can do that, I can find this out and I can find out deformation
in the member.
4
I am going to call this as vi, which is equal to Li upon (EA)i into Si I have just rewritten this
particular thing. Now, think about it: what does this relate to? This actually is a relationship
between the member force this is the force deformation relationship or if you want to call it
you can call it deformation force relationship. What does this particular quantity signify? This
quantity (Refer Slide Time: 21:35) is the deformation in a member due to a unit member force
deformation in member i due to unit Si; this by definition is called as the flexibility of member i.
What is flexibility of member i? It is equal to Li upon (EA)i. What is the flexibility? What is the
physical definition of the flexibility? The physical definition of the flexibility is the deformation
in member due to a unit member force. Once we have that, this implies that vi is equal to fi Si. In
this particular case, vi happens to be 1 by 1, Si happens to be 1 by 1 and therefore, fi is also 1 by
1, but I can actually relate it in this fashion: if I have more than one member force, then the
deformation corresponding to that force is going to be related through the flexibility matrix or in
this particular case, the flexibility coefficient I am just going to keep it in this fashion. Now
since this is true and I can find this out and note that I know this because I know this and this
loading in normally given, I can find this out. If I substitute this into this, what do I get?
You see that I get vi is equal to fi into bi into R I have just substituted Si equal to bi into R into
bi equal to fi Si; that is all I have done. What does this relate actually? In this particular case,
since I am dealing with a truss member, this is a 1 by 1, this is a 1 by 1, this is 1 by n and this is n
by 1. Here, the major point is. What does this give me? This gives me the member
deformation in terms of the member loads, where this is given by the force deformation
relationship of a member and this is given from equilibrium.
Now, what is the next step? Note what my ultimate goal is: given these loads, what are R1 and
R2? Note that I have already found out the member forces the first thing that I did was find out
the member forces, because they are related in terms of bi and you can always obtain bi by
5
actually solving the equilibrium equations for this statically determinate structure. The next step
is I am trying to find out R1 and R2. How would I find out R1 and R2? Well, we know that due to
this load, we have real deformations in the members. If I use the concept that I have these and I
want to find out R1, what would I do? I would apply a unit virtual force corresponding to the
degree of freedom R1, so let us do that: unit multiplied by R1. Note that since I have only applied
a unit virtual force corresponding to R1, the virtual force corresponding to R2 is 0 and therefore,
that will not do any work, so this is equal to the external virtual work.
What is internal virtual work? Internal virtual work is going to be equal to the Si these are the
virtual forces due to the unit applied load corresponding to R1 multiplied by vi (I am continuing
the fact that they are scalars). What does this give me? Virtual member force times the
deformation this is the work done in each member; this summed up over all the members is
going to give me the internal virtual work. What do I have to now compute? Now, I know this,
because I know this, this, this and I know this (Refer Slide Time: 27:07), so I know this. Now the
only thing I need to find out is the member forces due to a unit virtual force corresponding to R1.
Let us go back. Which was R1? R1 was this horizontal degree of freedom.
What is my force? Virtual force is this. What are the virtual member forces? The member forces
are these (Refer Slide Time: 27:41) these are the same. Whether this force is real or virtual, it
makes no difference the member forces are going to be identical. I know what those are, so I
just need to substitute those. What are those? These are equal to in this particular case b11 into 1
into v1. This is the virtual member force due to the unit moment times the deformation plus b21
into 1 into v2.
Think about this. What is this? This essentially means that this r1 is equal to b v1 into this thing.
The next equation; I need to find out is r2. I apply a unit virtual force corresponding to r2 and that
is my external virtual work. My internal virtual work is again going to be Si into vi, but this Si is
due to this degree let us see what that is. Look at this: R2 is this degree of freedom, so if I apply
unit load, it is going to be b1; whether this is virtual or real, it makes no difference, it is going to
be this; only thing is if this is virtual, these are virtual forces. If you look at this, this implies that
r2 is equal to b12 into 1 into v1 plus b22 into 1 into v2. If I write this, do you notice something? I
can write these two equations in matrix form. What would that become?
You will see that this implies r1, r2 I am just writing in a matrix form b11, b21, b12, b22 into v1,
v2. Note that I have all I have done is written these two equations in a simultaneous in a matrix
form that is all I have done; but note something: what does this look like? Remember that I had
Si is equal to bi R? You can look at this. This you will see is going to be equal to and if I take
Si equal to or I can write it as S equal to b R where S since I have two, if you look at it,
this is what I have written this one.
S is a 2 by 1 because I have two members. This is b into R; b is b11, b12, b21, b22; so in this
particular case it is 2 by 1 (Refer Slide Time: 32:08), 2 by 1, 2 by 2. What is b? It is equal to b11,
b21, b12, b22. If you look at this, what is this equal to? You will see that this is nothing but equal to
b transpose. This is very interesting: S is equal to b R, but if I use virtual work I get that r is
equal to b transpose v, where v is a 2 by 1 each member in the column vector corresponds to
the displacement and this is a 2 by 1, this is a 2 by 2. Virtual work gives this, this is equilibrium,
this is from virtual work. You notice something very interesting and now, what I am going to do
is I am going to now split it up. I am going to split it up and you will see that this can be written
in this format: r is equal to summation over all (bi transpose vi). What is this now (Refer Slide
Time: 33:54)? This is the deformation in each member. Let me see what this implies.
Here, it means that r1 is equal to b11 bi transpose; bi transpose would be equal to. What is bi? bi
is b11, b12, so this is going to be equal to. r11 is going to be equal to b11 v1 plus b21 v2 and r2 is
equal to b21 v1 plus b22 v2 this is a summation (Refer Slide Time: 35:21). If I put this, let us see
what happens here. What is vi equal to? It is equal to fi bi into R (Refer Slide Time: 35:37). I am
now going to plug that in and this r will then be equal to summation over all the members (bi
transpose fi bi) into R.
Note that R is outside, so actually, the thing that you have is this (Refer Slide Time: 36:09). Very
interesting, very interesting. This part comes from equilibrium, this is Si equal to bi R, then fi into
Si is vi and then, from virtual work, we know that r is equal to bi transpose vi; therefore,
ultimately, when you put it all together you get this. This means that this particular term is
summed up over all is. What is this term actually? What is this term totally? It relates the
displacement to load. This represents something like that if these were one, what would these
be? These would be the displacements given unit load, so this matrix I can call it as Fi and
summed up over all i is equal to F and I can write it in this fashion, where this is now the
structure flexibility matrix and this is the contribution of member i to F.
Now, I can find out the displacements because I know this (Refer Slide Time: 38:15), I know
this, I know this so I can find out this and summing it up over this, I can find this. Once I have
this, note one thing: nowhere do I need to find out anything. Once I have found this out, given
the load, I can always find out the displacements of the structure this is the beauty of the matrix
method. Once you go through the steps the steps are identical to what we have done earlier
using the virtual work principle, we have already discussed the virtual work principle; the only
thing that I have done is explicitly write down the virtual work, first equilibrium I have just
related it in a general format and ultimately I have come up with an equation that directly relates
the displacements and the loads through the structure flexibility matrix. This is the reason why
this is known as the flexibility approach matrix method because ultimately you really are finding
8
out the structural flexibility. And if you look, along the way you are actually finding out the
member flexibility term.
This entire procedure is set up very very easily. Of course, what is the assumption? The
assumption is that I have a statically determinate structure and of course, it is a truss-type
structure because the only forces in the members that I have written till now are truss because
axial force is only axial forces and axial deformations. Let me just illustrate this and then, I will
show you next time that exactly this same thing can be written down even for a statically
indeterminate structure. Let us see how we can find out the member forces by applying it to an
actual example structure.
Let us have a situation. This here is 4 meters, this is 3 meters, this is 3 meters, this is 4 meters.
Both of them have the same EA. Let me just put some loads here: 10 Kilonewton, let me apply
20 Kilonewton. The problem statement is to find the member forces and find displacements r1
and r2 this is what the problem statement is. This is a statically determinate structure, so I can
actually find out the member forces directly, so let me find out the member forces. The only
difference is I am not going to find out the member forces directly for the given load I am
going to be using the matrix method.
If I use the matrix method, then I do not solve the structure for the loads directly. What I do is I
put R1 is equal to 1 and find out the member forces. Let us see what I get. This is a 4, 3 slope and
this is a 4, 3 slope this way. Let me call this a, b, c. If I take sigma Fy is equal to 0 at joint b, you
will see that Fab which is my member 1 and bc is member 2 Fab and Fbc I will call this in my
notation as S1 and this as S2. These are the forces in the members and note that member forces
are positive when you have tensile forces that is the assumption that we make and we are going
to continue making that (positive tensile forces negative).
Once I do that, let us see what happens. The vertical component of S1 is going to be 3 by 5 S1.
Since this is positive, this is positive, you will see that S1 is downwards the vertical component
plus 4 by 5 S2. Since there is no vertical force it has to be equal to 0. What does this give me?
This gives me that S2 is equal to minus 3 by 4 S1. This is sigma Fy is equal to 0; sigma Fx is equal
to 0 gives me that minus 4 by 5 S1 (that is the horizontal component pulling in this direction)
plus this direction 3 by 5 S2 plus 1 is equal to 0, but note that S2 is equal to 3 by 4, so this
becomes 9 by 20 S1, 9 by 20 minus 16 by 20 is going to be minus; this implies minus 7 by 20 S1
plus 1 is equal to 0; this implies S1 is equal to 20 by 7 and S2 and this minus 3 by 4, this is going
to be minus 15 by 7. Satisfy yourself that this indeed does satisfy: take 3 by 5 of 20 by 7 it is
going to be equal to 12 by 7; take this minus 12 by 7 is equal to 0; and here (Refer Slide Time:
47:55), take minus 4 by 5 this is going to be equal to minus 16 by 7 plus 9 by 7 and that is
going to be equal to minus 7 by 7 so that is minus 1 plus 1 is equal to 0. But note that this is S1
due to R1, so what is this? This I have found out that b11 is equal to 20 by 7 and b21 is equal to
minus 15 by 7; these are what I have found out.
10
Let us move on to the other one, which is R2 is equal to 1 and let us find out this. I am going to
directly put that Fab if I put R2 is equal to 1 is actually b12 and Fbc is going to be equal to b22. If I
write it in this fashion, we will see that this is 3, 4 and this is 4, 3. sigma Fx is equal to 0 gives me
that 4 by 5 S1 (this is a horizontal component) minus 3 by 5 S2 equal to 0. This implies that S2 is
equal to 4 by 3 S1 this is sigma Fx is equal to 0; sigma Fy is equal to 0 gives me 3 by 5 S1 plus 4
by 5 S2 is equal to 1. If I plug this in, I get this is 4 by 3, so this becomes (3 upon 5 plus 16
upon 15) into S1 is equal to 1. Ultimately if you look at it, if you put these through, 9 and 25, 25
upon 15 is 5 upon 3, so that means S1 is equal to 3 by 5, that means b12 is equal to 3 by 5 and b22
is equal to plug it in and you will see that this is equal to 4 by 5. The horizontal vertical
component of this is going to be 12 by 5 and the horizontal component is 12 by 5 they cancel
each other out; the vertical component of this is 9 by 25, this is going to be 16 by 25, 9 plus 16
by 25 is 25 by 25 is 1, so this is okay. Ultimately, I have found out the b1.
11
Therefore, now I can say that S1 is going to be equal to b11 R1 plus b12 R2 and S2 is going to be
b21 R1 plus b22 R2. I know what R1 and R2 are, I am given R1 and R2 R1 is equal to 20 and R2 is
equal to 20. If I put in it, I can get my S1 and S2 as equal to 20 by 7, minus 15 by 7, 3 by 5, 4 by
5; this is 20 and this is 10 (Refer Slide Time: 53:08). From that, you can find out S1 and S2. The
first part of the problem is done and the second part of the problem I can actually find out; both
the members are 5 meters long.
12
Therefore, f1 is equal to 5 upon EA, f2 is equal to 5 upon EA and therefore, the contribution of
the first member is going to be equal to bi1 b12 so that is going to be equal to in this way 20 by 7,
3 by 5. What is this? This is equal to b1 (Refer Slide Time: 54:26), so F1 is equal to b1 transpose
f1 b1; this is going to be equal to 20 by 7, 3 by 5, 5 upon EA 20 by 7, 3 upon 5 and you can find
out F1. Similarly, you can find out f2.
Once you find out F1 and F2, then you have F is equal to F1 plus F2 and finally r is equal to F into
20, 10 and you have got your displacement.
We will continue with this particular problem in the next lecture and then I am going to extend
this to see how we can use the same procedure to solve statically indeterminate structures. We
are going to be using the flexibility approach to solve statically indeterminate structures also. I
hope I have been able to give you a brief background on the matrix method flexibility approach.
You will see that there is nothing new in what I have talked; it is all exactly what I have talked
before; it is just that I am using notation and putting it into a matrix format and then using matrix
algebra. I suggest that if you are uncomfortable with matrix algebra, you go back to any standard
matrix algebra book and study the concepts in the matrix algebra book. I am going to bring out a
few points in the next lecture, which will show that for the kind of flexibility matrix that I have,
there are specific forms of the flexibility matrix more on that in the next lecture.
Thank You.
13
Prof. Dr. P. Banerjee
Lecture 27
Good morning. We have started discussing about the matrix approach for structural analysis and
in the last lecture, we looked at the basic concept behind the flexibility approach for a statically
determinate truss. Today, we are going to continue that. Please understand that in the last lecture,
the problem that I looked at I was just looking at it; I did not solve the entire thing purely
because I was just explaining the concept; later on, we will have enough opportunities to look at
problems. Right now, the way I look at all my lectures is that I first look at all the concepts, use
problems only as illustrative purposes to explain the concepts and then, once the concepts are
described, I move on to solving problems in their entirety so that you know how to use the
method in applying it to problems. Today, we are going to be looking at the flexibility method
for beams and frames.
The essential difference between what we did for truss last time and what we are doing for
beams and frames is that they are flexural members as opposed to axial members in other
words, they are flexural deformations. Just like we saw that in a truss, you started off by looking
at the truss structure and then looking at the member forces in terms of the external loads and
from there, we moved on to solving the structural problem.
Now, let us look at a member a member that has flexural deformations. Now, what are the
member end deformations? We have already done this I do not want to go into this. This is
going to be theta1 and this is going to be theta2; these thetas are essentially due to moments M1
and moment M2. In addition to this, there may be thetas due to loads on the member; so if we
write down this, we will see that we have already written this down, I am just explicitly writing
it down theta1 is equal to L upon 3EI M1 I am actually going back to the old I am just
rewriting it in this form, and then plus theta10. What is the theta10 due to? Understand that theta1,
the rotation here would be related to M1, M2 as well as theta10, which is the rotation at the left
end due to the member loads; so this aspect is due to member loads. Similarly, theta2 is equal to
minus both of these are due to member loads. If there are no member loads, then these two are
0 this we have already seen. Now I am going to rewrite in this format. This is the format that I
am going to write it down in. Let us see what these terms are.
If you look at these terms, vi is equal to theta1 and theta2, so in other words, the member
deformations are the rotations at the two ends, these are the member deformations. I am using the
same notation that we are going to be consistently using in this approach. What is S i? Si is equal
to M1, M2. We know that vi0 is theta10, theta20 and vi0 is due to what I will call in a broader sense
member effects because we have already seen that rotations can happen not just because of
member loads but they can also happen due to temperature etc., etc. we have already discussed
that, so I am not going to go into the details. Finally, what is fi? If you look at it, it is L upon 3EI,
minus L upon 6EI, minus L upon 6EI, L upon 3EI. If these are defined in this fashion, then this is
this is valid because this we know is valid from earlier. This is what we are going to be going
with this is one relationship. The other relationship is.
What is this relationship? If you look at this, all it says is how we get the member end moments
corresponding to the loads apply at joints. Remember that loads applying on members are
already considered in the effect that we considered in vi0, so this is just member end loads and so
if you really substitute it in here, we will see that ultimately, we get displacements in exactly the
same way. I am not going to go into the details but once everything is set up, you will see that
this is equal to the same terms the way I wrote them. This is exactly valid because once you use
this, substitute the value of Si into vi and all those things, you will see here that this is one thing
and then there is plus one additional term sigma over i bi transpose vi0 this is a very important
term. These are member end rotations due to the loads applied on the members themselves.
Therefore, by using virtual displacement, you will see that r can be related in this form.
This in essence is the final form of r: r is equal to summation. Here, for a flexural member, it
is 2 by 2 for flexure, it is 1 by 1 for axial. If this is n by 1, then by definition, this is going to be
2 by n or 1 by n depending on whether in a flexural member you are finding out the member end
moments or in an axial member you are trying to find out the actual force in terms of this. This
correspondingly is going to be either n by 2 or n by 1, this is going to be equal to a 2 by 1 or a 1
by 1. Now, I have this question, I did not introduce this in the truss but I am introducing this
here: what is this effect? For a flexural member, we know what this is; this is due to member
effects, for example, member loads, temperature etc., etc. In a truss, where would this come
from? This would again come from member effects. What are the member effects that we have
already considered? These member effects may be temperature or elongation or contraction due
to temperature change or lack of fit remember that.
This is member effect and this one again is either n by 2 or an n by 1 and ultimately this is an n
by 1. This, in essence, defines the flexibility approach for structural analysis. How do we find out
the member forces? Member forces in terms of the external loads are this (Refer Slide Time:
14:14), so these are the member forces. Remember that all of this is for a statically determinate
structure. This in essence defines these are the member forces related to the structure loads and
these are the structure displacements corresponding to the loads. This, in essence, is the
definition of your structural analysis: find out the member forces and find out the deflections.
Now, this again is for a statically determinate structure. Once we have this, let us see how we can
apply it to a member and for a beam, it does not make sense, so I am going to take a frame.
Let us take a frame problem and see how we deal with this; of course, the frame has to be
statically determinate. Again, I am not going to go through the entire problem I am just going
to list out how to get what we want to. Now, let us put some loads on this structure, let me put
some numbers here. This is EI, this is EI, this is EI. Let us say that this is 5 meters this is 5
meters, so this is 5 meters all of them are 5 meters. This load is let us say 40 Kilonewton.
Here, what we want to find out is There are lots of things to find out but let me just tell you
that I want to find out the member end moments, because once I know the member end
moments, I can draw the bending moment diagram, so I want to find out the member end
moments and in this particular case, I want to find out how much let us put some other
number, let me put something else, let me put 10 Kilonewton here and I want to find out what is
the displacement of this joint.
In other words, the lateral note that since all members are axially rigid, automatically the
horizontal displacement of this will also give me the horizontal displacement of this (Refer Slide
Time: 17:55). In addition to that, let me find out how much this goes by. These are my two
questions: how much does the top displace by and how much does this this is my overall
statement and I also want. Therefore, the question is find member end moments. Once we find
out member end moments, you can actually draw the bending moment diagram. What else? The
values of r1 and r2. This is my problem statement. Now, I have to understand how I am going to
go about this. Look at this and you will see that this is essentially a statically determinate
structure, so let us see how we are going to go about it. Now, in the first step in this problem,
what do we find out? For each member, we find out its member end displacement. Let me
qualify this; I will call this member 1, this member 2, this member 3 and this is a, b, c, d; ab is
member 1, bc is member 2, cd is member 3.
The first step: for each one, we write down the member force deformation relationship for ab,
what is this thing? There is no load and therefore, vi, which are the two rotations at the two ends,
are equal to I am going to write this in this format L upon 6EI and L in this particular case is
5 and note see, 2 L upon 6EI is upon 3I, which is what you have, multiplied by Si, which are
Mab and Mba. Since there is no member load, for ab (which is member 1), v 1 is this. For member
2, this has a load of 40 Kilonewton acting here, so the first thing that we have to do understand
that the first part is this, but there is an additional vi0, v20 that we have to write down.
How do we find out V20? Under this, it is going to deflect like this and we need to find out this
(Refer Slide Time: 21:55) and this. How do you find that out? I think we have done it enough, so
I am just going to write down how to find it out. This is M, this is m1 and this is M2. Once we go
through the entire process, we see that theta10 is going to be equal to in this case 50 into 5 by 4EI;
by the way, minus because these are opposite to each other, so 50 into 5 by 4EI is this part and
for that, it is going to be equal to 2 by 3, then you will have minus 50 into 5 by 4EI into 1 by 3,
so essentially, theta10 is equal to minus 62.5 by EI and theta20 is equal to 62.5 by EI. Actually if
you look at it, it will work out to p L squared upon 48EI but that does not matter, we have
computed this.
For b2, it is going to be equal to 5 upon 6EI into 2, minus 1, minus 1, 2 into Mbc, Mcb plus 62.5
upon EI into minus 1, plus 1 and since there is no load on it, v3 is going to be equal to 5 upon 6EI
into 2, minus 1, minus 1, 2 into Mcd, Mdc. We have written down the force or deformation force
relationship for each member. This part is the f into Si and this part is the vi0; for 1 and 3, there is
no vi0 but for 2, there is a vi0, which we have computed.
Now, the question comes: what next? Next is evaluating Si this is very very important. How do
we find out this Si into bi R? What is R? In this particular case, what we are finding out? We are
finding out small r is equal to r1 and r2 those are the displacements at the top and at the hinge.
8
This is this way. Therefore, capital R will be R1 and R2 corresponding to the forces. If you look
at this particular concept in this particular case, what is R equal to? It is irrelevant what we are
getting. R you will see will be equal to 10 and 0 in this particular case. How do we find out bi?
What we do actually is this. This is 2 by 1 and this is 2 by 1, this by force has to be 2 by 2. Now,
this 2 by 2 has two columns. For each column, the way we find out the column vector is this: put
R1 is equal to 1 and find out the member end moments; put R2 is equal to 1 and find out the
member end moments.
If we look at that, this becomes 1. Then, how would we do it? If you look at this, you will see
that this gives rise to 1. The reactions: sigma Fx is equal to 0, sigma M is equal to 0, take
moments about this point and then sigma Fy is equal to 0 gives me all of these. Once I have all of
these, my member end moments become this. How I find out is The first column of b1, I am
finding out the first column for b1 what would that be? Let us look at it. If I look at this, I can
draw the bending moment diagram. How would the bending moment diagram look? It will look
like this (Refer Slide Time: 30:11). This would be 5 here, this would be going in this fashion and
on bc, it is going to be like this, 5 and 0 here, this is going to be and over here, it has to be 0
because this 1 does not take any moment. If you look at b1. At this Mab, what is the moment?
0, so this is 0; at this point, what is the moment? Anticlockwise 5, so anticlockwise 5 is plus 5.
Let us find out b20. It is clockwise at this end it is clockwise 5, so this is minus 5, here, it is 0,
so it is going to be 0. What about b3? You will see it is equal to 0, 0 because there is no moment
at this end. This is the first part for R1 is equal to 1.
Then, we are going to put R2 is equal to 1; let us put R2 is equal to 1; I am now putting R2 is
equal to 1. If you look at this, this will give rise to this and these two are going to be equal to 0.
How will the? This is 5, 5, 5. This is the bending moment diagram and therefore, if you look at
b1 due to R2 is equal to 1, this is going to be equal to 0 and this is again anticlockwise, so it is
going to be plus 5; b2 is going to be this is clockwise, so it is minus 5, here it is anticlockwise,
so it is plus 5; b3: here, it is clockwise, so it is minus 5, here, it is 0.
Ultimately, having done these, if we write down b1, this is going to be equal to 0, plus 5 this is
coming from R1 is equal to 1; the next one is coming from R2 is equal to 1. Similarly, R1 is equal
10
to 1, R2 is equal to 1 (Refer Slide Time: 34:36) in each column; b3: R1 is equal to 1 and R2 is
equal to 0. Now here, for b1, what are the things? Si is written in this way, Mbc, Mcb and this is
equal to Mcd, Mdc. This is S1, this is S2, and this is S3. If I have got these, now I can plug in. What
is my R? I know, I have already given my R: R is equal to plus 10, 0, so I can find out my Si very
easily that is the first step: find out the member end moments.
If you look at the member end moments, you will see that they are equal to If you put in, you
will get S1 equal to 0 plus 50, S2 is equal to minus 50 and 0, and S3 is equal to 0, 0. Satisfy
yourself that this is indeed the case for this particular problem. We will come to problems later
where we will see that this is not a complete picture of my Si in fact, Si actually has some
additional terms that we will look at later; right now, I am not too concerned with this. For this
particular problem, you will see that this is indeed how my S1, S2 are. Once I know my b1, b2, b3
and I have also got this, what is my R going to be equal to?
11
I am going to just put it down. My R is going to be equal to. If I just put it down in. You can
actually do the matrix multiplication later. This is v1 transpose into ((5 upon 6EI) into (2, minus
1, minus 1, 2) and then vi, which is (0, 5, 0, 5) plus (minus 5, 0, minus 5, plus 5) into (5 upon
6EI) into (2, minus 1, minus 1, 2) into (minus 5, 0, minus 5, plus 5) plus (0, 0, minus 5, 0) into (5
upon 6EI) into (2, minus 1, minus 1, 2) sorry, this would be (0, minus 5, 0, 0) (0, minus 5, 0,
0)) and this entire thing multiplied by 10, 0 plus (minus 5, minus 5, 0, 5) multiplied by minus 1,
plus 1 into 62.5 by EI. If you do all these computations, you will get your displacements R 1 and
R2. You have got your S1 and S2 and you have got your R and that in essence is.
What are these terms? This term is the contribution to the flexibility matrix of member 1, this is
F2, this is F3. Since I had F1, F2, F3 that gives me the complete flexibility matrix multiplied by 10
into 0. This gives me the structure flexibility matrix for this particular structure and you will also
notice something else. In a structure flexibility matrix, do you remember what the member
flexibility matrix was? It was 2, minus 1, minus 1, 2 into L upon 6EI. Did you notice something?
It was symmetric. When you do these computations, you will see that the structure flexibility
matrix is also symmetric. What does that mean? It means that bij is equal to bji or I am Fij is equal
to Fji.
What is this flexibility? What is f11? In this, you will get F11, F12, F21, F22 you can evaluate each
one. What do these mean? Let us look at this. What does F11 mean? It is the displacement at 1
due to a unit R1, capital R1; F22 is the displacement at 2 due to a unit load corresponding to R2.
What is F21? It is the displacement at 2 due to the unit load corresponding to R1. What is F12?
The displacement at 1 due to a unit R2. Now, think about this. Displacement at 1 due to a unit
load at 2 and displacement at 2 due to a unit load at 1 should they not be the same? Which law
is that? The MaxwellBetti's Reciprocal Theorem. If you look at that, then automatically these
two are equal to each other and you will see that this is also a symmetric matrix; a symmetric
matrix is one in which the transpose is equal to itself that is what a symmetric matrix is. This
additional term essentially comes from the effect of the local load. We have considered the load
12
effect of the structure loads and the effects of the member loads and obviously, this is going to be
a superposition of the two this, in essence, is structural analysis the way it is. Remember I said
that Si is equal to bi R in this particular case. When do we need to consider some additional
terms?
We will see that Si will not only be just bi R, there will also be an Si0. Where does this Si0 come
from? In other words, when you plug this in, you will see that ultimately, r will have three terms:
one term will be (summation over i of bi transpose fi bi) into R plus (summation over i of bi
transpose fi Si0) plus (summation of i bi transpose vi0 this, in essence, is the real thing. Where
does this Si0 come from? Let us take a particular problem that I am going to look at. Do not be
worried that I am not solving all the problems I am illustrating all the equations that I am
talking about. Let us take the same problem that I have.
13
Instead of applying the load here (Refer Slide Time: 45:41). I still have this 10 let it be and I
have a 40 Kilonewton load acting at the center. Note that there will be two aspects to the
problem: one is this member load. We are still finding out r1 and r2 and we are finding out the
S the member end moments. If you look at this particular case, there will be two aspects to this.
In this particular structure, one aspect of this load is going to be the effect of the local vi and
there will also be another if you look at the member end moments. Let us do the member end
moments through the traditional approach.
What will be the member end moments coming out to be? In the traditional approach, we take
the load and find out; this is a statically determinate structure and I can solve for it. Let us solve
for it. What will we get? Here, due to these load you will have sigma F x gives you that this is
equal to 50. Take moments about this point (Refer Slide Time: 47:32). What is this into 10 plus 5
into 40 going to give you? It is going to give you 200, 250 and 250 divided by 5, so this is going
to land up being equal to 50 and this one is going to be equal to 50. If we draw, if we find out the
member end moments let me draw the bending moment for this.
14
How will the bending moment look? It is 250 here. What kind of a moment is this is going to be?
A moment like this. Here, this is also 250 obviously from equilibrium considerations and you
have 50 over here, so we are going to be having 50 into 5, 250 and so this is going to be equal
to let us see. Over here, you have (Refer Slide Time: 49:19). If you look at this particular
point, what moment do you have? You will see that you will have 0 moment here. Let us look at
this. Here, what is the shear force? Here, you will see that the shear force is going to be 40, so if
you draw the bending moment. Let me just take you through this step by step.
15
This is going to see this (Refer Slide Time: 50:07) and on this side, it is going to be in this
direction this is 50, this is 50, here this is 50, on this side, this is going to be 50, the moment
over here is going to be in this direction and it is going to equal to 250 Kilonewton meter. Look
at this: here, this is 0, from here to here, there is no moment and then here, what do we have? We
have 40 here, 40 here and we are going to have 50 here (Refer Slide Time: 51:13). What is going
to be the moment at this point? You will see that since this goes this way, this will go this way
and it is going to be equal to.
The fundamental mistake: actually, this is where I made the mistake, this is 30 (Refer Slide
Time: 51:39) and this is 30 because if you take moments about this, 40 into 2.5, so that is 100,
100 plus 150 divided by 5 will give me 30. This is going to be 30, 30, 30, 30. Over here, what do
I have? I have 40 into 2.5, so that is 100 Kilonewton per meter. Let us look at this one: this one is
going to have 40 this way, 40 this way, 30 this way, 30 this way and the moment over here is
going to be 250 over here and so if you take moments about this point, this is going to be 30, this
is going to be 100 over here (Refer Slide Time: 52:53). If you look at the bending moment
diagram, it looks like this that is going to be the bending moment. Now, the problem that you
will have over here is. This is the bending moment diagram as we have it. Now, I am going to
show you
I am going to end my lecture over here, but next time I am going to be showing you how we are
going to get exactly the same bending moment diagram by using the concept of Si is equal to bi
into R plus Si0 and we will see what Si0 actually stands for, in general. Thank you for today. We
are going to continue looking at how to look at the flexibility approach for beams and frames;
beams we are doing simply supported, so simply supported statically determinate beam is not
an interesting problem, so that is why we were looking at a statically determinate frame. We will
continue with this particular lecture as we look at it next time.
16
Structural Analysis-I1
Prof. P. Banerjee
Lecture 28
Good morning. We were talking about the matrix analysis approach, especially the flexibilitybased approach right now.
In the last lecture, if you remember, I talked about the fact that Si was equal to bi into R where R
is the loading corresponding to the degrees of freedom, in which case it is essentially joints
because degrees of freedom are defined as joints. In addition to that, you have, as I was saying,
this (Refer Slide Time: 02:13) and we were discussing why we need this additional term which
we did not need. We were looking at this, we had solved this problem and we did not have
anything here but when I looked at this problem (Refer Slide Time: 03:20), I did not have an Sio;
remember, we did not use any Sio, we just used Si equal to bi R and solved the problem. As soon
as I came here, I said there is an Sio. What is the actual difference between the two? In this case, I
do not have the Sio and in this case I have the Sio. Let us actually look at this. Let us forget this
10 Kilonewton because 10 Kilonewton is coming here (Refer Slide Time: 03:48) and anyway its
superposition is valid and therefore you do not have a problem in putting it together.
Let us just consider the 240 Kilonewton loads and see what happens. In this case, we do not have
any Sio. Why? Let us solve it and you will see that you will get 20 Kilonewton, 20 Kilonewton,
this is equal to 0; therefore if you look at the bending moment diagram, for these two, the
bending moments are 0 and the bending moment over here turns out to be(Refer Slide Time:
4:55). Now remember these were all 5 meters in length every one of them was 5 meters. This is
how the bending moment diagram looks.
If you look at it, what are the member end moments over here due to this load (Refer Slide Time:
05:17)? You will see that the member end moments at all points are 0. Therefore, actually the
reason why we did not consider it was this was 0 for all the members. However, if you look at
this, you have 40 Kilonewton load here (Refer Slide Time: 05:56); now, let us try to see what
happens. You will see, first of all, that there is going to be 40 Kilonewton load here and due to
this 100, you will have.
Let us draw the bending moment due to this just this one alone and you will see that you get
something like this: this value is 200 Kilonewton meter, the same value over here and then, this
one reduces it; over here, it goes from 200 to 100 Kilonewton meter; this is in this fashion and
ultimately, here also, this is going to be 100 and this is going to be in this fashion, so over here,
100; this is going to be 0, so it is going to be 0 from here to here (Refer Slide Time: 07:55) and
from here to here, it is like this, this is going to be equal to 100.
This is the bending moment diagram due to this. Therefore, you will see that even if we do not
have any Ri, you are going to have Si, so if this is 1 (Refer Slide Time: 08:21), this is member 2,
this is member 3. You will see that for S10, it will look like 0, 200; for S20, it is going to be equal
to minus 200, plus 100, and for S30, it is going to be minus 100, 0 these are. This is the point
that I am trying to make: typically, you have bending moments; if you look at it in this particular
case, R is equal to 0 and therefore, Si into bi R is equal to 0. However, this is not 0 because you
have this and therefore, there has to be an additional term that takes care of this aspect; in this
particular case, we did not have it because this happened to be 0.
Once you put that in, we again come back to the fact that once Si is equal to bi R plus Si0, this
implies that r is equal to (summed up over all the members) bi transpose fi bi into R this part is
the F, the structure flexibility matrix plus (summed up over all the members) bi fi into Si0 plus
(summed up over all the members) bi transpose vi0. Now, understand that this (Refer Slide Time:
11:02) is relating the joint loads to the joint displacements this is directly given by FR. In
addition to that, this aspect only happens if you have a member load that gives rise to bending
moments at the member ends so this is the effect of the member; and this (Refer Slide Time:
11:28) is the additional effect of the member rotations due to the member load. All these together
account for and this is the total for the flexibility approach for statically determinate structures.
What you have in here is that we have set up how to get the member forces as well as the joint
displacements using the flexibility approach, but this is true for a statically determinate structure
because we said that you can actually solve for Si, given the loads. Now, how can we extend this
to include the effect of statically indeterminate?
We are now using the flexibility approach for statically indeterminate structures. How do we
apply this? In reality, it is actually very very simple; the reason behind it is that even in a
statically indeterminate structure, first you determine static indeterminacy and then identify the
redundant forces; then once you identify the redundant forces, redundant forces equal to 0
implies statically determinate base structure. Since you have that. all that we have developed
over here is the only additional thing. Let us see what happens.
It is exactly the same problem excepting that it is statically indeterminate. What is the level of
static indeterminacy? You will see that static indeterminacy is equal to 1, then identify redundant
4
forces and once you identify the redundant force let me say that my redundant force is 1, I will
identify this as my X1. In other words, this is the force that I am making redundant, so as soon as
I put that equal to 0, what is the structure that I get?
This is my base structure this is my base statically determinate structure. If you look at this
structure, this is identical to the structure that we have just solved in the last lecture, is it not? We
have taken a statically indeterminate structure. What is the difference? In this particular case is,
let us see what the difference is. In the previous case, what did we do? We took this load and we
found out the displacement at this point (Refer Slide Time: 18:32) we found out the horizontal
displacement at this point and the horizontal displacement at this point. In this particular case, I
am going to do exactly the same thing.
How would you solve this problem in the original situation? Due to these loads, you would find
out this displacement (Refer Slide Time: 19:20). Remember I called it r1 and r2 this time;
however, in this particular case, I will call this x1. What is the reason behind it calling it x1?
Because in this particular case, remember that this is the degree of freedom associated with the
redundant force and since it is the redundant force, I will give a different notation. I have just
given it a different notation, but as far as I am concerned, it is still the same r1 and r2 in the
previous case is identical to what I am calling as r1, x1.
You would find out x1 here, you would find out the displacement here that is the step; I am just
going through the steps; you would find out the displacement here. What would the next step be?
The next step would be to apply a unit force over here (Refer Slide Time: 20:27) and find out the
displacement at this point. Therefore, we can say that this would be x 10 and I would say that f1
X1. What is this (Refer Slide Time: 20:46)? This is the displacement at this point due to X1 is
equal to 1. Then, you would say f1 into X1 is the displacement at this particular point plus x10 is
equal to 0, right? This is the compatibility condition that you would satisfy in the normal force
method. In this particular case, understand that this problem once this is the case.
Then, note that in the original case we had R1, R2; in this case this becomes R1 and R2 is my
redundant force. In the previous case, this was equal to 10 and this was equal to 0 this is what
we knew because it was a statically determinate structure. In this particular case, this would still
be 10 but this is unknown but look at this: r1, r2 is equal to r1, x1; this is the statically determinate
case in the previous case, this is the current case (Refer Slide Time: 22:15). Here, this and this
both were unknown; in this case, this is unknown and this is equal to 0 because I am considering
the effect of x1 and r1, so this is 0. Now, I am not solving this particular problem because we
have all ready found out this thing for this.
In the earlier case, for each member what would you do? Find out Si is equal to bi into R plus Si0.
Understand that once I accept that x1 is a redundant load, it behaves exactly like an external load
but the only problem with x1 is that whereas for most of the situations we know the loads, here,
x1 is the load that we do not know. I can write this one in this fashion: I can write it as (S 1, S2)i is
equal to in the previous case., this would turn out to be equal to b11 , b11, no (Refer Slide Time:
24:00), b21 b22 into R1, R2 plus S1, S2, this is i0.
In this case, all that happened. Remember we computed these (Refer Slide Time: 24:26). How
did we compute this? We computed this by putting R1 is equal to 1 and finding out what the
member end moments were. We found this by putting R2 is equal to 1 and finding out what the
member end moments are. Therefore, in this particular case here, understand that R2 is x1, so
here all that would happen is (S1, S2) into b11, now b1x, bx1, bxx into (R1, X1) plus this still remains
the same because this is nothing but the effect of the loads. No difference absolutely these two
are identical to these, this is equal to this, this is equal to this; the only thing is that whereas in
the previous case since R2 was an external load for statically determinate structure, R2 was a
determinate load, now we have exactly the same statically determinate structure except that x 1 is
a redundant force and therefore unknown. If I go through the steps, I can say that this is equal to
bi R plus bix X plus Si0. Do you see this? Since this is nothing but R and X together, I can
actually write it in this fashion.
I am going to say that r is equal to FRR, FRX, FXR, FXX into (R vector, X vector) and then here this
is x vector plus I am going to put inside all the standard items, which is (bi transpose fi Si0)
plus sigma over i bi transpose vi0 note this equation that I am writing is for the base statically
determinate structure. This is identical to what we had for the statically determinate structure
excepting for a small problem and that is that in this particular case, we know these but we do
not know these. These of course are known all; we can find these out; since the flexibility
matrix is symmetric, you will see that this is true. I do not know this (Refer Slide Time: 29:11),
however, compatibility conditions give me that I know this, so here you will see this problem; I
8
know this, I do not know this; so if I rewrite this you will see that this bi will include both the
effect of R as well as X.
If you look at this particular thing, you will see that this can be written in this form. The top set
of equations will become r is equal to FRR R plus FRX X plus summation (now, these are just
corresponding to the R) fi this is nothing but the structure levels, so this does not have anything
to do plus summed over all the length only due to the R, so this is the part of the matrix that
belongs to only the Rs. In other words, bi can actually be segregated into biR and biX. This should
actually be [biR and biX]. This corresponds to the internal forces due to the external loads and
(Refer Slide Time: 32:04) these are the internal forces due to a unit redundant. This is equal to bi
transpose f1 vi0 and you will see that x is equal to FRX transpose into R plus FXX into X transpose
plus sigma biX transpose.
When you do transpose you will see that this implies that bi transpose is going to be equal to biR
transpose, bix transpose and that is how it happens biX transpose fi Si0 plus bix transpose into vi0.
Note that typically, this is a 0, typically this is the zero vector. Therefore, I can write this as
minus FXX X is equal to FRX transpose R plus plus.Note that I know this, I know this, I
know this, I know this, I know this, I know this, I know this and I know this (Refer Slide Time:
34:20), so I can solve this for Xi. Once I solve for Xi, I can substitute that in here and I can get
the displacements. What about my Si?
I have found out the displacements and my Si is given by biR into R plus bix into X plus Si0. Note
this is known, this is known, this is known, this is known and this is solved (Refer Slide Time:
35:19). Therefore, I can find out my member forces in a statically indeterminate structure. This is
known as the matrix force method. The beauty of the matrix approach is that the force method
becomes a part of the overall flexibility approach we are not deriving it separately. Therefore,
the redundant force is just treated as another additional load, of course, which we do not know.
However, we know the displacement corresponding to that load so we can always solve for it.
This is the major advantage that the compatibility equation that we get in the force method really
is part of the entire flexibility method in itself. Once you have understood this, it is very easy to
understand that there is absolutely no problem in using the matrix approach in the force method.
What I am going to now do is essentially start off a problem; I am not going to be able to solve
the entire problem we will continue into the next lecture, but I want to tell you to look at a
problem in its entirety and solve it.
10
Up till now, I have just only briefly discussed this problem but now I am going to actually solve
it; so let us take this problem and solve it in its entirety. This is my member 1, member 2,
member 3 (Refer Slide Time: 38:11) and this is a, b, c and d; all the members have the same EI
and all of them are 5 meters in length this is at the center all of them are the same EI. The
problem is one: find horizontal sway displacement of the top and two: find support reactions and
member end moments and bending moment diagram this is the problem statement and we are
going to be using the matrix force method, so we are going to be using the matrix force method.
11
Already, we know that this is a single degree of freedom and we have identified the redundant
force and therefore, I need to find out this displacement (Refer Slide Time: 40:45) and this
displacement, these are the two displacements. Why do I need to find out this displacement? It is
because this displacement is what I have been asked to find out. Why do I need to find out this
displacement? It is because this is the displacement that is going to be equal to 0 in the original
structure, and so now, we are finding this out. How are we going to find this out? We go through
the steps. The steps are first find out vi0 and Si0 for the loading; remember vi0 and Si0 are only
found out for member loads. Which is the only member load here?
The only member load is this (Refer Slide Time: 41:51); so this gives rise to this and this so we
know that Si0 is going to be equal to 0; we have already discussed this so all of them for i equal
to 1, 2, 3. Now we have to find out vi0; now, vi0 will only happen in the members that so vi0 is
equal to 0 for i equal to 1 and 3 and for v20, we find out; this is the member given this load. Here,
you have 50 and we have v; this displacement, this is going to be my (v1)20 and this is going to be
(v2)20; so, (v1)20 and (v2)20 is going to be equal to minus 50 into 2.5 divided by 2 EI and this is
going to be equal to plus 50 into 2.5 upon 2 EI (we have already found this out, I am just putting
them down). We have found out (v1)20 and (v2)20. What are the next steps?
12
The next steps are to find out the bis. If I put this, I get this equal to 1 and this is going to be this
way, so 1, 1. Therefore, if you will look at it, you will get that this is equal to 5 and this is equal
to 5, this one goes this way and this one goes this way (Refer Slide Time: 45:43) we have
already done this, so I am just putting down everything properly. Here, you will see that b1R is
equal to 0, 5 and b2R is equal to minus 5, 0 and b3R is 0, 0. There is nothing new, but I am just
going through the steps again because I want to solve the entire problem.
Remember that in the force method, we are doing everything to the base statically determinate
structure; therefore, always I am actually solving for this. What is the next step? Put X 1 is equal
13
to 1 and find out how the thing will look like. If you look at this, this is going to be like this,
these are going to be 0, 0. What you have is all these I am just putting them down as if I
have found it out, but if you cannot and if you do not know what I am talking about, I
recommend that you go through and compute the bending moment yourselves. This is the
bending moment for X1 is equal to 1 and therefore b1X is equal to 0, 5; b2X is equal to minus 5,
plus 5; and b3X is equal to minus 5, 0. Till now, you have not seen anything that is different from
what we have already done, so the only thing is that I have identified this as X and the other one
as R. Let us go through the steps.
Therefore just putting down everything, I have got Si0 are all of them for all i; vi0 is equal to 0 for
i equal to 1, 2 and v20 are equal to (minus 62.5 by EI, plus 62.5 by EI). I have got my Sio, I have
got my vio, now what I need is my b. b1R is equal to 0, 5; b2R is equal to minus 5, 0; b3R is equal
to 0, 0; b1X is 0,5; b2X is equal to minus 5, 5; and b3X is equal to minus 5, 0. Now what? Now,
you will see that r is equal to b1R; I am going back to the equation that I had written down:
remember that the equation that I had written down was that r was equal to (biR transpose fi
summed up over all the i biR) into R plus (biR transpose fi biX into X) plus bi transpose in this
case, Si0, so all I am left with is bi transpose, so this is going to be equal to biR vi0 and similarly, if
I put X here, I will get biX fi biR R plus biX fi biX X plus biX vi0; then, we put X is equal to 0 and
we can solve for the problem. I am going to actually solve this problem in the next lecture, so I
just wanted to introduce the entire concept and I hope you are beginning to appreciate all the
things that I have been talking about for now.
14
Structural Analysis- II
Prof. P. Banerjee
Lecture 29
Good morning. In the last lecture, I introduced you to how to extend the flexibility approach to
include statically indeterminate frames and how to solve the entire problem that is called as the
matrix force method. Then, we started looking at a particular problem. Just to review the
problem to you let us see what were the steps.
This was the problem; using the matrix force method you had to find out the horizontal sway of
the top and the support reaction and member end forces and the bending moment diagram.
(Refer Slide Time 02:26)
Then the way we went about it was that these were the two degrees of freedom that were defined
and ultimately you went ahead and found out first and foremost the displacements the S i0 and the
vi0 and once you found those out, then the next step was to find out the bi0 and biR and the biX
also we found out by doing equilibrium.
Then, I listed out all of these things and ultimately, this was the equation that you had to solve.
Here, r was equal to this (Refer Slide Time: 03:30) you can do this if you just rewrite it; this
into R plus biR into X, so this is FRR, this is FRX, this is FXR, you will find out that if you take the
transpose of this, you will get biR fi transpose biX; fi transpose is itself and therefore you see that
FXR is equal to FRX transpose and then you can find out FiX; these are the additional terms due to
the vi; I do not have the terms for Si0 over here because in this particular problem, we have
already seen that Si0 is 0 and therefore, I am just dropping those terms out. Now, all I need to do
is evaluate all these terms and then put them into the equation.
If you look at this, the first step is to find out the summations. Note that since vi0 is equal to 0
excepting for i is equal to 2, the summation term only has these two terms and I have found them
out; b2R transpose v20 is 312.5 upon EI and b2X v20 is 625 by EI. What is the next step? The next
step is to find out all those biR and biX. We are going to go through the steps and we will do it
member by member.
First and foremost, member 1: Let me find out biR transpose. What am I doing? I am actually
evaluating. I have evaluated this term because in this term, excepting for bi is equal to 2
everything else is 0; this term is equal to 312.5 by EI and this term is equal to 625 by EI we
have already done this.
The next step is to find out this (Refer Slide Time: 05:47), this and I am going to find out this. If
I put that in, I get for member 1 b1R transpose f1 into b1R. What is b1R? b1R transpose is going to
be (0, 5). What is f1? l upon 6EI, l is 5 upon 6EI into (2, minus 1, minus 1, 2) and here, we have
(0, 5). We are going to go through all the steps so that you know exactly how we are solving this.
If I look at this particular thing, you will see that you will get 0, 5 and inside minus 5, here 10
and this is equal to 5 upon 6EI and this is equal to 250 upon 6EI.
Let us look at b1R transpose f1 b1X you will see that since b1R and b1X are identical both will give
you identical results so you are going to have 250 on 6EI and b1X ; this (Refer Slide Time:
8:02) is equal to biX transpose f1 biR satisfy yourself that this indeed is the case; b1X is also 250
upon 6EI. I have found out for member 1. Similarly, I am going to find these for member 2 and
3.
For member 2, I have b2R f2 b2R; b2R is (minus 5, 0) and into the same all the fis are the same
because the lengths are the same, the EIs are the same and so here, this is equal to (minus 5, 0),
which is equal to (minus 5, 0) multiplied by (minus 10, 5); if you look at this, you will have
(minus 5, 0) into (minus 10, 5) into 5 upon 6EI and so this is equal to 250 upon 6EI. Now, b2R
transpose f2 b2X is minus , this is going to be different because if you look at this, b2X is equal to
(minus 5, 5) so this is going to be equal to (minus 5, 0) into minus 15 and here also this is plus 15
(plus 5 plus 10) into 5 upon 6EI (Refer Slide Time: 10:38) this is going to be equal to 375 upon
6EI; this by the way is going to be the same as b2X transpose f2 b2R, so if we do b2x transpose f2
into b2X, you will get it equal to (minus 5, 5) into 5 upon 6EI into (2, minus 1, minus 1, 2) into
(minus 5, 5), so this is equal to (minus 5, 5) into (minus 15, minus 15), which is equal to 750
upon 6EI so we have done it for member 2.
Now, finally, we have to do it for member 3. Let us see what we get for member 3. Once we
have done it for all three members you will see that we can add up and actually solve the
problem. Let us look at b3R f3 b3R. b3R is equal to (0, 0) into 5 upon 6EI into (2, minus 1, minus 1,
0) into (0, 0) and without much ado, we can see that this is going to be this (Refer Slide Time:
12:33). Similarly, you will see that b3R f3 into b3X even though b3X is not equal to 0, you will
see since this is 0 that this is equal to this and.. Finally, b3X transpose f3 b3X is going to be
equal to (minus 5, 0) into 5 upon 6EI into (2, minus 1, minus 1, 2) into (minus 5, 0) and this you
will see is going to be equal to 250 by 6EI. Now, having done all of this, we need to add for each
one and when we add them, this is what we get.
Now we are going to add up all of them. Thus you will have 250 plus 250 plus 0, so b1R is equal
to 500 upon 6EI R; note that this vector R is just a vector of 1 and this is what you will see that
this is nothing but r1, this is nothing but R1 (Refer Slide Time: 14:20) plus biR transpose fi X. If I
add all of them up, you will see 250 plus 375 plus 0, so this is going to be equal to 250 plus 375
is equal to 625, 625 by 6EI X1 plus biR 0, which if you remember was equal to 312.5 that was
positive, so plus 312.5 upon EI.
Similarly x1 is equal to 625 by 6EI X1 plus 250, 250, 500 plus 7 times 50 is 1250, so it is 1250
upon 6; this is R1 (Refer Slide Time: 15:40), this is X1 plus we have got it as 625 upon EI and
therefore, we have it as 625. Now, we know this is equal to 0, we know that this is equal to 0 and
we know R1 is equal to 10 so we have 6250 plus 6250 so this is going to be equal to 6875 upon
6EI plus 1250 upon 6EI X1 is equal to 0 and we solve for X1. X1 is equal to minus 6875 by 1250,
which is equal to minus 5.5 Kilonewtons.
We have X1 is equal to minus 5.5 and therefore, R1 is equal to 5000 upon 6EI minus 3437.5 upon
6EI plus 312.5 upon EI. Here there is a change; this is wrong here, this is 6250 plus 3725 (Refer
Slide Time: 18:28); 625 into 6 is equal to 6750, 6750 is going to be 10000 so this is going to be
10000; this goes, so this goes 5, 25, 40, this is minus 8 Kilonewtons and so when you do 8 over
here you get 200, 5000 and what you are left with is 312 upon EI that is your R1; you know
your R1, you know the sway we have found out your sway; ultimately, once you have found
out your sway, the next step is finding out the rotations.
If you look at the member end forces, the member end forces will be given by Si is equal to biR R
plus bix X. Let us look at S1: S1 is equal to plus Si0 (Si0 is 0 for all of them), so biR b1R is equal to
(0, 5) into 10 plus b1X into minus 8; this you see will turn out to be (0, 10) Kilonewton meter; S2
is equal to (minus 5, 0) into 10 plus (minus 5, plus 5) into minus 8 and what we finally get is
(minus 10, minus 40); and S3 is equal to (0, 0) into 10 plus (minus 5, 0) into minus 8, so I get
(40, 0) these are my member end moments.
8
If I have put these in, let us see what we finally get. This is my x1, minus means in this direction.
Then, if I take moments about this point, you will see that I will get 10 into 5, 50 plus 100, 150
divided by 5, this is your member end forces and if I were to plot the bending moment diagram,
it would look like this. Here this is in this fashion, here, it is in this fashion and here it is in this
fashion (Refer Slide Time: 24:21) and these are the member end moments that you already have.
This is your bending moment diagram. This one is this way that is only due to the load but
there is a superposition over here and that is 50 so this is going to be and 40, so at the center it is
going to be equal to 40 minus 10 upon 2, it is going to be 15 in the middle, so this is going to be
35, so it will go like this and like this (Refer Slide Time: 24:57) and here, I have this and over
here, this is going to be 35 Kilonewton meter.
This is the total bending moment diagram and also you know what R1 is, it is equal to 312.5
upon EI in this direction. That is solving this problem using the matrix force method. I suggest
that you solve this problem using your standard force method and satisfy yourself that whatever
answers we have gotten are the answers that you would get using the standard. I can assure you
that you would get exactly the same if you went ahead and did it that way. So much for the
matrix force method as used for frames. Now, these are using flexural members. Now, I want to
spend the next bit of time looking at a problem where you have truss, in other words, the axial
deformation is the key. Let us take a problem for that.
This is the problem and here, the question asked of you is the following: I need to know how
much this has gone by. So I want to find out the horizontal displacement at this roller and I also
want to find out what are the member forces. This is again static indeterminate it is the problem
that we have looked at earlier: the static indeterminacy is 1 and I am going to assume that this is
my x1 (Refer Slide Time: 27:41) the internal force is x1. How I am going to solve this problem?
All I am going to tell you is how I am going to set up the problem and that will give you the
details of how to solve the problem. The numbers I am not going to go into. What we are going
to do here is assume that this is 10, we will take this also as 10.
10
Therefore, let us take down the steps over here. First and foremost, what are the f, the flexibility
matrix? You will see that vi is nothing but delta it is just a single degree of freedom. Si is
nothing but Pi and fi is nothing but Li upon (EA)i. In this particular problem, we will assume that
the EAs are the same for all the members.
Then, if we put it in this fashion, a, b, c, d, let me put this member 1, member 2, member 3,
member 4, member 5, member 6. f1 is equal to f2 is equal to f3 is equal to f4 and all of them are
equal to 10 upon EA; f5 and f6 are equal to 10 root 2 upon EA these are the flexibility matrices.
11
Now, what is the next step? The next step is actually to find out the equivalent statically
determinate. This is the hinge, this is the roller. I have this member and this member exists but it
is cut, so the force in this member is equal to 0 (Refer Slide Time: 31:00). This is the point I was
trying to make: in the force method, typically in this particular case, it is very easy to solve
because you would have all the degrees of freedom. How many degrees of freedom does the
structure have? You will see that this is 0 (Refer Slide Time: 31:18), here, you will have 1, 2, 3,
4 and 5. 5 is the only one that you have been asked to find out; you have not been asked to find
out 1, 2, 3, 4.
In fact, when you actually solved it on the computer, you would consider all these five degrees of
freedom and therefore, these displacements would become joint loads. However, since I have
asked you to only find out this (Refer Slide Time: 32:03) and this hand calculation I cannot
deal with five degrees of freedom plus one which is the sixth redundant degrees of freedom and I
would have a 6 by 6 matrix that I would have to solve. I am not quite interested in that particular
case.
If this (Refer Slide Time: 32:24) is my only degree of freedom that I am interested in, understand
that these then become actually not joint loads technically they are joint loads, but in this
particular case, since I am only interested in finding this out and this out, so these do not come as
members and so I am going to solve these separately and I am going to say that these are going
to give me my Si0.
Remember I have said that you could solve it, that Si to be equal to bi into R plus Si0. In this
particular case, R is the only one corresponding to this is R1 here, of course here in this
particular case, there will also be biX. I am now going to club all these loads because they do not
correspond to this one as these loads member loads. I can find out the member loads to begin
with, so let me find out the member loads.
12
Until and unless our joint loads happen to be corresponding to a degree of freedom that you are
interested in finding out, you will never consider those loads as joint loads you will club them
together and forget about it. I am going to just solve this problem to begin with and get my Si0
for each one I am just going to forget it at that point of time. Therefore, let me solve for this.
What am I going to get? Here, this is going to be equal to 20. Then, if I take moments about this
point, I will get 20 into 10, 200 plus 200, 400 divided by 10 you get 40 and this is equal to 0. If
I have this particular situation, let us go to a specific point: since this is 0, I can find out that this
is going to be equal to 20, similarly, this is going to be equal to 20.
Then, what else do we have? Let us look at this particular joint. In this particular joint, the
horizontal of this is going to tie up with this, so this is going to be equal to 20 root 2 and
therefore, this is going to be equal to 40 and this is going to be equal to 0. You will see that at
this particular joint, 20 the vertical matches with this and the horizontal matches with this, so this
is 0, this is 0; here (Refer Slide Time: 36:20), there is only vertical, so this horizontal is 0 and
this is anyway 0. Now, here I get that S10 is equal to minus 20, S20 is equal to minus 20, S30 is
equal to minus 40, S40 is equal to 0, S50 is equal to 20 root 2 Kilonewton and S60 is 0 because that
happens to be the redundant force.
In other words, what I have done actually is to include all these loads as this thing, because the
only one that I am interested is here. Next, what do I need to find out? I need to find out biA. By
the way, will there be any vi0? There will not be any vi0 because there are no member effects; the
only time that you will get vi0 is if you have a temperature stress or you have a lack of fit; in this
particular case, there is none, so vi0 is equal to 0 and I need to find out biR and biX.
How do I find out biR? This is the displacement, I need to apply R1 is equal to1 (Refer Slide
Time: 38:08) and find out all the member forces just for R1 is equal to 1. Obviously, this is going
to give me this, you will see that this has to be equal to 0, this has to be equal to 0, this has to be
equal to 0, this has to be equal to 0 and so the only thing that you have is 1. b1R is equal to 0, b2R
13
is equal to 0, b3R is equal to 0, b5R is equal to 0 and b6R is equal to 0 all of them are equal to 0;
the only thing that you have is b4R which is equal to plus 1.
Now, put X1 is equal to 1 so you essentially have X1 is equal to 1 and find out all the net forces.
You will see that the external reactions are all 0 because this is an equal and opposite force it is
an internal force and therefore, what are you going to have over here when you put X1 is equal to
1? This is equal over here. This is going to be 1 over root 2; this one is going to be 1 over root 2;
similarly over here, you will see that this will be 1 over root 2 and you will get this equal to 1;
you will get that this is equal to 1 over root 2.
Therefore, b1X is equal to b2X, b3X, b4X is equal to minus 1 over root 2 because they are all
negative and b5X is equal to b6X is equal to 1 both of them are plus 1. Now, I have found out my
Si0, I have found out my b1R and b2R.
14
Next, what is the solution? If you look at the solution, you will see that r1 is equal to (summation
(biR transpose fi biR)) into R1 plus (summed up (biR transpose fi biX)) into X1 plus now you will
see Si0 come in so this is going to be equal to summation over the entire biR transpose fi Si0 and x1
(which by the way is the separation, so it is equal to 0) is equal to summation over biX transpose
fi biR into R1 plus summation biX transpose fi biX into X1 plus summation biX transpose fi Si0.
Now, let me first find out these two quantities (Refer Slide Time: 43:37) and then I will find out
these two quantities.
15
My b1R is equal to 0 and I am only left with b4R is equal to 1 and then biX, b3X, b4X are all 1 over
root 2 and b5X and b6X are plus 1 that gives me my bs. Then, what do I have? My Si S10 is
equal to S20 which is equal to minus 20; then I have S30 equal to minus 40, S40 is equal to 0, S50 is
equal to 20 root 2 and S60 is equal to 0 Kilonewton. Remember that this is identical to what I am
doing this is the problem that I had solved earlier too. The f1 equal to f2 equal to f3 equal to f4
equal to 10 upon EA, f5 and f6 are equal to 10 root 2 upon EA. Once I have all of these things
together, now all I need to do is put together the terms that I have to get. If I put together those
terms, you will see that they turn out to be equal to. These are all single, single terms.
If you look at it, in biR, since all the ones are 0 excepting for b4R (Refer Slide Time: 46:23), you
will see that only b4R will contribute and so this r1 will become equal to 10 upon EA R1 plus now
if you look at it, biR, and because all the biR transpose are equal to 0 the only one that will come
in will be b4 so we need to get b4X and that is minus 1 over root 2 so this is going to be 10 into
R1; and this is going to be equal to minus (10 over root 2 EA) into X1 plus now here this is the
term that we need to evaluate this is the term (Refer Slide Time: 47:18) and we are going to
evaluate this term.
Now, let me just put this together: you will see this will become minus 10 over root 2EA into R1
and ultimately, we will see that in this particular case, you will see that this term all the b4Xs
are root 2, root 2, so 1 over root 2, so you will get one half and one half into 10 EA is going to be
5, 5 by 4 times is going to be 20, 20 and then you are going to have 20, and then you are going to
have one 1 into 10, so it is going to be 20 root 2 upon EA into X1 plus this term. Now, this term
is the one that we have to sit and evaluate for each member and let us do that for each member.
16
Member 1: note that in all these cases, biR transpose is equal to 0 excepting for b4R, so the only
one that I have to actually do is to find out b4 S4 but you see S40 is equal to 0, so you will see that
summation biR fi Si0 for this particular case is equal to 0; now for the biX transpose, now this is
going to get this thing, so here I am going to do for member 1; member 1 biX is minus 1 over root
2 into 10 upon EA multiplied by S10 is minus 20; for this one, this is going to be equal to 200
upon root 2EA.
Member 2: for member 2, you will find that it is minus 1 over root 2, S2 is also the same, so for
member 2 also, this is 200 over root 2EA. For member 3, you will see that b1, so it is 1 over root
2 into 10 upon EA multiplied by minus 40, so you have plus 400 over root 2EA. Then, you have
member 4 and member 4 is going to have 0; S4 is 0, so this is 0. Then, you have member 5 and
member 5 is going to be equal to b5 is 1 and this is multiplied by 10 root 2 upon EA and this is
going to get multiplied by 20 root 2, so this is going to be equal to 400 upon EA. For member 6,
it is going to be 0. We add it up and you will get summation biX transpose fi Si is equal to (400
plus 800 root 2) all over EA.
17
If we come back to this particular problem, you will see that this is 0 and this is (400 plus 800
root 2) upon EA (Refer Slide Time: 52:04). Now, what is R1? R1 in this particular case is 0, so all
we get is that (20 plus 20 root 2) upon EA into X1 plus (400 plus 800 root 2) upon EA is equal to
0 and from this, we can solve for X1 and we get r1 is equal to minus 10 upon root 2EA into X1.
Finally, my Si: Si are going to be equal to biX X1 plus Si0. You can solve for X1 in here: plug this
in, you have got this displacement, you have known this, you substitute these and you get your
member end forces solved.
What I have tried to explain to you today is how to use the matrix force method to solve the
statically indeterminate approach. The entire concept is built on the flexibility approach and
whether it is statically determinate or indeterminate, the approach is identical excepting that in a
statically determinate problem, you know all the capital Rs and you do not know the
displacements, you know all the loads, you do not know the displacements.
For the statically indeterminate problem, you know some of the loads for which you do not know
the displacements and you do not know some of the loads, which are the redundant forces, for
which you know the displacements because those are the 0 compatibility displacements that you
have so that is the only difference.
Up till now, I have been looking at the flexibility approach. Next, I am going to be moving on to
the stiffness approach of the method which is also the matrix displacement method although we
never say matrix displacement method we call it the stiffness method. Thank you very much.
18
Prof. P. Banerjee
Lecture 30
Good morning. Till the last lecture, we had been looking at the matrix flexibility approach,
which essentially included the matrix force method. Today, we are going to be starting off on
what I call as the matrix stiffness method, which is essentially the matrix analysis approach for
the displacement method. Let us review what the displacement method talks about.
The displacement method essentially goes as follows: define degrees of freedom that is done
very easily; then, the next approach is the member force deformation relations; third is the
kinematic relationship between member deformation and displacement corresponding to degrees
of freedom; four is the virtual displacement principle to relate loads to displacements for degrees
of freedom. I think this is the overall scope of the displacement method and this will essentially
be the stiffness method; the reason behind it will shortly be understandable.
Here, let me call this member 1, member 2, member 3. This is just an illustrative example. Let us
say you have this and you have these loads (Refer Slide Time: 04:49). I am not putting down
right now what the loads are etc we are just looking at it in defining the entire problem. How
many degrees of freedom do you have in this particular case? Assuming axial rigidity, we know
that you have three degrees of freedom. We will call that r1, r2, r3 these are the degrees of
freedom. Therefore, I can say that the r vector is equal to r1, r2, r3.
In the matrix method, everything is either a vector or a matrix. Therefore, I am writing down the
displacement vector corresponding to the degrees of freedom in this way and those are my
degrees of freedom. The next step is to relate the member force deformation relations and what
are those? I will write the slope deflection equations down using the rotations from the chord, so
in that case I do not have EI so it is going to be 4EI by L thetaab (this is from the chord to the
tangent) plus 2EI by L thetaba plus (FEM)ab. These are the slope deflection equations (Refer Slide
Time: 07:03). I am going to write this in a different form: I am going to use the same notation
that I have been using.
I will say that vector v, which is equal to v1 and v2 is equal to thetaab, thetaba. Similarly, S which
is equal to S1, S2 is equal to (Refer Slide Time: 07:51) If I write them in this fashion, you will
see that I can write S1, S2 as and I am going to call this fixed end moments as. These are the
member end moments for the kinematically determinate structure, so this is a kinematically
determinate structure, so that is 0 because in the displacement method, the base structure is the
kinematically determinate structure, where all degrees of freedom are restrained the fixed end
moments essentially comes from that. Now, it is interesting to note that this can then be written
in this format: S is equal to K v plus S0, where K is equal to.
K is 2EI by L into (2, 1, 1, 2) this is of course for a member i and this would be this (Refer
Slide Time: 09:50). This is the stiffness matrix of course flexural member stiffness matrix. If
you have an axial member then you will see that Ki is equal to (EA)i upon Li. The interesting
point to note is that Ki is an inverse of the flexibility matrix. We will see that that is true for the
flexural member; for this it is obvious because invert of this is L upon EA which you already
know; for this (Refer Slide Time: 10:59), you just need to invert it 2 by 2.
This is actually the way we had set up the slope deflection equations, if you remember, right at
the beginning so there is nothing new in this. This relationship at the member level exists as long
as you define the degrees of freedom to be thetaab and thetaba, which are known as deformation
degrees of freedom. I am not going to go into those details because that comes in a much later
course where you relate different degrees of freedom for a member. We are being consistent and
we will continue with this approach of defining the degrees of freedom.
Essentially, what we have done is; Si is equal to Ki into vi plus Si0 this gives us the member force
deformation relationship. What is the next step? The next step is our kinematic relationship, so
let me take the example that I have.
Just broadly looking at it, how do we relate the member force deformation relationship? The way
you do it is; first and foremost exactly the same way that you got the load member force
relationship in the flexibility approach. You put r1 is equal to 1 and the other two equal to 0; if
you look at the displacement pattern, this will be 1, 1; if we look at this, this is 1 by L, same
thing here. If I am going to call this member 1, member 2, member 3, the way we do it is we say
vi is equal to ai into r this is the relationship we are looking for each one. The way we get it
is again, this will become v1, v2 for each member is related to here, we have three degrees of
freedom r1, r2, r3 and so what we have is a11, a12, a13, a21, a22, a23 and a31, a32, a33 each one we
get by putting r1 is equal to 1, this one we get by putting r2 is equal to 1 and this one we get by
putting r3 is equal to 1 (Refer Slide Time: 14:52). This is known as the kinematic relation.
4
Therefore, what we do is we put each one and for each one, we find out its column. Here, what is
v1? It is just a21, this is a 3 by 1, this has to be a 2 by 3, so you get this to be 2 by 1 (Refer Slide
Time: 15:44), so you get a11 and a21. Let me find out a11 and a21 for the first one. What is it?
From the chord to the tangent, so that is positive 1 over L and a21 from the chord to this thing,
anticlockwise positive 1 by L. What is a11, a21 of 2? It is 0, and 0. What is a11, a21 of 3? You will
see again from the chord to the tangent, so it becomes 1 upon L, 1 upon L. We have got the first
column. For the next column, what do we need to do?
Put r2 is equal to 1; r2 is equal to 1 says this is 1, this is 1 (Refer Slide Time: 17:18), this is r2 is
equal to 1. What will my a12 and a22 for the first one look like? It is going to be 0, 1; a12, a22 of 2
is 1, 0; a12, a22 of 3 is 0, 0.
Finally, put R3 is equal to 1, so this is 1 and this is 1 (Refer Slide Time: 18:29) and therefore, a31
and a32 of 1 is equal to 0, 0; (a31 a32)2 is equal to 0, 1 and a31, a32 of 3 is equal to 1,0.
In this way, what we have done is we have evaluated ai for each i. I have made a mistake here:
this is (Refer Slide Time: 19:42) one three, two three, one three, two three and one three, two
three. The first one relates to the degrees of freedom of the member and the last one refers to the
degrees of freedom at the global level. Therefore here (Refer Slide Time: 20:07) this is two three,
first one member level, second one and so on. We have evaluated this for every member and
that gives us the kinematic relations which essentially means vi is equal to air.
6
If we now put this into our entire equation by substituting, we get Si is equal to Ki ai r plus Si0, Si0
being the fixed end moments. So now this is incorporating the kinematic relationship into the
member force deformation relationship, so you get the member forces in terms of the.
The next step is virtual work and in virtual work, what do you do?
You use the principle of virtual displacement to relate R and Si this relationship would
typically be a kind of an equilibrium relation and the principle of virtual displacement actually
replaces the equilibrium relation. How do we do it? What we do is, we actually apply a virtual
displacement pattern, an arbitrary virtual displacement pattern that can be given in this fashion.
The virtual displacement pattern is given in this manner (Refer Slide Time: 22:53) this is the
virtual displacement pattern.
Then, what is the external virtual work going to be equal to? You will see that this is equal to the
work done and if you look at this, this is nothing but summation of Ri ri this is basically force
into the displacement corresponding to it; this I am writing in matrix form in this manner this is
all I am doing: the displacement corresponding to each degree of freedom multiplied by the load
corresponding to that degree of freedom; this (Refer Slide Time: 23:39) can be written in exactly
this form. Therefore, the external virtual work is virtual displacement into real loads.
What would be the internal virtual work done? You will see that the internal virtual work will
be each member will be undergoing and again, this one (Refer Slide Time: 24:15) is
effectively is summation of, so this is summed up over all the members i is a member so
this is the internal force.
Now, the question here is how is this related to this? There are some additional terms I will bring
a little bit later. This does not include all the internal work terms. Especially if you have member
loads there are certain loads that are not included in the internal work if you put it in this. But
right now, I am just bringing it in a broad framework; later on, I will bring in all the details.
7
Understand that this is the internal virtual work, this is the external virtual work done at every
member level; this again you will see is nothing but you know v1i into S1i plus v2i into S2i which
is essentially the moment into the virtual rotation at that particular joint summed up over all the
members. Now, the important thing over here is that obviously if you if the system is in
equilibrium which is under the loads, the structure deforms and so sets up forces.
What happens is the virtual work principle says VWI is equal to VWe. Now, since VWI is equal
to VWe, it stands to reason that this into this (Refer Slide Time: 26:02) is equal to summation
over i into this, but this does not help us because this is something, this is something else, but can
we relate the virtual rotations at the member levels from the displacement? Sure, we can. We
already know that vi is equal to air this is of course when we have real, but whether we have a
real displacement or a virtual displacement, it does not matter, so this is also equally valid: the
virtual rotations are given in terms of the arbitrary virtual displacement in terms of the kinematic
relationship.
The kinematic relationship remains the same whether the displacements are real or virtual
because the boundary conditions are exactly the same and so we go through exactly the entire
steps all over again. Therefore, if you look at it, this implies that vi transpose is equal to r
transpose ai transpose. Now, how we got from here (Refer Slide Time: 27:27) to here is basically
matrix algebra. When you take transpose, you interchange the order, you can go back to any
matrix algebra book to understand this.
This relationship we are going to incorporate in this. Ultimately, what does this become? It
becomes r transpose R is the summation over all the members r transpose ai into Si. Now, this
does not depend on i, so I can actually rewrite this as r transpose summation ai transpose Si.
Therefore, what we ultimately have is that the virtual work equation lands up being this this
becomes the virtual relationship, virtual work equation. Note here that this (Refer Slide Time:
28:44) has to be equal to this for any arbitrary r.
8
Since the r prime transpose is appearing in both the cases, we see that this implies that R is equal
to summed up over all the members ai transpose Si. This looks like any equilibrium relationship
but it is actually a virtual work equation. It looks like I have got R in terms of Si, but we have
used virtual displacement and never forget that we have to take into account all the work done
up till now in the internal work, we have only included this aspect but later on we will see that
there are other terms that come into play. We will see that it is important to
Right now, I am just writing down the basic equations and therefore if we put that together, this
is valid (Refer Slide Time: 29:54) if we only have member loads that do not have reactions that
do work joint loads are not a problem when subjected to a virtual displacement pattern; this is
very important. Later on we will include this effect. Therefore, this is true. Now, all I am going
to do is substitute the Si in here and what do I get?
I get R is equal to summed up over all the members ai transpose and what is Si? It is equal to Ki
ai r plus Si0. If I rewrite this, this becomes summed up over all the members ai Ki ai r and note
that since r is this, the summation is only here plus summation over all the members of ai
transpose Si0. This is nothing but the structure stiffness matrix and therefore, R looks like K r
plus We will see later on there are some other terms in here, but this in essence represents my
solution. Therefore, what do we normally know? We know this, we can find this out, we know
this and we can find this out and therefore we solve for r. Once we solve for r, this is my Si, so I
substitute r and I get my member end forces and that in essence is the member structural
analysis.
You have found out the displacement corresponding to the degrees of freedom and you have
found out the member forces this overall is the basis for the matrix stiffness method. Now, I am
going to spend quite a few lectures because there are certain issues that I have not I put dot dot
dot and this dot dot dot includes some terms that come in under specific conditions. What I am
going to do for the rest of the time in today's lecture is actually take a simple problem and
illustrate the methods that we have done. I am going to solve a lot of problems using the stiffness
method; you will understand why because in today's computer application where we do analysis
by computer software packages, this is the method the stiffness method is really the method
that is used for solving problems.
Till now, I have always done only illustrative problems because they are methods that are useful,
but since this method is the one that is used the most in today's world, I am going to solve a lot of
problems. I shall start off by looking at simple problems and then get more and more
complicated till we have tackled all kinds of problems that you can come up with.
We are going to look at first structures with joint loads, then we are going to look at structures
with member loads, then we are going to look at structures where the members are not only
subjected to loads but also to temperature and slowly, we will continue on and look at all the
10
variety of problems so that you are exposed to the spectrum of problems. Of course, one of the
things that you have to appreciate is that since I will be doing everything by hand, I am only
going to use at the most one or two degrees of freedom in a structure. Let us start off by looking
at some simple problems and then go on from there.
Let me take one of the problems that I have been looking at thematically for a long time. Let me
put it this way: this is 5 meters, this is 3 meters, this is 4 meters, this is EI, this is EI and I am
going to say that this is being subjected to a load of let us say 10 Kilonewtons (Refer Slide Time:
36:33). We have to find out the displacement at all points, in other words, the deformed shape of
the body under this load and also find out the member end moments so that we can draw the
bending moment diagram for this structure.
First and foremost, no point, we have dealt with this particular issue long enough to realize that
this is a two degree of freedom structure. I have done enough of how to compute the kinematic
indeterminacy or the number of degrees of freedom of a structure, so from here on, if you do not
understand how I have taken these, how there are two degrees of freedom and why I have taken
these two, please go back and review from the past few lectures in this course. These are the two
degrees of freedom.
11
First and foremost is the force deformation relationship for each member. If you look at the force
deformation relationship, both Si are equal to Ki into vi. plus Si0 (Refer Slide Time: 38:05), the
fixed end moments. If you look at the fixed end moments, since there are no member loads this is
equal to 0 and for both of them, for i equal to 1 and 2, since they are identical Ki is equal to 2EI
by 5 into (2, 1, 1, 2) that is my Ki and gives my relationship. This (Refer Slide Time: 38:39) is
the force deformation relationship for both the members, they are identical because the member
stiffness matrix is identical because EI is the same and the length is the same.
What I am going to do is I am going to do r1 on a separate sheet of paper but I am going to put r2
here because you people know that r2 is relatively simple. So r2 would imply this and therefore, if
you look at a12, a22 of 1, it is going to be equal to 0, 1 and a12 and a22 of 2 is equal to this is my
v1 that is thetabc so that is going to be equal to 1, thetacb is my v2, so this is equal to this. This is
the force deformation relationship and this is the kinematic relationships, so we are looking
through them reasonably fast.
12
Note that we have already done this, but I want to go through this again; kinematic relationship
there is never any end to kinematic relationships. This point cannot go vertically because this
member would have to elongate so this will only go this way so it has come 1 over here. If this
member was left to go its own way, this would have gone exactly the same one here, but it
cannot, so now it has to move, it is going to be moving along this line. It has to move
perpendicular to it so that this is the point where C is. And what is this? (Refer Slide Time:
41:42) This is 90 degrees. Let us see how much this entire thing has gone up by. This one, if you
look at it, is going to be. How we are going to determine this? What we have to determine is
that if this is 1, how much is this going to be equal to? Let us look at that. We do not know this,
we know this is 1 and we have to find out this this is this one (Refer Slide Time: 42:25). Let us
look at it.
What kind of relationship is this? Since this is theta, it is tan inverse of 3 by 4, so this is this theta
and this is the theta; that means that this is 3, this is 4, so this is going to be 4 by 3 and this is
going to be 5 by 3. Now, we know that this is 5 by 3. What will the displacement pattern look
like? The displacement pattern will look like this. Note this tangent over here (Refer Slide Time:
43:28) has to be this way, the tangent over here has to be this way and the tangent here because
r2 is equal to 0 and here, you have a fixed. Once you have that and the tangent over here of
course has to be this way, the chord is this, the chord is this (Refer Slide Time: 43:50), so this
angle and this angle are going to be 1 upon 5 and this is also going to be 1 over 5; this is 5 over 3
and this is 5, so this is going to be 1 over 3 and this is going to be 1 over 3. Therefore, my a11, a21
of 1 is going to be equal to from the chord to the tangent positive 1 by 5, 1 by 5; my a11, a21 of
2 is equal to from the chord to the tangent negative minus 1 over 3, minus 1 over 3.
13
Having got that, now we can put that v1 is equal to 1 over 5, 1 over 5, this is going to be equal to
0, 1 into r and v2 is equal to minus 1 by 3, minus 1 by 3, 0, 1 into r you have got the kinematic
relations. From virtual displacement, we know that capital R is equal to summation over i ai
transpose Ki ai into r; note that here since Si0 is 0, this is the only thing (Refer Slide Time: 46:40).
Now, what is capital R equal to? This is R1 and R2 and if you look at this you will see R1 is equal
to 10 Kilonewton and R2 is equal to 0 Kilonewton meter. Therefore, R vector is 10 and 0. This is
my a1 vector, this is my a2 vector, so all I need to do is just go through these steps. Let us go
through these steps it is instructive to go through the steps.
14
Now ai transpose Ki ai r if you look at it, this part can be actually written as the contribution of
the ith member to the structure stiffness matrix and note that Ki ai also has a specific aspect to it,
because you will see that Si is equal to Ki ai r plus Si0 and in this particular case, this is 0 (Refer
Slide Time: 48:08), so Si. Therefore, this one I will say is Ti this gives me directly Ki ai is
equal to Ti directly gives me S, the member end forces in terms of this, so I am going to actually
compute these as a step in the whole process.
Let me first do Ki ai; K1 a1 which is equal to T1 will be equal to 2EI by 5 into (2, 1, 1, 2)
multiplied by this thing for the first, which is 1 by 5, 1 by 5, 0, 1 and this is going to be equal to
2EI by 5 I keep it outside, inside 2 by 5 plus 1 by 5 is 3 by 5, here 1 by 5 and 2 by 5 is 3 by 5,
here 2 into 1, 1, here 2, so this is my T1.
Next, I am going to compute T2. T2 is K2 a2, which is equal to 2EI upon 5 into (2, 1, 1, 2)
multiplied by minus 1 by 3, minus 1 by 3 and then 1, 0; let me look back at my old thing and
then I will get back to you it is 1, 0 so this is equal to 2EI by 5 into minus 2 by 3 minus 1 by 3
is equal to minus 1, minus 1 by 3, 2 by 3 is minus1, here I have 2, here I have 1, so this is my T2.
15
Next is computation of K1 which is essentially equal to a1 transpose T1; a1 transpose is equal to (1

by 5, 1 by 5, 0, 1) times 2EI by 5 (3 by 5, 3 by 5, 1, 2). If you look at this, if I put 2EI by 5
outside, inside I get 3 by 25 plus 3 by 25 is 6 by 25, here 0, 0, 3 by 5, here 1 by 5 plus 2 by 5 is 3
by 5 and this is 2 (Refer Slide Time: 52:18). Note that it is symmetric it has to be, has to be
symmetric. K1 is the contribution of the first member to the structure stiffness matrix is this.
Similarly, K2 is equal to a2 T2, which is equal to a2 is (minus 1 over 3, minus 1 over 3, 1, 0)

and T2 is 2EI by 5 (minus 1, minus 1, 2, 1) and I will take 2EI by 5 outside and inside, you get 1
16
over 3, 1 over 3, so you get 2 over 3, here minus 1, so minus 1, minus 2 by 3 minus 1 by 3,
minus 1 and here 2 that is K2. Now, if we add K1 and K2, what do we get?
K becomes equal to 2EI by 5 and inside, I have 6 by 25 plus 2 by 3 and here, I have 3 by 5 minus
1, 3 by 5 minus 1 and 2 plus 2. If I look at this, it is 2EI by 5 and this becomes 75, if I take 75, I
get 18 and here I get 50, so I will get 68 upon 75, this becomes minus 2 over 5, minus 2 over 5
and this becomes 4 that is my structure stiffness matrix.
17
Now, since I know R is (10, 0), 2EI by 5 into (68 by 75, minus 2 by 5, minus 2 by 5, 4) into r1 r2;
this way, I can solve for r I just take invert this, multiply this and you can get r and then, my S i
are equal to Ti r; I can find out my Si and once I find out my Si, I have solved the problem. You
can do the numbers yourself, it is very simple. I have evaluated each and every term and you just
need to go through the steps to evaluate them. Once you got your Si, since there are no member
loads, you can draw the bending moment diagram very easily. Once you draw your member end
this thing, you can also get the support reactions and everything. This, in essence, illustrates
briefly the displacement method.
Go through the numbers yourselves and please solve this problem and get through it. Next time
we will see what the answers are to this one.
18
Structural Analysis- II
Prof. P. Banerjee
Lecture 31
Good morning. We have been discussing the displacement method and in the last few lectures I
have established the basic concepts behind the displacement method. I had promised that we will
spend about three to four lectures looking at various examples in the application of the
displacement method. Today, we are going to be starting off by looking at one particular
example. The kind of way I have chosen the examples has been essentially to establish the key
principles of the application of the displacement method or actually the matrix displacement
method which is the stiffness method as I had already stated.
In the next few lectures, we are going to do that. Of course, it is impossible to do all types of
problems, so I am only going to be taking up representative problems and along the way, I will
illustrate a few detailed notes about the applications of displacement method, which I have not
set up in the this thing, for example, how do we consider it if we have a support settlement? How
do we consider temperature stresses in the displacement method or the stiffness method? I will
be taking up all these things as examples as I go along during these particular few lectures, next
few lectures. Let us start off with an example.
I am going to start off with simple examples and then go on to slightly more complicated
examples. Let me take absolutely the simplest example, so this is the structure. This is EI, 4
meters (Refer Slide Time: 03:59) and this is 2 EI, 5 meters. This is the problem and here, the
problem is essentially to figure out its displaced shape and also to figure out the bending
moment, member end moments and bending moment diagram for this particular structure. First
and foremost, how many degrees of freedom? I am going to just state it and I will expect you to
1
find this out. This is a two degrees of freedom structure; use the procedure that I have developed
and establish that these are two degrees of freedom. Of course, all of them are axially rigid. The
displacements are r1 and r2. Once we have those displacements, the next step is first to find out
the fixed end moments. Fixed end moments: this is a, b, c.
The fixed end moment for ab the beam that I consider for ab is this one with this (Refer Slide
Time: 05:48) and due to this, you are going to have a (FEM)ba; the (FEM)ba is going to be equal
to 100 of course, the entire thing is minus, so 100 into 4 square divided by 12 plus half into
100 into 4 square upon 12 that is equal to... 3 by 2 is going to be equal to 108, if we look at this,
it is going to be 100 divided by 4, it is equal to 8 upon. so it is going to be minus 8, 200 and
this is going to be 300 so this is going to be equal to minus 300 Kilonewton meter. (FEM)bc is
equal to (FEM)cb which is equal to 0, because there is no member force in this particular
member. Having done that, what happens is, the next step is going to be getting the displacement
pattern for each one, so let us have a look at that.
It is going to be r1 is equal to 1, this is going to go in this way, come like this simple
displacement pattern that you have. If you look at it, what is thetaba equal to? Thetaba is equal to
0, thetabc is equal to 1 upon 5 from, so that is positive 1 upon 5, thetacb is positive 1 upon 5,
then we have the rotation the rotation is going to give me this (Refer Slide Time: 09:42), so this
is going to give me thetaba. This is r1 is equal to 1, this is r2 is equal to 1. thetaba is equal to 1,
thetabc is equal to 1 and thetacb is equal to 0.
Therefore, if we look at it, what we have is for member ab, a is going to be equal to 0, 1 and a for
member bc is going to be equal to (1 by 5, 1 by 5, 1, 0). What is a? If you look, it is essentially
3
vi is equal to ai r, so here, what I have done is I have actually found out the two displacements the
two kinematic relationships: one that relates the member end rotation (Refer Slide Time: 11:01).
Note that since I have a modified member, I only have thetaba as my member rotation and the
relationship with the degrees of displacement corresponding to the degrees of freedom.
Having found out this, what do we have then? We know that Si, which in this case is Sab is
actually equal to Mba because Mab is equal to 0, is equal to (Refer Slide Time: 12:06) this is a
transpose so it is going to be equal to 0, 1 into r1, r2 minus 300; Sbc, which is equal to (Mbc and
Mcb) is equal to this (Refer Slide Time: 12:43) into Kab. Note that Si is equal to Ki vi plus Si0. In
this particular case, the K is equal to 3EI by L and L in this particular case is 4; for this, the Mab
is equal to 4EI by L and 4EI, now EI is 2 EI, so this is going to be (8 EI upon 5 and 4EI by 5, 4EI
by 5, 8 EI by 5) into (1 upon 5, 1 upon 5, 1, 0) into (r1, r2) plus 0.
This is our relationship and ultimately using the virtual displacement, we can find out that capital
R is equal to sum of ai transpose Si; this is summed up over all the members (Refer Slide Time:
14:43) plus summation over i ai (this I am going to put it as transpose) into ni these are the
reaction terms that come from the member forces, the reactions only because those are not
considered in this specific thing. In this particular case, this will only exist for the summation a i
into ni. In this particular case it is only in ab, because that is the only member force that you have
and the reactions that you have for this particular case are 100 Kilonewton, which is 50 and 50
and if you look at these the work done by these 50, the aab transpose is actually equal to in both
the cases (0, 0, and 0, 0).
In other words, what we are really trying to find out is given r1 is equal to 1, how much does this
point (Refer Slide Time: 16:21) move up by and how much does this point move up by? If you
go back to the displacement pattern for r1 is equal to 1, how much does this point go up by? 0.
How much does this point go up by? 0. In this particular case, how much does this point go up
by? 0. How much does this point go up by? 0. That is essentially what you get over here.
4
Therefore in this particular case this term is equal to 0, because this is going to be equal to this
into your ni in this particular case are equal to 50 and 50, so that is equal to 0. Now, what we
need to do is essentially look at the other terms.
If you look at it, we get R equal to summation over i ai transpose Ki ai into r and the other term is
0. I need to find out for ab what is ai transpose Ki ai. Now, if you look at it, ai is equal to. aab is
equal to (0, 1); Ki is equal to 3EI upon 4. Therefore, what we have here is since Ki is a scalar, I
can just do ai transpose into (0, 1) into 3EI upon 4 and if you look at that, that becomes (0, 0, 0,
1) 3EI by 4. Therefore, contribution of ab is equal to (0, 0, 0, 3EI by 4). Similarly, we can find
out for bc.
Kbc is going to be equal to abc transpose Kbc abc. Now, abc is equal to 4EI upon 5 into (2, 1, 1,
2) that is Kbc and abc is equal to (1 by 5, 1 by 5, 1, 0). Therefore, the first thing is Kbc into abc
and this is equal to 4EI by 5 into (2, 1, 1, 2) into (1 by 5, 1 by 5, 1, 0) and this is going to be
equal to 4EI by 5; you have 2 by 5, so 3 by 5, here also you have 3 by 5, here you have 2, here
you have 1.
The next step is to find out Kbc. It is going to be equal to (1 by 5, 1 by 5, 1, 0) into 4EI upon 5
into (3 by 5, 3 by 5, 2, 1). This is going to be equal to 4EI by 5, this is going to be 3 by 25, 3 by
25, so it is going to be equal to 6 upon 25, this is going to be 3 upon 5, this is going to be 3 upon
5 and this is going to be 2. What is the total K?
The total K is going to be equal to 24 by 125, K is going to be equal to 24EI upon 125, this is
going to be 12 EI upon 125, this is going to be 12 EI upon 125 and this is going to be equal to EI
into 8 by 5 plus 3 by 4 this is my K. Therefore, what is my R? If you look at R, R is going to be
equal to (50, 0) and we have not yet done the ai transpose into this thing and so the ai transpose
into Si0 is only there for ab and this is going to be equal to [0, 1] (Refer Slide Time: 22:53) into
minus 300, so this is going to be equal to 0, minus 300.
Ultimately, my equation looks like this. It is going to be (50, 0) is equal to (24EI upon 125, 12 EI
upon 125, 12 EI upon 125, 47 upon 20 EI) into (r1, r2) minus (0, 300). I can take this on this side,
so this implies that if I put EI outside, I get (24 upon 125, 12 upon 125, 12 upon 125, 47 upon
20) into (r1, r2) is equal to (50, 300). Therefore we solve this for r1 and r2 from which we can find
out our Si because it is equal to Ki ai into r. And once we find out the member end moments we
can always draw the bending moment diagram. This is a fairly simple specific problem and
therefore there is not too much in it.
Let me change this problem a little bit and then we can solve another problem with it, because
the essential thing that happens in this particular case is that we need to be able to find out how a
change can be incorporated in the entire problem statement. Let me change this problem and let
me give it as a slightly different problem and the problem that I will give you is this one this is
another example.
Here, this is the same, excepting that now what I have done is I have now done this (Refer Slide
Time: 26:08), so this particular one that was vertical here I have inclined it, this is same.
Here, you see this point can actually move in this direction, so instead of the horizontal, I am
going to define the vertical as my displacement degree of freedom you can define it horizontal
also, there is no problem this is 2 EI and this is still 4 meters, the only difference here is that
this is going to be equal to 3 meters and this is equal to 4 meters, so the total length is still 5
meters it is just inclined. If we do that, what do you get?
The fixed end moment still remains the same and the fixed end moment is going to be equal to
100 multiplied by 4 and this (Refer Slide Time: 27:25) is going to be equal to 100 into 4 PL
upon 8, it is PL upon 8, this is minus plus half of PL upon 8; please make the correction even in
the previous problem so that instead of 300 what you have over here is minus 75 Kilonewton
meter; this is ba; please make the change that in the previous case also, this is minus; the same,
there is no difference in the fixed end moment.
In other words, by changing the structural configuration, the fixed end moments in the two
members are not changed. The only thing is that in the previous case when I had done minus 300
which is wrong: actually, what I had done was I had made a mistake, I had taken 100 to be the
intensity rather than a load; it is PL upon 8, so this is PL upon 8, half carryover from the other
side and therefore this is minus 75 and the frequencies are different.
Now, we come to the Ks. They are also the same. K of ab is equal to 3EI by 4 and Kbc is equal to
4EI by 5 because EI is 2 EI into (2, 1, 1, 2) that is the stiffness for each member.
Now, we find out the as. For finding out a, what do I do? I have this and then I have this. Here, r1
is equal to 1. Now if you look at this, the only way this can move is along this line perpendicular
to this and therefore, for it to get 1 vertical, this will have to be 4 by 3 and this will have to be 5
by 3; therefore, this moves by 4 by 3, so this comes here and this 1 remains identical; this is
going to be like this and this is going to be like this; then, when I put r2 is equal to 1, I get the
10
same as in the previous case. Note that here, there is no moment restraint, so it can take any
arbitrary; here, this is equal to 1, this is equal to 1, this is r2 is equal to 1.
Therefore, we have aab is equal to. If we look at the first one, the aab is equal to. Let us take
stock of this: it is going to be like this (Refer Slide Time: 31:51) and this is going to be straight,
so this is going to be clockwise, so aab is going to be equal to 1 upon 4 and if you look at it, it is
going to be clockwise, so it is going to be minus 1 upon 4 and for this one, it is going to be equal
to 1. abc is going to be equal to for the first one, the total displacement is 5 by 3 and if you look
at it, anticlockwise, so this is 5 by 3 divided by 5 is 1 by 3, so it is going to be 1 by 3, 1 by 3 and
for this one (Refer Slide Time: 32:45) it is going to be equal to (1, 0).
Note that we also need to find out in this particular case, how much does this point move by? I
am just going to do aab transpose which is to see how much these move up by (Refer Slide Time:
33:14), so these move up if you look at it, the aa is equal to 0 (this moves up by 0) and this
moves up by 1. If we look at it now, we have got the fixed end moments, we have got the Ks, we
have got the as and all we need to do now is to find out the contribution of each one.
I will write it down: Sab is equal to Ka into (minus 1 by 4, 1) into (r1, r2) plus if you look at this,
this is going to be minus 75 and Sbc is equal to 4EI upon 5 into (2, 1, 1, 2) into (1 over 3, 1 over
3, 1, 0) into (r1, r2) plus 0. This is Mbc and this is Mcb (Refer Slide Time: 35:40). This relationship
will come out to be important. Here, this part I will call as Tab and this part I will call as Tbc so
Tab is equal to minus 3EI by 16 and 3EI by 4.
In other words, this into the r1 r vector will give me Sab. Tbc becomes 4EI upon 5, in this twothird plus one-third is equal to 1, here, you will have one-third and two-third, 1, this way you get
2, this way you get 1. This is Tab and Tbc.
11
Once we have that, we can find out Kab as essentially aab into Tab and similarly, Kbc is equal to
this. If we find out this, we get it equal to What is ab transpose? You will see this is equal to (1
upon 4, 1) and Tab is equal to (minus 3EI by 16, 3EI by 4), so this is equal to this way you get
3EI by 64, this way you get minus 3EI by 16, here you get minus 3EI by 16 and this way, you
get 3EI by 4 that gives the contribution of member ab to the structure stiffness matrix. We have
evaluated the contribution of the member ab to the structure stiffness matrix. The next step is
finding out the contribution of the member bc to the structure stiffness matrix.
12
This is equal to abc transpose Tbc, so that is equal to (1 over 3, 1 over 3, 1, 0) multiplied by 4EI
upon 5 into (1, 1, 2, 1); this is equal to 4EI upon 5, I get one-third, one-third, two-third, here I get
1, here I get 1 and here I get 2 this is the contribution of the member bc to the structure
stiffness matrix. If we put it together, let us see what we get.
We are going to put it together to get the K matrix and the K matrix is going to be equal to 3 by
64 plus 8 by 15, I am going to put EI outside, so this is going to be 3 by 64 plus 8 by 15, then we
have minus 3 by 16 plus 4 by 5, then we have minus 3 by 16 plus 4 by 5 and then we have 3 by 4
plus 8 by 5. What we get over here is K is equal to EI into 0.5802 and this is going to be 0.6125
plus, this is also plus, this is 0.8 minus 3 by 16, so this becomes that way, this is plus 0.6125 and
this one becomes 1.6 plus 0.75 is 2.35 that is your K. Now let us see what happens to ai, the ai
n.
13
One part is computing aab into (Sab)0, which is the fixed end moment. If we do this part, aab
transpose is equal to (minus 1 over 4, 1) and Sab is equal to minus 75, so this is equal to on this
side 75 by 4 is 18.75 minus 75. This is one and the other one that we have is the work done by
the ai, so this is going to be aab transpose into (nab)0 and if you look at it what was aab? We had
computed aab, do you remember? We will just go back to that; that was equal to (0, 1), so this is
0, 1 and then 0, 0 for r2 this is going to be 0, 0 so this multiplied by 50, 50 so what we get is
corresponding, this is corresponding to the second one and this is 0, 0. So what we get over
here is that this is equal to 50 and this is equal to 0 (Refer Slide Time: 44:03).
14
Now, I am going to put everything together and so what I have over here is r which is 50, 0 is
equal to K; K is equal to EI into (0.5802, 0.6125, 0.6125, 2.35) into (r1, r2) plus (18.75, minus
75) plus (50, 0).
What we have over here is the fact that in this particular case, you will see that when I do on the
other side, I get it equal to EI into (0.5802, 0.6125, 0.6125, 2.35) into (r1, r2). I made a mistake
here let us look at this. This is not 50. Why? Because even this one, I need to find out; my r1 is
my vertical, so the 50 is horizontal, so I need to find out how much this has moved by, so this is
going to be 4 by 3. When you have 4 by 3, you are going to have it equal to this is going to be
equal to 50 by 3, minus 18 by 5, so this is going to be equal to minus, this is 68.5 and here, you
have 50 by 4, so this is equal to minus 50 by 3 minus 18 by 5, 50 by 3 is equal to 20, so this is
going to be equal to just bear with me, 16.67, so this is going to be equal to 2.08 and this is
going to be 75.
We can solve, so r1 and r2 turns out to be equal to 1.0118 upon EI into (2.35, minus 0.6125,
minus 0.6125, 0.5802) into (minus 2.08, 75). If we do that computation, we get it equal to minus
51.434EI and this one works out to be 45.321 upon EI. We have got r1 and r2 and now if I am
going to plot this one, how would it look? r1 is negative which basically means that it actually
comes down and anticlockwise, so if I were to show that to you, it would look something like
this.
15
In other words this is of course exploded. This vertical (Refer Slide Time: 49:25) has actually
moved down; in other words, it cannot move down, it will have to move this way also, so this
point has actually come here where the vertical displacement is equal to 51.434 upon EI (Refer
Slide Time: 49:45). From that we can find out this displacement and furthermore this has rotated
in an anticlockwise manner. In other words, it has gone this way, so what we have over here is
this. This will have moved over here (Refer Slide Time: 50:13) by 4 by 3 times this, so this has
moved by 4 by 3 into 51.434 by EI, this is this distance, and this one is going to look like this.
This rotation from the horizontal is equal to 45.321 by EI and if you were to look at it from this,
this is also equal to this from the original. This gives you the displaced shape of the structure.
Finally, once you have that, you can plug in the r1 and r2 values into the expressions and we get
Mba is equal to minus 31.365 Kilonewton, Mbc is equal to 31.365 and Mcb is equal to minus 4.89
Kilonewton meter.
16
The last part of this is to draw the bending moment diagram of this. We know the member end
moments, so this is clockwise; clockwise means this way so I am going to draw it on the tension
side, this would be this way (Refer Slide Time: 52:09) and this one if it is this way is this, this
one also goes here. We have 4.89, 31.365, the bending moment is in this way and then if we go
here, this is also 31.365 and here, we have 0, so we have this and then, I superpose on that the
100 so this I get this way and this way, this one is in this manner, this one is in this manner.
Therefore, what we have here is this value you can check it will be equal to 84.32. This is the
bending moment diagram and that, in essence, is the complete solution of this method.
I have gone through all the steps in this particular case. I would like you to go back and revise
these steps so that you can be confident how to apply the displacement method. Thank you very
much. See you in the next lecture.
17
Structure Analysis - II
Prof. P. Banerjee
Lecture 32
Two examples of how to apply the stiffness method in analyzing structures today, we are going
to continue with it, we are still going to look at only loads on the structure. From the next lecture
onwards, we will start looking at other effects other than loads and what effect they can have on
the structure, or in other words, how to consider those in the application of the stiffness method.
As I said, I am going steadily into more and more complicated problems but ultimately, as I told
you, I am never going to exceed more than two degrees of freedom in all these examples for the
simple reason that hand computation is not possible. However, the method that I develop is valid
for all degrees of freedom you can have 150 degrees of freedom if you so desire but the only
thing is that at the end of it, you are going to get 150 by 150 matrix, which you will only be able
to solve using a computer or maybe a program over calculator in today's world. Let us look at the
example that I am going to present in today's lecture.
Do understand this: when I say I am steadily getting into more complicated problems, I am
essentially really referring to complexity in these kind of problems, which can only come in the
complexity of the kinematic relationships rather than any other complications because once you
apply the matrix stiffness method, everything else is just one step after another.
The advantage of this method is that each member by member, you can look at first what are the
fixed end moments first, you decide what is the member type that you consider; next, you
compute the fixed end moments; third, you compute the relationship with the member end loads
and member end deformations.
Next, you compute the kinematic relationship between the member end deformations and the
structure degrees of freedom. These are the steps that you follow there cannot be any further
complications in this. The only complication that I can bring in is how to draw kinematic
relations and in fact, in the entire steps over here, the only major thing that really has a role to
play is the kinematic relationship.
In this particular problem, I have A, B, C, and D and the loads are the following: I have 50
Kilonewton load here, r1 here, r2 here, this is EI, this is also EI and this is also EI (Refer Slide
Time: 05:42); this length is 15 meters, this by the way is the same point, 20 meters, from B to C
is 15 meters; furthermore, there is UDL on this and this intensity is 4 Kilonewton per meter. This
is a two degree of freedom structure and these are the degrees of freedom satisfy yourself that
this is a two degree of freedom structure. I am not going to going into finding out the kinematic
indeterminacy or the number of degrees of freedom by now, I expect that you should be able to
do that and I have chosen these two as my degrees of freedom in this particular problem.
Furthermore, I am given here that EI is equal to 2 into 10 to the power 6 Kilonewton meter
squared (Refer Slide Time: 06:54). The question here is to find a), the horizontal displacement of
point B that is why I have taken that as r1 and the rotation of point B, which is why I have
3
taken that as my r2 and part b) is member end moments and support reactions. These are the
things that you have been asked to find out in this particular problem.
Obviously, we are going to solve this using the matrix approach of the displacement method,
which is essentially the stiffness method. Now, let us proceed on to the solution. What is the first
step? You have already completed the first step figured out how many degrees of freedom and
identified the degrees of freedom. The next step is to find out the fixed end moments for each
member.
First, member AB: what is the type of member AB? The type of member AB is we are going
to take it in this form, this is A and this is B. Therefore, in this member, what are the fixed end
moments? You know this is a modified member, so the only fixed end moment is going to be for
BA and that is going to be equal to 0. Furthermore, what is KAB? Member it is equal to 3EI by L,
L is 20 in this particular case, which if I substitute becomes 3 into 10 to the power of 5 and the
vab is equal to thetaba. These are my identifications for each member.
Then, I will go into computing for BC. What kind of a member for BC do I consider? For BC, I
consider the member to be in this fashion, where this is equal to this, this is, B this is C and the
loading that we have on the member is 4 Kilonewton meter (Refer Slide Time: 10:26). vBC is
equal to thetabc only. KBC is obviously going to be 3EI upon L, L is 15 in this particular case so
this is equal to 4 into 10 to the power of 5 Kilonewton meter by radian and (FEM)BC is going to
be equal to (4EI into 15 squared by 12) plus (half into 4 into 15 squared by 12), so this is going
to be equal to 112.5 Kilonewton meter. I have done this for BC now and now, finally CD.
For BD, the form is: this is B, D and therefore, v of BD is going to be equal to (thetabd, thetadb)
and KBD is going to be equal to 2EI upon L so this is going to be equal to 2EI upon 25 into (2, 1,
1, 2) which is equal to 16 into 10 to the power of 4 into (2, 1, 1, 2) that is KBD and (FEM)BD
which is essentially S0 of BD, is equal to 0, 0; there is no load, so 0 Kilonewton meter. Therefore
5
now I have written down degrees of freedom for each member, these are the member end
deformations (that depends on what is the element that you are using), then you have to write
down the stiffness matrix (that also depends on what are the degrees of freedom) and I have
finally evaluated the fixed end moment for each member. This, in essence, is everything done for
each member. The next step is to find out the kinematic relationships, which relate the member
end deformations, in other words, member degrees of freedom to the structure degrees of
freedom.
I am going to draw it slightly enlarged. How do we find that out?

I am going to be giving r1 is equal to 1 which essentially means that this point (Refer Slide Time:
15:35) has to move horizontally by 1, but if it moves horizontally by 1, if you note, this
particular member is fixed here, so the only way this can move is in this plane. These can move
this way, so what we essentially have is that this point comes; if this is 1, this is 4 by 3 and this is
5 by 3; this point comes here, this point moves here by 1, this point moves here by 1 so what we
have these come here.
note that I am always drawing this exaggerated to essentially established the point. By the way,
these have to remain in this way because there can be no rotation at this point. So once we have
this we can draw it in this way it does not matter; here, this one goes this way and here, this one
goes this way. Let us join the chords: this is the chord for AB, this is a prime, this is my b prime,
this goes actually here, this is my C prime and my D does not move anywhere; this is the chord
of AB, this is the chord for BC and this is the chord for BD.
What are the rotations? This is this way, this is this way, this is this way. So the rotations here
are going to be this, this is going to be equal to 4 by 3 into By the way, this is not going to go
here (Refer Slide Time: 18:07), this one is actually aligned this way, these cannot go this way,
this has to move along this direction, so if we see this direction, this is given as 4 here and 3 here.
If this moves, this moves by 1 and this moves by 3 by 4, this actually is this way, so the total
movement is going to be equal to 4 by 3 plus 3 by 4 that is the total distance moved divided
by 15 (Refer Slide Time: 19:01). If you look at that, this is 4 by 3, this is going to be equal to 12,
so 12, 16 plus 9, it is 25 upon 12 and 25 upon 12 is going to be equal to you will have 5 upon
3, so this in essence becomes 5 upon 36 this is this rotation. How much is this rotation? This is
going to be equal to 5 by 3 divided by 25 which is equal to 1 by 15; this (Refer Slide Time:
19:55) is going to be equal to 4 by 3 divided by 20 which is equal to 1 upon 15 and this of course
is the same, 1 upon 15. This gives you all the member end rotations, that: thetaba is equal to 5 by
3 of r1, this is equal to 5 by 36, these are 1 by 15, 1 by 15.
Now we are also going to look at r2 is equal to 1 and I am going to do that. r2 is equal to 1 is
going to give me that this point cannot move anywhere because that would imply that these
points are here too and r2 is equal to 1 would imply this (Refer Slide Time: 21:04), this, this,
where this is 1, this is 1 and this is 1, r2 is equal to 1.
We can essentially now put together the aab is equal to r1 and what is r1? 1 by 15, in this it is 1,
because remember this is thetaba in terms of r1 and r2. Then, abc is going to be equal to again
this is thetabc in terms of r1 and r2, so thetabc for r1 is equal to 1 is 5 by 36 and for r2 is equal to 1
is this. Finally, aBD is both thetabc and thetacb and this, if we look at it from the chord to this, this
is negative, from the chord to the tangent, this is negative and here this one turns out to be from
the chord, positive, 1 upon 15 and this one is going to be equal to 1, 0, 1, 0, B, D. These are my
as.
I have already got my Ka, everything etc so now, I can put that my R vector and this is the
modified R vector you will see what the modification is. R is equal to in brackets over all i ai
Ki ai into r plus summation ai transpose Si0 (what is Si0? these are fixed end moments) plus
summed up over all of them the ai transpose ni0 which are the reactions that come from
Now let us see what R is. These are the nodal loads that are not included in the member loads;
the only nodal load is the 50 Kilonewtons, so let us find out what that is. So what we have to find
out is how much that displaces by for r1 is equal to 1 and how much does it displace by for r2 is
equal to 1. Therefore, what we have is 50 multiplied by whatever, (Refer Slide Time: 25:00).
Let us look first at r2. When r2 is equal to 1, how much does it move vertically by? 0. That means
0. Now, for r1 is equal to 1, how much does this move vertically by? 4 by 3. This is going to be
50 multiplied by 4 by 3 and it does not move down, so it does do positive work. This is going to
be equal to 200 by 3 and 0 Kilonewton this is R prime. Now, let us find out what the ai
transpose Si0 are. For AB, what is Si0? Let us take for each one.
For AB, let me find out what ai transpose Si0 is. We have evaluated; for AB, the fixed end
moment is 0. What is ai transpose? ai transpose is equal to 1 by 15, 1 into 0, so this is going to be
equal to 0, 0. Note that since there is no member load, the ni0 are also going to be 0 and therefore
it does not matter what you do you are going to be having it equal to 0, but what should be the
ai transpose? The ni0 are 0, 0 but ai transpose if you look at it is going to be equal to. how
much does this move up and down by? 0. This is going to be equal to 0 and how much does this
move down by is 4 by 3, it was upward so this would be minus 4 by 3 and for this one how much
does this move by? 0 0. Therefore this is going to be equal to 0, 0. I am going to actually put
those in so that you know exactly how to compute it.
For bc, let us look at ai transpose Si0. What is ai transpose for this? Again, ai transpose is going to
be equal to minus 5 upon 36 and 1. What is Si0? We have computed fixed end moment for bc.
Where did we compute it? 112.5, so 112.5 and if we compute that, what do we get? We get it
equal to minus 15.625 and 112.5. What about ai transpose ni0? ai transpose are going to be equal
to. Let us look at how much these move up and down by due to this thing. The left-hand side
moves down by minus 4 by 3. This is going to be minus 4 by 3. How much does the other side
move up by? It moves up by 3 by 4. This is going to be up 3 by 4. Then for r2 equal to this, we
are going to see that both are going to be 0 and then what is my a? You will see that for this
particular case, what is my? Let me put those down. My n, which are the reactions at these points
they are going to be equal to due to this 4, it is going to be equal to 4 into 15 is 60, so this is
going to be 30 30. What I have ultimately is 30 30 and if you look at the work done, this is going
to be equal to minus 40 and this is going to be equal to 22.5 I have found these two out.
10
Finally for BD, there is no load, so I am not going to belabor the point. Well, anyway, let us
belabor the point, there is no problem. What is ai transpose? ai transpose is going to be equal to 1
by 15 1 by 15 1 0. What are the member end moments? 0 0. What is this? 0 0. Let us do a i
transpose ni0. What is it? Let us look at that. Again for r1 is equal to 1, how much does B move
by? It moves down by so it is 5 by 3, this is 0 and for r2, these are going to be 0 0. What are
the reactions? 0 0. What do we get? 0 0. We have got it we have got all the members' ai, etc.
We can find out the contribution of ab to K and you will see that this is going to be equal to a i
transpose and what is ai transpose? It is going to be equal to 1 by 15, 1 and what is the EI? Kb is
equal to 3 and 10 to the power of 5 into a; a is 1 upon 15, this and so this is going to be equal to
4000 by 3, this is going to be equal to 20000, this is going to be equal to 20000 and this is going
to be equal to 3 into 10 to the power of 5 that is my Kab.
11
My KBC is going to be equal to minus 5 by 36, 1, this is 4 into 10 to the power of 5 into minus 5
by 36, 1, so this is going to be equal to 7716.049, minus 55555.56, minus 55555.5 6 and this is
equal to 4 into 10 to the power of 5 so this gives me KBC. Then, what else do we have?
Finally, KBD, which is equal to (1 upon 15, 1 upon 15, 1, 0) into K and K for this thing is going
to be equal to 16 into 10 to the power of 4 into (2, 1, 1, 2) into (1 upon 15, 1 upon 15, 1, 0), this
is going to be equal to (1 upon 15, 1 upon 15, 1, 0) into 16 into 10 to the power of 4, this is going
to be 3 by 15, which is 1 upon 5, this is going to be 1 upon 5, this is going to be 2, this is going to
be 1. Ultimately, this is going to be equal to 4266.67, 32000, 32000 and 3.2 into 10 to the power
of 5.
12
When we put all of these together, what you ultimately get is that r is going to be equal to 200 by
3 and this was 0 multiplied by. when we add all of them up, we get 13316.04, minus 3555.56,
minus 3555.56 and this is going to be equal to 10.2 into 10 to the power of 5 into (r1, r2) plus the
only one that comes in; both BD and AB are going to be 0, 0, so what we have is minus 15.625,
112.5 plus (minus 40, 22.5) I have made a mistake here.
This is ai transpose and the transpose would be 3 by 4 and this would be 0, 0, here, I would get 0
and this one would turn out to be equal to minus 17.5.
13
This is actually minus 17.5 and 0 (Refer Slide Time: 38:50).

Ultimately, what we are left with is this: (13316.04, minus 3555.56, minus 3555.56, 10.2 into 10
to the power of 5) into (r1, r2) is equal to 32 and 66, so 99.79 and over here, it is minus 112.5 and
you can solve this (Refer Slide Time: 39:57) for r1 and r2; r1 and r2 in this particular case is equal
to 7.469 into 10 to the power of 3 and minus 9.342 into 10 to the power of 5 radians. This is the
procedure that we use to solve for r1 and r2. Once we have solved for r1 and r2, the next step is to
find out the member end moments.
14
There you put in; Si is equal to Ti into r plus Si0. Once we do that we are going to get that Mba is
equal to 121.3 Kilonewton meter, Mbc is going to be equal to minus 339.9 Kilonewton meter,
Mbd is going to be equal to 209.1 Kilonewton meter and Mdb is 214.1 Kilonewton meter. I will
leave it to you. Once you find out the member end moments, then you can find out the reactions
at every end and from there, you can find out the support reactions. We have found out the
member end moments.
Here, you have it as this way, 0 (Refer Slide Time: 43:13) and then here, you have it going
anticlockwise and here also, you have it going anticlockwise. From this, you can find out the
reactions: they will be in this manner and they are going to be equal to 6.06, this is also going to
be 6.06, here, this is going to act in this way. Once you put this, you are going to get this is equal
to 4 into 15, 60 Kilonewton, so you will have two parts to it: one part will be this aspect, so here,
15
you get 52.66, 39.49 you can compute the forces there and this will be 7. 34, there is going to be
this way, this way.
Ultimately, from this, we can find out what this and this are and then we can find out what
reaction comes over here and what reaction comes over here. The reaction that comes over here
is directly 39.49 which is this one and the reaction over here which is a vertical reaction is going
to be 65.8; this is a roller, roller on an end gives you both vertical and horizontal; then you can
find this out, this is going to be equal to 52 and 62, it is going to give you this way, so
essentially, what we have over here is going to be this way and this way the load is going to be
equal to 8.92 and the moment over here is going to be 224.1.
For this support, there is a fixed support, this is for this roller support and this is for this roller
support and that gives you all the support reactions that you have. The overall point is, in this
particular problem, if you look at it, was probably that you had to consider different kinds of
issues in this and if you do it step by step as I said, what you need to do essentially in this
particular kind of situation is this: start the problem by considering member by member.
First is of course the degrees of freedom, then. The first step is for each member, what do you
find out? First, what type of element do I use? Do I use the standard fixed-fixed element or do I
use the modified element with a hinge at one end and fixed at the other end? Once you do that,
the next step is automatically; define the member-end degrees of freedom.
In other words, if you have both the members fixed, then the rotation at both the ends would be
the degrees of freedom; if you had pinned at one end and fixed at the other end, then only the
fixed end rotation becomes the member-end degree of freedom. Three: define the stiffness matrix
corresponding to this degree of freedom in other words, if you have both ends, then the
member stiffness matrix will become 4EI upon L, 2EI upon L, 2EI upon L, 4EI upon L; if you
only have a pin at one end and fixed at the other end, then there is only that this thing, then you
have Mba is equal to 3EI upon L into thetaba that defines the matrix. Finally, find the fixed end
moments, which are essentially Si0 if member effects are present. That is the first step. For each
member, you have to figure out what type of element to use, define the member-end degree,
16
define the stiffness corresponding to the degrees of freedom and then find out the fixed end
moments.
Once you have done that, the next step is kinematic relationships: solve the geometry problem
for each ri is equal to 1 and all others 0 solve the geometry problem; provide r1 is equal to 1
and all others equal to 0 and find out what the displaced shape looks like, put r2 is equal to 1 and
all the others 0 and find out what it looks like; this way, you would solve the kinematic
relationship.
Once you have that, then define ai and a prime i; ai is essentially vi in terms of ai r and these are
(Refer Slide Time: 50:12) the displacements corresponding to the end reactions; these are the
standard ones, defining the member end deformation in terms of this thing. Once you have this,
17
the fourth is find R prime, the effective load vector; the effective load vector becomes the loads
at the nodes multiplied by the corresponding displacements for each degree of freedom that is
what you would find out.
Once you do that, the final thing is solve R is equal to K r plus summation (ai transpose Si0) plus
summation (ai transpose ni0) for r. Once you find r, six, find Si and once you find out Si, the
problem becomes a statically determinate problem for which you can solve for anything. This, in
essence, is the entire problem.
In the next lecture, I am going to solve one more problem where I am going to introduce the
factor that one of the members may be flexurally rigid how does the kinematic relation look in
that particular case? Then, I am going to introduce all the member load effects like temperature
etc. Thank you very much.
18
Structure Analysis - II
Prof. P. Banerjee
Lecture 33
Good morning. Today, we are going to continue looking at the matrix stiffness method and the
specific problem. Till now, we have been looking at a member that is essentially subjected to
flexure because we have neglected both the shear and the axial deformations in the member.
Today, I am going to take up an example in which you will have a member that is essentially
subjected to only actual forces and we cannot neglect the effect of axial deformations. What do
we do in such a situation? Today again, we are looking at matrix stiffness method application
with respect by taking examples. Let me show you the problem statement and then we will see
how to tackle it.
This is the problem that we are looking at: what you have is a frame and this is more in terms of
a braced frame; this is a bracing (Refer Slide Time: 03:20). Bracings typically are pinned at two
ends because essentially if you look at what happens to bracings, bracings in a frame are
essentially to stiffen the frame and to take axial forces only. Bracing members are not normally
designed for flexure. This is a braced frame a typical braced frame.
Let us look at the typical loading that you would have: you would have this kind of a loading and
this kind of loading (Refer Slide Time: 03:59). Let me give some numbers: this would be EI, this
would be EI, this would be EI; these are all frame members; note that this pin is only the
2
connection of this brace it is pinned at both ends, so essentially this is only subjected to axial
and so, the only thing we have to do is to define its axial rigidity. Let me define some... this
could be 4 meters, this could be 4 meters, this could be 4 meters this could be 4 meters, this
could be 100 Kilonewtons and this could be 50 Kilonewtons. Let me say that EI is equal to 1 into
10 to the power of 5 Kilonewton meter squared and EA is equal to 1 into 10 to the power of 8
Kilonewton. These are just numbers that I am fitting in to ensure to explain what this is. Let me
put A, B, C, D. Here if you look at it, we have to look at the members.
Now, ab is a member defined in this way and its Kab is defined by 2EI upon L into (2, 1, 1, 2).
This is a, this is b, v is going to be thetaab, thetaba and S are going to be Mab, Mba and since there
is no loading, Si0 is going to be 0, 0. This is our definition of member ab. By the way, let us go
back here.
How many degrees of freedom does this structure actually have? Essential, in other words, there
is a degree of freedom which is thetaD but since moment at that point is equal to 0, this is
essentially a three degree of freedom structure, which are one, r1, two, r2 and three, r3 (Refer
Slide Time: 08:13). This is member ab.
Similarly, member bc is continuous over both b and c and therefore, what we have over here is v
is equal to thetabc, thetacb, S is equal to Mbc, Mcb and Si0 is equal to.... Here, there is a loading
(Refer Slide Time: 09:11) and the loading is 100 Kilonewton per meter and this is going to be PL
upon 8; PL upon 8 is going to be plus 50, minus 50 and Kbc is going to be 2EI upon 4 into (2, 1,
1, 2). So much for member bc.
Member cd: This member is continuous over c but pinned at d and therefore its v is equal to
thetacd, its S is equal to Mcd, its Si0 is 0 and its Kcd is 3EI upon L that is for member cd. Now we
have another additional member bd. What kind of a member is it? It is actually a member of this
type; pinned at both ends and only subjected to deltaab (Refer Slide Time: 11:03). Its v is equal to
the axial deformation, its S is equal to Pab; since there is no axial force in the member, its Si0 is
going to be 0 this is b, this is d, so this is deltabd, deltabd, deltabd and K of bd essentially relates to
this and this and you already know that since deltabd is equal to Pbd into L upon EA, so Kbd,
which is Pbd in terms of deltabd is going to be equal to EA upon L, and L here in this particular
case is 4 root 2. (Refer Slide Time: 12:24) This is an axial member.
In fact, let me just point out that this is something that we had not discussed and I am introducing
today, but the concept is the same whether it is a flexural member or it is an axial member; the
only thing that happens is the member degree of freedom for flexure is the rotations from the
chord at the two ends whereas for an axially loaded member, it is actually equal to delta, which is
the axial deformation and the load is the axial force in the member. In fact, I will just expand
this; this, of course, is only a member that is subjected to axial; you might have a situation where
you have a member that is subjected to both axial and flexure. How do we consider that?
I am going to divert a little bit, come back to this problem a little bit later and treat how we may
have a problem where we have a member. I am going back to the basics: Starting from flexibility
and then going off from there, because you can always invert the flexibility matrix and get the
stiffness matrix for a member I have already shown that many times, so let us go back to this.
Here, what we have is Mab, Mba and Pab. This is the situation and therefore here the flexural
rigidity is EI and the axial rigidity is EA. This is a situation where we are considering both axial
deformations as well as flexural deformations in a member. This typically has to be done
although we tend to neglect it in a general situation, but suppose we want to consider it, how do
we write down the v?
If you look at it, the v in this particular case are equal to thetaab (I am just saying a, b), thetaba and
deltaab; deltaab is here (Refer Slide Time: 15:47); this is the deformation the change of length of
member ab axially; this is different from the earlier situation. S is equal to Mab, Mba, Pab. If S is
equal to Kv... sorry, flexibility this is what we find out by actually going through the process. If
you look at this, you will see that if I give a force P, you are not going to generate any bending
moment and therefore, you will see that thetaab and thetaba are equal to 0 if Mba and Mab are equal
to 0.
Putting that in, we can see actually that v is equal to (L upon 3EI, minus L upon 6 EI, 0, minus L
upon 6 EI, L upon 3EI, 0, 0, 0, L upon EA) into Mab, Mba and Pab. This is relatively easy to prove
because v is nothing but equal to ...(Refer Slide Time: 18:36) and for a straight member, Pab does
not give rise to thetaab and thetaba and Mab and Mba do not give rise to any delta that is the
reason why these are 0s. Therefore, deltaab is equal to PL upon EA we already evaluated earlier.
This is all that happens. If you have a member where both flexural and axial deformations have
to be included then all you have is this term but you will see that in a straight member these two
terms are 0, so actually the axial deformations are uncoupled from the flexural deformations. But
this in essence is what happens. Then, of course, you have to find out Si0, which are essentially
(FEM)ab, (FEM)ba and fixed end axial force in ab (FEP)ab.
Remember I talked about temperature? If you have temperature, if you have the neutral axis
expanding, then you actually have developed a fixed end force that you can compute easily. This
in essence is the overall but by and large, since these are uncoupled, we do not consider flexure
and axial together and that is the reason why if I go back to the problem that I was looking at, I
have one member that is an axial member and therefore, the axial member only has axial force
and this is EA upon L. We have written down, we have four members, three degrees of freedom,
we have written down all the relationships for all the members. What is the next step? The next
step is kinematic relationships take every degree of freedom, put displacement equal to one and
then find out what the member deformations are under that particular loading.
I am going to give r1 is equal to 1. What is going to happen? r2 and r3 have to be equal to 0 so this
will also come here. Note that these members are axially rigid so they cannot deform, but this
member can deform and it does deform, in fact. The only way it can go is this way and so, if I
look at it, this is how it goes. We have here this. By the way, this one does not go this way, this
one goes this way because here you have 0, so we have this; this is my theta, this is going to be 1
by 4, again this is going to be 1 over 4, this is going to be 1 over 4.
Now, the other thing that we have to find out is how much has this member shortened by. To
find out how much it has shortened by, we need to draw the perpendicular because note that
whatever it has moved perpendicular by, that does not change length and so when we drop a
perpendicular here, this is the amount of change of length, this is 45 degrees, this is 1, so this is
going to be 1 over root 2, this is 90 degrees, so this is sine of 45 degrees and sine of 45 degrees is
1 upon root 2. That is the most important thing and then of course, we have theta2 and theta3
equal to 0, so you can plug that in. Those are relatively easier.
You get theta, you get this, you get this (Refer Slide Time: 24:51) that is for r2 is equal to 1.
Note that this point goes nowhere and therefore this does not change length. The only thing that
you have is this is equal to 1 and this is equal to 1. When you put r3 is equal to 1, all you get is
this, this member remains the same. Therefore, if we were to write down our as, aab is equal to...
corresponding to the first degree of freedom, this is 1 by 4, 1 by 4, corresponding to the second
degree of freedom, this is 0 1, corresponding to the third degree of freedom this is 0 0.
If we look at abc, corresponding to the first degree of freedom it is 0, 0, corresponding to the
second degree of freedom it is 1 0 and corresponding to the third degree of freedom, it is 0 1.
Then, we have acd: corresponding to the first degree of freedom it is from the chord to the tangent
is one-fourth, second one, 0, third one, 1, and abd: what is abd? This is the axial shortening or
lengthening due to the unit displacement and in this particular case what you have is (Refer Slide
Time: 27:01) a shortening due to r1 and shortening is defined as minus, so you have minus 1 over
root 2. Now what is the shortening in these two? It is 0 0.
You will see essentially the overall concept, whether you have an axial member or whether you
have a flexural member or whether you have a member that has both flexural and axial, all that
happens is that you have to consider both the effects together. That is why in this particular case
since bd is only an axial member, this (Refer Slide Time: 27:47) corresponds to the axial
deformation in the member bd due to r1 is equal to 1, due to r2 is equal to 1 and due to r3 is equal
to 1. In essence, there is no difference in the entire scheme of things whether it is an axial
member or whether it is a flexural member. Let us go through some of the numbers. I am not
going to be solving this entire problem because all I wanted to introduce was the concept of this.
Ultimately, what we have is; Si is equal to Ki ai into r plus Si0. In that way we can find out for
each member and for the flexural members ab, bc and cd; let me do it; for ab what we have is Ki
is equal to 2EI upon 4 into (2 1, 1 2) and my ai is equal to 1 by 4, 1 by 4, 0 1, 0 0 (Refer Slide
Time: 29:33). This implies that K into a is equal to I am going to keep EI upon 2 outside and
the inside then becomes 2 by 4 plus 1 by 4 is 3 by 4, this way 3 by 4, this way 1, this way 2 and
this way 0 and 0. Similarly, we can do it for bc. For bc we have 2EI by 4 into (2 1, 1 2) into abc,
which is 0 0, 1 0, 0 1 so this implies that Ka is equal to EI upon 2, the first one is 0 0, the second
one is 2 1, the third one is 1 2. This is Ka for bc and this is Ka.
10
Let us write it down for cd. For cd, Ki is equal to 3EI by 4 and cd is 1 by 4, 0 1. This basically
becomes 3EI by 16, 0 3EI upon 4 that is for cd, this is Ka for cd. For bd, this is equal to EA upon
4 root 2 multiplied by minus 1 over root 2, 0 and 0 so this implies Ka. Note that whether it is a
flexural member or whether it is an axial member, the concept is still the same, only thing is
here, we have EA and so if I look at it, this becomes minus EA upon 8, 0 0. We know the values
of EI and EA, so we can plug those in.
Now, once we have that, ultimately, we have this: R prime is equal to summation ai transpose Ki
ai summed over all the members into r plus summed over all the members of ai transpose Si0 plus
summed up over all the members ni0. Let us first find out this (Refer Slide Time: 33:22). We
know this one only exists for bc this only exists for bc it is 0 for all the others and the major
thing is that the ni0 are the reactions.
Again, if we look back at the shape corresponding to this, there is no vertical displacements of
these points and in r2 and r3, there are no vertical displacements. In this particular case, we can
see that this is going to show up as a zero vector this does not exist and this will only show up
for bc. Let us do this for first bc; for bc, this is the only one: ai transpose Si0 is going to be equal
to ai transpose would be 0 0, 1 0, 0 1 and Si0 we have already seen was equal to 50 and minus 50
and if we look at that this becomes equal to 0 then 50 and minus 50.
11
Now, if we look at the R and we find out the work done by the nodal forces, we will see that in
this particular case, this becomes 50 and these do not do any work. Next, we have to find out the
ai transpose into Ki ai for each member and I will first do it for the members which are the
smaller ones and for those, just trying to find where I have written down the... here, Ka.... If I
pull these together, I have for cd, ai transpose is going to be equal to 1 by 4 and 0 1 and here we
have 3EI upon 16 0 3EI upon 4 and if I do that you will see that this becomes essentially 3EI by
64 0 3EI by 16, then we have 0 0 0 and then we have 3EI upon 16 0 3EI upon 4 this is the
contribution of cd to the structure stiffness matrix (Refer Slide Time: 36:59).
12
Similarly, we can find out for bd. Then, ai transpose becomes minus 1 over root 2 0 0 and here
we have minus EA upon 8 0 0 and if you look at this, this essentially becomes EA over 8 root 2 0
0 and everything else is 0 (Refer Slide Time: 37:43). Why do you think that is? That is the
contribution of bd to the stiffness matrix. If you look at it, what does this mean? It means that it
only provides rigidity towards r1. That is true because when you have r2 and r3, the bracing, the
member does not deform.
In fact, the bracing member is only provided to provide an additional rigidity to the lateral
displacement that is the whole purpose of the bracing member; it is not there to take vertical
loads; when there is lateral displacement, it is to provide and therefore, it is not surprising that it
adds to this. We have done for cd and bd; can we do it for ab and bc?
For ab it is going to be equal to ai transpose into Ki ai; ai transpose for ab is going to be equal to
one-fourth one-fourth, 0 1, 0 0 that is just ai transpose and then, Ki was equal to EI upon 2 into (3
by 4 3 by 4 1 2, 0 0). If we multiply those we get the following: this implies the contribution of
ab to the stiffness matrix (Refer Slide Time: 39:39) is equal to if you look at the first one, it is
going to be 3 by 16 plus 3 by 16 that is going to be 3 by 8 so this is going to be 3EI by 16 the
first one; the second one: this way is going to be 0 into 1 by 3 so this is going to be 3EI by 8; this
one 0 so this going to be 0; now we do the second one: this is going to be 1 by 4 plus 2 by 4 so 3
by 4 so 3 by 8, so it is going to be 3EI by 8; next one, 0 into 1 and 1 into 2 so this is going to be
EI; the next one is going to be 0 0 so this going to be 0; and the third one is going to be 0 0 and 0.
In other words, ab does not contribute to the stiffness corresponding to r3 that is not surprising;
there is no displacement in ab due to theta r3 and therefore it will not contribute. Whenever there
is no deformation in a member due to a particular degree of freedom it does not contribute to that
degree of freedom. We are now done with ab.
13
We move on to bc. For bc, ai transpose is nothing but 0 0, 1 0, 0 1 and over here, we have EI by
2 into (0 0, 2 1, 1 2) and let us see what happens. This first one is 0 into 0, so it is 0; 0 into 0 is 0,
0, 0 into 0 is 0 the first one is here; the second one: this is 0, this one is 2, so this is EI, third: if
you look at it, it is 1, so this is EI over 2; third one: this is 0, this is EI by 2 and the fourth one is
2, so this is EI we have got Kbc. Let me put up all the four here and add it up.
Since all the EIs are the same, my stiffness matrix, my K is going to be equal to 12 over 16, so
this is going to be 15EI upon 64 plus EA by 8 root 2, the second one is going to be 3EI upon 8,
the third one is going to be 3EI upon 16; then we again look through this, we get 3EI upon 8, we
14
look through this we get 3EI upon 16 and if we look at the third one we get EI, EI and 0 so we
have 2EI, we have EI by 2 0 0, so this is EI by 2, here EI by 2 0 0, so it is going to be EI by 2 and
the last one is going to be EI plus 3EI by 4, so that is going to be 7 EI by 4 this is my K; my R
I am just repeating it is going to be 50 0 and 0 and my summation ai transpose Si0 is going to
be 0 then 50 then minus 50.
If we look at the final solution, what we get over here is going to be equal to 50 0 0 is equal to
(15 EI by 64 plus EA by 8 root 2 then 3EI upon 8 3EI upon 16 3EI upon 8 2EI, EI upon 2, 3EI
upon 16, EI upon 2, 7 EI upon 4) into (r1, r2, r3) plus 0 50 and minus 50. Ultimately from this we
can solve for r1, r2, r3. I am not going to solve this because if you look at it, you will see that in
this particular case it is a 3 by 3, you can plug in the values of EI and EA.
What is the value of EI? The value of EI it is going to be equal to 1 into 10 to the power of 5 and
the value of EA is equal to 1 into 10 to the power of 8.
If we look at this particular value (Refer Slide Time: 46:36), do you notice something? You will
see that this stiffness value that you get is essentially overpowered by the value of the bracing. In
fact, that is the whole point. The point why the bracing is provided is to provide lateral stiffness
and that is where it has provided stiffness because r1 corresponds to the lateral motion of the
structure.
Once you have found out r1 and r2, then you can always go back and find out and show what
those would be. How would I find out my Si0? My Si0 would be this. Since I know r and I have
Ka for all the members, I can find out Si. And for the bracing member, the Si that will come in
will only be the axial force. In other words, the point I tried to make today was that it does not
matter whether a member has flexural deformation, whether it has axial deformations or a
combination of flexural and axial because all that happens is the S, the v and the K for the
member is different for each case, but it does not matter.
15
I am just putting down the steps. The first step is given the member, a priori we have to know
whether the member is flexural, whether it is axial or whether it is flexural plus axial. if I have a
frame member where you neglect axial deformation, then it is a flexural member; if it is a truss
member, in other words pinned at both ends, then it is only axial member; if we have a frame
member where we do not neglect axial deformations, then it is a flexural plus axial. Once you
have that, then all you know is once you have this, you define your v, you define your S and you
define your Ki (Refer Slide Time: 49:33). Once you have this, the next step is vi is equal to ai r;
in this, the kinematic relationship... this is the kinematic relationship; once you know what this,
you can always find out what this is by giving each displacement corresponding to each degree
of freedom.
16
Then you see the third step becomes Si is equal to Ki ai r plus Si0. This I know, this I know, this I
know, this I can find out. In essence, the point I am trying to make here is that in the stiffness
method, the only difference it makes, whether you have a truss member, whether you have a
frame member where you neglect axial deformation or you have a frame member where you do
not neglect axial deformation, all that it does is it only affects this nothing else; everything else
comes out automatically.
Therefore, the beauty of the stiffness method lies in the fact that it does not really matter what
kind of member you are dealing with all you need to know is when you give each individual r
equal to 1, you have to find out the ai that gives the member degrees of freedom in terms of the
displacement degrees of freedom. I am going to stop here today; I have introduced to you the
concept that axial deformations are included, not included does not matter. We are going to
continue looking at more problems and each time, each example that I look at, I will highlight
one important aspect of the application of the stiffness method. In general, the stiffness method is
easy; what I am going to do is specific cases.
Next time, I am going to be introducing you to rigid members. Suppose I have a member that is
both flexurally and axially rigid, what happens then? How do we consider that effect? Then, we
are going to look at.... We have only looked at member effects where Si0 depend only on loads.
What happens when temperature becomes an issue? You have temperature effects. How do we
include those? Those are all the things that I am going to look at in the next few lectures and
hopefully, these will illustrate how the stiffness method, which is essentially the matrix
displacement method, how that is utilized to analyze structures.
17
Prof. P. Banerjee
Lecture 34
Good morning. We are continuing by looking at examples related to the application of the matrix
displacement method or the stiffness method and that is our today's lecture: matrix displacement
stiffness method with examples. Let us continue looking at more examples and today I am going
to introduce to you a situation where maybe one of the members is rigid.
In other words, its flexural rigidity is significantly higher than the other two members. How do
we deal with such a problem? Let us look at the problem that we are looking at. This is the
problem: a, b, c, d where a is fixed, b is continuous, ab has EI flexibility, bc has EI flexibility and
cd is the rigid member. This is actually a two degree of freedom structure and the r1 and r2 that I
have defined over here (Refer Slide Time: 03:01) the displacement and rotation at b are my
degrees of freedom.
What has been asked of you is the typical structural analysis problem: Find out the displaced
shape and the support reaction and the bending moment given this load (Refer Slide Time:
03:19) this is the only load that is there on the structure. Let us now look at it.
What I will do is I shall look at the solution process for this problem to begin with and then, I
will change this problem around a little bit to introduce to you another concept of the application
of the stiffness method. What is the first step? Identifying the degrees of freedom and all the
other forces etc. The next aspect is to find out the degrees of freedom for each member; first of
all, find out the type of element; remember that there are only two types of elements that we have
discussed: one is the standard fixed-fixed element and the second one is the modified element
with fixity at one end and hinge at the other end, pinned at the other end.
We have to decide which one to take and then proceed. Note over here that any member that is
rigid is not considered a member in the stiffness method itself. How does the effect of that
member come into the picture? The effect of that member actually comes into the kinematic
relationship. Although cd is a member and we have to find out the bending moment diagram for
cd also, in the stiffness method since it only refers to flexible members cd will not be considered
a regular member in the analysis. In effect, there are only two members in this particular
problem: ab and bc.
Let us look at ab. If you look at ab, a, b this is the standard element (Refer Slide Time: 5:35)
because a is fixed and b is continuous, therefore, you have this element. The degrees of freedom
are thetaab and thetaba these are the two member end deformations that we are interested in. My v
is going to have thetaab and thetaba. What about K? Obviously then, K of member ab is the
standard 2 EI upon L in this particular case, L is 4 meters into (2 1, 1 2). This is my Kab and
finally I should find out the fixed end moments. Since there is no member load, fixed end
moments are both 0. This is for member ab. Let us look at member bc.
Member bc is rigid. At b it is continuous and it is continuous at c also so the element is of this

form, this is b, this is c. The degrees of freedom are thetabc, thetacb and therefore v of bc is equal
to thetabc, thetacb. Note that this one has a load in the member. Kbc is the same standard one: 2EI
upon L and L again is 4, so it is 2 L (2 1, 1 2). The fixed end moment is PL upon 8 (FEM)bc we
know that so it is going to be 100 into 4 divided by 8, that is 50 Kilonewton meter which is
(FEM)cb (Refer Slide Time: 08:31). Thus, we have evaluated what the degrees of freedom are,
what the stiffness matrix is and what the fixed end moments are. The next step is the kinematics.
Here r1 is equal to 1. This point where will it go? Since this is fixed, it can only move in this
direction (Refer Slide Time: 09:22), so r1 is equal to 1 means b moves here. Now, this point also
has to move horizontally by 1, but note that since this is an inclined member it cannot go this
way so it has to move along this, so it comes over here whose horizontal component is 1 and then
from geometry you will see that this is equal to 3 by 4 and this is equal to 5 by 4.
This is the point that I was trying to make: Look at it, what happens over here is that this member
which is rigid is going to go in this way. What is this angle? This angle is equal to 5 by 4 and the
length is 5 meters, so this angle is 1 by 4. Note that since this has rotated clockwise by 1 by 4, to
keep continuity, this also, the tangent has to go up by 1 by 4 (this is 1 by 4 the tangent) so that
this continuity is maintained this is an important point that I cannot highlight. The rigidity of
this ensures that if this rotates by 1 by 4 then the tangent at this point (Refer Slide Time:
11:01) remember that this is a continuous member, this is the original position, so if this goes
clockwise by 1 by 4 so will this have to go to keep continuity because originally, it is this way
this and this; now since this has rotated by 1 by 4, this also has to rotate. If we look at the
displaced shape over here this is how it looks. This one will go by 1 and if we look at it from the
chord, these are the tangents and this tangent is equal to 1 divided by length and so is this: 1 by
4. Here, remember this tangent is this way, so this tangent is this way, so this one will go in this
fashion but over here, the tangent will have to be off by 1 by 4. If we look at this and we join
this, then this is the chord connecting c, b, a, d. If you look at this particular one, how much is
4
this angle? This angle has to be equal to 3 by 4 divided by 4, so this is going to be equal to 3 by
16.
What about this (Refer Slide Time: 12:50)? From the tangent if you look at this particular
situation, what is this angle equal to? Between the horizontal and this; between horizontal and
this, this is going to be equal to the same because this is the same angle, so this is 3 by 16 but
from here, the tangent is 1 by 4, so from to the tangent is equal to 3 by 16 plus 1 by 4, which is
equal to 7 by 16. This one we need not consider. Therefore thetaab is equal to 1 by 4, thetaba is
equal to 1 by 4. Are these anticlockwise? From the chord to the tangent both are anticlockwise,
therefore you have plus 1 by 4 and plus 1 by 4. Over here from the chord to the tangent, this is 3
by 16, so thetabc is equal to minus 3 by 16 and thetacb is equal to from the chord to the tangent
clockwise so it is minus 7 by 16. That is all we need cd does not come into the picture because
cd is a rigid member.
You look at the kinematics. The kinematics is very, very important when you have a rigid
member. The thing is that if this had not been a rigid member, then this would have gone
something like this (Refer Slide Time: 14:33) and this would have remained the same. But since
this is a rigid member as soon as this has to rotate, from the tangent this one turns out to be 1 by
4 because it is rigid and as soon as that is 1 by 4, this also has to rotate this is an important
point that all of you should understand in the kinematic relationship wherever rigid members are
considered.
Today, my focus is going to be on the effect of rigid members in the analysis of frames. For r 1 is
equal to 1, this is the displacement pattern. Now, let us look at r2 is equal to 1.
Here r2 is equal to 1 is rotation. Always remember that rotation is always easier to do, this is the
rotation let us look at it. Can this joint (Refer Slide Time: 15:45) rotate without causing any
displacement? Sure, it can it can go in this fashion and I will show you. Please note one very
5
important thing and that is that I always show displacements as exaggerated and it is obvious that
this is not. In other words, the tangential motion does not increase the length of this.
Therefore, even though I write 3 by 16 etc it is not really 3 by 16, it is 3 by 16 of r1, r1 itself is
very small; this is 1 into r2, r2 itself is very small; it is just that I am taking r1 and r2 equal to 1 to
essentially establish. Here, what is the rotation? 0. Rotation here. This is b, a, c and d, so ab
is 0, ba is 1, bc is 1, cb is 0. Now, can we write down the a matrix for this particular structure?
For ab, the a matrix is equal to 1 by 4, 1 by 4 and for r2, it is equal to 0, 1; for bc, a is equal to
minus 3 by 16, minus 7 by 16 and 1, 0 for r2 this is my a matrix (Refer Slide Time: 17:54).
The next step is relating Si; Si is equal to Ki into ai r plus Si0. What we have to establish is Ki for
ai; for ab, Ka is equal to 2 EI upon 4 into (2, 1, 1, 2) into one-fourth, one-fourth, 0, 1; this is
equal to 2 EI upon 4 into 2 by 4 plus 1 by 4, so this is 3 by 4, 2 into 1, so this is 1, this is 1 by 4
plus 2 by 4, 3 by 4, and this is 2 this is for ab and Si0 is 0, 0 this is for ab and let us find out
that for bc too.
If we find out for bc, you will see that Ka is equal to 2 EI by 4 into (2 1, 1 2) multiplied by a,
which is (minus 3 by 16, minus 7 by 16, 1, 0). If you put this together, what you get is 2 EI by 4
and minus 6 by 16, minus, so this is minus 13 by 16; similarly, this minus is going to be minus
17 by 16; this is going to be equal to 2 and this is going to be equal to 1. S i0 is going to be equal
to 50, minus 50 Kilonewton meter. We have written these down.
The next step now is to find out; R effective is equal to summation over all the members ai
transpose Ki ai into r plus summed over all the members that have fixed end moments plus (Refer
Slide Time: 22:09) and in this particular case, I just wanted to tell you that How do we get ai
trans a bar? You will see that a bar essentially is for member bc, which is the loaded member,
we need to find out what is the vertical displacement of b that is 0. So we can write down for
bc because bc is the only member that a bar is equal to bc that is 0, then the vertical
displacement of c which is 3 by 4, so this is 3 by 4 and the displacement corresponding to r2 is
vertical 0 0, so we can put down 0 0.
An important point to note over here is that. What are nio? Since it is 100 at the center, the
reactions are going to be 50 50 upwards (Refer Slide Time: 23:22). This is what we have and
therefore for ab, since there is no fixed end moment and there is no fixed end reactions
obviously, therefore these two terms do not exist for ab and the only term that exists is ai
transpose Ki ai. For ab, Ka is already given, so a transpose becomes 1 upon 4, 1 upon 4, 0 1
that is a transpose multiplied by 2 EI upon 4 into what we have already computed; 3 by 4, 3 by 4,
1 2 and this is equal to 1 by 4, 1 by 4, I will put EI upon 2 outside, so you have 1 by 4 into 3 by 4
that is 3 by 16 plus 3 by 16 that is equal to 3 upon 8, here you have 3 upon 4, this way you have
1 by 4 plus 2 by 4, 3 by 4 and this way you have 2. That is ai transpose Ki or we can call this as
the contribution to the structure stiffness matrix of ab. Now, what we need to do is we need to
find out the same thing for bc.
Of course, for bc we also have to find out the ai transpose Si0 as well as ai prime transpose into
ni0 because bc has member loads. What do we have here? First is ai transpose Ki ai. Now, a
transpose for this will be minus 3 by 16, minus 7 by 16, 1, 0 and the Ki ai has already been
obtained, that is equal to minus 13 by 16, minus 17 by 16, 2 1. If we look at that, this becomes EI
over 2 and what do we have? 39 upon 256 and 119, 119 and 39 is 158 upon 256, so 1, sorry 39
and 119 is 158, so it will become 79 upon 128, this one is going to be equal to minus 13 by 16,
this is going to be minus 6 plus minus 7, so that is going to be minus 13 by 16 and this one is
going to be 2. That is the contribution to the structure matrix of bc. I also need to find out ai
transpose nio, so if I do that, ai transpose Si is going to be equal to minus 3 by 16, minus 7 by 16,
1, 0 multiplied by Si, which is 50, minus 50 and if we look at it, it is going to be equal to minus
50, so I can take it as 1, so it will be minus 10 upon 16 into so it will be minus 500 upon 16,
500 upon 16 will become equal to minus 125 upon 4 and on this side, I will have 150, so this is
going to be 50. This is going to be ai transpose Si.
Finally, I need to do ai prime transpose into ni0. ai prime transpose is going to be equal to 0, 3 by
4, 0, 0 into 50, 50 is going to be equal to 150 by 4 and 0. By the way, what is R prime equal to?
You will see that there are no nodal loads, so R prime has to be equal to 0, 0. Let me put it all
together; I am doing the summations.
I have 0 0 equal to now I need to do Kab it is given by this (Refer Slide Time: 30:08) and Kbc
is given by this, so this is equal to. I can put EI by 2 outside and inside, I will have 79 plus 48,
79 by 48 is 127 upon 128, here I have 13 by 16 and 12 by 16, so I will get minus 25 upon 16 (13
by 16 plus 12 by 16, minus 25 by 16) and here also, we will get minus 25 upon 16 and over here,
10
you have 2 plus 2, which is equal to 4 and this times r1, r2 plus of course that is minus 125 by 4,
plus 50 plus my ai transpose, which is 150 by 4, 0. This is the relationship that I have and if I
plug in this, what will we get?
I will put this thing down, put this on this side and put the EI outside and then what we get over
here is 127, so this becomes 0.4961, this one is 2 and this one is 25 upon 32; let me go back
and check whatever we have written down; this is equal to 3 by 8EI r1 plus EI upon 2 that is
Mab and Mba is equal to 3 upon 8EI r1 plus EI r2 correct; right now, I am just going back and
checking the statements that I have made, so Ka I have 13 upon 32EI into r1 plus EI r2 plus 50
and in the bottom, we have minus 17 upon 32 plus EI.
If we look at this, this is minus 13 upon 16 (Refer Slide Time: 33:41) and Kab is 3 upon 4, so
actually if you look at this, when you add the two, you do not get minus 25, you get 12 and you
get minus 13 plus 12, so you get minus 1 upon 16 and this is also minus 1 upon 16 that is what
I was checking and then this equation becomes minus 0.03125, minus 0.03125 is equal to and
on this side, we have it equal to minus 125 and plus 150; just let me check this over again let us
do all the checks properly.
What we get over here is minus 3 upon 16 into 50 plus. Again, here is a mistake: minus 3 into
50 (Refer Slide Time: 35:17) and here we get plus, so basically you will get 4 upon 16 positive,
so this is going to be plus 4 upon 16, which is so this is going to be plus 50 upon 4. Here, we
have plus 50 upon 4, so plus 50 upon 4 plus 150 by 4 is equal to 200 by 4, which is 50, so 50 and
50, so what we get on this side is minus 50 and minus 50, of course into sorry r1, r2. This is the
final equation that we get.
11
If we look at this equation, the solution turns out to be equal to you can solve for this: r1, r2 is
equal to 1.0089 upon EI and here, I have (2, 0.03125, 0.03125, 0.4961) into (minus 50 and minus
50) is equal to minus 102.46 upon EI, minus 26.60 upon EI this is my r1 and r2 (Refer Slide
Time: 37:14). I know what my r1 and r2 are and once I have got my r1 and r2, can I draw my
displaced shape? Let us try to do that; we will draw the displaced shape. How do we draw the
displaced shape? Well, let us look at it.
This is this way, this is this way and this is the undisplaced shape where this is rigid. We know
that the r1 which is this is minus, which basically means r1 was positive in this direction (Refer
12
Slide Time: 37:59), so this will have moved by 102.46 in this direction. Furthermore, the
anticlockwise is considered as positive, so minus 26.6 is that if this was straight, then it has
moved clockwise, so this is your tangent at this point for this and therefore, this also will have
moved 90 degrees.
Note another thing that, this point would have moved by 102.6; therefore, this will have moved
at 90 degrees to this; this is my point and if I look at this particular point and I were to draw the
displaced shape, the displaced shape would look like this: this one will have moved and note that
this is rigid this is rigid and so it has to move in a straight line.
How much is this? If we look at this, this is equal to 102.46 divided by EI and this one will turn
out to be 3 by 4 of that, so 3 by 4 of that is equal to 76.845 upon EI and this is equal to then
128.08 upon EI. If this is 128, then this angle turns out to be equal to divided by 5, which is
equal to 25.615 by EI; if this go this way, note that this is the straight line that means this is
also equal to and therefore this one will also have moved, this has moved anticlockwise, so
this will also have moved anticlockwise and this should be it, where this is equal to 25.615 by EI
to the horizontal.
If we look at this, this is going to look like this, this and this, where this angle and this angle
both of them are equal to 26.60 upon EI and this displacement is equal to 102.46 by EI. This is
the displaced shape. This is straight, the angle here is 26.6, this one is 78 point and note that
since this is this way, this will actually go this way and then go in this manner and this. This is
the displaced shape that we have for the structure. Once we have the displaced shape and we also
have our r1, now I have got my Si for both ab and bc.
Therefore, Sab is going to be equal to 2EI upon 4 into we have calculated this earlier this is
Ki ai, this is equal to this into r1 and r2, r1 and r2 are 102.46 upon EI, minus 26.60 upon EI, there
13
is no Si0, so that is 0, so Sab is equal to minus 51.72 and minus 65.02 Kilonewton meter this is
Mab, this is Mba. Having done that for ab we can find for Sbc.
Sbc is equal to 2 EI upon 4 into (minus 13 by 16, minus 17 by 16, 2 1) into minus 102.46 upon
EI, minus 26.60 upon EI plus Si0 which is 50 and minus 50. These imply that in this particular
case, Mbc, Mcb are equal to plus 65.02 and minus 6.87 Kilonewton meter. Once you have got the
member end moments we can now take the free body to be able to get the support reactions.
14
Now note one thing. What are the things that we know? We know that this is equal to 51.72
(Refer Slide Time: 45:01), we know that this is equal to 65.02, so what we know is we know this
and therefore, we will know the shears here because there is no other load and the shears are
going to be in this direction and they are going to be equal to this plus this divided by 4, which
works out to be 29.19, 29.19 these we know. Furthermore, what else do we know?
For this one, we do not know what the vertical reaction here is; there will be a vertical reaction
we do not know what that is. Therefore, what we have to now do is we have to look at this
particular member it has a 100 Kilonewton force acting here, we have a 65.02 moment acting
here (Refer Slide Time: 46:04) and we have a clockwise moment of 6.87 acting here. Therefore,
you have one aspect, which is this minus this divided by 4 that is one and the other is 50 50
both sides, so when you combine the two you get that this is equal to 64.04 and this is equal to
35.96.
Once we know that, from this we know that this is going to be 64 and so over here we have
64.04 Kilonewton, this is Kilonewton meter, this is Kilonewton. I know the reactions at this end.
Now let us look at this (Refer Slide Time: 47:03). If this acts here, this has to act here, this has to
act here that is going to be 29.19. Therefore, if we look at this by this side, we will have to
have the situation that this will be 6.87 and over here, we have this and this; this is going to be
equal to this and this is going to be equal to this. Therefore, over here, what kind of reactions do
we have?
If we go through with this, you will get 35.96 and over here, this is going to be acting in this
direction 29.19 and moment over here, if you check these out, you will see that the moment will
turn out to be 0, so this is the reaction at this support and this represents the reaction at the fixed
support at a, this is d; these are the reactions at the hinge and these are the. This is the way we
can find out the reactions and once we have these member end moments and these we can now
draw the bending moment diagram.
15
Once you know that, this is the final one: drawing the bending moment diagram. I will draw the
final bending moment diagram without going into how I have done it I will leave it up to you
to check it out. This is 51.72 (Refer Slide Time: 49:02), this is 65.02, so what we have over here
is this, this is equal to this kind of so that tension at this side, compression at this side and over
here, you will have just the opposite way this way (Refer Slide Time: 49:32), here you will
have the same and here, you will have 6, so if we do this, this is going to be 100, so if we draw
this, it will look like this, this is going to be the same 65.02, this you can compute is 63.06, this is
6.87 and over here, we have the same.
Of course, here, this is the bending moment diagram but that does not give rise to any flexure for
the simple reason that this is a rigid member, so you do have a situation where you have this, but
although this looks like this, the bending moment looks like this, there is no flexure; on these,
you have flexure. I have got the bending moment diagram, I have got the reactions and I have got
the displaced shape complete problem solved. I am going to leave you with this problem and I
want you to think about it.
Same structure: this is EI, this is EI, this one is now rigid, this is a, b, c, d; all I have done here is
that in the previous case, this was rigid (Refer Slide Time: 51:44) and this was flexible. Now,
this is a more realistic kind of situation where the two columns are flexible and the beam
connecting the two columns is rigid this is a more realistic kind of structure. Why is it
considered rigid? Because typically, you might have a situation where the columns may be EI
but the beam because it is a T-beam, it may be something like 10 or 15EI for all practical
purposes, you can consider it to be rigid. This is the same 100-ton load, ab, bc think about this
problem.
I am going to leave you today with this think about this and we will take up this problem very
quickly next time before we move on to the other kind of load effects or other kind of member
local effects, which are important. Thank you very much. See you next time.
16
17
Structure Analysis II
Prof. P. Banerjee
Lecture 35
Good morning. Today, we are going to be continuing with our exposition on the stiffness method
where we are going to be looking at a few more examples. I will look at one example and then I
will essentially establish how to consider thermal stresses in the stiffness method.
Today's lecture is on the stiffness method and we are going to be looking at examples. The first
example was the one that I had presented in the last lecture at the end.
This was the problem that I had solved in the last lecture: a, b, c, d. this (Refer Slide Time:
02:38) member is now flexible, this member now becomes rigid. This is EI, this is EI and this is
rigid. This is a more realistic problem where the columns are flexible and the beams are normally
monolithic with the slabs and therefore, these are T-sections and the EI values for beams are
typically much larger than those of columns. One of the simpler ways to deal with it when one
member has a significantly larger EI than the others is to assume that it is flexurally rigid
because after all, the rotations over there are going to be small and we might as well neglect it.
How many degrees of freedom do we have? If you look at it, let us go through it. This was the
problem. How many degrees of freedom? It will still hopefully remain two; the only factor that
becomes that it is two is that this is a degree of freedom. Think about it: can you have a rotation
at this point at b without this? Think about it: if this rotates, then this has to go up, if this goes up,
then the only way this can go up is by this moving and if this moves, it moves horizontally.
Therefore, the two degrees of freedom actually are this and this but this is no longer a degree of
freedom, so let us write down, prove to you that this is actually a single degree of freedom
system and for that, let us go through the steps.
How many joints? 4 into 3 is 12. Then, how many restraints? 1, 2, 3, 4, 5 so five restraints. How
many constraints? It is 5; 1, 2, 3, 4, 5. That means there are two degrees of freedom but one of
them happens to be this end rotation, where anyway we know that the moment is there, so I have
only one degree of freedom essential degree of freedom; there is a second degree of freedom
here but I do not want to find out the displacement right now. This goes as a single degree of
freedom system and how do we solve this particular problem? Let us see. First and foremost, ab
and cd are the members.
For ab, what is it? The member is fixed at a and continuous at b and therefore, v is equal to
thetaab, thetaba that is v. What is S? S is Mab, Mba. (FEM)ab, (FEM)ba is equal to S0 is equal to 0,
0 and Kab is equal to 2EI and this is 4 meters, this is 2 1, 1 2 this is your member ab. Now, bc is
a rigid member and therefore you do not consider bc as part of the members.
Therefore, the next consideration would be for cd; cd is continuous at c so it is fixed and at d it is
hinged; this is c, d and therefore, v in this particular case is thetacd, S is equal to Mcd, S0 is equal
to 0 (there is no load) and Kcd is equal to 3 EI by length which is 5. I have found out for each of
the members. Now, I need to find out to how to get the member degrees of freedom related to the
3
kinematic relationship between the member degrees of freedom and the structure degrees of
freedom for which I have to do the kinematic relationship.
Therefore, if we look at this particular situation we have, now note that this is the rigid one
(Refer Slide Time: 08:49) and then we have this which is continuous and here I have pinned. If I
give it a unit displacements r1 is equal to 1, this would imply that this would go here, this point
would also move here but it cannot move only that way it has to move perpendicular to this
and this perpendicular is if this is 1, this is 3 by 4 and this is 3 by 4 and the displacement
pattern looks like this: if this has gone here and this point is here the rigid member is going to
remain straight that is the most important thing. What is this angle (Refer Slide Time: 09:59)?
This angle is equal to 3 by 4 and upon 4 this becomes 3 by 16.
Once that happens, note that if this is the thing, this also has to move by 3 by 16, so the tangent
becomes like this and this member comes like this. Note also that this tangent also has to rotate
because this has rotated and therefore this goes this way and these rotations are 3 by 16, this
rotation over here is also 3 by 16, this is hinged. We are always taking theta from the chord, so
this angle since this is 1 and this is 4, this is going to be 1 over 4, this is going to be 1 over 4
and this is going to be 1 over 4 and 5 by 4 upon 5 because that is the length of the member and
this here is going to be 1 over 4. Since this is my only this thing, aab is going to be equal to ab is
1 by 4, ba is 1 by 4 plus 3 by 16, that is from the chord to the tangent, so if you have that, that is
7 by 16 and note that these are anticlockwise so they are positive so you have 7 by 16 and acd is
equal to 3 by 16 plus 1 by 4 from the chord and this is also counterclockwise so this is 7 by 16.
That is it. Note that this load over here how much is that displaced by? That point, the central
point how much has it displaced by? The central point, by the way, has moved by 1. How
much is this? This is going to be half of three-fourth, so this is going to be 3 by 8, so the load
point has moved up by 3 by 8 this is going to play an important role in our calculations a little
bit later. I have computed the aab.
Note that in both the cases, the fixed end moments are 0 and therefore, all I have is directly Sab is
equal to aab into Kab into r, which happens to be a 1 by 1. If I look at this, this becomes 2EI upon
4 into (2, 1, 1, 2) into (1 by 4 into 7 by 16) into r1. This is equal to I will make it 2 by 4 plus 7
by 16, 8 by 16, so this becomes 15 by 16, so EI upon 2, this is going to be 15 by 16 and this is
going to be 14 by 16 will become 7 by 8, 7 by 8 plus 1 by 4 is going to be 9 by 8, this is into r 1
that is Sab. The contribution to Kab is going to be equal to one-fourth, 7 by 16 into EI upon 2 into
(15 by 16, into 9 by 8). If you look at this, this becomes 63 upon 128 and this becomes 30 upon
128, so 63 plus 30 is 93 upon 128, so 93 upon so this is going to become 93 EI upon 256,
because 128 into 256 that is Kab. Now, let us look at cd. For cd, Scd is going to be equal to Kbc
into abc into r, which is going to be equal to 3 EI upon 5, so it is going to be equal to 3 EI upon 5
into abc is 7 by 16, so this is going to be equal to 21 EI upon 80 into r1. Kbc is going to be a
transpose a transpose is again the same so it is going to be 7 by 16 into 21 EI upon 80, which is
equal to 147 upon 1280 EI (Refer Slide Time: 17:24). When we add both of them up, what do we
get? We get this, the following.
R is equal to K into r plus summation ai transpose Si0 plus. The major point to be noted here is
that there are no Si0 and ni0, so we do not need to worry too much about that. Therefore, the K is
going to be equal to 5 as it is going to be 465 plus 147, so it is going to be 465 plus 147 upon
1280 EI, which is equal to if you look at it, this becomes 612 upon 1280 EI and if we look at
this, this can go through by 4, so this becomes equal to 153 upon 320 EI that is K. What is R?
How do I? Note this is R prime.
How do I get R prime? For R prime, note that in the previous case, even though in this particular
case, I have a load over here (Refer Slide Time: 19:04) which looks like a load that is on a
member but note that bc is not a flexible member and since bc is not a flexible member, any load
on a rigid member is considered equivalent to a nodal load and therefore, since the effect is not
on a member, here, the members are ab and cb, since those are not there, you have to consider
this not as a member load because bc is not a member note that bc is a rigid member and
therefore it is not one of the considered members. What we have to do is Find out the work done
by the 100 Newton and this 100 is going to do negative 3 by 8, so the work done would be 100
into negative 3 by 8, which basically becomes minus 75 by 2. This then says that r 1 is equal to
minus 75 by 2 multiplied by 320 upon 153 EI. This is equal to whatever this value comes out to
be (I am not going to do it) that is my r1. As soon as I know my r1, I know my deflection pattern
and since I know r1, I can also find out what Mab, Mba and Mbc are and these can be immediately
evaluated. Once I evaluate those, for example, what is Mab equal to?
Let me just put it down. Mab is equal to 15 by 32EI into minus 75 by 2 multiplied by 320 upon
153 EI; EI, EI cancels, 320 cancels 10, 10 cancels 5, so it is going to be equal to minus 75
squared upon 153 Kilonewton meter. Mba is equal to 9 by 16 EI multiplied by minus 75 by 2 into
320 upon 153 EI; EI, EI cancels out, this goes 20, this goes 10 and so, what we have is 775 into 9
becomes 675, 6750 upon 153 Kilonewton meter. Mbc is equal to 21 EI upon 80 into minus 75 by
2 multiplied by 320 by 153 EI; EI, EI cancels, this goes 4, this goes 2, so what we have is 1575
into 2, 1575 into 2 is equal to minus 3150 upon 153 Kilonewton meter. Therefore, these are my
Mab, Mba and Mbc. Once I have found out my r1 and Mab, Mba and Mbc, now how do I solve this
problem? Displaced shape no problem, all you have to do is. This is my displaced shape,
only thing is that it will be in the opposite direction, so this will come down, this will go in this
direction, this value I know and therefore, I can find out all the other values automatically and
that is what you would expect: since the load at this point would come down and therefore, this
would have to move in this direction and in the opposite direction. Now, onto the support
reactions and the bending moment diagram.
I am just going to again put it. Here, what do I know? I know what Mab is. This is going to be
minus, so this is going to be clockwise (Refer Slide Time: 24:43) and clockwise approximately
about 73 Kilonewton meter this is what it comes out to be approximately; this is also going to
be clockwise and this is going to be equal to about 4 into 15, 60, so it will be about 44
Kilonewton meter. I am giving you approximate numbers, I do not know anything about this one
but I know this one (Refer Slide Time: 25:23).
What is it? This value is going to be equal to again clockwise, so this is going to be clockwise
and the value is going to be approximately 20, so it is going to be 20, so this is going to be 306,
90, so it is going to be about 21 Kilonewton meter 20 point something. Here, this is 0, I know
that this is 0, the displacement here is going to be 0. Now, let us go through the steps. I know
nothing about this because this is not a member. This is a, this is b, this is b, this is c, this is c,
this is d. There is no moment at this particular point. Since there is no moment at that particular
point, for bc we can find out the member end moments purely from equilibrium. Since I have 44
over here, from equilibrium this will have to be 44; here again from equilibrium this has to be 21
Kilonewton meter and this has to be 44 Kilonewton meter. I know that these are going to be the
moments.
What else do I know? Since I know this, I can find this, I can find this (Refer Slide Time: 27:03)
and this and since this is 117, 117 divided by 4 , I am again doing approximate numbers is
going to be equal to 117 this way divided by 4 is going to be equal to 29 so 29 Kilonewton
meter, 29 Kilonewton meter. Once I know this, again this has to be 29. Now note that there is
this 100 force, so this has to be this way 29, so therefore, here I need to have a force this way.
Since I have force this way, this is going to be 29 and therefore I am going to have a reaction
here which is equal to 29 I know this reaction.
How much is this going to be? This is going to be in this fashion (Refer Slide Time: 28:05), so
this 29 into 4 is 117 plus 21 is going to be 138, 138 divided by 3 because this one is this way, so
8
this one has to generate this way (Refer Slide Time: 28:33), it is going to generate this way and
this is going to be equal to 117 plus 21 which is 138 by 3 and 138 by 3 is going to be equal to 46,
this is 46 Kilonewton. Since this going to be 46, by definition this has to be equal to 46 and this
is 46 this has to be equal to 54; if this is 54, then this has to be 54 and therefore, this has to be 54.
Now the question is: does this check? Let us look at this.
What we generate over here from the 100 is 50, 50 but for this, it is going to generate an
additional one, so this is going to be equal to 44 and 21 is 44 and 21 is going to be equal to 65,
65 divided by 4 is 15. Let us look at this again because there has to be equilibrium without
equilibrium, you cannot satisfy this particular problem. There is no moment here, so if I take
moments about this point, this 46 into 7 has to be equal to Let us check let us go back and
check what we are doing here. This is 117 this way. Essentially, what happens now is therefore if
you look at the reactions, the reactions are this way: the reaction over here is 29, 54, 73 the
reaction here is 46 and 29.
I have found out all the reactions and once I know these, I can draw my bending moment
diagram very easily. I think this particular example was being presented to you essentially for the
last point that I wanted to make as far as low defects were concerned that if a load is on a
member that is rigid, then it is not a loaded member because bc, which is a rigid member, will
not have any effect and we have to consider the equivalent load on the left-hand side.
I hope that after this, you should not have any problem by applying the stiffness method for the
solution of any structural problem as far as we have only considered effects of loads on the
structure.
Let us look at certain other effects now and the only other effects that I am going to spend any
time. Note that support settlement is not a major problem in this particular case because all
that means is if there is support settlement, that means there is a degree of freedom
corresponding to that support settlement direction and the only thing that happens is that we do
not know the reaction at that point but we know what is the left-hand side, I know the
displacement. I can always solve for it because after all, I just get one additional degree of
freedom and that degree of freedom I just need to solve it to be able to get any this thing. Now,
what I want to consider is the last thing, which is how we consider temperature in structures and
that is going to be in the stiffness method because we already know how to consider it when you
have, when you use the force method, the flexibility approach. Let us now move on to that
quickly.
Once we know temperature, how to consider temperature, we will have understood the complete
concept behind the stiffness method. Let us now look at how to consider the stiffness method.
Again, since it is a member load, I am going to first consider, how to consider it I am going to
first talk about how to consider it at the member level and once we consider at the member level,
then considering at the global level becomes a trivial issue because after all, again, all it does is
that S becomes Kab a a r plus Si0 where 0 is the fixed end moment, so all I need to find out is for
temperature, how do I compute the fixed end moment? Let us go back.
Let us review the situation as we had already dealt with it earlier, let me see what happens. Let
me take a situation with both the top and the bottom. Note that here since we are only
considering flexure, we are going to assume that the neutral axis does not suffer a temperature
differential; if it does, all it means is that there is an axial deformation in the member and that is
all there is to it. Let us consider the situation where you have deltat; this is the top fiber it has
delta Tt and the bottom fiber has delta Tb. We have already seen that this gives rise to a curvature
beta which we defined as 2 alpha into (delta Tbottom minus delta Ttop) upon d where d is the depth
of the beam, the difference between the top fiber and the bottom fiber this is d (Refer Slide
Time: 35:43) to the depth of the beam.
We have already done this and we saw that this gives rise (Refer Slide Time: 35:48), so due to
this curvature, you are going to have theta and thetaab this is a, this is b is equal to minus
alpha now this is length l, so it is going to be alpha L upon d. thetaba is equal to alpha. I have
already established these in the force method and the flexibility approach that we looked at
earlier. This is my thetaab and thetaba due to the curvature. This is assuming that the neutral axis
delta T at neutral axis is equal to 0. If it is not, then all it does is that in addition to this, if delta
TNA is not equal to 0, all that means is that in addition to thetaab we also have a deltaab, where
deltaab is equal to delta TNA into L that is all. What does that mean?
10
But then, you see, our structures, our basic member that we have is this. Then, I will talk about
the modified member a little bit later and here, we have delta Tt and delta Tb. Now, how do I
compute the fixed end moments as well as in this particular case, if there is an extension, you
also have these forces that come into picture. Therefore, what is that equal to?
We find out the fixed end moment exactly the way we find out that that remember all that we
knew is that this one may be due to a load or may be due to a temperature. How did we compute
the fixed end moment? Look back, think back at how we computed fixed end moments for
members and you will see that all we need to do is find out the fixed end moment such that the
other direction is going to give you that theta minus of thetaab and thetaba.
The (FEM)ab and (FEM)ba are going to be equal to minus into Ka, Ka is 2EI upon L into (2 1, 1 2)
into thetaab and thetaba. These are my FEMab (Refer Slide Time: 39:55) and this is my (FEM)ba,
this is my P, this is my P. We can show that P is equal to nothing but this is minus because it is
opposing but then you know I am just proving it, this is equal to EA delta T NA; I am sorry, EA by
L into delta TNA L so L, L cancels, this becomes EA delta TNA. I am not going to go into this but
let us see what this comes out to be. This comes out to be equal to I am going to plug in the
values of this thing.
11
Therefore, (FEM)ab is equal to minus 4 EI by L into thetaab minus 2EI by L thetaba but then,
thetaab is minus of thetaba so this becomes equal to 2EI by L into thetaba which is equal to alpha
delta into (Tb minus delta Tt) L upon d; L, L cancels and so, this becomes 2EI alpha upon d into
(delta Tb minus delta Tt). If you look at the other one, this is going to be (FEM)ba is going to be
equal to minus (FEM)ab these are the fixed end moments due to the rotation. If we have this
kind of a situation, how how do we solve this problem? Let me now just take you through the
problem that we solved last time and instead of the load of 100 Kilonewtons I am going to put it
equal to temperature differential of 100 Kilonewton. Let me take the problem that we were
talking about.
12
I am looking at the last week's lecture's problem. This was the problem: EI, EI, rigid, a, b, c, d.
Now, the only thing that I have over here is that this is minus 20 degrees and this side is plus
(Refer Slide Time: 44:04). The reason why I am doing this plus and minus is to ensure that delta
NA at 1 is equal to 0. How do we solve this particular problem? Two degrees of freedom;
without going into the detail, we have done it already r1, r2.
All that happens is you see, ab is still a member without any loads, so I am not going to go into
the computation of ab. The only thing that happens in bc is that (Sbc)0. In the previous case, we
computed it for 100 Kilonewton load at the center and we computed PL upon 8, PL upon 8 and
went through and went through with that. In this particular case, all that happens is that Sbc is
equal to Mbc, Mcb is equal to Now, note that in this particular case, we can take alpha equal to
1.2 into 10 to the power of minus 5 meter per meter per degree Centigrade (Refer Slide Time:
45:42).
If we plug that in we get that the fixed end moments, these are the fixed end moments that are
equal to 2 into EI into 1.2 into 10 to the power of minus 5 into delta Tbottom minus delta Ttop is 40
divided by d this is my fixed end moment. Note: what are the units of this? Per degree
centigrade. This becomes a dimensionless value. What is this (Refer Slide Time: 46:32)? Newton
meter square. Newton meter squared divided by meter is Newton meter and that is the units of
the moment. This consistent unit and this one becomes 2 into EI into 1.210 to the power of 540
upon d, so there is consistent units and this one becomes 2 into EI into 1.2 into 10 to the power
of minus 5 into 40 upon d. The only thing that happens is that in this particular case instead of
the 100 Kilonewton force, you essentially have this moment. Therefore, once the moment, you
know I can plug in a value of EI and d I have not put that in, you will actually get it in let us
say if units of EI are in Kilonewton meter square, let me say EI is equal to 1 into 10 to the power
of 5 Kilonewton meter squared (Refer Slide Time: 47:38) and let me say that the depth of the
beam is 50 centimeters, so it is 0.5 meters. If we plug that in, we will put those values in, what do
you get? You will see that this is 1 into 10 to the power of 5 and this is 0.5, 0.5, 1 into 10 to the
13
power of 5 (Refer Slide Time: 48:06). What you get is 48 into 2, 96, so you get 128, so these
become 128, minus 128 Kilonewton meter. Once you have 128 Kilonewton meter then all that
you have to do is that this is my Sbc and everything else continues in exactly the same way. What
are the steps?
Once you have done this, you have both for ab and bc, you have Kab, aab, Kbc, abc and so the Si is
going to be equal to Ki into ai. Where do I get ai from? I get ai from the kinematic relationship. I
am not going into the details of the kinematic relationship because I have already solved this
problem. The only difference between this problem and the previous problem that I have already
solved in the last lecture is that instead of the 100 Kilonewton force I have just considered it to
be a temperature; therefore, there is no difference in (a, Ka) and all of those kinds of things: Kab,
aab, Kbc, abc all these are identical to what we have already seen. I am only trying to show you
what effect the thermal has. I have already shown that the only way to consider thermal is to
compute the fixed end moments and I have already computed the fixed end moments. All that
happens is once you have this, this becomes this (Refer Slide Time: 50:15) where this now
includes effects of temperature, which means thermal stress, this includes the effects of
temperature and I can always get this.
14
Then, I can also get ultimately this which is K into r note that in these (Refer Slide Time:
50:55) there is no change and summation ai transpose Si0. Note now that ai bar transpose ni0 is
0 for thermal load because when you have a thermal load, you do not have any additional
reaction; there is no reaction there is only an equal and opposite moment that is developed at
the fixed ends and those are essentially to account for the curvature that comes into picture
because of thermal. This of course includes thermal load, no difference for thermal load. This is
of course no effect of thermal.
If you look at this particular problem, the only part where the thermal effects at the member
comes into picture is only in this, which is the fixed end moment, and that we have already
included in our formulation. Therefore, if you really look at it, all we have to do right now is if
you were to review it that wherever you have a temperature load effect and you are looking only
at the flexural part of it, all it does is that you have to account for it and the fixed end moments;
for fixed end moment, I have already given you the effects; this is the fixed end moment (Refer
Slide Time: 53:10) where this is the temperature differential at the bottom, temperature
differential at the top fiber and this is the depth of the beam, this is the coefficient of thermal
expansion, this is EI. That is all there is to it. Of course, there is an additional effect here but if
EA is equal to infinity, then all you have is that this effect has no effect whatsoever all that
happens is that you have a expansion and therefore, you have an additional displacement because
of the expansion of the particular member; there is no effect on the structural composition as
such.
So much for the displacement method and the stiffness method, which is the matrix displacement
method and I hope that over the last many lectures, I have solved quite a few problems in the
stiffness method and at the end of this, I hope you shall be able to apply the slop deflection
method, the displacement method and the stiffness method, which is the matrix approach to the
stiffness method.
15
From the next lecture, I am going to look at a completely different topic and that is influence
lines, which is equally important. By the end of this lecture, I hope you will be able to solve any
planar problem, planar frame problem using either the force method or the displacement method
as well as both their matrix methods associated with them; the matrix method associated with
force method is called the flexibility method and the matrix method associated with the
displacement method is known as the stiffness method.
Thank you.
16
Prof. P. Banerjee
Lecture 36
Good morning. Today, as promised last time, I am going to be starting off by looking at
influence lines. Therefore, we are going to be discussing influence lines in today's lecture.
What are influence lines? Influence lines is the name of a topic where we consider the effect of
moving loads.
Essentially, influence lines are there to consider the effect of moving loads. What happens? Let
us just take a simple beam. Let us say I have a fixed load of 100 Kilonewton acting at L by 2. I
know everything: I know what is the reaction here (Refer Slide Time: 02:54), I know what is the
reaction here, I can find out this is going to deflect like this, the maximum bending moment will
be at the center and I can find out the maximum bending moment, reactiona, reactionb, this is my
designed load and I can compute my reactions, bending moment, shear force, deflections
everything; unique; this is for a static load. Think about it. Here, what is going to happen?
Suppose this load was not a static load.
Suppose this was the load which on the same beam could be here (Refer Slide Time: 04:18), it
could then move here, move here (Refer Slide Time: 04:23), move here and ultimately, go here
and then move off. This load is a moving load, it is moving with a particular velocity. It may be
moving very slowly or may be moving very fast. It is the same load by the way 100
Kilonewton. Now, the question is: what is the maximum reactions? What is maximum bending
moment? What is maximum shear force? Right now, I am just going to stick to the forces. How
do I know?
If the load was here, this would be 100, this would be 0 and bending moment would be 0, would
it not? If the load was here, this would be 100, this would be 0 and bending moment would be 0.
If it were here, this would be something else, this would be something else and the question here
is that the question over here is that for this moving load, what is the maximum bending
moment?
In other words, what is the maximum? Before we answer that question, we need to answer this
question: which is the position at which you are going to get maximum load? This is a moving
load and you can obviously understand just by looking at this, if I were looking at this particular
reaction, the position where this is maximum is when the load is here but the load at that point
here is. However, the load is here, the maximum is 100; therefore, you can understand that for
this 100 Kilonewton moving load, the maximum reaction here is 100, the maximum reaction
here is 100 and bending moment etc. How do we do this? The only idealistic way of doing this
is, let me do this.
Let me put the load at a position x from the left end. Now this load is P (Refer Slide Time:
07:09) it could be any load, put it equal to x and then find out the reaction here, the reaction
here. What would be the reaction over here? The reaction over here would be Px upon L, the
reaction over here would be P into (L minus x) upon L. What would be the bending moment at
the center? The bending moment at the center span would be equal to Px assuming that the load
3
is on this side it would be if x is less than L by 2 then bending moment at this would be into L
upon 2 which would be Px upon 2 and if x is greater than L by 2, bending moment at center span
would be P into (L minus x) by 2.
The question here then becomes what would be the maximum? For that, I would have to put x is
equal to 0 and from x is equal to 0 I would vary x is equal to L, so 0 is less than x is less than L
vary it and find out the maximum value. If you put it in here, put x is equal to 0, this is equal to P
(Refer Slide Time: 09:16); put x is equal to L, this is P, put x equal to L by 2, this will be P by 4,
put x equal to less than that and you will see that this will always be less.
Therefore, x is equal to 0 would give you the maximum (Refer Slide Time: 09:40), x is equal to
L would give you the maximum here and if you are looking at bending moment at center span x
equal to L by 2 would give you the maximum value of.. Now, you can do this easily by putting
P is equal to x. Suppose my point of interest was this reaction then, my influence line actually
becomes, what is this value as x moves from 0 to L?
In other words, the way to influence line would be that, suppose I were to draw the influence line
for reaction at a. You always draw the influence line for a particular response parameter
influence line cannot be done for everything. The point is that you can only do things for a
particular parameter in the influence line. What is the influence line? The influence line is this
that, I draw my x-axis going from 0 to L. What does every point over here represent? Every point
over here represents a position on the beam of the load, of the vertical load. What is the value at
x is equal to L? You will see that this is equal to P. What is it? If you look, if you look at the
expression, is this a linear in x? It is linear in x. What is it when x is equal to L? You will see that
this is equal to 0.
Now, what is the value of the reaction when x is equal to 0? This is the value. This is the
reaction. What I am actually doing is in this structure, this is my Ra, I am just drawing this. If the
4
load was at this point what would be the reaction Ra? It would be this. If the load was here, what
would be the reaction Ra? It will be this. Therefore just by looking at the ordinate corresponding
to any value of x, I can find out the value of the reaction at a. Now, what is done is the influence
line is for P is equal to1. In fact, if P is equal to a unit load, then this ordinate is known as the
influence line, so this is 1, this is my influence line the influence line for Ra. Note that the
influence line is for particular response parameters please understand that. Now, how would I
find out the influence line?
The best way and the easiest way is equilibrium direct approach. Suppose this is my reaction
(Refer Slide Time: 13:46), find it out and then plug in the values of x and find out this value. Of
course, here P is equal to 1 you have to put unit load vertical. If my bending moment in the
center span was my this thing, I would put if x was less than equal to 0, this is less than L by 2
this would be the value, this would be the value if it was greater than L by 2.
If I were to plot it, it would look like this, unit load and this would be the influence line for the
bending moment at the center span. If the load was here, I could find it out for any value of x.
Therefore, what this ordinate means is that if the unit load was here, what would be the bending
moment at the center span would be given by this ordinate that is the definition of the influence
line for a particular response parameter, for a particular structure. Now, if you had a statically
determinate structure, then equilibrium is one way it is the direct approach and you can obtain
it as absolutely directly. However, what you tend to do is you tend to not use the equilibrium
approach for statically determinate structures.
Why? Because there is another very, very useful approach known as the MllerBreslau
principle. What does the MllerBreslau principle say? I am trying to find out the influence line
for Ra. Now, the way the MllerBreslau principle works is it says the influence line for Ra is
obtained by releasing, so instead of the hinge, you release this (Refer Slide Time: 17:23) and
give a unit displacement corresponding to this and then what will the displaced shape look like?
This is a hinge, this is a hinge, so if it was two hinges, how would this member look? This
member would look like this the deflected shape of the member would look like this and the
MllerBreslau principle says that this deflected pattern is directly equal to the influence line.
Note: you can see obviously that this does look like the influence line for Ra, which we have
already done.
Let me just first try to understand where the MllerBreslau principle actually comes from. The
MllerBreslau principle actually works on the virtual work principle. What it says is this:
suppose I had load at this point (Refer Slide Time: 18:35), the load is at this point, then I could
find out the reaction. How? Well, put unit displacement. This is the method of virtual
displacement put displacement corresponding to the support reaction that you are trying to find
out and then the work done, the external work done is equal to reactiona into 1 plus P, now P is
downwards, P is not this, it is 1 into what is this value? Let us call it x. 1 into minus x because
the load is downwards, this is going upwards is equal to 0. Ra is then equal to x and that is the
MllerBreslau principle. It is based on the virtual work principle.
Remember I had said originally that the method of virtual displacement can be used to replace
the equations of equilibrium for a structure and this is what the MllerBreslau principle actually
does. What is says is whatever you want to find out the influence line for whichever response
quantity, release the restraint corresponding to that response quantity, give a unit displacement
corresponding to that response quantity and then the deflected shape would represent the
influence line because remember that the load that you have is only a unit vertical downward
6
load that you are finding it out for that is the basis for the MllerBreslau principle. Let me
look at this particular case.
Note that whenever we are drawing influence lines, we do not have any load because what is the
assumed load? The assumed load is unit vertical load at a point acting downwards. When we are
looking at influence lines, we do not have a load at all because it is always unit vertical load at a
point acting downwards.
Now, how do we find out? Let us say I have this member and I want to find out the bending
moment, the influence line for the bending moment at the center span. MllerBreslau principle:
what is the restraint corresponding to the bending moment? Remember: the corresponding
bending moment is that the rotation has to be continuous that is how you generate a bending
moment; if rotation was not continuous, then you will not generate a bending moment.
Therefore, how do I remove the restraint? I put a pin here (Refer Slide Time: 22:44). As soon as
I put a pin here, what happens is that the bending moment gets released.
Always you have to define positive. So I am going to define this as my positive bending
moment. If you look at it, then this is for this side, so it will be like this (Refer Slide Time:
23:15) and this will be for this side. This is the action that you do. If you do that action, what
happens? Center, this and this and what is the unit? The unit is the relative rotation, so this has to
be unity. Note that if this has to be unity and this is L by 2, you will see that this is equal to half
and this is equal to half and if you look at this, half into L by 2 is equal to L by 4. I have my
influence line for the bending moment at center span using the MllerBreslau principle. Today,
what I am going to do is I am going to solve several problems on this aspect. I have looked at
reaction, I have looked at bending moment; in bending moment, whatever is the positive, you
give a reaction in that way and get unit relative displacement.
Suppose I were asked to find out shear force, how would I do shear force? Note that shear force
actually tries to get relative displacement of one edge corresponding to other and the only way
this is stopped because the continuity and that is what develops a shear force. If I make it relative
to each other but I have to ensure that the bending moment is not released I am only finding out
the for the shear force, so what has to happen is that, MllerBreslau principle says unit
relative displacement, so the displacement between the two edges has to be equal to 1 but the
slopes have to remain the same so that there is no relative rotation between the two ends.
Therefore what I am doing is I am introducing a fictitious cut to release the restraint
corresponding to the shear force; however, I am not making a hinge at that point, so it looks
8
weird but this is the whole effect. How I ensure that there is no hinge is by ensuring that there is
continuity of slope. You are maintaining continuity of slope without maintaining continuity of
the beam this is a very funny situation.
Let me say that I want to find out shear force at the center span. My definition is plus, minus this
way (Refer Slide Time: 26:35). If you look at this way, this implies that this goes up and this
comes down because if you look at this side, this is the one that is going to be pulling it down, so
do this, then you get something like this and here you get something like this. The only thing that
you have to ensure is that since this becomes any restraint removal that you do in a statically
indeterminate structure makes it a mechanism and in a mechanism you only get straight lines, so
all you have to ensure is that this theta and this theta are equal. That will ensure that the tangent
here and the tangent here are equal to are equal to each other and this is equal to 1.
If you look at this, this is equal to let us say that this is L1, this is equal to L2 (Refer Slide Time:
27:43). Now, these have to be theta, so by definition by equal triangles, you will see that these
ordinates are half and half, these are 0, 0, so that becomes the influence line for the shear force at
the center span. Suppose I were to ask you to find out the influence line for the bending moment
at quarter span again, for the simply supported beam then how would I do it.
Quarter span; I have half, quarter, three-fourth, quarter span. Again, I need to do this. When I
need to do this, this one will go this way and this way, where this angle has to be equal to 1 the
relative. How do we obtain the ordinates? Let us look at it. If this is 1, this is a constant length L,
this ordinate is L, so this has to be L divided by 1 by 4 and this has to be equal to L divided by 3
by 4. This theta1 is equal to this, theta2 is equal to this. What does that mean? This means theta1
is equal to 4 l where l is this; this is the unit so I will call it x; this is going to be l upon 4, this is
l upon 4, this is x upon 3 l upon 4, this basically becomes 4x upon l and theta2 becomes 4x by 3 l
but theta1 plus theta2 is equal to 1 and this implies that if 12x plus 4x upon 3 l is equal to 1, so
9
16x upon 3l is equal to 1, x equal to 3 l upon 16, so this becomes 3 l upon 16 and that becomes
your ordinate for one-fourth span.
Suppose I wanted to find out the shear force at one-fourth span, how would I do it? Well, shear
force at one-fourth span, l by 4, how would I do it? This is x1, this is x2, this is theta and this is
theta by definition because the slope here and the slope here have to be the same. So what we
have is x1 plus x2 is equal to 1 because relative displacement has to be equal to 1 so x1 plus x2 is
equal to 1 as well as x1 upon l by 4 is equal to theta, x2 upon 3 l by 4 is equal to theta. So what
essentially it means is that this one is equal to this, so what we get here is that 3 x1 is equal to x2.
However, x2 is equal to 3x1 so if we substitute in you will get x1 equal to 1 by 4 and x2 is equal to
3 by 4; this is one-fourth and three-fourth and that is the influence line for the shear force at onefourth span. Do this and satisfy yourself that you can actually get both of these by the direct
approach. Satisfy yourself that the influence lines that I have drawn for all the specific
characteristics actually come out to be them. The main advantage is that the whole idea of
finding the influence line, actually it is an equilibrium problem but by using the Mller
Breslau principle we change it into a geometry problem and geometry is always much easier to
do than equilibrium. Today, I am only going to be looking at statically determinate structures just
to introduce you to the concept of how to apply the MllerBreslau principle. Let us look at
various structures for which the MllerBreslau principle can be used.
10
Let us look at this. This is my structure this is a statically determinate structure and I want to
find out the influence line. Note that the vertical load can only go from here to here this is a
factor that is very important; it is a unit vertical load, so it can only go from here to here. Now, I
want to find out the influence line for this the moment Ma and I am going to assume that this is
positive. What do I do then? Take this structure and I release this and I give it a unit. Note that
I have only released it I have not let it move up and down. All I have done is released it and I
have given it a unit rotation here I give it a unit rotation here. What does the beam become?
Unit rotation here, this is unity, so what does this value become? It becomes l. If you look at this,
the influence line for Ma is this, where each ordinate over here gives me the value of the moment
due to the load being at that point. Think of the load at this point (Refer Slide Time: 36:10).
What would Ma be equal to? Unit load at this point of length l, so Ma would be equal to l.
Suppose in this particular problem I wanted to find out the bending moment at this point (Refer
Slide Time: 36:25), what would I do? This is positive. What would I do? I would put a restraint
here and find out how this would go, so think about it; introduce a hinge and do this. Note this
part is fixed over here, so if you look at this when I introduce a hinge over here, the simplest
thing that would happen would be this and then unit, this would be unit because that is the
relative rotation between the two sides, I have introduced a moment here and this would be l
upon 2.
Think about it and you will see that it is indeed minus l upon 2 if you have the load here and
this you have taken as positive, this is sagging and you would see that if you put load here, you
would get hogging, which is basically minus l upon 2. What does this mean? This means that
when the influence line for this portion is 0 and only kicks in here. Why? Put the load here. What
would be the bending moment here? You will see that the bending moment everywhere beyond
this point is 0 and therefore you would not have any bending moment and that is reflected by the
MllerBreslau principle.
11
Suppose for this same structure I wanted to find the shear force at this point (Refer Slide Time:
38:22) and my positive is this way, essentially what I am doing is I am going to force them to do
this. Now, how am I going to get the shear force at this point? Think about it. The only way I
would get shear force is in other words, I am just pulling it this way. Now this point cannot do
anything.
Therefore, if you look at it, the diagram would look like this: unity, unity at the point at which
you have this; this slope and this slope are the same. Is that clear? Because this part as soon as
you make a cut over here, this bit goes by this and this does not have to rotate. Remember you
always go for the simplest displacement pattern that can satisfy your kinematic conditions, so
this is what happens. When you make a cut here and you have to keep this slope and this slope
same, the easiest way to do is keep this and move this down; then the slopes remain the same,
this becomes 1.
Again, if I put a load here (Refer Slide Time: 39:52), what is the shear force here? You will see
that the load shear force will be negative 1 and it will remain negative 1 till it goes beyond this,
which I meant becomes 0. This is my V at point C influence line for the displacement at point
C. Now, let us get slightly more complicated. We have done fairly simple things and I hope you
have figured out how to get it for support reactions that are linear, support reactions that have
moments and also draw the influence line for the bending moment and shear force. Now, let me
take an example that would be slightly more complicated.
12
Let me consider a situation like this: this is the frame this is a statically determinate structure.
Where does this thing move? Think about it. I am going to draw the influence line. Actually, this
is a crane moving over here a crane goes from here to here and for that, I need to draw the
influence line because ultimately for the crane, I need to know what would be the maximum
support reaction, I would need to know what would be the maximum bending moment all of
those kinds of things I need to find out. Therefore, the vertical load is moving from here to here
only.
Let me say that in this particular case, I would like to find out the influence line for reaction at a,
so let me say a, b, c, d, reaction at a. Now, note that the hinges are here and here, so I release this
and I give a unit displacement corresponding to this. If I give it a unit displacement
corresponding to this, this will go here, this will go 1 here but it is very important to understand
that I cannot draw it this way. Why can I not draw? What is the problem of drawing it like this?
Think about it. In other words, I am saying why cannot the reaction at this point be the same as
what we have done earlier, which is this? There would be a problem. Why? Because if this
member went up like this and this thing went, this would imply that this, which is a continuous
joint, would be hinged and that cannot be true because the only thing that we are doing is
releasing this and making it become something like this.
For this, how will it be satisfied? The only way that it would be satisfied would be if this went
this way (Refer Slide Time: 43:51) so that here some l, so that this would be l, so that you would
have a continuity over here and then, this would also have to remain and then you have
something like this. This would be the displacement pattern. How much would this move by? Let
us see. Let me say that this is l, this is l and this is l.
Now, we have to ensure that this theta (Refer Slide Time: 44:38), this theta and this theta are the
same. You will see that this one, this displacement is l theta, so that means if this is 1, theta has
13
to be 1 upon l and if this one upon l, this is 1, this is 1 and this is the displacement pattern
exactly; this one remains here and this one goes up, but what is the influence line here?
Is this displaced shape the influence line? No. Understand that the only thing that we are
interested in is the influence line is the displaced shape of the part of the member on which the
load acts. On which member does the load act? The load acts only on the beam, not on the
columns. Therefore, if you look at that, what does that look like the influence line? The
influence line looks like this (Refer Slide Time: 45:51). Look at that influence line that
influence line looks like this; 1, 0. Is there any difference between the influence line for this and
the influence line for the simply supported beam? No. Why? That is because this one (Refer
Slide Time: 46:34) and one shift has absolutely nothing to do with that because it is only the
vertical ordinate that actually tells us what the reaction is when the load is placed in various
positions. Therefore, the point here is that you will see that this Ra whether it is for a simply
supported or whether it is for this frame, it makes no difference. That is for Ra.
Let me ask you the same problem. Let me say I would like to find out the influence line for the
bending moment at this point. What do I need to do? All I need to do is I need to put a hinge here
and rotate it so that I get unity. Let us assume that this remains this way and I have this but now,
can this go like this? It cannot for the simple reason that if this has to go in this fashion, what is
going to happen? This will have to move down and there is no way that this can move down then
this becomes an issue. Therefore, obviously, this (Refer Slide Time: 48:51) cannot remain
straight and this cannot go this way because this would ensure that you have something like this.
What would you have to do? Think about it. The only way is this has to be moved here (Refer
Slide Time: 49:06). Therefore, take this and move this such that you get a situation that this does
not move; this cannot rotate because if this point rotates, this point cannot go anywhere so this
point cannot go anywhere vertically; if this point cannot go anywhere vertically (Refer Slide
Time: 49:32), then the only way is that this can move is only in this direction it cannot move in
14
any other direction because this cannot move up and down, this cannot move up and down, the
only way is this can go up and down.
If you really look at it you will see that ultimately the bending moment looks like this: this
member is going to look like this, this has to be equal to 1, so this is 1; if this is 1, this becomes l
over 2, this becomes l over 2, this becomes l over 2, this has to remain straight so that this
remains straight and if this remains straight, this also remains straight.
My displaced shape is this but again understand one critical parameter and that is that this is the
displaced shape but what is the influence line for this bending moment? The influence line for
this bending moment is this (Refer Slide Time: 51:13) and what is that influence line? If you
look at it the influence line is 0. What does that mean? That means if the load moves here there is
never any bending moment at this point and you can satisfy yourself that these two members
actually do not deform. This is the bending moment. Suppose I wanted to find out the shear force
at this particular point. Let me do that. This is the final thing that I am going to be doing in
today's thing that same frame.
It is continuous here (Refer Slide Time: 52:07) it cannot have a hinge there, continuous over
here, this is a, b, c, d, e and I want to find out the influence line for shear force at e. How do I
find that out? I have to give a relative displacement. The relative displacement has to be such that
this goes like this and this goes like this, so that is going to be the thing. However, this is not
possible because this again would entail. How would I have to do it? I would have to move this
this way, move this this way and then, and this point also would move this way.
If I were to draw the deflected shape, it would look like this (Refer Slide Time: 53:33) where this
angle and this angle are the same and then, this angle and this angle have to be the same to
ensure and so, this angle and that angle has to be the same. All the thetas have to be the same
and the only way that you can get those thetas is that this theta is equal to. What is this equal
15
to? This (Refer Slide Time: 54:17) is equal to l by 2 theta, this is equal to l by 2 theta. Now, l by
2 theta plus l by 2 theta is equal to 1, so theta is equal to 1 upon l.
If you have 1 upon l, then what do you get? You will be getting half, half and if you look at
this, it has no role to play, if I look at the it would go like this, half, half and as far as the shear
force at this point is considered, it does not make a difference what I have in the frame or not.
Again, the entire thing boils down to the fact that influenced lines are obtained through kinematic
relationship. The MllerBreslau principle lays stress on getting kinematic relationships that
would be compatible. Therefore, you always have to give a unit displacement corresponding to
the degree of freedom or to the release that you are going to do and if you have a statically
determinate structure and if you release, it becomes a first-order mechanism, so you are bound to
get only straight lines. You have to get only straight lines as influence lines for a statically
determinate structure.
I hope I have been able to establish the concept of influence lines as well as the concept of the
MllerBreslau principle. Next few times, I am going to be spending more time on the
application of the MllerBreslau principle first for statically determinate structures and then,
we will be moving on to statically indeterminate structures.
16
Prof. P. Banerjee
Lecture 37
Good morning. We have been looking at influence lines and the concept behind influence lines.
In the last lecture, I reviewed the concept for statically determinate structures, which I think you
have already been exposed to it was just a review and also a review of the MllerBreslau
principle to show the application of the MllerBreslau principle for statically determinate
structures to lead into what we are going to be doing today, which is influence lines for statically
indeterminate structures. Today, we are going to be looking at influence lines for statically
indeterminate structures and how do we look at the influence lines for statically indeterminate.
For determinate structures, the MllerBreslau principle itself application since for a statically
determinate structure, when you remove the restraint corresponding to whatever it is that you are
finding out, you make it into a first-order mechanism. Therefore, all influence line for a statically
determinate structure are straight lines and since they are straight lines, you can find out the
ordinates at every point and therefore, you know exactly what the values are for a statically
determinate structure. You do not need to follow any other approach to determine what the
influence line values are at different ordinates for different positions of the moving loads.
Today, we are going to be looking at statically indeterminate structure and we will see that for
statically indeterminate structure, to draw the influence line there are two approaches. One: to
draw the qualitative influence line, to get a feel for how the influence line looks, we use the
MllerBreslau principle. However, in statically indeterminate structures, even if you remove
one restraint, it does not make it a mechanism.
1
Therefore, typically, influence lines for a statically indeterminate structure are not straight lines
and therefore, the MllerBreslau principle application will only give us a qualitative form of the
influence lines. To actually get the ordinates, we need to go back to our direct approach, which is
using equilibrium to obtain the coordinates. For the next three or four lectures, this is what I am
going to be looking at. Today, I am just going to introduce the concept and see how to
qualitatively evaluate or qualitatively determine the influence lines for a statically indeterminate
structure.
Let us first look at this. This is the structure and the moving load moves from a to b or b to a it
does not matter, it moves between a and b. How do we draw the influence line for a particular
response quantity? Let us look at some of the response quantities. Let us say that I want to find
out the influence line for Ma, the vertical reaction at a, the vertical reaction at b and at the center I
want to find out bending moment at c as well as shear force at c. We want to find out the
influence lines for all of these.
Let us start one by one. I want to find out the influence lines for this. Ma what do I do to find
out for Ma? I release influence line Ma I release and give a unit rotation. If I give a unit rotation
in this direction (Refer Slide Time: 07:49), this is how the influence line looks, where this is 1.
What have I done? MllerBreslau principle: release and then given a unit rotation
corresponding to the influence line and this is the shape. What are the ordinates? I do not know.
However, qualitatively, this is how it will look. In other words, when the load is here, it is going
to be 0, when the load is here, it is going to be 0 and in between, it is going to reach a maximum
somewhere between the half-way point and this and it is always going to be positive. In other
words, the Ma is always in this direction and the value itself I can always find out. I do not know
the ordinates, but. Qualitatively, qualitative influence line for Ma. Next, I want to draw the
influence line for Va. For Va, what do I do?
2
I have released that and then I give a unit displacement in the direction of the force and the
influence line looks like this. The slope here is 0, this is 1, and here it is 0 that is it. Again, the
values I do not know; I do not know the values, so what do I do? I release this but note that it is
still fixed, so it is going to go up 1, so this is going to remain horizontal the tangent, then it is
going to go in this way. I am going to look at the others by and by, but what have we learnt here
in these two influence lines?
The influence lines are that the MllerBreslau principle can be applied to get a sense of the
ordinates we can get a sense of the ordinates. However, we do not know what the ordinates are
because the lines are no longer straight. For a statically determinate structure, when you release
one restraint, it becomes a mechanism, we get straight lines and the MllerBreslau principle
directly gives me the influence line, the type of the influence line as well as the ordinates of the
influence line, but in statically indeterminate structures, we only get a qualitative sense of the
influence line; however, we know nothing about the ordinates because we do not know what the
ordinates are going to be using the MllerBreslau principle alone.
But let us continue looking at it because even when we compute (we will see how we compute
later on), we need the qualitative assessment to be able to check that our computations that we
have carried out are correct. It is very important to use the MllerBreslau principle even for
statically indeterminate structures even though you do not directly get the ordinates but you get a
feeling of how the influence line is going to look; you also get a feeling of where it is going to be
positive, where it is going to be negative etc.
Getting back to this particular we have done it for Ma and Va and now we want to do it for Vb.
How would we do it for Vb? For Vb, we release that; when we release that, this becomes a
cantilever and we give it a unit displacement. This is how it looks; the influence line actually
resembles the cantilever shape function. One of the approaches is not a very popular approach,
but you can see that I can put a unit load here, find out this shape and I could find out this shape
3
for a cantilever and that shape would directly give me the ordinates then, but I am not going to
follow that method for this approach at all.
Let us look at it. For the bending moment at c, how will the structure look? For bending moment,
I release the relative rotation and then I do this. If I do this, what happens? This part is straight
this part is a straight line, this is a curved line and the tangent, this is equal to 1, this is the
tangent at this point and this straight line the angle is 1. How much is this value? I do not
know; also, I do not know about others. All I know is that this relative rotation is 1. Again, you
will see that if we define bending moment as positive this way, then this is the sense of the
bending moment diagram at c.
Finally, let us look at the shear force at c. How do I do it? If this is to be the positive sense, then
this one on the right will have to move up and this will have to move down, so it is going to look
something like this. This angle theta (Refer Slide Time: 18:20) and this angle theta have to be
identical and this one will be 1. All I know is that if this is x and this is y, because this is a
cantilever and this theta has to be equal to this theta, obviously y will be greater than x that
much I know, but what will be the values? I have no idea. This is the qualitative influence line
for shear force at c. This is for bending moment at c, shear force at c, this is for the reaction at a,
reaction at b and the moment at a. We can find these out. This is the pure application of the
MllerBreslau principle.
Essentially, just to write down what the MllerBreslau principle does. One: identify a response
for which IL is required. Two: remove the restraint corresponding to response. Three: deflect the
structure, ensuring a) all geometric restraints are satisfied and b) unit displacement
corresponding to unit displacement corresponding to response. Once you do that, you have got
the influence line for a statically determinate structure, the MllerBreslau principle gives you
not only the qualitative shape, it also gives you the quantitative values at every point of this
thing; if you have a statically indeterminate structure, the MllerBreslau principle only gives
you the qualitative values of the influence lines.
We see that the MllerBreslau principle is a very useful tool for influence lines, but for
statically indeterminate structures, it essentially gives you only a qualitative and just getting
qualitative influence lines is not good enough because you need to know what the values are.
Why do you do influence lines? Because ultimately, we will show that a little bit later that the
reason why you draw the influence lines is actually to find out the values of a particular response
quantity given a particular load. Therefore, to find out the value of a response quantity, you need
to know the quantitative values also. For a statically determinate structure, the MllerBreslau
principle gives it. How do we do it for the statically indeterminate structure?
For that, we actually need to go back to our first principles. That is, put the load here at a
distance x I am still dealing with the same problem. Can I use equilibrium for this? This is a
statically indeterminate structure. How do I do it? Can I find out something for this? Does this
propped cantilever remind you of something? (Refer Slide Time: 25:38) Look back to the
displacement method, the member. Does this remind you of a particular kind of member? Does
this not remind you of the modified member with a fixed end at one end and roller at the other
end? Now, do we know what the fixed end moment is? We know the fixed end moment for this,
do we not? We have already done this. This is unit load (Refer Slide Time: 26:28) what is this
(Refer Slide Time: 26:33)? This becomes L minus x. If you look at this, this one is given by (L
minus x) squared into x upon L squared. What is this? This is this way. This is equal to x squared
into (L minus x) upon L squared. We have already done this, do you not remember? Now for this
one, to get the modified fixed end moment, how does it go? Well, release this, put an opposite
one, carry over half. This fixed end moment is equal to (L minus x) squared into x upon L
squared plus half x squared (L minus x) upon L squared. I can find out the fixed end moment
directly using this.
Once I find out the fixed end moment, think about it, this problem then becomes this problem
with unit load, where I know this (Refer Slide Time: 28:34). This, which I can call as Ma, is
going to be equal to because this is a fixed end, so that is Ma, which is equal to (L minus x)
squared x upon L squared plus half x squared into (L minus x) upon L squared. If you look at
this, is this not a statically determinate structure, where I know this moment? Actually, I know
both the member end moments this moment is equal to 0 and I know the value of this moment.
Given length L and given the position of the load, I know this value; if I know this value and
since this is a simply supported structure, I can find out all the others using equilibrium. Let us
go through the process. What is that?
One of the things I wanted to find out was Ma I have already found that out directly. The next
was finding out Va. What is Va going to be equal to? Let us look at it. This is x, this is Va, this is
Vb (Refer Slide Time: 30:14), then, of course, there is Ma. If you look at it, Va is equal to 1 minus
x by L that is the part of this load that comes here. How did I find that out? Take moments
about this particular point and you will see that this will be Va into L is equal to 1 into L minus x;
this is L minus x remember that; so when you did that, this becomes this plus now if you look at
this, this is doing this, so this is going to be this, it is going to be plus Ma by L. Now, I know Ma,
I know Va. What is Vb going to be equal to?
Vb is going to be equal to x upon L minus Ma upon L. In this particular case, I know Ma, I know
Va, I know Vb where the load is x. Therefore, the main thing over here is that I I have found
these three. The next thing that I wanted to find out was the moment and shear force at the
center. Let us see how we can solve for that.
Let us look at this situation, let us assume that x so, the shear force and bending moment are
this way, this way, this is Va (Refer Slide Time: 33:11), this is Ma, this is Vb and this is L over 2
this length, because this is point c, this is shear force at c and this is bending moment at c. If x
is less than L by 2 then shear force at c is equal to look at it Vb directly, bending moment at c is
equal to if you look at this, this does this, this is equal to Vb into L over 2.
If x is greater than L by 2, then look at what happens. If x is greater than L by 2, then this comes
on this side (Refer Slide Time: 34:35) and if you look at the shear force, the shear force is equal
to minus Va because shear force plus Va because this is now over here this unit load is now
over here, so it is going to be shear force plus Va is equal to 0 because the 1 is here right now, x
is greater than L by 2, so this going to become Va plus SFc is equal to 0, and bending moment is
going to be equal to let us look at it: this is opposing it, it is going to be (Va into L over 2)
minus Ma. The values of shear force and bending moment are different for x is less than L by 2;
the position of the load is different. Now, the point is that I have found out the expressions for all
the quantities that I wanted and let me now draw.
Let me just say 1, 2, 3, 4, 5, 6, 7, 8 and 9. First is x value and next Ma. Then, once we get Ma, the
next is Va and Vb and once we get Va and Vb, the next row are bending moment at c and shear
force at c. What is my x by L? This will be 0, L over 8, L over 4, 3 L over 8, L over 2, 5 L over
8, 3 L upon 4, 7 L upon 8 and L. With these expressions that I have got, can I plug in the values
of M? I am now going to put in the various values of x into the expression for Ma that is here, Ma
and I am then going to go through the steps, so let us look at it. x is equal to 0 plug that in and
you will see that Ma is equal to 0.
If you look at L over 8, you are going to see that this is going to become 7, let me do the
computations. It is going to become 7 by 8 the whole squared into 1 by 8, so it is going to be 49
by 64 into 8 is going to be 512, so this is going to be 49 upon 512 plus if you look at it, it is
going to be 1 upon 8 squared into 7 upon 8, that going to be 7 upon 15, so that is going to be 7
upon 512, that will be 7 upon 2, so this is going to be 98 plus 7, 98 plus 7 is going to be equal to
105, so 105 upon 1024, this is 105 upon 1024 (Refer Slide Time: 40:21). Can I find out my Va
for that? Let me see what the Va is going to be.
9
Va is going to be Ma upon L. By the way, this is into L here (Refer Slide Time: 40:39). Now, Va
is going to be 1 minus x upon L, x upon L is 1 upon 8, so this is going to be 7 upon 8 plus M a
upon L, so 7 upon 8 plus this. If you do this, this is going to be 128, so it is going to be 7 into
128, 7 into 128 is equal to into 8, so this is going to be 7, this is going to be 56, 14, 19, 19 is
going to be 896, so 896 plus 105, 896 plus 105 is 1001, so it is going to be 1001 upon 1024
that is Va. If you look at Vb, it is going to be equal to 1 minus that, so this is going to be 23 upon
1024. Then, if you look at shear force, Vb is going to be 23 upon 1024 and if you look at the
bending moment, that is going to be multiplied by 2, half, so it is going to be equal to 23 upon
2048 L. Since this L by 8 is less than L by 2, these two are kicked in.
Now, let us look at the qualitative values that we have drawn. Let us have a look at that. Let us
see if these seem to make sense to you. It is positive value at L by 8, positive. What about for
Va? It is almost 1. That makes sense. For Vb, it is almost 0, makes sense. For Va, very small
value and here also, a very small value, makes sense. This is beginning to make sense.
Now, I will just put down for the L by 2 value. This way, you can continue on, so let us just do a
couple of more values. If you do L by 4, what will happen is this becomes 3, let us see what
happens here. L by 4, this becomes 3 L by 4, so 3 by 4 squared becomes 9 by 16, 9 by 16
multiplied by 1 by 4 is 9 by 64, so this becomes then 9 by 64, so this one is 9 by 64 and this one
will become 1 by 16 and this one will be 3 by 4, so this will be 3 by 64, so this becomes 9 plus
half of 3 by 64, so this becomes essentially 18 by 128 plus 3 is 18 plus 3 is 21, this is going to be
equal to 21 upon 128 L.
If that is the case, then let us look at what Va is going to be. Again, plug in 1 by 4, this becomes 3
by 4, this one is 21 upon 128, so 3 by 4 becomes 32, 32 into 3 is 96, 96 plus 21 is 117 upon 128.
This becomes 11 upon 128, this becomes then 11 by 256 L, this becomes 11 by 128. Let us look
at L by 2; at L by 2, what does this become? At L by 2, this one will become 1 by 4 into 1 by 2,
that is 1 upon 8, 1 upon 8 plus 1 upon 16 basically becomes 3 upon 16, so this is 3 upon 16 L.
Then if you look at this, 1 minus half becomes half, 50 by 16 becomes 11 by 16. V b becomes
then obviously 5 by 16. What will happen is that I have to actually do. For bending moment,
there is no problem you will see that you will get the same, but for shear force at L by 2, there
will be an L by 2 minus and there will be an L by 2, there will be two values: one will be for L
by 2 minus, one will be for L by 2 plus. The reason behind it is that the shear force just when the
load is to the left of c and when it is just to the right of c are going to be two different values.
Let us look at that. We have already got these values. When x is equal to exactly L by 2, we
cannot define this. Let us look at the bending moment. The bending moment is going to be Vb
into L by 2 so this is going to be 5 by 32 L. Let us find out the bending moment using this
approach: Va into L by 2 where Va is 11 by 16 into L by 2 is going to be 11 by 32, 11 by 32
minus 6 by 32 is going to be 5 by 32. Whether you use this (Refer Slide Time: 48:13) or this, the
bending moment turns out to be the same. However, the shear force is different. If it is just to the
left, it is equal to Vb and the value is 5 by 16; if it is just to the right, it is going to be 11 by 16
this is L by 2.
Let me just do it at 3 L by 4 and I am going to stop it at that and of course L, so let us look at
what happens. Let me do first for L, this is going to be equal to 0, so this is 0. What about this
10
particular one? Va is going to be equal to 1 minus x upon L, so 0 plus 0, this is going to be 0. By

the way, this is 1, this is 0, bending moment is 0, this is 0. Now, Vb is going to be equal to 1, now
this here we have to put this in, shear force is minus VFa, so this is 0 and V is Va into L by 2
minus Ma, Va into L by 2 minus 0 is 0.
Let us just find out one particular value in the middle that is at L by 4. If we look at 3 L by 4,
what do we get? Plug in the value of 3 L by 4 (Refer Slide Time: 50:01), you get one-fourth, so
one-sixteenth, one-sixteenth plus 3 by 4 is going to be equal to 3 by 64, 3 by 64 and this becomes
9 by 64, this is 3 by 64 plus half, so this becomes 6 plus 15, so this becomes 15 upon 128 L.
If we look at this, this is going to be 1 by 4 plus 15 by 128, so this is going to be 1 by 4 is going
to be 32, 32 plus 15, this is going to be 47 upon 128; if this is 47 upon 128, this is going to be
equal to 81 upon 128. If we look at this, this becomes minus 47 and this one, if we look at it, V a
into L by 2, it is going to be 47 upon 256 and 47 by 256 minus 15, so this is going to be 30, so 47
minus 30 is going to be equal to 17, so this is going to be 17 upon 256 L.
I am not going to do the other ones, but let us look at the qualitative values. We are going to end
here today with this, let us look at this. This started off large this is about 0.1, this is about
0.18, this one is less, so if you look at this, this is half (Refer Slide Time: 52:29), half is actually
3 by 16 and this is 21 by 8, this one is only 15 by 128, 0.1, this 0.1 at this point is almost the
same as L upon 8, so you can see the shape indeed is the way we have shown it. Then, if you
look at the influence lines for these, Va starts off very slow and then starts dropping really, really,
fast and if you look at this, it is almost 0.7 and this is less than it is about 0.3. Therefore, it is
really dropping fast, which is how this is.
On the other hand, if you look at this (Refer Slide Time: 53:20), this starts off really slow, here,
this is about 0.3, whereas if you look at this, this is 0.7. In fact, this is going this way and if you
11
look at it, it is 1 here, it is 0 here. The qualitative and the quantitative show similar tendencies
and then if you look at this, we will see.
At this, what is this value? Now we know what this value is 5 by 32 L. What is this value? We
know this value is now 5 upon 16. What is this value? 11 by 16. Add the two of them up, you get
1. You look at this theta, you will see that this is the same theta and over here, 0.7, you will see
that this is exactly half of that, you will see this is exactly half, this is linear. All these things that
we saw qualitatively, we see quantitatively and just to ensure, how did we do the quantitative
influence line? The only way we have done qualitatively is that we knew how to find out the
fixed end moment. Once we know the moment at the end, then it becomes a statically
determinate structure for which we used equilibrium and solved it. The only way to get
quantitative is by the direct approach.
Therefore, you see that for influence lines for static indeterminate, you have a combination of the
MllerBreslau principle and the direct approach that we have discussed already for statically
determinate structures. Thank you very much. I am going to continue with this and show you
how the direct approach can be used. Here, we could get Ma directly because it is a fixed end
moment. Later on, you will see that this is not obvious we will have to use some other method.
Thank you. See you next time.
12
Prof. P. Banerjee
Lecture 38
Good morning. We have been looking at influence lines for the last couple of lectures and today
we are going to continue looking at influence lines. If you remember, in the first lecture I
introduced you to the concept of influence lines, what an influence line was. Then we looked at
how to get the influence line for structures and we saw that there were essentially two methods.
One was the direct approach in which you use equilibrium to obtain the influence line. The
disadvantage of this method is that you need perforce a statically determinate beam or frame,
whatever you want to call it, before you can use equilibrium.
Secondly, it is a more time-consuming procedure obviously because you have to evaluate every
response quantity and you have to get the expression for that response quantity as the load moves
along the path that it is supposed to move. Therefore, we saw that the MllerBreslau principle is
based on the method of virtual displacement. The MllerBreslau principle was an easy way of
getting the influence lines.
We also saw that the MllerBreslau principle can both give a qualitative and a quantitative
assessment of the influence line of any response quantity for a statically determinate structure but
when we came to a statically indeterminate structure, we saw that the MllerBreslau principle
can only be used to get the influence line in a qualitative fashion. The MllerBreslau principle
cannot give the influence line quantitatively. Therefore, we saw that to get a qualitative
assessment, the MllerBreslau principle is okay but to get a quantitative assessment, we have to
go back to the direct method.
The direct method is based on equilibrium and we saw an example last time as to how for a
specific structure, because we could evaluate, we knew how to evaluate the fixed end moment
and the fixed end moment directly was the bending moment, we could get a statically
determinate structure and we could solve it. Unfortunately, that was a very specific case because
we looked at a specific problem where we had a single-span beam with one fixed and one
pinned. Therefore, we knew the fixed end moment and the fixed end moment was directly the
moment itself bending moment at that particular point. Therefore, from that, we could directly
get static equilibrium. However, it is not always true that this is valid.
Today, we are going to look at influence lines and it is going to be statically indeterminate
structures and somewhere where you could not directly get the statically determinate beam by
finding out fixed end moment. In a single-span beam, it was not a problem; as soon as we go in
to multi-span beams, you cannot do it, but then the direct approach is the only approach to get a
quantitative assessment of the influence lines. Therefore, we have to figure out how to get a
statically determinate structure where equilibrium can be used for multi-span beams or more
complicated structures let us look at this. By the way this is the thirty-eighth in the series of
lectures for this particular course that I have done. Let us look at the problem.
This is a two-span beam and we have to find out the influence line for various parameters. I am
going to define some responses for which we will have to find out the influence line and today,
we are going to see how this can be applied to get the influence line both qualitatively as well as
quantitatively. Let me put few of the response quantities. Obviously, in this particular beam, the
moving load is going to go from a to b to c. Let us look at Ra, let us look at Rb, let us look at Rc
and let us look at (BM)b. Let us just do these four.
This kind of bending moment actually implies that it is this way (Refer Slide Time: 07:59). This
is the internal positive and that is why this is also positive because this is the right phase, this is
this way, this is the left phase, this is this way. This is the internal force, this is the external force
that balances the internal force. Note that whenever we do, we have to externalize all forces
before we can apply the MllerBreslau principle. Let us see how we solve this particular
problem.
First, qualitative: We are going to do qualitative influence lines and we will be using the Mller
Breslau principle. We draw this. For Ra, this becomes this way pinned and then you have to give
a unit displacement here and then this is for Ra (Refer Slide Time: 9:53) Let me do it for Rb.
Understand that all we are doing is giving unit displacement. It is the same principle whether it is
a statically determinate structure or whether a statically indeterminate structure. As Mller
Breslau principle says release the restraint corresponding to the response you are trying to find
out, give it a unit displacement and the deflected shape of the structure is going to represent the
influence line.
Therefore, this represents influence line. Note something interesting: in this particular case, you
will see that this is going to be symmetric this side and this side are going to be identical. Here
this is going to be 1, there are going to be lot more displacements here than are going to be here.
Here, this displacement, this displacement, this displacement are all the same, this is equal to
this, this equal to this, they are the same, it is symmetric. That comes from the boundary end
3
conditions being symmetrical; you are giving a displacement at the point, obviously, it is going
to be symmetrical. Now, let us look at Rc and bending moment.
Remove the restraint at c, this is for Rc, give it a unit displacement and find out the deflected
shape. You will notice that Ra and Rc have just the opposite, they are just the opposite; we will
see that later when we do a quantitative analysis. I am just showing it to you because again this
and this are just the opposite of each other. Therefore, what we got for Ra which was this and this
way. Between ab, whatever the deflections are and between bc, whatever the deflections are for
Ra, those are going to be between ab which is what in between b and c and it is just the opposite.
Finally for the bending moment: for the bending moment a, b, c and what we have to do is give it
a pinned and make it go this way.
Again, this is going to be symmetrical about point b. I do not know what this value is. However,
I know that this has to be equal to 1 because again Mb right, so (BM)b and therefore, I have to
give a unit displacement corresponding to the response quantity and so therefore, it goes this
way. Why does it go this way? Because if you look at it, the moments do this, so it is going to go
this way and this relative between the two tangents have to be equal to 1 at this point. The
MllerBreslau principle gives you the qualitative assessment. Now, the question comes: how do
I quantitatively assess it?
Quantitative assessment: What do I need to do for the quantitative assessment of influence lines?
Remember that the only way that you can do quantitative assessment is by using the direct
equilibrium approach. For direct equilibrium approach, note that if I know in this particular case
a, b, c for both member ab and bc, for example, if I knew the (BM)b, would both ab and bc be
statically determinate? Note that if I knew the (BM)b, ab and bc would become statically
determinate. Therefore, entire thing hinges around finding out the (BM)b. But how can I find out
the (BM)b? The only way I can find out the (BM)b is to analyze the statically indeterminate
structure and find out the bending moment at b. Which are the approaches we know? The
flexibility approach, the stiffness approach or call them the force method and the displacement
method or the moment distribution method.
Note that in both the force method and the displacement method what do you find out first? Do
you find out the bending moment directly? No. You first find out actually the displacements
corresponding to the degrees of freedom in both the force method and the displacement method;
in the force method, you find out redundant force and from that solve the statically determinate
and in the displacement method, you find out the displacement.
In other words, you do not directly find out any quantity without first finding out something else
and then deriving it, but in this particular case, we do not need to know that under the load what
is the displacement, we do not need to know under the load what would be the support reaction
at c; I may not want to find out the influence line for support reaction at c you understand the
point. You have to understand another point that you actually have to solve this using the force
method and displacement method you would have to actually solve it for different positions of x
and find out the redundant or the displacements so it is actually solving many force methods and
displacement methods. You can do it, I am not saying you cannot do it but it is not an efficient
way of finding out an influence line because in an influence line.
All I am interested in is to find the ordinates of the influence line so that I can find out the
particular response quantity and the best method which directly finds out the (BM)b without
5
finding out something else is the moment distribution method. Therefore, the whole thing over
here is to make it a statically determinate structure. We know here that if we knew the (BM)b it
would be a statically determinate structure, so we use the moment distribution method. How do
we use the moment distribution method? Let us see.
In the moment distribution you will actually fix this, do you not (Refer Slide Time: 20:59)?
These you do not fix because these are ends, remember? We can do the modified 3 by 4 I upon L
and use the modified member, if you remember. In this particular case the load could be between
a and b. Suppose the load were between a and b, what would happen? You would get (FEM)ba
and what would be the (FEM)bc? It will be equal to 0. You get a fixed end moment and then what
would you do? You would release it, you put this fixed end moment, find out etc. While the
member is between a and b, we could actually find out the (BM)b by moment distribution by first
finding out (FEM)ba and then doing it.
Now, suppose it is between b and c now, it is not here (Refer Slide Time: 22:13), it is here. Now,
what happens? In that particular case, the second case, between ab and now the member is
between bc, then (FEM)ba is equal to 0 and (FEM)bc is what you calculate. The point here is, if
you could somehow find out the (BM)b in terms of the fixed end moment either ba or bc
depending on where the load is, then you see, can I find out the fixed end moment given a load?
Fixed end moment I can find it out. If this is x, then this will be in terms of x and L etc; this is
L, this is L, this also would be in terms of x and L etc.
In other words, given a position of the load, I could find out the fixed end moment and if I could
find out bending moment in terms of this and this, then I could find out this in terms of x and L
and then, that is it, we have solved the problem because once I know bending moment for a
given position, I can find out everything else.
The whole point really boils down to this. All we need to find out is for example, possibly if we
could get something like this is equal to c1 (FEM)ba plus c2 (FEM)bc, if we could get an
6
expression like this and we knew c1 and c2. Note that when the member is in ab, I know that
(FEM)bc is 0, this would drop out and since I know this and I can find this in terms of x by L, I
can find out bending moment and make both ab and bc as statically determinate structures.
Suppose it is between b and c, then this would be 0, this would kick in, so all I need to do is
really find out c1 and c2. If I can find out c1 and c2, I have solved my problem and therefore, we
do now is the following.
What we do is we find out c1 and c2 separately. How? All I need to do is apply 100 units fixed
end and find out what the (BM)b is using moment distribution and whatever I find this (Refer
Slide Time: 25:43), then c1 is equal to (BM)b upon 100. Next, this is separate, I give 100 units
(FEM)bc and find out the bending moment at b using moment distribution and then, c2 is going to
give me bending moment upon 100. This is the first case, this is the second case. Two different
loads and two different moment distributions so let us do that.
I am going to do the moment distribution moment; distribution is simple, go back and look at
your thing. You will have ab, ba, bc, cb. For this, it will be 3 by 4 I by L, 3 by 4 I by L, 0.5, 0.5.
What is the fixed end? First case, fixed end moment, I am applying plus 100, 0, 0, 0 (Refer Slide
Time: 27:26). I am only applying a (FEM)ba because my whole goal here is to actually find out
c1, which relates (BM)b to the (FEM)ba. How would I? First is going to be distribution: minus
50, minus 50 and look at this, nothing is going to happen here because no carryover, no
carryover, they are it is already 3 by 4 modified, so what we get is plus 50, minus 50.
In other words, the (BM)ba is equal to 0.5 of the fixed end moment; instead of writing that down,
I am going to write it in this fashion mba and Mbc is equal to 0.5 (FEM)ba. Let us now do the
same thing, another moment distribution where ab, ba, bc, cb and now, I am doing the (FEM)bc. I
will do plus 100, this is 0, this is 0, this is 0, again the half, half, this is going to be minus 50,
minus 50 and that has done it, minus 50, plus 50 and therefore, this is going to be minus 0.5 fixed
end moment and this is going to be plus 0.5.
Note these are the (BM)ba (Refer Slide Time: 30:09) and bc, these are the fixed end moments, in
other words, we found out the moments in terms of the fixed end moment at ba and bc, so now
we have this expression. We have this situation. Wait a minute, this is this way, a, b; if the load
is at x, if x is less than L, that means load on ab, then the (FEM)ba is equal to think about it, it
is going to be x squared into (L minus x) upon L squared, this is minus because it is
anticlockwise, this is also minus half (L minus x) squared x upon L squared. You do it; I am
not I have done enough of this to let you know how you can find this out directly, so I am not
going to .
This is the fixed end moment if x is less than L. If x is greater than L, then (FEM)bc. Note that
here I am going to be taking x, this is b, this is c and x is actually from this end, so this is going
to be equal to x minus L x is from a and this is going to be 2 L minus x because x, remember,
is going from 0 to 2 L. For this, fixed end moment is going to be equal to again, b is going to be
8
positive, so it is going to be (2 L minus x) the whole squared into (x minus L) all over L squared
plus half (x minus L) the whole squared into (2 L minus x) all over L squared. We have got the
fixed end moments etc. I am going to write them down again here.
If x is less than L, that is load on ab, then (FEM)ba is going to be equal to minus x squared (L
minus x) over L squared minus half (L minus x) the whole squared x upon L squared and
(FEM)bc is equal to 0. If x is between L and 2 L, that is load on bc, then (FEM)ba is equal to 0
and the (FEM)bc is equal to (2 L minus x) the whole squared into (x minus L) upon L squared
plus half into (x minus L) the whole squared into (2 L minus x) upon L squared. Finally, we also
know that the (BM)ba is equal to 0.5.
This I am denoting as mba (Refer Slide Time: 35:27), this I am denoting as mbc, this I am
denoting as mba and this I am denoting as mbc. I am writing down the reason I am doing this
it is not for anything else, it is still the fixed end moment; it is just that writing down long
expressions does not serve a purpose; I am using a smaller notation, that is all, there is nothing,
there is no other reason for this. Bending moment at bc is equal to minus 0.5 mba plus 0.5 mbc.
Now, we have these and finally, if I know the moments, then let us see what we can do, let us
look at it.
Essentially, this is member ab where we know Mba. Now if I know Mba, let us look at what I can
find out. By the way, let us assume that the load is in. If we look at this, then you can see that Ra
and I will call this Vb (Refer Slide Time: 37:24) because this is an open section where. This is
the support b, so you have made a cut and then we have bc; here, if I have like this, then this
goes this way and this is Mbc, this is Rc, this is Vb minus this is just to the other side, this is Vb
plus and if you look at it, this one is going to be positive this way, this one is positive this way,
therefore this one comes down here and this one goes up here. Let us see that Ra directly is equal
to Ra.
We can find out the expression for Ra if x is between 0 and L, then Ra is equal to one aspect of
it is 1 minus x over L that is directly from the load, the other one is from the bending moment
and this is going to be this upon L, so that is Ra. Let us look at what happens to Vb minus. Vb
minus is equal to x by L minus Mba upon L. Let us look at what happens here. What is Rc equal
to? Rc is equal to minus just directly so Rc is going to be minus Mbc by L. Vb plus is equal to
Mbc by L and since Rb is equal to Vb minus minus Vb plus, Rb is equal to Vb minus, minus Vb
plus, it will become x by L minus Mba by L minus Mbc by L. We will come to it. Also, if you
look at (BM)b, the (BM)b is equal to Mba because this is positive (Refer Slide Time: 41:28).
The (BM)b is equal to Mba or minus of Mbc; in fact, Mba is equal to minus of Mbc and you will see
that since Mba and Mbc are minus, this actually lands up being x over L (Refer Slide Time:
41:54). Let us now proceed. If the load is between a and b, these are the expressions. We have
found out everything we have found out Ra, Rb, Rc and the (BM)b the expressions.
10
If L is between x, what happens here is this; as far as a and b are concerned then we have this, so
this remains as Mba, this is Vb minus, this is Ra (Refer Slide Time: 43:00), this is Rb, this is going
to be this way, this is going to be this way, this is going this way, this is going this way and on
this side, I have Mbc, this is Mba (Refer Slide Time: 43:20), Mbc, Rc (Refer Slide Time: 43:24)
and now this position is such that this is 2 L minus x. This is when x is between L and 2 L. Then,
what would Ra be equal to? You will see Ra is directly equal to Mba by L, Vb minus would be
minus Mba by L and then, Rc would be equal to this would be x L (Refer Slide Time: 44:25),
so Rc would be x by L minus 1 and this (Refer Slide Time: 44:44) is from the loading itself and
then, the other part is minus Mbc by L that is Rc and then, Vb plus, that is this one Vb plus, Vb
plus would be equal to 1 minus this, so it would be essentially 1 minus this will be 1 minus this
will be 2 minus x by L that is from this part plus Mbc upon L and this implies that Rb is equal
to, here it is Vb minus this so this becomes minus (2 minus x upon L) and then minus Mba by L
minus Mbc upon L which basically means minus (2 minus x over L) which is essentially x over L
minus 2 and (BM)b still continues to be Mba or minus Mbc. Now, we have all the expressions and
I will just illustrate what these would be for a particular value.
11
You can obviously find out more details as we go along, but I am just going to do this for
specific quantities.
This is going to be x and I am going to say 0, there is no L by 4 so I will just do it L by 2, then L,
3 L by 2 and 2 L. The reason why I am only doing these of course, you just have to evaluate it.
You should be able to evaluate that. You can find it as one-eighth of L, you can just try to find it
at one-eighth of L that can be an exercise for you to do so that you can get a finer thing on the
influence line I am just doing it grossly.
The first thing we need to find out is the fixed end moment and I am going to put that as mba,
then I am going to make fixed end moment at mbc. From that, we can find out the moments at the
ends ba and member ab, we can find out Mbc and once we find out these, then we can find out.
What can we find out? The quantities that we are interested in Ra, then Rb, Rc and then I will just
draw another line, (BM)b.
The first step is to find out these values. I am going to spread out all these in front of me so that I
can find out each one of them directly. When this is between 0 and L, this, this and this are 0
(Refer Slide Time: 49:19) and then, when it is between L and 3 L by 2, this is 0 this we have
already done. For this, I know that this is also equal to 0 and here, this is also equal to 0. Let me
find out for L by 2. If I find it out for L by 2, what would be the value? Just plug in x equal to L
upon 2, substitute it in and you will see that this becomes 3 L upon 16, then plug in over here too
minus 3 L upon 16 because it is anticlockwise and this is going to be plus 3 L upon 16. Let us
now do this. Mba we know in terms of the fixed end moments at ba and bc, so this is going to be
0 0 (Refer Slide Time: 50:27).
What is the expression for Ra when x is between L and L? Just plug it in Mba is 0, so you get x is
equal to L, so Ra is equal to 1 and Rb is x by L, so that is 0, Rc is minus Mbc by L, that is 0 and
the bending moment is Mba which is 0. Let us now find out for this (Refer Slide Time: 50:57).
12
Obviously, Mba and Mbc are equal to 0. Ra is equal to 1 minus x by L, x by L is 1 so this is 0, Rb

is equal to x by L, so it is equal to 1 and Rc is equal to minus Mbc by L so this is 0 and bending
moment is also 0. Let me find it out for here. This is going to be 0 and 0. Let me see what R a is,
it is Mba upon L, so this is 0. What about Rb? Rb is equal to x upon L minus 2 x upon L is going
to give me 2 minus 2 so this is going to be 0 and Rc is going to be equal to x upon L minus 1 so
this is going to be 1, this is going to be 0.
Let me evaluate it here. If we look at this, Mba is half of this (Refer Slide Time: 52:02), so this
becomes minus 3 L upon 32. What is Mbc? It is minus Mba, so this is going to be 3 L by 32 and
Ra is going to be 1 minus x by L, 1 minus x by L is 0.5, 0.5 plus M ba by L, Mba is minus and so,
this is going to be equal to 0.5 minus 3 by 32, that is going to be 16, so this is going to be 13 by
32 positive (Refer Slide Time: 53:02), Rb is going to be x by L, so x by L is going to be equal to
half and R.
Now, Vb plus is actually this way (Refer Slide Time: 53:35) and this is this way, so Vb plus and
this is going to be this plus this, so you are going to have plus this and so, this is not x upon L
(Refer Slide Time: 53:50), this is plus and plus. The other point that I would like to make is that
Mbc is this way, Rb is equal to this and then, this is proper. Actually, this is equal to not just half,
it is going to be this plus 6 upon 32, that is 3 upon 16, so it is going to be 11 upon 16 and R c is
going to be minus Mbc upon L, so that is going to be equal to 3 by 32 negative because here
(Refer Slide Time: 54:52), you have this so it is minus Mba upon L and the bending moment is
going to be equal to 3 L upon 32. Now let us go over here. This is going to be 3 by 16, so Mba is
going to be minus 3 L by 32, this is going to be 3 L by 32 and therefore, you get exactly the same
values, 13 by 32, 11 by 16, this becomes minus 3 by 32, this becomes 13 by 32 and this is 3 L
upon 32 this is the end.
If you look at this, you will see Ra is 1. At the center point, it is 13 by 32 which is less than half
as you expected it; over here, It is minus 3 by 32, much less than this this is what you expected;
it is 0 here, 0 here and 1 here that is there. If you look at Rb, it is symmetrical 0, 11 by 16, you
would expect that this is more than half and 11 by 16, symmetric and this one is just the opposite
and this is how it is. If you look at the bending moment, you will see that it is indeed correct. In
other words, we have illustrated how to use moment distribution to get the influence lines
qualitatively for a particular structure.
I will stop here. I will continue over the next few lectures looking at other examples of this.
Thank you.
13
Prof. P. Banerjee
Lecture No. 39
Good morning. As you know already, we have been looking at influence lines for statically
determinate structures. To begin with, we introduced the concept of the MllerBreslau principle
and then, we have spent the last two lectures looking at influence lines for statically
indeterminate. Again, to reiterate, we said the last time we spoke of qualitative influence lines,
MllerBreslau principle, quantitative only the direct approach and in the direct approach, we
saw that the moment distribution method gives us the quickest way of computing.
Let us now continue looking at influence lines. Hopefully, we will try to look at one or two
examples in this particular lecture. I am going to be moving a little bit faster because by now, I
do not have to reiterate every small step that I take why am I releasing, why am I doing this; by
now, you should be able to be fairly comfortable with that. Let us take an example.
Now we are going into real examples. The load only moves in a, b and c vertical loads only
move between a, b and c. The flexural rigidity for this is EI, the flexural rigidity for this is EI and
the flexural rigidity of this is 2 EI (Refer Slide Time: 03:42). We are given the fact that bending
moment positive is this way and this is shear force positive shear force and this is bending
moment. This span is 4 meters (Refer Slide Time: 4:31), this is 4 meters and this is 8 meters
height.
We have to find out influence line for Ra, for Md, for Rd and for the bending moment and shear
force bending moment at e and shear force at e; this is the center point (Refer Slide Time:
05:32), it is 2 meters. This is the problem: both qualitatively find out the influence lines for this,
so for these are qualitative and quantitative, find out Ra and bending, only quantitative is Ra and
BM; qualitative all of the others. Let us go through these step by step. We can do this
reasonably quickly. Qualitatively, let us look at what happens.
For Ra, what we have to do is. This is the system and for Ra this goes up (Refer Slide Time:
06:35), this point goes nowhere and so, what we get is. Note that this goes and this is fixed so
this has to be like this, goes like this (Refer Slide Time: 07:01), fixed goes; and this is going to
be less because this also has to go. However, the influence line is only over here. This is the
influence line for Ra. If we want to do it for bending moment Mba, this is for Mba, let me draw it
this way. For Mba, what happens is that this goes this way and this goes this way where this and
this has to be 1. That is the bending moment this is for Mba. Then, we need to find it out for Rd.
For Rd, what do we do? Think about it.
We release this and push this up by 1 (Refer Slide Time: 08:16), push this up by 1, this goes up
by 1. What is going to happen is that this is going to go and this is going to be 1 and this is a, b,
c. Note that always a, b, c, we do not care what happens to this part because the load does not
travel on that. Is that clear? Therefore, even if you have a frame, the influence line is only to be
drawn the shape that the part on which the load travels, that is the only part that comes into the
influence line remember that, never forget that.
Even if I have shown a frame here, I am only drawing the influence line for abc, because that is
the only position, that is the only kind of extent the load travels we have been given that. Then
finally for Md; for Md, this has to go, this is going to be this way and so, our thing will be this
way and the influence line, this has to be equal to 1 but note that the influence line is this; even
though we have drawn this part, it has no role to play. This is my influence line remember that.
This is for Md although I have given this (Refer Slide Time: 10:31). I am not going to draw
anymore; okay, let me just draw the bending moment at e for your benefit.
Bending moment and shear force let me draw that. If I have bending moment at a, then this is
going to go straight and then this is going to go in this fashion, where this is a straight line and
this is going to be equal to 1. That is going to be bending moment at e. For the shear force at e, it
is going to be this and the shear force, positive is this being pulled down and that being pulled
up, so it is going to be this way; of course, this is going to go this way. Again, this is the only
part that matters; this is going to go this way, it is going to be less than if this did not exist, it is
because this stiffness reduces that, that is all. This is for the shear force at e. Qualitative influence
lines is fine.
Now, we need to go into computations of this for this thing. The thing that we need to know is
from which position does this go? Let us look at that. Quantitatively, we need to know this goes
ab, bc and therefore, when it goes ab, (FEM)ba, when it goes bc, it is going to be fixed bc and cb,
but cb is anyway fixed, so it does not matter whether cb is fixed or not. Therefore, we do not
apply a fixed end moment that fixed end moment is going to go directly and at ab, we do not
provide because Mab is always going to be equal to 0. Therefore, the only ones we need to
provide are bc and cb.
I am just going to show you notionally, I am not going to do it precisely. I am applying and
remember: this one, there is nothing (Refer Slide Time: 13:48), this one this and here, I have this,
for this, we have this side and this, this is 1.0. Let me go back and have a look. This is EI and 4
meters. This is going to be I upon l, three-fourths of I upon l, so this side, we have three-fourths
of I by 4; this side, we have I upon l, so we have I upon 4; this side, we have 2 I upon 8; so this is
going to be 3 by 16 I plus I upon 4 and I upon 4. What we have then is 4, 4, 11 by 16, 11 by 16, 3
upon 16 divided by 11 upon 16 is 3 by 11 and this is 4 by 11, 4 by 11. What we have is 0.272,
this is going to be 0.364 and 0.364, this is going to be 0.728, yes 0.364 these are the
distribution factors that you have over here.
The first one is going to be plus 100 here, 0 here, 0 here and of course, this is 0 and so is this. We
need to distribute it and we distribute it. This becomes minus 27.2, minus 36.4, minus 36.4, this
one goes here, minus 18.2, this one goes here, so this becomes this, this becomes minus 18.2 and
that is it. Now, we go on to the next one. It is the same thing that we have. It is going to be 0.364,
0.364, 0.272 and this time, this is plus 100, this is 0, this is 0, this is 0, and so is this. When we
do this, this becomes minus 27.2, minus 36.4, minus 36.4 and this becomes minus 18.2, close,
close, close. Nothing goes here and here, we get minus 18.2 (Refer Slide Time: 17:40).
If we look at this, we see that Mba is equal to 0.728 of mba and minus 0.272 of mbc this is the
fixed end moments. Then, Mbc is equal to minus 0.364 mba plus 0.636 mbc. Mcb this is the
moment at that end is going to be equal to minus 0.182 mba minus 0.18 mbc plus mcb. Note that I
have not done this moment distribution. Why? Because at a fixed end moment if I apply a fixed
end moment, what is going to be the moment? The fixed end moments, so that is going to be
directly this, it is not going to get distributed anywhere else. These are the expressions that I get
from my moment distribution because if you look at this, if this is 100 (Refer Slide Time: 19:43),
then this is 72.8, this is minus 36.4.
By the way I have not done the others because those do not In this particular case, I have only
asked you to find out Ra and bending moment at e. If I had asked you to find out Md, then you
would of course have got it in terms of minus... let me do that, let me do that, let me find out Md
so that I would need to find out Mdc would be equal to minus 0.182 mba minus 0.182 mbc that is
what Md would be, so we will find that out. I have got these. Once I have got these, then the next
step is of course the same procedure that we have already gone through and without much ado.
We have already done this in the previous case.
We know that if 0 is less than x is less than L, then (FEM)ba is going to be equal to minus x
squared into (L minus x) upon l squared minus half (L minus x) square into x upon L squared
and (FEM)bc is equal to 0. If L is less than x is less than 2 L, (FEM)ba is equal to 0, (FEM)bc is
equal to (x minus L) into (2 L minus x) squared upon L squared plus half into (x minus L) the
whole squared into (2 L minus x) upon L squared. Note that the only thing that we have been
asked to find out are three quantities.
Those three quantities if you look at it is Md. Md at all times for 0 less than x is less than L by 2
is equal to Mdc. What are the other ones? The other two things that I have been asked to find out
7
are Ra I need to find out Ra, so let us see what Ra is going to be. I have over here Mba and I am
only interested in finding out Ra and the bending moment at e. When a load is between 0 and x
by L, Ra is equal to (1 minus x by L) plus Mba by L we have already done this, I am not going
to go into that.
When x is between L and 2 L, Ra is equal to Mba by L. That gives me the expressions for Ra.
Now, for the bending moment at a, when load is between 0 x and L by 2, bending moment at E,
let us look at what bending moment at E would be like. You would have a bending moment in
this fashion. If we look at it, Mba and the load is over here and so x is between and you have Ra
here (Refer Slide Time: 24:49) and this is the bending moment. If it is between this and this and I
take moments about this point, what do we get? We get Ra into L by 2 that is being this thing,
so it is Ra into L by 2 and here, we have this one going the opposite way, so this is going to be
minus 1 into L upon L minus x. This is the bending moment and you can see that this is also
equal to Mba minus x by L into L by 2, x by L into L by 2 is minus x by 2, so we can actually put
it in this fashion. If you substitute into this, you will see that this is exactly the same as this, so
there is nothing new in it. We will probably use this this is the easier one to use. What happens
when x is between L by 2 to 2 L? Bending moment is just equal to R a into l upon 2. Once I have
this, then I can write it down.
This time, I will do it with L by 4 so that we can actually find out more details. This is going to
be x (Refer Slide Time: 27:26) and the overall distance that it travels is from 0 to 8. I will have 0,
1, 2, 3, 4, 5, 6, 7, 8. The first ones we have to find out are mba and mbc these are the fixed end
moments and those are the first things that you have to evaluate. Once you evaluate that, if we
look at it, we need look at this so all we need to do is to evaluate Mdc and Mba because these are
the only ones that we need. We are going to do that. I am going to evaluate Mdc and Mba. I do not
need to evaluate any of the others. That will give me directly Md, Ra and the bending moment at
e, the bending moment at e. Let me now do this.
8
We have done this now. We know that mbc is going to be 0 while the member goes between 0
and 4. We also know that mba is going to be 0 when it travels between 4 and 8 and of course,
when it is at the end, these are going to be 0. All we have to do really is evaluate these and this
one we know, it is 3 by 16 (Refer Slide Time: 30:03) and this is negative and this is going to be
positive 3 by 16, here plus 6, this is not correct, this is plus 3 by 16 at the center span. Let us find
out at three-quarter span.
I am going to plug in the value of x equal to one-fourth in this (Refer Slide Time: 30:39). When I
plug in one-fourth, I get 1 upon 16, this is going to become 1 minus, so this is going to be threefourths and this is going to be 9 upon 16 and this is going to be one-fourth. What we have over
here is minus 3 upon 64 and then we have minus half of 9 upon 128, so this is going to be equal
to 6, so minus 15 upon 128. The other one with three-fourths is going to be just the opposite, so
this is going to become minus 9 upon 64 and the other will become 3 upon 128 and so when we
do this, 18, we get minus 21.
What you have here is that since we are doing this thing, this is going to be 18 and 3, plus 21,
this is going to be plus 21 upon 128 and this is going to be plus 15 upon 128. If you look at this,
this turns out to be as the load comes in, the bending moment increases and this is more and
this is less. Now, let us look at directly Mdc. This one if you really look at it, this into L, so this is
going to be 60, so this is going to be 4. I will have to do this again, this is going to be 32 this is
into L, L is 4 here, this is 3 upon 4, this is going to be 21 upon 32, 21 upon 32, 3 upon 4, 21 upon
32.
If we write those down in numbers, you will see that you get 0.4 on 28, 12, 12, so that is going to
be 4.44, this is going to be 0.75, this is going to be equal to 0.6, 6 will make it 192, that is going
to give me 12, so it is going to be 0.64; this is going to be minus, minus, minus, this is going to
be minus 0.64, this is going to be minus 0.75, this is going to be minus 0.44. When we plug those
9
values in here in Mdc, you get minus 0.182 into 0.44 and if you look at that, that basically
becomes minus, so it is going to be plus, so plus 0.072, so this is going to be 0.08.
Then if you do this into 3 by 4, it is going to be 0.75, this is going to be 3 by 4, 0.55, 0.55, 3, so
this going to be 0.55 divided by 4, that is going to be 0.14 and in 0.64, so this is going to be 0.11,
this is going to be equal to 0 and this is going to be minus, note that these are plus, plus, plus, so
this Mdc is going to be minus, it is going to be minus of 0.11, this way, it is going to be minus
0.14 and minus 0.08 and 0. Let us look at for Md. This is exactly how it looks. This is Md and the
point that I am trying to make is this way you can find out mba, plug it in and you will get mba
and then, you can plug in these. Once you get Mba, you can find out Ra values and you can find
out moments.
Once you know this, Mdc is directly this (Refer Slide Time: 36:28). Now the point that we are
trying to make over here is that once you put this factor in, you can get Ra and bending moment,
you can get Mba. I am not going to do all of this. The overall procedure boils down to three steps.
The first step is to determine where exactly your loads are. In this particular case, for example,
load is going from a to c.
Now, a is hinged, it cannot have a fixed end moment. Therefore, when the load is between a and
b, you have a fixed end moment at ba. Therefore, you need to do one moment distribution with
ba. Then when it is between b and c, you need to do the moment distribution. There is going to
be a (FEM)bc as well as cb, so you need to do a moment distribution with bc and cb. But note that
the there is no need to do Mbc because that is a fixed end and you do not release that point, so
there is no fixed end moment, so it only comes into M into M, the moment at cb these are
important points. Therefore, once you know the levels of moment distribution that you have to
do, the next step is once you have done the moment distribution and you have to now write down
what are the member end moments in terms of the fixed end moments those are the coefficients
you have found out from the moment distribution.
10
Once you have written those down, then you need to do the equilibrium of each member and get
it from there. Is that clear? This is the step and now I am going to quickly look at another
problem so that I can illustrate one other type of problem to you. Right now, we have set up
the entire step and now all we need to do is essentially put down all the values that you are going
to get so that we do not need to do anything anymore. We were done with this, this is it. Just let
me put my papers together and then we will quickly move on to the next one. The next problem
is, let me take a particular example, let us take this example.
This is A, this is B, this is C and this is D; for this, I have I, for this, I have I, for this, I have I;
this is 4 meters, this is 3 meters, from here to here is 4 meters and from here to here, it is 4
meters. The vertical load only goes between B and C and D this is important; it does not go on
AB. Is that clear? The vertical load is only in this zone. What does that mean? When we draw the
influence line, we draw the deflected, use the MllerBreslau principle and get the deflected
shape. We only need to draw the deflected shape of BCD for the influence line.
Let me now say that the influence lines for Ma, for Ra, then at E, which is the center point, I need
for the shear force, shear force, the reaction at C and Ra, Ma, VE, Rc and MD. We have five for
which we need to find out I leave this as an exercise for you you are going to do it
yourself, qualitatively over this entire scope at 1 meter intervals, you are going to find out VE. I
leave this to you as an exercise I will give you the answer but I will not go and solve the
details. However, what I will do is I shall use the MllerBreslau principle and show you how
this thing can be done.
11
First, Ra: I am going to make it a question that is going to become very very important and that is
the reason why I have given you this particular problem, nothing else.
Here, note that now going to become a fixed roller (Refer Slide Time: 43:25) which is going to
move in this direction, so this way it has to be equal to 1. What does that mean? Ideally, this
would go (Refer Slide Time: 43:46), 1. However, this cannot move, so therefore this has to now
move perpendicular and how much will the perpendicular be in turn, think about it. If this is 1
(Refer Slide Time: 44:01) then the perpendicular is going to be 3 by 4 so that this then becomes
5 by 4. this is going to go like this and I am not going to draw this, this is going to go like this.
Therefore, this is going to rotate in this direction if it has to move up; the tangent is not going to
be this way, the tangent is going to be this way, so it is going to go like this, go here and then go
this way and this is going to be 5 by 4. Note that this is for Ra.
Note this very importantly and that is the whole reason why I have given this particular problem
and that is why I have given Ra in this direction instead of vertical. Vertical would have been
very easy. But I gave it in this direction for you to understand that when this has to go this way,
you have to satisfy the boundary conditions. Remember that the MllerBreslau principle has to
satisfy all the boundary conditions that you have in your specific problem this is the point, this
was the whole point, everything else is now trivial. However, I am going to go through with it.
Let us look at the next one which is Ma. I am going to Ma.
12
Ma means I rotate it so this becomes now a hinge and rotate it here. How will it look? It will look
like this. When this goes this way, this would tend to do a little bit and so this is where it
becomes interesting. We knew that this is 1 (Refer Slide Time: 46:44) and this is the influence
line. However, qualitatively we know nothing over here. This is for Ma. Now, let me do it for Rc.
For Rc this is going to remain this, this is going to remain this, we are going to remove this and
move it up and up by 1. How will this thing look? You will see that when this goes up, this is
going to go like this, so this is going to go a little bit like this and so this is going to get a little bit
13
like this (Refer Slide Time: 47:45), probably this will have an angle this way but this in essence
gives you for Rc.
Now, Md, let us make it for Md. For Md, now what do I do? I release this, make it into a hinge,
this is anyway a hinge (Refer Slide Time: 48:36), this is fixed, make it into a hinge and then
rotate it. What is going to happen here? You are going to see that this is going to go in this
fashion and this is going to go this way and this way, so this is going to be where I know that this
slope is equal to 1. This is for Md. Finally, for the shear force also I am going to draw this
because this is important. This is quantitative.
14
Then we have to actually get the quantitative also. This is fixed, this is hinged, this is fixed.
Therefore, what we need to do is here, get it up and here, get this to go down; when you make
it go down, this this will go in this fashion and go this way, this is going to go in this fashion
this (Refer Slide Time: 50:10) and this have to be the same, so that this is going to be something
like this; this does not matter. It is this, this and this and this total is going to be equal to 1. This
is for Ve. Finally, as I said, I am not going to show you how I have done it but I am going to put
down for x, so x is moving from b to c to d at 1 meter intervals. It is 1 meter (Refer Slide Time:
51:03) here, here, here, here, then here, here and here. At 1 meter intervals, I am plotting it
starting from here my x starts from here.
For 0, 1, 2, 3, 4, 5, 6, 7 and 8 I am putting down this as x, I am putting down the derived values
of Ve for you you are going to actually calculate it. For 0 this is 0, for 1 it is plus 0.2235, for 2
there is going to be a 2 minus and then there is going to be a 2 plus; for 2 minus this is plus
0.5112 and for 2 minus it is minus 0.4888 this is for 2; for 3, it is minus 0.2068, for 4, it is 0,
for 5, it is plus 0.0803, for this, it is plus 0.0733, this is 0.0273 and for 8 it is 0.
I am going to leave this with you; you have to solve this problem and get these Ve use this.
Next time, I am going to do it in detail and show it to you. Do not wait for that try to get these
values using the procedure that I have developed for you. Thank you. See you next time.
15
Prof. P. Banerjee
Lecture 40
Good Morning. Today we are going to be looking at influence lines and its applications.
Last time I had given you a problem and I had given you the answers. Today we are going to
take up that same problem and work through it. It will be the last problem I will look at and then
I will look at how to apply it. We will see how to apply it. Hopefully at the end of by this lecture
we will know what to do with influence lines. We are going to be looking specifically at the
applications. Let us go back to the problem that we were looking at.
This is A, this is B, this is C and this is D. We know the qualitative. I am only looking at the
quantitative. The quantitative is the center point of the VE. For that this is I, this is I and this is I,
this is 3 and 4 this is 4 and this is 4. The vertical load goes from B C D. If we look at it what is
the first step? By the way we have drawn the influence line for this. I just put it together for you.
If you remember this one - this theta and this theta are the same and this one of course will go
like this. This is the influence line for E. As the load proceeds from B to C to D let us see. When
the load is between B and C you set up a fixed end moment at BC and CB and then when you go
from C to D you set up a bending moment at C and D.
We have to tackle 3 problems. We are not going to tackle DC because it is a fixed end. The fixed
end moment BC is going to land up being there. It is not going to be a moment distribution but
we need to do 3 sets of moment distribution and I am going to do all 3 sets together. The first set
is here, the second set is here and the third set is here. This is going to be AB. There are 3 of
them - BA, BC, CB, CD and then DC. This one again is going to be the same AB, BA, BC, CB,
CD and DC and finally the last one AB, BA, BC, CB, CD and DC. In this particular case we are
going to first tackle plus 100, 0, 0, 0.
In the second case we are going to handle 0 plus 100 0 0 and the last one we are going to handle
0 0 0 plus 100 and 0. These moment distributions we have to do for fixed end moment at BC CB
and CD, the DC we do not need to do that because that is the fixed end that is going to show up
directly at DC. The distribution factors both of them are continuous. We are going to have I upon
5 and I upon 4. I upon 5 and I upon 4 is going to give you this is I upon 5 this is I upon 4, this
one is going to be I upon 4 I upon 4 so it is going to be point 5, point 5 and this is going to be 1,
this is going to be 1. Here same thing 0.44 0.56 this is going to be 0.5, 1 here, 1 here 0.44, 0.56,
0.5, 0.5 and 1.0.
There is going to be carry over in all the directions. We know how this entire thing is going to
happen. I am just going to put down some of the numbers. So you know how to look at it minus
2
44 minus 55, this is cancelled, (Refer Slide Time: 9:20) here you get minus 22.2, here you get
minus 27.8, this 27.8 gets distributed 13.9 13.9, this carries over here to 7.0, this is carried over
here to 7.0, this gets minus 3.1 minus 3.9, this gets carried over to minus 1.6, this gets carried
over to minus 1.9 and then this gets done as plus 0.9 plus 1. We will stop there because we have
reached less than 1 percent and close it at this point. This becomes minus 23.8, this becomes
minus 47.5, this becomes plus 47.5, this becomes minus 14.9 plus 14.9 and plus 7.5.
Similarly, we do it for this (Refer Slide Time: 10:49). This is going to be minus 50 minus 50
break carry over minus 25 minus 25 gets done with plus 11.1 and plus 13.9. This one comes as
plus 6.9, this goes as plus 5.6, this comes out as minus 3.4, minus 3.5, minus 3.5 goes as minus
5.17, we then come over here this is plug this is minus 1.7, this becomes plus 8 plus 0.9, this
carries over to plus 4. We close it here and whether this I have just done the so this becomes plus
6 this becomes plus 11.9. This becomes minus 11.9. This becomes plus 53.5. This becomes
minus 53.5 and this becomes 26.7. This one goes exactly the same way; minus 50 minus 50 - this
goes minus 25 minus 25 plus 11.1 plus 13.9, this carries over to 6.9, this becomes minus 3.4
minus 3.5 carry over 1.37 here 1.7, this carry over this becomes 0.8 and 0.9, this carries over 0.4
close and this becomes plus 6.0 plus 11.9 minus 11.9 minus 53.5 plus 53.5 and this 46.5 (Refer
Slide Time: 13:03) and this becomes minus 26.7.
Essentially we are writing down that MBC is equal to we are not interested in AB and BA
because a load does not travel on AB and BA. Those are always going to be equal to 0. We are
only interested in the once that whether fixed end moments are non zero. We have 0.475 MBC
minus 0.119 MCB and then minus 0.119 MCD. Then we have MCB is equal to minus 0.149 MBC
plus 0.535 MCB minus 0.465 MCD and MCD is equal to 0.149 MBC minus 0.535 MCB plus 0.465
MCD and finally MDC is equal to 0.075 MBC minus 0.267 MCB minus 0.267 MCD plus MDC this is
the fixed end moment at DC. These are our expressions. When 0 is less than x is less than 4 MBC
is equal to 4 minus x squared into x all upon 4 squared. MCD and MDC are equal to 0. When you
have between 4 x and 8 MBC and MCB are equal to 0 and MCD is equal to 8 minus x the whole
3
squared into x minus 4 upon 16 and MDC is equal to x minus 4 squared into 8 minus x upon 16.
These are the fixed end moments.
What is the next step? Once we have figured out from the moment distribution the relationship
between the member end moments and the fixed end moments we also have the expressions for
the fixed end moments while the load is traveling from B to C to D. We have to write down all
the expressions for it. If we look at it now the only thing that we are doing is VE. Understand that
we are only interested in VE.
Let us say load is between 0x and 4. One of the things that we do know - let us look at this. This
is MBC (Refer Slide Time: 18:24) this is MCB. In addition we have the load 1 at a distance x.
Therefore if you look at this one I will call it as VBC and I will call this as VCB. If you look at this
in this way VBC is going to be equal to 4 minus x upon 4. From this part and the other part is plus
MBC plus MCB upon 4 and VCB is equal to x upon 4 minus MBC plus MCB upon 4. Once we know
these now our E point is here, when 0 is between 2 minus that means the load is between B and
E. That means the shear force definition will have this and will have this this is VE. VE is going
to be equal to VCB and when the load is between this VE is equal to we will see that this is VBC
and this is the positive VE then this is going to be minus V so for this V is equal to this and for
this - these are the expressions for VE when the load is between B and C. We now need to just
find out what is going to be the VE.
If the load is between 4 and 8 then BC becomes MBC MCB the load is no longer there. We have
VBC and VCB and we see that VBC is equal to MBC plus MCB by 4 and VCB is equal to minus of
MBC plus MCB by 4. If we look at it then VE is equal to minus BC or which is also equal to VCB.
This is minus and this is minus (Refer Slide Time: 22:23). Now let us look at how to compute.
What we do is we compute.
For computation again we draw this. We are going to have x and we are going to have MBC MCB
MCD we get small m and capital M, then we get VC and finally we get VE. This is x, this is the
fixed end moment, this is (FEM)cb, this is (FEM)cd we do not need anything else for MBC and
5
MCD because VE only depends on those VBC VCB and VE. then we put 0 1 2 minus 2 plus 3 4 5 6 7
8. Let us draw these lines. Why 2 minus and 2 plus? When the load is just to the left and just to
the right of the E the sheer is different which we already computed.
Having done this just substitute x we get this 0, this 0, this 0, this is 0, this is 0, this is 0, this is 1,
this is 0 and this is 0, this is the exponent (Refer Slide Time: 24:39). Then for 1 we compute and
this becomes 2.25 by 4, this becomes 0.75 by 4, this is 0 because MCD is 0 when it is 0 up to here
up to 4 and MBC and MCB are 0 when between 4 and 8 because, this is also 0, this is 8. When we
compute this becomes point 290, substituting these MCB becomes minus 0.184 so VBC turns out
to be 0.7765, this turns out to be 0.2235 and therefore VE is 0.2235 because up to 2 minus it is
equal to VCB.
Plug this in (Refer Slide Time: 25:49). This becomes 2 by 4, this becomes minus 2 by 4 so when
you compute this becomes 0.297, this becomes minus 0.342, this becomes 0.488, this becomes
0.5112, this is now 0.5112. At this point 2 plus all of these are identical because the moments
cannot change, VBC also is the same but now at 2 plus VE is equal to minus of, so this is going to
become minus 0.4888 and then this becomes 0.75 by 4. Now you keep putting in these values
and you will get this is equal to 0.156, this is equal to minus 0.329, this becomes 0.2068, this
becomes 0.7933, this becomes minus 2 0 6 8. At 4 all of these are 0 now here this becomes 2.25.
This is 2 by 4 and this is 0.75 by 4.
Substituting those in we get minus 0.067 minus 0.060 minus 0.022. This is of course 0, 0, 0 and
0. Once we substitute these we get minus 0.262. This becomes minus point 0823, this becomes
.08 plus 0.823 and this one is minus 0.233. This is minus 0.087. This becomes minus 0.733. This
becomes plus 0.733. This one is minus 0.0273. This becomes plus 0.273. These are the numbers
that I had done. I just evaluated the procedure for getting these.
Steps for quantitative influence lines: first identify load path, identify which fixed end moments
come into play. When you know the load path you know exactly from which fixed end moments
are going to be non zero as the load moves from which ever point to which ever point we wanted
to move. Identify which come into play. For each FEM equal to 100 do moment distribution. As
many number of points at which you develop fixed end moments you are going to get that many
moments to distribute.
For example, in this particular case we had when it was between B and C we had BC and CB
operating. When it is between C and D we had CD and DC, DC was the fixed end so we did not
do a moment distribution for that but for the other 3 BC CB and CD we have to do the moment
distribution. Once you do all of those then you get member end moments Mij in terms of fixed
end moments. Find expressions for fixed end moments as load moves.
In other words, for each member that you have, you put it at a point x and you find out the
expression for the fixed end moments. Once you get the find out the expression for fixed end
moments you get the expressions for Mij as the load moves. Then do equilibrium or should I say
use equilibrium to compute any response quantity in any member in terms of Mij and unit load.
Once you do that then you have got it and then you can put in different values of x and evaluate
whichever response you need at that point and that gives you the influence line for the for the
particular response quantity. For example, in that problem that I have just done the response
quantity of interest was Ve and so therefore we only obtain the expressions for VE in terms of
MBC and MCB and when the load was in bc also in terms of the load.
When the load moved away from BC and the load did not come in, it was only in terms of MBC
and MCB. So much for how to get influenced lines both qualitatively and quantitatively for any
structure, be it statically determinate. In the statically determinate structure the Muller Breslau
principle gives us both the qualitative as well as quantitative because whenever we remove the
restraint corresponding to the response quantity of interest a statically determinate structure
becomes a first auto mechanism and all displacements are straight lines. Therefore, we do not
need any method we also get the quantitative. When you have a straight line and you know the
end you always can find out anything as we have already seen. Therefore, the Muller Breslau
principle gives us both the qualitative as well as the quantitative for a statically determinate
structure.
For a statically indeterminate structure we can still continue using the Muller Breslau principle
but it only gives us the qualitative sense of the influence line. To get a quantitative sense you
have to go through the direct approach in which you have to make each member into statically
determinate and then solve for it. So much or influence line. Now the question arises - see till
now whatever we have done there has been a reason for it. This is the first time I looked at a
problem when I say look let us first look at the influence line and then we will see what the
applications are.
Where do we apply influence line? It is obviously moving load. What kind of moving load can
we have? We can have something like this that over some a we have some load intensity which
is given uniformed intensity Pa so that total load is P into a. When does this kind of a thing
happen? This happens when you have a truck or a car traveling over a bridge. There is an area
over which the load is distributed.
On the other hand, you may have a situation like this; effectively point load where you have a
load which is P this is capital P this is small p (Refer Slide Time: 35:50). Load is P and it is
acting at a particular point when you can have this kind of a thing. This can happen when you
have a train on a railway track or a crane. Essentially the load is transferred through a steel wheel
which has in a sense a point contact. These are the two kinds of moving loads that we have. How
do we tackle this? You see the overall point here is this. Let me just take this particular example
that we have just done to look at.
We have this influence line and now we know the values also. This is (Refer Slide Time: 36:45)
point 511 and this is point 489 what is this do actually. This tells us to get the maximum value of
VE which is the position of the load we have. This is the position we have.
For maximum value of VE we need to put it just to the left. So, x equal to 2 minus is VE max and
the value for a unit load is equal to plus point 511. When we have this, for this load, this moving
we know what the maximum VE is going to be, we know where the load has to be applied, it has
to be applied at this point and once we apply the load at this point you can do analysis and you
can get VE max is equal to 0.51 P.
9
I am just giving you the example VE but this you can do for anything, any response quantity. For
example, what is the maximum support reaction? We know what the maximum support reaction
is. We can also find out the position where it is, we also know what the maximum value is for the
unit load. So when you have a train or a crane all you need to do is whatever value you get for
the response quantity you multiply it by P. When you are moving a load you can never find out
and that is the reason why we have influence line, but now we know for a point load how to
apply the influence line to get the maximum response for a moving point load.
What happens when you have something like this? Let me look at VE.
I have VE, I am illustrating with VE mind you, it is not if I am choosing, and so this is 0.511 and
this is 0.489. This is what we have evaluated then we have evaluated for every point. Now the
point here is, if we have a load which has a particular pressure over a particular area. This is a
typical class AA loading if you look at the Indian Roads Congress code IRC: 6 then you
will see that you have given this kind of a load. Of course what you have given is many loads
together. The point that I am trying to make is that, once you have for one you can always place
it any which way you want because once you have the load for one position you have all the
other load positions automatically.
Now the total load P is equal to Pa. How do we solve this problem? Note that all we need to do is
find out for this load what the maximum is, we need to find out the position. Which position do
you think is going to be the position of the load? Automatically if you look at this you will see
that this is the maximum. So automatically one edge of the load should be here and if the load
sits like this you will get the maximum influence of the load on VE. This is a and what we do is
we find out the area under this curve. So the area under the influence lines over the distance a
when the load is placed at the maximum over the x multiplied by P gives me, this is the pressure
so this pressure multiplied by the influence line gives me the pressure at this point and then when
we integrate it we get VE max for this load.
10
How do we integrate it? We actually do not integrate it. What we do is we discretize it and then
numerically integrate it. You got your VE max. What might happen is that once you place one
load the other load may come here, all you need to do is sum up the influence lines. Once you
place one load at the maximum the others will come automatically. The distances are fixed, the
wheels, axels, loads are fixed. This way we find out VE max. I am illustrating it with VE max.
Therefore you see now, how for a given moving load how we get the maximum response in this
particular VE, we could also do it for Ra RD MD Ma anything for this particular problem.
For your own particular problem you can always solve the specific problem and find out its
effect. This is the reason why it is actually an analysis problem. We are actually trying to find out
the maximum value of a particular response by sheer, the sheer force at a particular point,
bending moment at a particular point, the reaction at a particular support, the moment at a fixed
support so these are the things that we are interested in. And for whatever we are interested in,
we draw the influence line for that. What we always do is that whatever load you have you
always place the load where you get the maximum influence. Once you place the load at the
maximum influence if it is a point load the value of the influence multiplied by the load directly
gives you the maximum value of the response under the moving load and then if you are doing
for distributed load like a tyre load then this is the procedure that you use (Refer Slide Time:
44:30).
In essence the overall use of influence line is; 1) to determine the position of the load - wherever
influence on the response - that is the reason why we call it influence line, the Influence on the
response is maximum. Once we determine the position; if point load then design response is
equal to influence line into load; if distributed load then design response is equal to integral over
the distribution area or length into influence line into pressure into dx where p into a is equal.
Basically P is equal to P upon a that is the pressure, therefore here what we do is load upon a
multiplied by influence line over the length a you integrate it and that gives you the design
response. This in essence is the overall thing.
11
Just to ensure that I leave you with some problems to do I am going to give you two or three
problems to do for your own self. I am going to define one problem, this is the same problem as
what I have just done.
In this problem this is my e point which is 2 meters identical. This is the same problem in fact for
M I I I this is 3, this is 4, this is 4 A B C D and I ask you to find out RA, the sheer VE and the
bending moment at B, quantitative influence line for bending moment at B R at a and V at E.
You find these out I have already done this I will give you the answers for your computation.
12
If you look at it this is x 0 1 2 3 4 5 6 minus 6 plus 7 8 and Ra VE bending moment at b this is

going to be 1 0 0. This is going to be 0.7855 minus 0.038 minus 0.294 0.503 minus 0703. This is
going to be minus 0 3 0; for 3 it is 0.219 minus 0.71 minus 0.161, for 4 all these are 0 (Refer
Slide Time: 49:55), for 5 this is minus 1, this is 0.62 point 0.383. At 6 we have 0.125, this is
0.399 0.95 and for 6 plus these two remain the same, this one is minus 0.061; for 7 this is 0.078
minus 0.313 0.060 ultimately 0 0 0.
I have given you the answers you have to solve this problem yourself. Then I would like you to
find out the maximum values for a point load of 100 kilo Newtons.
So find out the maximum value or the designed responses for all these 3 for 100 kilo Newtons.
I am going to stop here I hope that the lectures on the influence line has given you a little bit of
an insight on how to obtain influence lines for structures and then how to apply them for actual
problems to be solved.
Thank you very much. It has been a pleasure interacting with you. I hope that my lectures on the
force method, the displacement method and then the matrix flexibility approach, the matrix
displacement method, the moment distribution method and finally the influence line has given
you some idea of structural analysis for statically indeterminate structures. Thank you very
much. Bye.
13

Lecture PDF IIT Bombay

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Lecture PDF IIT Bombay

Загружено:

Авторское право:

Доступные форматы

Structural Analysis - II

(Refer Slide Time: 09:59)

(Refer Slide Time: 24:31)

(Refer Slide Time: 40:43)

(Refer Slide Time: 43:24)

(Refer Slide Time: 46:06)

(Refer Slide Time: 14:43)

(Refer Slide Time: 19:18)

(Refer Slide Time: 28:55)

(Refer Slide Time: 32:27)

(Refer Slide Time: 45:36)

(Refer Slide Time: 47:35)

This is a propped cantilever. It is a statically indeterminate structure, because it has 4 support

(Refer Slide Time: 49:09)

(Refer Slide Time: 52:16)

(Refer Slide Time: 08:55)

What is the next step?

(Refer Slide Time: 37:03)

(Refer Slide Time: 11:22)

(Refer Slide Time: 28:36)

Force in ab is equal 3 by 5 X1(X1 is equal to one and this is compressive; assuming X1 to be

In the original structure

(Refer Slide Time: 34:37)

(Refer Slide Time: 39:40)

(Refer Slide Time: 39:56)

(Refer Slide Time: 40:46)

(Refer Slide Time: 50:29)

(Refer Slide Time: 54:20)

(Refer Slide Time: 05:36)

(Refer Slide Time: 06:38)

(Refer Slide Time: 13:04)

(Refer Slide Time: 31:14)

(Refer Slide Time: 34:09)

(Refer Slide Time: 40:22)

(Refer Slide Time: 06:21)

(Refer Slide Time: 10:44)

(Refer Slide Time: 12:04)

If I put X1 is equal to 1, then Fab is equal to 3 by 5, 4 by 5, 1, 3 by 5, 4 by 5 and 1.We have

(Refer Slide Time: 20:54)

equal to length of cd is equal to 3 meters; length of ad is equal to length of bc is equal to 4

(Refer Slide Time: 30:46)

(Refer Slide Time: 35:17)

(Refer Slide Time: 43:32)

(Refer Slide Time: 46:37)

(Refer Slide Time: 09:45)

(Refer Slide Time: 38:19)

(Refer Slide Time: 41:27)

(Refer Slide Time: 41:49)

(Refer Slide Time: 44:22)

(Refer Slide Time: 48:47)

(Refer Slide Time: 09:14)

(Refer Slide Time: 43:44)

(Refer Slide Time: 46:34)

(Refer Slide Time: 02:11)

(Refer Slide Time: 02:19)

(Refer Slide Time: 02:36)

(Refer Slide Time: 02:50)

(Refer Slide Time: 02:54)

(Refer Slide Time: 04:05)

(Refer Slide Time: 08:37)

(Refer Slide Time: 10:41)

(Refer Slide Time: 13:22)

(Refer Slide Time: 16:10)

(Refer Slide Time: 18:40)