21

CHAPTER 7

Fatigue Failure Resulting from Variable
Loading





7.1 Introduction to Fatigue in Metals

When machine parts are subjected to time varying loading, their behavior is entirely
different from what they could behave when they are subject to static loading.
For example, a particular fiber on the surface of a rotating shaft subjected to the
action of bending loads undergoes both tension and compression for each revolution
of the shaft, is stressed in tension and compression N times each minute if the shaft is
part of an electric motor that is rotated at N rev/min.
Machine members are found to have failed under the action of repeated or fluctuating
stresses: yet the most careful analysis reveals that the actual maximum stresses were
well below the ultimate strength of the material, and quite frequently even below the
yield strength.
A fatigue failure has an appearance similar to a brittle fracture, as the fracture
surfaces are flat and perpendicular to the stress axis with the absence of necking. See
Fig. 7.1 page 307.
The fracture features of a fatigue failure, however, are quite different from a static
brittle fracture arising from three stages of development.
Fatigue failure is due to crack formation and propagation. A fatigue crack will
typically initiate at a discontinuity in the material where the cyclic stress is a
maximum. Discontinuities can arise because of:

o Design of rapid changes in cross section, keyways, holes, etc.
o Elements that roll and/or slide against each other (bearings, gears, cams, etc.)
under high contact pressure, developing concentrated subsurface contact
pressure that can cause surface pitting (hole) or spalling after many cycles of
the load.
o Carelessness (lack of care) in locations of stamp marks, tool marks, scratches,
and burrs; poor joint design; improper assembly; and other fabrication faults.
o Composition of the material itself as processed by rolling, forging, casting,
extrusion, drawing, heat treatment, etc.




22
7.3 Fatigue-Life Methods

There are three major fatigue life methods used in design and analysis:

1- Stress-Life Method

o It is based on stress levels.
o The least accurate approach, but most used method, since it is the easiest to
implement for a wide range of design applications.


2- Strain-Life Method

o Involves more detailed analysis of the plastic deformation at localized regions
where the stresses and strains are considered for life estimates.


3- Linear-Elastic Fracture Mechanics Method

o It assumes a crack is already present and detected.
o It is then employed to predict crack growth with respect to stress intensity.
o It is most practical when applied to large structures in conjunction with
computer codes and a periodic inspection program.


In our study we will focus only on the first method (The Stress-Life Method).


7.4 Stress-Life Method

To determine the strength of materials under the action of fatigue loads, four types of
tests are performed: tension, torsion, bending, and combinations of these.
In each test, specimens are subjected to repeated forces at specified magnitudes while
the cycles or stress reversals to rupture are counted.
For the rotating-beam test, a constant bending load is applied, and the number of
revolutions (stress reversals) of the beam required for failure is recorded. The first test
is made at a stress that is somewhat under the ultimate strength of the material. The
second test is made at a stress that is less than that used in the first. This process is
continued, and the results are plotted as an S-N diagram.
The figure blow is the S-N diagram for typical UNS G41300 steel.






23
S
ut

0.9S
ut

S
e

Low Cycle High Cycle
Finite Life
Infinite Life
Number of stress cycles, N
F
a
t
i
g
u
e

S
t
r
e
n
g
t
h
,

S
e

10
1
10
2
10
3
10
4
10
5
10
6
10
7
10
8
10
9
10
0






















It should be noted that from the N-S diagram for the case of steel:

o A knee occurs around N=10
6
and beyond this knee failure will not occur, no
matter how great the number of cycles.
o The strength corresponding to the knee is called the endurance limit S
e
, or the
fatigue limit.
o For nonferrous metals and alloys, the graph of S-N diagram will never be
horizontal.

Meaning of N:

o A stress cycle (N=1) constitutes a single application and removal of a load and
then another application and removal of the load in the opposite direction.
o Thus, N=0.5, means the load is applied once and then removed, which is the
case with the simple tension test.
o A body of knowledge available on fatigue failure from N=1 to N=10
3
cycles is
generally classified as low-cycle fatigue.
o A stress cycles greater than N=10
3
, is classified as high-cycle fatigue.
o The boundary between the finite-life region and the infinite-life region can be
defined only for a specific material such as steel as shown above in which it
lies somewhere between N=10
6
and N=10
7
.

S-N method does not work well in low-cycle application, where the applied strains
have a significant plastic component.
For more information about the previous sections, read sections 7.1, 7.3, and 7.4.
24
7.7 The Endurance Limit

Figure 7.18 suggests that the endurance limit ranges from about 40 to 60% of the
tensile strength for steel up to (1460 MPa)
About 1460 MPa, the scatter appears to increase, but the trend seems to level off at
'
e
S =740 MPa.
Mischke has analyzed a great deal of actual teat data from several sources and
concluded that endurance limit can be related to tensile strength.
For Steel:
o
'
e
S =

'

MPa
S
ut
740
504 . 0

MPa S
MPa S
ut
ut
1460
1460

Where
ut
S : Minimum tensile strength

'
e
S : Rotating-beam specimen itself

For Steel Microstructures, the endurance limit ratio
ut e
S S /
'
is given in table 7.3.
Aluminum and Magnesium alloys do not have an endurance limit, and the fatigue
strength
f
S is usually based on 5(10
8
) cycles of stress reversal and it is given in table
A-24.
Endurance limits for various classes of cast irons, polished or machined, are given in
table A-24.
We could also use for Cast Iron and Cast Steel:

o
ut e
S S 45 . 0
'
for 600
u
S MPa
o 275
'

e
S MPa for 600 >
u
S MPa


















25
S
ut

fS
ut

S
e

Low Cycle High Cycle
Number of stress cycles, N
F
a
t
i
g
u
e

S
t
r
e
n
g
t
h
,

S
e

10
3
10
6

S
f
=aN
b

7.8 Fatigue Strength

The purpose of this section is to develop methods of approximating the S-N diagram
for steels when the results of the simple tension test are known, in the high cycle
region.
High-Cycle fatigue extends from N=10
3
to N=10
6
(the endurance limit life N
e
)
Let the equation for the S-N line be written as:

b
f
aN S (1)























At N=10
3
, equation (1) becomes:

( ) ( ) ( )
b
b
f
a a S
3 3
10
10 10
3
(2)
Also,
( )
ut f
fS S
3
10
(3)


Where f = the fraction of S
ut
represented by ( )
3
10
f
S

In high-Cycle Fatigue and stress level below the proportional limit.

( )
b
f a
N 2
'
(4)
26
Where
'
f
= fatigue strength, for steel H
B
500, 345
'
+
ut f
S MPa
2N = reversals to failure ( 1 rev. = 0.5 cycle)
N= number of cycles
b = fatigue strength exponent

Now for S-N diagram, we can define equation (4) as:

( )
b
f a f
N S 2
'
(5)

At N = N
e
=10
6
we have S
f
= S
e
, then equation (5) becomes:

( )
b
e f e
N S 2
'
(6)

Solve equation (6) for the parameter b we have:

( )
( )
e
e f
N
S
b
2 log
/ log
'

(7)

Hence, we can now find the value f either from equation (3) or (5)

From equation (3):
( )
ut
f
S
S
f
3
10
(8)
From equation (5):
( )
b
ut
f
S
f
3
'
10 . 2

(9)

If we need to use equation (1) directly to find endurance limit or the life cycle, we need
two points to determine the value of a and b, then

At N =10
3
S
f
= fS
ut

At N =10
6
S
f
= S
e


Substitute these two points in equation (1), we have two equations with two unknowns a
and b.
Solve these two equations, we have:

( )
e
ut
S
fS
a
2
(10)
( )
( ) 10 log 3
/ log
e ut
S fS
b (11)
27
Example 7.1

A 1050 HR steel an ultimate tensile strength of S
ut
=724MPa and a yield strength of
414MPa.
a. Estimate the rotating-beam endurance limit at 10
6
cycles
b. Estimate the endurance strength for a polished rotating specimen corresponding to
10
4
cycles to failure.
c. Estimate the expected life under a completely reversed stress of 379MPa.


Solution:
Problem Given: S
ut
= 724MPa, S
y
= 415MPa
Problem Required:
a.
'
e
S
b. ( )
4
10 N
f
S
c. N

a. From equation 7.8

S
ut
1460MPa MPa S S
ut e
9 . 364 504 . 0
'


b.
( ) ( )
( )
( )
( )
( )
( )
( ) ( )
( ) ( )
( ) ( ) MPa a S
MPa
S
fS
a
S
f
N
S
b
MPa S
S
f
MPa S S
a aN S
b
N
f
e
ut
b
ut
f
e
e f
ut f
b
ut
f
e e
b
b
N
f
513 ) 10 ( 10 3 . 1015 10
. 3 . 1015
10 9 . 364
10 724 8407 . 0
8407 . 0 10 . 2
724
1069
10 . 2
0741 . 0
10 . 2 log
9 . 364 / 1069 log
2 log
/ log
1069 345
10 . 2
9 . 364
10
0741 . 0 4 6 4
10
6
2
6
2
0741 . 0
3 3
'
6
'
'
3
'
'
4
10
4
4

c. ) 10 ( 96 . 5
3 . 1015
379
5
0741 . 0
1 1

,
_

,
_

b
a
a
N

cycles

28
7.9 Endurance Limit Modifying Factors

In previous section, endurance limits is determined based on rotating-beam specimen
that is used in the laboratory under control condition.
It is unrealistic to expect the endurance limit of a mechanical member to match the
values obtained in the laboratory. Some differences includes:
o Material ( composition, variability, etc )
o Manufacturing ( method, heat treatment, etc )
o Environment ( corrosion, temperature, etc )
o Design ( size, shape, life, stress concentration, etc )
To account for the most important of these conditions, we employ a variety of
modifying factors, each of which is intended to account for a single effect.
Hence, the endurance limit can be written as:

'
e f e d c b a e
S k k k k k k S (12)

Where
S
e
=endurance limit at the critical location of a machine part in the geometry
and condition of use.
k
a
=surface condition modification factor
k
b
=size modification factor
k
c
=load modification factor
k
d
=temperature modification factor
k
e
=reliability factor
k
f
=miscellaneous-effects specimen endurance limit

Surface factor k
a


o Surface modification factor depends on the quality of the finish of the actual
part surface and on the tensile strength of the part material
o Surface factor can be calculated using the following formula:

b
ut a
aS k (13)
Where a and b are to be found in table 7.4

Surface finish Factor a Exponent b
ground 1.58 -.085
Machined of cold-drawn 4.51 -.265
Hot-rolled 57.7 -.718
As-forged 272 -0.995





29
Size factor k
b


o For axial loading k
b
=1
o For bending and torsion:

'

<

mm d
mm d d
k
b
254 51 51 . 1
51 79 . 2 24 . 1
157 . 0
107 . 0
(14)

o If the round bar in bending is not rotating, or when a noncircular cross section
is used we use the definition of the effective dimension. Therefore,

1. The effective size of a round corresponding to a non-rotating solid or
hollow round: d
e
=0.37d where d is the diameter

2. for rectangular cross section d
e
=0.808(hb)
0.5


3. For wide-flange beam cross-sectional area, see table 7.5












5 . 0
95 . 0
95 . 0
0766 . 0
2 2 025 . 0 05 . 0
1 1 1 . 0

,
_

'

>

A
d
axis a t ab
axis at
A
e
f
f
(15)


4. For channels beam cross-sectional area, see table 7.5







30
( )
5 . 0
95 . 0
95 . 0
0766 . 0
2 2 1 . 0 052 . 0
1 1 05 . 0

,
_

'

A
d
axis x b t xa
axis ab
A
e
f
(16)


Loading factor k
c

'

torsion
axial
bending
k
c
59 . 0
85 . 0
1
(17)

For axial load we could also use the following formula:

'

>
MPa S
MPa S
k
ut
ut
c
1520 923 . 0
1520 1
(18)

Temperature factor k
d


If the rotating beam endurance limit is known at room temperature, then

RT
T
d
S
S
k (19)
where
S
T
= tensile strength at operating temperature
S
RT
= tensile strength at room temperature
And it can be found in table 7.6

If the rotating beam endurance limit is not given, then compute it using equation
of (S
e
=0.504S
ut
) with S
ut
at the operation temperature and use k
d
=1.

Reliability factor k
e


a e
z k 08 . 0 1 (20)

where z
a
is the transformation variate and it is given in table 7.7

Miscellaneous-Effects factor k
f





31
7.10 Stress Concentration and Notch Sensitivity

Most mechanical parts have holes, grooves, notches, or other kinds of discontinuities
which alter the stress distribution.
The fatigue stress concentration factor can be defined as:
K
f
= maximum stress in notched specimen / stress in notch-free specimen
To use this factor in endurance limit formula, we write:

'
1
e
f
e
S
K
S (21)
or increase the stress load by:

o f
K
max
or
o f
K
max
(22)

Notch sensitivity q is defined by the equation:

1
1
1
1

ts
fs
shear
t
f
K
K
q or
K
K
q (23)
where
q is usually between zero and unity and stress concentration factor
K
t
= theoretical stress concentration factor which is given in table A-15

when
q = 0 K
f
= 1 material has no sensitivity to notches.
q = 1 K
f
= K
t
material has full notch sensitivity.

In design work:

1. find K
t
for different geometries, given in table A-15
2. find q from the charts given:

For bending or axial loads and the material are steels and 2024
aluminum alloys use figure 7-20
For shear load and the material are steels and 2024 aluminum
alloys use figure 7-21
For cost iron, the sensitivity is varying from zero to 0.2,
depending upon the tensile strength. And for conservative side,
it is recommended to use the value q = 0.2 for all grades of cast
iron.

3. K
f
can be found from K
f
=1+q ( K
t
-1 )

For combined load stresses: multiply each stress by it own K
f


32
Example 7.2

An AISI 1015 cold-drawn steel bar has a diameter of 25.4mm.
a. Estimate the endurance limit.
b. Estimate the endurance limit for reversed bending without rotation.
c. Estimate the fatigue strength at N=70(10
3
) cycles at an operating temperature of
287.77
o
C

Solution:
Given: Cold-drawn steel

a. Se=k
a
k
b
k
c
k
d
k
e
k
f
S
e


From table A-20 S
ut
=390 MPa
From equation 7.8:
S
ut
< 1460 MPa then
Se=0.504S
ut
=0.504(390)=196.56 MPa
k
a
= a(S
ut
)
b

from table 7.4 for cold-drawn
a = 4.51, b = -0.265, k
a
= 0.928

Unless specified, k
b
= 1.24d
-0.107
= 1.24(25.4)
-0.107
=0.877, for 2.79 d 51mm

Unless specified, the system works under the following condition: load factor k
c
=1,
room temperature k
d
=1, reliability k
e
=1, k
f
=1.

Hence, S
e
= (0.928)(0.877)(1)(1)(1)(1)(196.56)=159.97MPa

b. In this case, k
b
will change calculate based on non-rotating formula

d
e
=0.37d=0.37(25.4)=9.398mm
For 2.79 d
e
51mm
Hence, k
b
=1.24d
-0.107
= 1.24(9.398)
-0.107
=0.9756
Other parameters are the same, then
S
e
=(0.928)(0.9756)(1)(1)(1)(1)(196.56)=177.97MPa

Note: non-rotating reversed bending means: the bending force changes direction (+
or -). It is not the same as one way or push-push bending.

c. The temperature factor in this part will change because the operation temperature is
287.77
o
C.
From table 7.6, k
d
= S
T
/S
RT
= 0.981 by using the interpolation between T=250 &
300
o
C.
Hence,
S
e
= (0.928)(0.877)(1)(0.981)(1)(1)(196.56)=156.93MPa.
Then,
33
( ) ( )
( )
( )
( )
( )
( ) ( )
( ) ( )
( ) ( ) MPa a S
MPa
S
fS
a
S
f
N
S
b
MPa S
a aN S
b
N
f
e
ut
b
ut
f
e
e f
ut f
b
b
N
f
14 . 203 ) 10 70 ( 10 77 . 665 10 70
. 77 . 665
10 93 . 156
10 390 8288 . 0
8288 . 0 10 . 2
390
735
10 2
1064 . 0
10 2 log
93 . 156 / 735 log
2 log
/ log
735 345 390 345
10 7
1064 . 0 3 6 3
10
6
2
6
2
1064 . 0
3 3
'
6
'
'
4
10 7
4
4

+ +

Example 7.3
A rotating shaft supported in a ball bearings at A and D and loaded by the non-rotating
force F as shown in figure below. If all fillets are 3-mm radius, the shaft rotates, the load
is stationary and the material is machined from AISI 1050 cold-drawn steel, estimate the
life of the part. The bending moment diagram to the given problem is shown in figure b.
















Solution:

Failure will probably occur at either B or C. If the cross-section of the shaft is constant
then, the failure will occurs at C. But because the cross section is changeable across the
shaft, we need to check which of these two points will have the highest bending stress
and hence will fail first.
The bending stress can calculated from
I
Mc


At point B, M
B
= 695 N.m, I/c = 3.22x10
-6
m
3
,
B
= 216MPa
34
At max. point, M
max
= 904 N.m, I/c = 5.38x10
-6
m
3
,
max
= 167.8MPa
At point C, M
c
= 502 N.m, I/c = 4.21x10
-6
m3,
C
= 119.3MPa
Hence, point B has the critical section. Therefore our analysis will be based on point B.

Se=k
a
k
b
k
c
k
d
k
e
k
f
S
e


From table A-20 S
ut
= 690 MPa, S
y
= 580MPa
From equation 7.8, S
ut
< 1460 MPa
Thus, S
e
= 0.504S
ut
= 0.504(690) = 347.8MPa

k
a
= a(S
ut
)
b


From table 7.4 for cold-drawn
a=4.51, b=-0.265, k
a
=0.798

Rotating shaft, k
b
= 1.24d
-0.107
= 1.24(32)
-0.107
=0.856 For 2.79 d 51mm

For bending, the load factor k
c
=1
Unless specified, the system work under the following condition, at room
temperature k
d
=1, reliability k
e
=1, k
f
=1.

For stress concentration factor K
f


From table A-15-9: r/d=3/32=0.0938, D/d=38/32=1.188 K
t
= 1.65
From fig. 7.20, q = 0.82
Hence, K
f
=1+ q(K
t
-1) = 1.53

Method 1: Include the stress concentration factor in the endurance limit.

Therefore, S
e
= (0.798)(0.856)(1)(1)(1)(1)(1/1.53)(347.8)=155.3MPa
(
B
=216MPa) > (S
e
=155.3MPa), thus the shaft will have a finite life.

Therefore,
( )
( )
( )
( )
( ) ( )
( ) ( )
cycles
a
N
MPa
S
fS
a
S
f
N
S
b
MPa S
b
a
e
ut
b
ut
f
e
e f
ut f
75775
2 . 941
216
. 2 . 941
10 3 . 155
10 690 5542 . 0
5542 . 0 10 2
690
1035
10 2
131 . 0
10 2 log
3 . 155 / 1035 log
2 log
/ log
1035 345 690 345
131 . 0 / 1 / 1
6
2
6
2
131 . 0
3 3
'
6
'
'

,
_

,
_

+ +

35
Method 2: Include the stress concentration factor in the bending stress:

B
=K
f

B
=(1.53)(216)=330.5MPa
S
e
= (0.798)(0.856)(1)(1)(1)(1)(347.8)=237.6MPa
(
B
=330.5MPa) > (S
e
=237.6MPa), thus the shaft will have a finite life.

Therefore,

( )
( )
( )
( )
( ) ( )
( ) ( )
cycles
a
N
MPa
S
fS
a
S
f
N
S
b
MPa S
b
a
e
ut
b
ut
f
e
e f
ut f
38893
13 . 965
5 . 330
. 13 . 965
10 6 . 237
10 690 694 . 0
694 . 0 10 2
690
1035
10 2
1014 . 0
10 2 log
6 . 237 / 1035 log
2 log
/ log
1035 345 690 345
1014 . 0 / 1 / 1
6
2
6
2
1014 . 0
3 3
'
6
'
'

,
_

,
_

+ +

Which method to select when you deal with stress concentration factor:

o For simple loading, it is acceptable to reduce the endurance limit by either
dividing the unnotched specimen endurance limit by K
f
or multiplying the
reversing stress by K
f
.
o For combined loading, which may involve more than one value of fatigue-
concentration factor, the stresses are multiplied by K
f
.



7.11 Characterizing Fluctuating Stresses

Fluctuating stresses in machinery often take the form of a sinusoidal pattern
because of the nature of some rotating machinery.
It has been found that in periodic patterns exhibiting a single maximum and a
single minimum of force, the shape of the wave is not important, but the peaks on
both the high side (max.) and low side (min.) are important.
Thus, F
max
and F
min
in a cycle of force can be used to characterize the force
pattern
2
,
2
min max min max
F F
F
F F
F
a m

+
(24)
where
F
m
: midrange component of force
F
a
: amplitude component of force
36
F
max

F
min

F
max

F
min
= 0
Many time in design the stresses fluctuate without passing through zero
The following relationships and definitions are used when discussing mean and
alternating stresses:

min
= minimum stress

max
= maximum stress

r
= stress range

s
= steady or static stress

a
= stress amplitude
a
= (
max
-
min
) / 2

m
= midrange or mean stress
m
= (
max
+
min
) / 2
R= stress ratio =
min
/
max

A=amplitude ratio =
a
/
m




Fatigue under different loadings:

1. Fluctuating Loads

Example: spring & ball

a m
a

+
+

min
min max
min max
2
2









2. Repeated Loads (zero to maximum)

Example: Gear

1 , 0
2 2
2 2
max min max
max min max

A R
m
a

37
3. Completely Reversed Loads (fully reversed)

A R
m
a
, 1
0
2
2
min max
max
min max

4. Pure Torsional loads

2
2
min max
min max

m
a


Note:
The steady or static stress is not the same as the mean stress. In fact, it may have
any value between
min
and
max

The steady state exists because of a fixed load or preload applied to the part, and
it is usually independent of the varying portion of the load.
Example: A helical compression spring is always loaded into a space shorter than
the free length of the spring. The stress created by this initial compression is
called the steady state or static component of the stress.


7.12 Fatigue Failure Criteria for fluctuating Stress

Varying both the midrange stress and the stress amplitude, or alternating component, will
give some information about the fatigue resistance of parts when subjected to such
situations.
Three methods of plotting the results of such tests are in general use and are shown in
figures 7.24, 7.25, and 7.26.

1. Figure 7.24 (Modified Goodman Diagram)


m
plotted along the x-axis
all other components of stress plotted on the y-axis
The modified Goodman diagram consists of the lines constructed to S
e
(or S
f
)
above or below the origin.
F
F
38
S
y
is plotted on both axes, because S
y
would be the criterion of failure if max
exceeded S
y
.
Useful for analysis when all dimension of the part are known and the stress
components can be easily calculated. But it is difficult to use for design when the
dimension are unknown.





















2. Figure 7.25 (Fatigue Failures for Mean Stresses)

The x-axis represents the ratio of the midrange strength S
m
to the ultimate strength
The y-axis represents the ratio of the alternating strength to the endurance limit.
The line BC represents the modified Goodman criterion of failure
Existence of midrange stress in the compressive region has little effect on the
endurance limit.













39
3. Figure 7.26 (Master Figure Diagram)

It displays four of the stress components as well as the two stress ratios (
max
,

min
,
a
,
m
, R, A)
A curve representing the endurance limit for values of R beginning at R=-1 and
ending with R=1. Begins at S
e
on the
a
axis and ends at S
ut
on the
m
axis.
Constant life curves for N =10
5
and N =10
4
cycles have been drawn too.
Any Stress state, such as at the one at A, can be described by the minimum and
maximum components or by the midrange and alternating components.
Safety is indicated whenever the point described by the stress components lies
below the constant-life line.
When
m
is compression, failure occurs whenever
a
= S
e
or whenever
max
= S
yc


























The tensile side of figure 7.25 is again redrawn in terms of strengths in Figure 7.27,
instead of strength rations, with the same modified Goodman criterion together with 4
additional criteria of failure.






40
S
y

S
yc
S
yt
S
ut

a




















Such diagrams are often constructed for analysis and design purposes; they are
easy to use and the results can be scaled off directly.
There existed a position which divided safe from unsafe combinations of
a
and

m

In design work, the force amplitude and the mean force can usually be calculated
or determined


Example 7.4:

Working with bending moments or tensional moments, one can usually calculate
the mean moment and the amplitude of the moment.
The stress amplitude and the mean stress are related to these thought the
dimensions which are to be found.
The ratio
a
/
m
is the same as F
a
/F
m
or M
a
/M
m
, so a line from the origin through
point A can be drawn and the limiting values of
a
or
m
founds as the projections
of this points on the two axes.










41
Back to the figure 7.27, the five criteria of failure that are going to be used in
fatigue design are:

1. Soderberg Line:
1 +
yt
m
e
a
S
S
S
S
or
n S S
yt
m
e
a
1
+

(29)

2. Modified Goodman
1 +
ut
m
e
a
S
S
S
S
or
n S S
ut
m
e
a
1
+

(30)

3. Gerber
1
2

,
_

+
ut
m
e
a
S
S
S
S
or 1
2

,
_

+
ut
m
e
a
S
n
S
n
(31)

4. ASME-elliptic
1
2
2

,
_

,
_

y
m
e
a
S
S
S
S
or 1
2
2

,
_

,
_

y
m
e
a
S
n
S
n
(32)

5. Langer First-cycle yielding
1 +
y
m
y
a
S
S
S
S
or
n S
n
S
y
m
y
a
1
+

(33)


The theories of failure criteria are used in conjunction with a load line, r = S
a
/S
m
=

a
/
m
. and Principal intersections are tabulated in Tables 7-9 to 7-11


7.13 Torsional Fatigue Strength under Fluctuating Stresses

The existence of a torsional steady-stress component not more than the torsional yield
strength has no effect on the torsional endurance limit, provide the material is ductile,
polished, notch-free, and cylindrical.
The torsional fatigue limit decreases monotonically with torsional steady stress when
the material has stress concentration, notches or surface imperfections
In constructing the Goodman diagram

ut su
S S 67 . 0 (34)

Also, from chapter 6, S
sy
= 0.577S
yt
from distortion-energy theory, and the mean load
factor k
c
is given by equation 7-25 or use 0.577 for torsion.

42
7.14 Combinations of Loading Modes

In section 7.9 the load factor k
c
is used to obtain the endurance limit and hence that
the result is dependent on whether the loading is axial, bending, or torsion.
In this section, the following question will be answered:
How do we proceed when the loading is a mixture of axial, bending, and torsional
loads?
In addition to the complication introduced by the fact that a separate endurance limit
is associated with each mode of loading, there may also be multiple stress-
concentration factors, one also for each mode of loading.
To answer this question: Assuming that all stress components are completely
reversing and are always in time phase with each other,

1- For the strength, use the fully corrected endurance limit for bending, S
e

2- Apply the appropriate fatigue stress-concentration factors to the torsional stress,
the bending stress, and the axial stress components.
3- Multiply any alternating axial stress components by the factor 1/k
c,axial

=1/0.85=1.18.
4- Enter the resultant stresses into a Mohrs circle analysis or use the theory to find
the principle stresses.
5- Using the results of step 4 to find the von Mises alternating stress
a

6- Compare
a
with S
a
to find the factor of safety.

If the midrange stresses are also present, then steps 4 and 5 can be repeated for them
and the resulting steady Von Mises stress component
m
used amplitude components
are augmented by K
f
or K
fs
stress-concentration factor.


Example 7.5:
A 38 mm diameter bar has been machined from AISI 1050 CD steel. This part is to
withstand a fluctuating tensile load varying from zero to 71.2kN. Because of the ends and
the fillet radius, a fatigue stress-concentration factor K
f
is 1.85 for 10
6
or larger life. Find
S
a
, S
m
, and the factor of safety guarding against fatigue (n
f
) and first-cycle yielding (n
y
)
using (a) Gerber method and (b) Goodman method.

Solution:
From table A-20 Steel 1050 CD, S
ut
= 690MPa, S
y
= 580MPa.

MPa S
k k k
load axial for k
load aixal for k
aS k
MPa S S
e
f e d
c
b
b
ut a
ut e
59 . 235 ) 76 . 347 )( 1 )( 1 )( 1 )( 85 . 0 )( 1 )( 797 . 0 (
1
85 . 0
1
797 . 0 ) 690 ( 51 . 4
76 . 347 504 . 0
265 . 0
'

43
S
e
=235.6
S
y
=580 S
ut
=690
S
y
=580
Load Line r
r
crit

Gerber
Langer Line
S
m

S
a

a

A
B
D
0
C
MPa 78 . 57 ) 23 . 31 ( 85 . 1
MPa 78 . 57 ) 23 . 31 ( 85 . 1
MPa 23 . 31 kN 6 . 35
2
MPa 23 . 31 kN 6 . 35
2
min max
min max



+

mo f m
ao f a
m
mo m
a
ao a
K
K
A
F F F
F
A
F F F
F

a) Using Gerber Theory: form table 7.10
91 . 3
2
1 1
2
1
12 . 213
12 . 213
2
1 1
2
1
78 . 57
78 . 57
1
2 2
2
2 2
2

1
1
]
1

,
_

+ +

,
_

1
1
]
1

,
_

+ +

,
_

+
a ut
e m
e
a
m
ut
f
a
m
ut
e
e
ut
a
m
a
m
a
ut
m
e
a
S
S
S
S
n
MPa
r
S
S
MPa
rS
S
S
S r
S
S
S
r line load
S
S
S
S

44
S
e
=235.6
S
y
=580 S
ut
=690
S
y
=580
Load Line r
r
crit

Goodman
Langer Line
S
m

S
a

a

A
B
D
0
C
The factor of safety guarding against first cycle yielding is
( )
( )
02 . 5
78 . 57
290
290
1 1
) 580 )( 1 (
1

y
y
langer a
a
langer a
y
n
MPa
r
rS
S
S
OA
OC
n

To find the point of intersection (point D) between Gerber and Langer failure theories:
317 . 0
63 . 139 37 . 440 580
37 . 440 1
2
1 1
2
2
2

1
1
]
1

,
_

,
_

m
a
crit
m y a
e
y
ut
e
e
ut
m
m
a
crit
S
S
r
MPa S S S
MPa
S
S
S
S
S
S
S
S
S
r

r =1 and r
crit
< r , then primary threat from fatigue.

b) Using Goodman Theory: from table 7.9



















45
04 . 3
1
63 . 175
63 . 175
1
78 . 57
78 . 57
1

+
ut
m
e
a
f
a
m
e ut
ut e
a
m
a
m
a
ut
m
e
a
S S
n
MPa
r
S
S
MPa
S rS
S rS
S
S
S
r line load
S
S
S
S

To find the point of intersection (point D) between Goodman and Langer failure theories:
( )
11 . 0
03 . 57 97 . 522 580
97 . 522

m
a
crit
m y a
e ut
ut e y
m
m
a
crit
S
S
r
MPa S S S
MPa
S S
S S S
S
S
S
r

r =1 and r
crit
< r , primary threat from fatigue.
n
y
> n
Gerber
> n
Goodman


Example 7.6:
A steel bar undergoes cyclic loading such that
max
= 413MPa and
min
= -138MPa, for
the material, S
ut
= 551MPa and S
y
= 448MPa and a fully corrected endurance limit of
S
e
=275.5MPa and f = 0.9. Estimate the number of cycles to a fatigue failure using:
1. Modified Goodman criterion
2. Gerber criterion

Solution

( )
( )
( )
085 . 0 log
3
1
MPa 891
MPa 5 . 237
2
138 413
2
MPa 5 . 275
2
138 413
2
2
min max
min max

,
_

e
ut
e
ut
m
a
S
fS
b
S
fS
a

46
S
e

S
y
=448 S
ut
=551
S
y

Load Line r
Langer Line
S
m

S
f

A
B
0
Goodman
S
a

No. of cycle N to failure using Goodman
8 . 0
1

+
f
f ut
m
e
a
n
n S S



Since fatigue failure is going to occur, the part has a limited number of cycles before it
fail. Hence, defined the new linear equation that it pass through points (0, S
f
), (
m
,
a
),
and (S
ut
, 0) which can be done by change the term S
e
with S
f
in Goodman equation as
shown below:
( )
cycles N
MPa
S
S
n with
n S S
ut m
a
f
ut
m
f
a
33912
891
11 . 367
11 . 367
/ 1
1
1
085 . 0 / 1

,
_

No. of cycle N to failure using Gerber theory

( )
cycles N
MPa
S
S
n with
n S S
ut m
a
f
ut
m
f
a
466247
891
8 . 293
8 . 293
/ 1
1
1
085 . 0 / 1
2
2

,
_

,
_

47
Example 7.7:
A rotating shaft is made of 42X4 mm AISI 1018 CD steel tubing and has a 6 mm
diameter hole drilled transversely through it. Estimate the factor of safety guarding
against fatigue and static failures using the Gerber and Langer failure criteria for the
following loading conditions:
a) The shaft is subjected to a completely reversed torque of 120N.m in phase with a
completely reversed bending moment of 150N.m
b) The shaft is subjected to a pulsating torque fluctuating from 20 to 160N.m and steady
bending moment of 150N.m.










Solution

From table A-20, S
ut
=440MPa, S
y
=370MPa
42X4 D = 42mm, t = 4mm

MPa S
k k k
load bending for k
d for d k
aS k
MPa S S
e
f e d
c
b
b
ut a
ut e
87 . 165 ) 222 )( 1 )( 1 )( 1 )( 1 )( 8313 . 0 )( 8988 . 0 (
1
1
51 79 . 2 8313 . 0 ) 42 ( 24 . 1 24 . 1
8988 . 0 ) 440 ( 51 . 4
222 440 504 . 0 504 . 0
107 . 0 107 . 0
265 . 0
'

Theoretical stress-concentration factors

From table A-16 For Bending

( )
366 . 2 798 . 0
143 . 0
42
34
, 143 . 0
42
6
3310
32
3 4 4

t
net
net
K A where
D
d
D
a
mm d D
D
A
Z
Z
M

a
d

D

M M
T
T
48
From table A-16 For Torsion
( )
75 . 1 89 . 0
143 . 0
42
34
, 143 . 0
42
6
15500
32
2
3 4 4

ts
net
net
K A where
D
d
D
a
mm d D
A
J
J
TD

From Figures 7.20 and 7.21 with r =3mm
( ) ( )
( ) ( ) 72 . 1 1 75 . 1 96 . 0 1 1 96 . 0 1
07 . 2 1 366 . 2 78 . 0 1 1 78 . 0 1
78 . 0 , 78 . 0
+ +
+ +

ts fs
t f
T b
K K
K K
q q


Part A: completely reversed cycle:
0
96 . 27
2
8 . 93
, 0
, , , ,
max , ,
max , ,
max




m xy m y m x a y
net
fs xy a xy
net
f x a x
a m
MPa
J
TD
K
MPa
Z
M
K






Using Von Mises theory
( )
0
56 . 105 3
'
5 . 0
2
,
2
.
'

+
m
a xy a x a
MPa

Finding n
f
using Gerber Theory
57 . 1
56 . 105
166
0
1
'
'
2
' '

,
_

+
a
e
f m
f ut
m
e
a
S
n
n S S

Finding n
y
using first cycle factor (Langer Theory)
5 . 3
56 . 105
370
0
1
'
'
' '

+
a
y
y m
y y
m
y
a
S
n
n S S

49
Part b: Fluctuated torque and steady bending moment

0
8 . 93
, . 150
21
2
3 . 16
2
. 90
2
20 160
, . 70
2
20 160
, , ,
,
,
,
,

m y a y a x
net
f m x
m x
net
m
fs m xy
net
a
fs a xy
m a
MPa
Z
M
K
m N M
MPa
J
D T
K
MPa
J
D T
K
m N T m N T

Using Von Mises theory
( )
( )
28 . 0
6 . 100 3
2 . 28 3
'
'
5 . 0
2
,
2
,
'
5 . 0
2
,
'

+

m
a
m
a
m xy m x m
a xy a
S
S
r
MPa
MPa

Finding n
f
using Gerber Theory, From table 7.10
04 . 3
04 . 3
2
1 1
2
1
1 . 306
7 . 85
2
1 1
2
'
2 2
2
2 2

1
1
]
1

,
_

+ +

,
_

1
1
]
1

,
_

+ +
a
a
f
a ut
e m
e
a
m
ut
f
a
m
ut
e
e
ut
a
S
n or
S
S
S
S
n
MPa
r
S
S
MPa
rS
S
S
S r
S

Finding n
y
using first cycle factor (Langer Theory)
873 . 2
1
' '
' '

+
m a
y
y
y y
m
y
a
S
n
n S S




50
306
S
e
=166
S
y
=370 S
ut
=440
S
y
=370
r
=
0.28
Gerber
Langer Line
Part a
289
D
0
r
crit=
0.4
Part b