Академический Документы
Профессиональный Документы
Культура Документы
Introduction
SimXpert MD Explicit is a preprocessor for graphically preparing input data for MD Nastran Explcit, an
explicit dynamic software, used in applications such as crash, crush, and drop test simulations. Use of
this workspace allows users to work within one common modeling environment with other SimXpert
workspaces such as Structures. Thus, for example, a model originally prepared for NVH, linear, or
implicit nonlinear analysis can be easily used in explicit applications (e.g. crash). This dramatically
reduces the time spent to build different models for implicit and explicit analysis and prevents you from
making mistakes because of unfamiliarity between different programs.
Theory
A detailed theory of explicit analysis is outside the scope of this guide. However, it is important to
understand the basics of the solution technique, since it is critical to many aspects of using the SimXpert
MD Explicit workspace. If you are already familiar with explicit methods and how they differ from
implicit methods, you may disregard this section.
Method of Solution
Although MD Nastran Explicit Nonlinear uses the Explicit methods, a brief overview of both the Implicit
and the Explicit Methods for the solution of dynamic response calculations is given below.
Implicit Methods
Most finite element programs use implicit methods to carry out a transient solution. Normally, they use
Newmark schemes to integrate in time. If the current time step is step n , a good estimate of the
acceleration at the end of step n + 1 will satisfy the following equation of motion:
ext
Ma' n + 1 + Cv' n + 1 + Kd' n + 1 = F n + 1
where:
2 2
d'n + 1 = d n + v n Δt + ( ( 1 – 2β )a n Δt ) ⁄ 2 + βa'n + 1 Δt
v' n + 1 = v n + ( 1 – γ )a n Δt + γa'n + 1 Δt
or
2
d'n + 1 = d *n + βa'n + 1 Δt
v' n + 1 = v n* + γa'n + 1 Δt
The terms d n* and v n* are predictive and are based on values already calculated.
2 ext
Ma' n + 1 + C ( v* n + γa' n + 1 Δt ) + K ( d* n + βa' n + 1 Δt ) = F n + 1
or
2 ext
M + CγΔt + KβΔt ]a'n + 1 = F n + 1 – Cvn* – Kd
residual
M*a' n + 1 = F n + 1
This is analogous to decomposing the stiffness matrix in a linear static analysis. However, in dynamics,
mass and damping terms are also present.
Explicit Methods
The equation of motion
ext
Ma n + Cv n + Kd n = F n
can be rewritten as
ext int
Ma n = F n – F n
–1 residual
an = M Fn
where
ext
Fn = vector of externally applied loads
int
Fn = vector of internal loads (e.g., forces generated by the elements and hourglass forces)
= Cv n + Kd n
M = mass matrix
The acceleration can be found by inverting the mass matrix and multiplying it by the residual load vector.
In LS_DYNA, like any explicit finite element code, the mass matrix is lumped which results in a diagonal
mass matrix.
Since M is diagonal, its inversion is trivial, and the matrix equation is a set of independent equations for
each degree of freedom, as follows:
residual
a ni = F ni ⁄ Mi
The position, forces, and accelerations are defined at time level n , while the velocities are defined at time
level n + 1 ⁄ 2 . Graphically, this can be depicted as:
Introduction 5
Overview and Definition
v n + 1 ⁄ 2 = v n – 1 ⁄ 2 + a n ( Δt n + 1 ⁄ 2 + Δt n – 1 ⁄ 2 ) ⁄ 2
d n + 1 = d n + v n + 1 ⁄ 2 Δt n + 1 ⁄ 2
The Leap-frog scheme results in a central difference approximation for the acceleration, and is second-
order accurate in Δt .
Explicit methods with a lumped mass matrix do not require matrix decompositions or matrix solutions.
Instead, the loop is carried out for each time step as shown in the following diagram:
Grid-Point Accelerations
Element Stresses
in the mesh. Typically, explicit time steps are 100 to 1000 times smaller than those used with implicit
codes. However, since each iteration does not involve the costly formulation and decomposition of
matrices, explicit techniques are very competitive with implicit methods.
Because the smallest element in an explicit solution determines the time step, it is extremely important
to avoid very small elements in the mesh.
Courant Criterion
Since it is impossible to do a complete eigenvalue analysis every cycle to calculate the timestep, an
approximate method, known as the Courant Criterion, is used. This is based on the minimum time which
is required for a stress wave to cross each element:
Δt = SL/c
where
Δt = Timestep
c = E⁄ρ
where,
E = Young’s modulus
ρ = density
Explicit solutions have a greater advantage over implicit solutions if the time step of the implicit solution
has to be small for some reason. This may be necessary for problems that include:
Introduction 7
Overview and Definition
• Material nonlinearity. A high degree of material nonlinearity may require a small time step
for accuracy.
• Large geometric nonlinearity. Contact and friction algorithms can introduce potential
instabilities, and a small time step may be needed for accuracy and stability.
• Those analyses where the physics of the problem demands a small time step (e.g. stress wave
effects as in crash, crush, and impact analyses).
• Material and geometric nonlinearity in combination with large displacements. Convergence
in implicit methods becomes more difficult to achieve as the amount of nonlinearity for all
types increases.
Explicit methods have increasing advantages over implicit methods as the model gets bigger, and
bigger.
Typical Applications
Some of the typical structural applications which are well suited for the SOL 700 explicit analysis are:
Units
SimXpert interprets all dimensions and input data with respect to a system of units. It is important to set
the appropriate units prior to importing any unitless analysis files (such as a Nastran Bulk Data file) or
creating materials, properties, or loads. You can control the system of units by selecting Units Manager
from the Tools -> Options->Options Editor menu. If you import a file that contains units, SimXpert
will convert them into those specified in the Units Manager.
Creating geometry
In the first release SimXpert has very limited geometry creation capabilities. It is possible to create curves
and very simple surfaces. For the most part you will be importing geometry from an external source. The
imported geometry can be edited in SimXpert
Importing geometry
If the geometry of the part is available in a CATIA, parasolid, IGES, or STL file, it can be directly
imported into the SimXpert MD Explicit Workspace.
coordinate system with one of its axes normal to the inclined plane needs to be created and used to specify
the fixity (SPC) of the displacement component along the direction normal to the inclined plane.
CONSTRAINT
Local coordinate systems can be in cartesian, cylindrical or spherical systems. Coordinate system created
in SimXpert are represented by the following icons, corresponding to the method selected.
Spherical
Cylindrical
Cartesian
You can create local coordinate systems by selecting Cartesian, Cylindrical, or Spherical from the
Coordinate System group under the Geometry tab. There are numerous methods to create local
coordinate systems in SimXpert:
10
Parts and Geometry
1. 3 Points: Three points are used to define the coordinate system. The first point corresponds to the
location of origin. The second point defines the point on a specified axis and the third point
defines a point in a specified plane.
2. Euler: Creates a coordinate system through three specified rotations about the axes of an existing
coordinate system.
3. Normal: Creates a coordinate system with its origin at a point location on a surface. A specified
axis is normal to the surface.
4. Two Vectors: Creates a coordinate system with its origin at a designated location and two of the
coordinate frame axes are defined using vectors
5. Advanced: Location and orientation can be independently defined. There are 4 different ways to
define the location of the origin of the coordinate system: Geometry, Point/Node, Coordinate
System, and Center of Part. Further, the orientation can also be defined 3 ways: Global, Two
Axes, and Coordinate System.
Elements 11
Elements
12
Elements
Elements
The heart of a finite element program lies in its element library which allows you to model a structure
for analysis. MD Nastran has a very comprehensive element library which lets you model 1-D, 2-D, or
3-D structures. This section gives some basic definitions of the element types available in MD Nastran
Explicit Nonlinear. For a more detailed description of all the available elements in MD Nastran Explicit,
please refer to MD R2 Nastran Quick Reference Guide.
Element Types
Each element has five definitive characteristics that determine its behavior:
• Class
• Number of Nodes
• Interpolation
• Degrees of Freedom
• Integration Method
Class
The type of geometric domain that an element represents determines the class of the element. Listed
below are the classes of elements in the MD Nastran Explicit Nonlinear element library.
• Beam Elements - a 3-D bar with axial, bending, and torsional stiffness.
• Shell Elements - a curved, thin or thick structure with membrane/bending capabilities.
• Plate Elements - a flat thin structure carrying in-plane and out-of-plane loads.
• Continuum Elements - Solid elements used to model thick sections.
• Plane stress - a thin plate with in-plane stresses only. All normal and shear stresses associated
with the out-of-plane direction are assumed to be zero. (All plane strain elements lie in the global
x-y plane.)
• Plane strain - a region where there is no out-of-plane motion and the normal and transverse
strains are zero.
• Generalized plane strain is the same as plane strain except that the normal z-strain can be a
prescribed constant or function of x and y.
• Axisymmetric elements are describe in 2-D, but represent a full 3-D structure where the
geometry and loading are both axisymmetric.
• 3-D solid - a solid structure with only translational degrees of freedom for each node (linear or
quadratic interpolation functions).
• Truss Elements - a 3-D rod with axial stiffness only (no bending).
• Membrane Elements - a thin sheet with in-plane stiffness only (no bending resistance).
• Concentrated mass/Springs/Damper Elements -
• Rigid Constraints -
Elements 13
Elements
Number of Nodes
The number of nodes for an element define where the displacements are calculated in the analysis.
Elements with only corner nodes are classified as first order elements and the calculation of
displacements at locations within the element are made by linear interpolation. Elements that contain
midside nodes are second order elements and quadratic interpolations are made for calculating
displacements. MD Nastran SOL 700 do not support second order elements.
In MD Nastran the number of nodes is designated at the end of the element name. For example, a
CQUAD4 has 4 nodes.
Interpolation
Interpolation (shape) function is an assumed function relating the displacements at a point inside an
element to the displacements at the nodes of an element. In MD Nastran, three types of shape functions
are used: linear, quadratic, and cubic.
Degrees of Freedom
Degrees of freedom is the number of unknowns at a node. In the general case, there are six degrees of
freedom at a node in structural analysis (three translations, three rotations). For example there are three
(translations) for 3-D truss element; six (three translations, three rotations) for a 3-D beam element and
only three translations for the 3-D solid elements.
Integration
Numerical integration is a method used for evaluating integrals over an element. Element quantities –
such as stresses, strains, and temperatures – are calculated at each integration point of the element. Full
integration (quadrature) requires, for every element, 2-D integration points for linear interpolation and
3-D points for quadratic interpolation, where scalar “d” is the number of geometric dimensions of an
element (that is, d = 2 for a quad; d = 3 for a hexahedron). Reduced integration uses a lower number of
integration points than necessary to integrate exactly. For example, for an 8-node quadrilateral, the
number of integration points is reduced from 9 to 4 and, for a 20-node hexahedron, from 27 to 8. For
some elements, an “hourglass” control method is used to insure an accurate solution.
Table 0-1 summarizes the elements available in MD Nastran SOL 700. This section describes the
characteristics of the elements. For shell element CTRIA3 and CQUAD4 shell elements,
PARAM,DYNSHELLFORM is used to control the behavior.
Preliminaries
Consider the body shown in Figure 0-1. We are interested in time-dependent deformation in which a
point in b initially at X α ( α = 1, 2, 3 ) in a fixed rectangular Cartesian coordinate system moves to a
point x i ( i = 1, 2, 3 ) in the same coordinate system. Since a Lagrangian formulation is considered,
the deformation can be expressed in terms of the convected coordinates X α , and time t
x i = x i ( X α, t ) (0-1)
x i ( X α, 0 ) = X α (0-2)
x· i ( X α, 0 ) = V i ( X α ) (0-3)
Governing Equations
We seek a solution to the momentum equation
σ ij n j = t i ( t ) (0-5)
x i ( X α, t ) = D i ( t ) (0-6)
-
( σ ij+ – σ ij )n i = 0 (0-7)
Elements 17
Preliminaries
along an interior boundary ∂b 3 when X i+ = x i- . Here σ ij is the Cauchy stress, ρ is the current
density, f is the body force density, x·· is acceleration, the comma denotes covariant differentiation, and
n j is a unit outward normal to a boundary element of ∂b .
x X
3 3
x X
2 2
∂
t = 0 b
b
∂B
B0
x1 X1
ρV = ρ 0 (0-8)
where V is the relative volume, i.e., the determinant of the deformation gradient matrix, F ij ,
∂x i
F ij = -------- (0-9)
∂X j
·
E· = Vs ij ε ij – ( p + q )V· (0-10)
18
Preliminaries
is integrated in time and is used for equation of state evaluations and a global energy balance. In Equation
(0-10), s ij and p represent the deviatoric stresses and pressure,
s ij = σ ij + ( p + q )δ ij (0-11)
1 1
p = – --- σ ij δ ij – q = – --- σ kk – q (0-12)
3 3
respectively,q is the bulk viscosity, δ ij is the Kronecker delta ( δ ij = 1 if i = j ; otherwise
·
δ ij = 0 ) and ε ij is the strain rate tensor. The strain rates and bulk viscosity are discussed later.
We can write:
-
+ ( σ ij+ – σ ij )n j δx i ds = 0
∂b 3
where δx i satisfies all boundary conditions on ∂b 2 , and the integrations are over the current geometry.
Application of the divergence theorem gives
x i ( X α ,t ) = x i ( X α ( ξ ,η ,ζ ) ,t ) = φ j ( ξ ,η ,ζ )xi ( t )
j
(0-17)
j=1
where φ j are shape (interpolation) functions of the parametric coordinates ( ξ ,η ,ζ ) , k is the number of
nodal points defining the element, and x ij is the nodal coordinate of the jth node in the ith direction.
δπ = δπ m = 0 (0-18)
m=1
and write
n
ρx··i Φi dυ + σij Φi,j dυ – ρfi Φi dυ – ti Φi ds
m m m m m
= 0 (0-19)
m = 1 υm υm υm ∂b 1
where
Φ im = ( φ1 ,φ 2 ,… ,φ k ) im (0-20)
n
m
υ
ρN t Na dυ +
B t σ dυ –
ρN t b dυ –
N t t ds
= 0
(0-21)
m=1 m υ m υ m∂b 1
a x1
x··1 a y1
x··2 = N = Na
…
x··3 a yk
a zk
(0-23)
b is the body force load vector, and t are applied traction loads.
fx tx
b = fy t = ty
fz tz
(0-24)
Elements 21
CHEXA Solid Elements
x i ( X α, t ) = x i ( X α ( ξ ,η ,ζ ), t ) = φj ( ξ ,η ,ζ )xi ( t )
j
(0-25)
j=1
1
φ j = --- ( 1 + ξξ j ) ( 1 + ηη j ) ( 1 + ζζ j ) (0-26)
8
where ξ j ,η j ,ζ j take on their nodal values of ( ± 1, ± 1, ± 1 ) and x ij is the nodal coordinate of the j th
node in the i th direction (see Figure 0-2).
ζ
5
8
6
Node ξ η ζ
7 1 -1 -1 -1
2 1 -1 -1
3 1 1 -1
4 -1 1 -1
5 -1 -1 1
η 6 1 -1 1
4 7 1 1 1
ξ 1
8 -1 1 1
3
Figure 0-2 Eight-node Solid Hexahedron Element
φ1 0 0 φ2 0 … 0 0
N ( ξ ,η ,ζ ) = 0 φ1 0 0 φ2 … φ8 0 (0-27)
0 0 φ 1 0 0 … 0 φ8
∂
----- 0 0
∂x
∂
0 ----- 0
∂y
∂
0 0 -----
B = ∂z N (0-29)
∂ -----
∂
----- 0
∂y ∂x
∂ ∂
0 ----- -----
∂z ∂y
∂ ∂
----- 0 -----
∂z ∂x
In order to achieve a diagonal mass matrix the rows are summed giving the k th diagonal term as
∂φ i ∂φ ∂x ∂φ i ∂y ∂φ i ∂z
------- = -------i -----
- + ------- ------ + ------- ------
∂ξ ∂x ∂ξ ∂y ∂ξ ∂z ∂ξ
∂φ ∂φ ∂x ∂φ i ∂y ∂φ i ∂z
-------i = -------i ------ + ------- ------ + ------- ------
∂η ∂x ∂η ∂y ∂η ∂z ∂η
∂φ i ∂φ ∂x ∂φ i ∂y ∂φ i ∂z
------- = -------i -----
- + ------- ------ + ------- ------
∂ζ ∂x ∂ζ ∂y ∂ζ ∂z ∂ζ (0-31)
which can be rewritten as
Inverting the Jacobian matrix, J , we can solve for the desired terms
∂φ i ∂φ i
------- -------
∂x ∂ξ
∂φ i = J – 1 ∂φ i (0-33)
------- -------
∂y ∂η
∂φ i ∂φ i
------- -------
∂z ∂ζ
Volume Integration
Volume integration is carried out with Gaussian quadrature. If g is some function defined over the
volume, and n is the number of integration points, then
is approximated by
24
CHEXA Solid Elements
n n n
g jkl = g ( ξ j, η k, ζ l ) (0-36)
n = 1
wi = wj = w k = 2 (0-37)
ξ1 = η1 = ζ1 = 0
and we can write
g dv = 8g ( 0 ,0 ,0 ) J ( 0 ,0 ,0 ) (0-38)
∂φ 1 ∂φ ∂φ ∂φ
-------- = – --------7 --------3 = – --------5
∂x i ∂x i ∂x i ∂x i
(0-39)
∂φ ∂φ ∂φ ∂φ
--------2 = – --------8 --------4 = – --------6
∂x i ∂x i ∂x i ∂x i
at ξ = η = ζ = 0 reduces the amount of effort required to compute this matrix by more than 25
times over an 8-point integration. This cost savings extends to strain and element nodal force calculations
where the number of multiplies is reduced by a factor of 16. Because only one constitutive evaluation is
needed, the time spent determining stresses is reduced by a factor of eight. Operation counts for the
Elements 25
CHEXA Solid Elements
constant stress hexahedron are given in Table 0-2. Included are counts for the Flanagan and Belytschko
[1981] hexahedron and the hexahedron used by Wilkins [1974] in his integral finite difference method,
which was also implemented [Hallquist 1979].
.
Flanagan-Wilkins
It may be noted that 8-point integration has another disadvantage in addition to cost. Fully integrated
elements used in the solution of plasticity problems and other problems where Poisson’s ratio
approaches .5 lock up in the constant volume bending modes. To preclude locking, an average pressure
must be used over the elements; consequently, the zero energy modes are resisted by the deviatoric
stresses. If the deviatoric stresses are insignificant relative to the pressure or, even worse, if material
failure causes loss of this stress state component, hourglassing will still occur, but with no means of
resisting it. Sometimes, however, the cost of the fully integrated element may be justified by increased
reliability and if used sparingly may actually increase the overall speed.
Hourglass Control
The biggest disadvantage to one-point integration is the need to control the zero energy modes, which
arise, called hourglassing modes. Undesirable hourglass modes tend to have periods that are typically
much shorter than the periods of the structural response, and they are often observed to be oscillatory.
However, hourglass modes that have periods that are comparable to the structural response periods may
be a stable kinematic component of the global deformation modes and must be admissible. One way of
resisting undesirable hourglassing is with a viscous damping or small elastic stiffness capable of stopping
the formation of the anomalous modes but having a negligible affect on the stable global modes. Two of
the early three-dimensional algorithms for controlling the hourglass modes were developed by Kosloff
and Frazier [1974] and Wilkins et al. [1974].
Since the hourglass deformation modes are orthogonal to the strain calculations, work done by the
hourglass resistance is neglected in the energy equation. This may lead to a slight loss of energy;
however, hourglass control is always recommended for the under integrated solid elements. The energy
26
CHEXA Solid Elements
dissipated by the hourglass forces reacting against the formations of the hourglass modes is tracked and
reported in the output files MATSUM and GLSTAT. This is only provided if the PARM,DYENERGYGE
is included.
It is easy to understand the reasons for the formation of the hourglass modes. Consider the following
strain rate calculations for the 8-node solid element
8 ∂φ
· 1--- k · k ∂φ k · k
ε ij = -------
- x + -------
- x (0-1)
2 ∂x i j ∂x j i
k = 1
·
ε ij = 0 (0-3)
due to the asymmetries in (0-39). It is easy to prove the orthogonality of the hourglass shape vectors,
which are listed in Table 0-3 and shown in Figure 0-2 with the derivatives of the shape functions:
8
∂φ k
-------- Γ αk = 0
∂x i
i = 1 ,2 ,3 α = 1 ,2 ,3 ,4 (0-4)
k=1
α = 1 α = 2 α = 3 α = 4
Γ j1 1 1 1 1
Γ j2 -1 1 -1 -1
Γ j3 1 -1 -1
Γ j4 -1 -1 1 -1
Γ j5 1 -1 -1 -1
Γ j6 -1 -1 1 1
Elements 27
CHEXA Solid Elements
α = 1 α = 2 α = 3 α = 4
Γ j7 1 1 1 -1
Γ j8 -1 1 -1 1
The hourglass modes of an 8-node element with one integration point are shown [Flanagan and
Belytschko 1981] (Figure 0-3). A total of twelve modes exist.
Γ1k Γ2k
Γ3k Γ4k
Figure 0-3 Hourglass Modes
x· i Γαk
k
h iα = = 0 (0-1)
k=1
k
are nonzero if hourglass modes are present. The 12 hourglass-resisting force vectors, f iα are
k
f iα = a h h iα Γ αk (0-2)
where
28
CHEXA Solid Elements
c
a h = Q hg ρν e2 ⁄ 3 --- (0-3)
4
in which ν e is the element volume, c is the material sound speed, and Q hg is a user-defined constant
usually set to a value between .05 and .15. This is defined by DYHRGQH PARAM. The hourglass
resisting forces of (0-2) are not orthogonal to rigid body rotations; however, the approach of Flanagan
and Belytschko [1981] is orthogonal. This is controlled by using the DYHRGIHQ PARAM.
Material subroutines add as little as 60 operations for the bilinear elastic-plastic routine to ten times as
much for multi-surface plasticity and reactive flow models. Unvectorized material models will increase
that share of the cost a factor of four or more.
Instead of resisting components of the bilinear velocity field that are orthogonal to the strain calculation,
Flanagan and Belytschko resist components of the velocity field that are not part of a fully linear field.
They call this field, defined below, the hourglass velocity field
x· ik = x· i – x· ik
HG LIN
(0-4)
where
· ·
x· ik = x i + x i ,j ( x jk – x j )
LIN
(0-5)
8 8
1 · 1
x ik x· i
k
x i = --- x i = --- (0-6)
8 8
k=1 k=1
Flanagan and Belytschko construct geometry-dependent hourglass shape vectors that are orthogonal to
the fully linear velocity field and the rigid body field. With these vectors they resist the hourglass velocity
deformations. Defining hourglass shape vectors in terms of the base vectors as
γ ak = Γ ak – φ k ,i x i Γan
n
(0-7)
n=1
Elements 29
CHEXA Solid Elements
and setting
x· i γαk
k
g ia = = 0 (0-8)
k=1
k
f iα = a h g iα γ αk (0-9)
The hourglass forces given by (0-2) and (0-9) are identical if the, hexahedron element is a parallelepiped.
The default hourglass control method for solid element is given by (0-2); however, we recommend the
Flanagan-Belytschko approach for problems that have large rigid body rotations since the default
approach is not orthogonal to rigid body rotations.
A cost comparison in Table 0-2 shows that the default hourglass viscosity requires approximately 130
adds or multiplies per hexahedron, compared to 620 and 680 for the algorithms of Flanagan-Belytschko
and Wilkins.
Hourglass stabilization for the 3-D hexahedral element is available. Based on material properties and
element geometry, this stiffness type stabilization is developed by an assumed strain method [Belytschko
and Bindeman 1993] such that the element does not lock with nearly incompressible material. This is
activated by using the PARAM,DYHRGIHG,6. When the user-defined hourglass constant Q hg is set to
1.0, accurate coarse mesh bending stiffness is obtained for elastic material. For nonlinear material, a
smaller value of Q hg is suggested and the default value is set to 0.1.
where φ modifies the normal strains to ensure that the total volumetric strain increment at each
integration point is identical
vn + 1 ⁄ 2
Δu, Δv , and Δw are displacement increments in the x, y , and z directions, respectively, and
(xn + xn + 1 )
x n + 1 ⁄ 2 = ---------------------------- (0-13)
2
(yn + yn + 1 )
y n + 1 ⁄ 2 = ---------------------------- (0-14)
2
( zn + zn + 1)
z n + 1 ⁄ 2 = ---------------------------- (0-15)
2
Elements 31
CHEXA Solid Elements
To satisfy the condition that rigid body rotations cause zero straining, it is necessary to use the geometry
at the mid-step in the evaluation of the strain increments. The explicity formulation uses the geometry at
step n + 1 to save operations; however, for calculations, which involve rotating parts, the mid-step
geometry should be used especially if the number of revolutions is large.
Since the bulk modulus is constant in the plastic and viscoelastic material models, constant pressure solid
elements result. In the thermoelastoplastic material, a constant temperature is assumed over the element.
In the soil and crushable foam material, an average relative volume is computed for the element at time
step n + 1 , and the pressure and bulk modulus associated with this relative volume is used at each
integration point. For equations of state, one pressure evaluation is done per element and is added to the
deviatoric stress tensor at each integration point.
The foregoing procedure requires that the strain-displacement matrix corresponding to (0-10) and
consistent with a constant volumetric strain, B , be used in the nodal force calculations [Hughes 1980].
It is easy to show that:
B n + 1 σ n + 1 dv n + 1 = B n + 1 σ n + 1 dv n + 1
t t
F = (0-16)
vn + 1 vn + 1
N 1 ( r, s, t ) = r
N 2 ( r, s, t ) = s
(0-17)
N 3 ( r, s, t ) = 1–r–s–
N 4 ( r, s, t ) = t
1
2 r
If a tetrahedron element is needed, this element should be used instead of the collapsed solid element
since it is, in general, considerably more stable in addition to being much faster.
1
N 1 ( r, s, t ) = --- ( 1 – t )r
2
1
N 2 ( r, s, t ) = --- ( 1 – t ) ( 1 – r –
2
1
N 3 ( r, s, t ) = --- ( 1 + t ) ( 1 – r –
2 (0-18)
1
N 4 ( r, s, t ) = --- ( 1 + t )r
2
1
N 5 ( r, s, t ) = --- ( 1 – t )s
2
1
N 6 ( r, s, t ) = --- ( 1 + t )s
2
If a pentahedron element is needed, this element should be used instead of the collapsed solid element
since it is, in general, more stable and significantly faster. Selective-reduced integration is used to prevent
volumetric locking; i.e., a constant pressure over the domain of the element is assumed.
4
6
r
2
s
1
5
Figure 0-5 Six Node Pentahedron
34
CBEAM - Belytschko Beam
Co-rotational Technique
In any large displacement formulation, the goal is to separate the deformation displacements from the
rigid body displacements, as only the deformation displacements give rise to strains and the associated
generation of strain energy. This separation is usually accomplished by comparing the current
configuration with a reference configuration.
The current configuration is a complete description of the deformed body in its current spatial location
and orientation, giving locations of all points (nodes) comprising the body. The reference configuration
can be either the initial configuration of the body (i.e., nodal locations at time zero) or the configuration
of the body at some other state (time). Often, the reference configuration is chosen to be the previous
configuration, say at time t n = t n + 1 – Δt .
The choice of the reference configuration determines the type of deformations that will be computed:
total deformations result from comparing the current configuration with the initial configuration, while
incremental deformations result from comparing with the previous configuration. In most time stepping
(numerical) Lagrangian formulations, incremental deformations are used because they result in
significant simplifications of other algorithms, chiefly constitutive models.
A direct comparison of the current configuration with the reference configuration does not result in a
determination of the deformation, but rather provides the total (or incremental) displacements. We will
use the unqualified term displacements to mean either the total displacements or the incremental
displacements, depending on the choice of the reference configuration as the initial or the last state. This
is perhaps most obvious if the reference configuration is the initial configuration. The direct comparison
of the current configuration with the reference configuration yields displacements, which contain
components due to deformations and rigid body motions. The task remains of separating the deformation
and rigid body displacements. The deformations are usually found by subtracting from the displacements
an estimate of the rigid body displacements. Exact rigid body displacements are usually only known for
trivial cases where they are prescribed a priori as part of a displacement field. The co-rotational
formulations provide one such estimate of the rigid body displacements.
The co-rotational formulation uses two types of coordinate systems: one system associated with each
element; i.e., element coordinates which deform with the element, and another associated with each node;
i.e., body coordinates embedded in the nodes. (The term ‘body’ is used to avoid possible confusion from
referring to these coordinates as ‘nodal’ coordinates. Also, in the more general formulation presented in
Elements 35
CBEAM - Belytschko Beam
[Belytschko et al., 1977], the nodes could optionally be attached to rigid bodies. Thus, the term ‘body
coordinates’ refers to a system of coordinates in a rigid body, of which a node is a special case.) These
two coordinate systems are shown in the upper portion of Figure 0-6.
Y ^
Y
b2
0
e2
0
e1
^
b1 X
I J
X
(a) Initial Configuration
^X
J
Y
^Y
b2
e1
e2
X
I
b1
(b) Rigid Rotation Configuration
^
Y
Y
b2 ^
X
e2
e1 J
0
e1
b1
X
(c) Deformed Configuration
Figure 0-6 Co-rotational Coordinate System
36
CBEAM - Belytschko Beam
The element coordinate system is defined to have the local x-axis x̂ originating at node I and
terminating at node J ; the local y-axis ŷ and, in three dimension, the local z-axis ẑ , are constructed
normal to x̂ . The element coordinate system ( x̂, ŷ, ẑ ) and associated unit vector triad ( e 1, e 2, e 3 ) are
updated at every time step by the same technique used to construct the initial system; thus the unit vector
e 1 deforms with the element since it always points from node I to node J .
The embedded body coordinate system is initially oriented along the principal inertial axes; either the
assembled nodal mass or associated rigid body inertial tensor is used in determining the inertial principal
values and directions. Although the initial orientation of the body axes is arbitrary, the selection of a
principal inertia coordinate system simplifies the rotational equations of motion; i.e., no inertial cross
product terms are present in the rotational equations of motion. Because the body coordinates are fixed
in the node (or rigid body), they rotate and translate with the node and are updated by integrating the
rotational equations of motion, as will be described subsequently.
The unit vectors of the two coordinate systems define rotational transformations between the global
coordinate system and each respective coordinate system. These transformations operate on vectors with
global components A = ( A x, A y, A z ) , body coordinates components A = A x, A y, Az , and
ˆ ˆ ˆ ˆ
element coordinate components A = A , A , A which are defined as:
x y z
Ax b 1x b 2x b 3x Ax
A = A y = b 1y b 2y b 3y Ay = [ λ ] { A } (0-19)
Az b 1z b 2z b 3z Az
where b ix, b iy, b iz are the global components of the body coordinate unit vectors. Similarly for the
element coordinate system:
ˆ
Ax e 1x e 2x e 3x Ax
ˆ
A = A y = e 1y e 2y e 3y ˆ
Ay = [μ]{A} (0-20)
Az e 1z e 2z e 3z ˆ
Az
where e ix, e iy, e iz are the global components of the element coordinate unit vectors. The inverse
transformations are defined by the matrix transpose: i.e.,
Elements 37
CBEAM - Belytschko Beam
T
{A} = [λ] {A} (0-21)
ˆ T
{A} = [μ] {A} (0-22)
The following two examples illustrate how the element and body coordinate system are used to separate
the deformations and rigid body displacements from the displacements.
Rigid Rotation
First, consider a rigid body rotation of the beam element about node I , as shown in the center of
Figure 0-6b; i.e., consider node I to be a pinned connection. Because the beam does not deform during
the rigid rotation, the orientation of the unit vector e 1 in the initial and rotated configuration will be the
same with respect to the body coordinates. If the body coordinate components of the initial element unit
vector e 10 were stored, they would be identical to the body coordinate components of the current element
unit vector e 1 .
Deformation Rotation
Next, consider node I to be constrained against rotation; i.e., a clamped connection. Now node J is
moved, as shown in the lower portion of Figure 0-6, causing the beam element to deform. The updated
element unit vector e 1 is constructed and its body coordinate components are compared to the body
coordinate components of the original element unit vector e 10 . Because the body coordinate system did
not rotate, as node I was constrained, the original element unit vector and the current element unit vector
are not co-linear. Indeed, the angle between these two unit vectors is the amount of rotational deformation
at node I ; i.e.,
e 1 × e 10 = θ e 3 (0-23)
Thus the co-rotational formulation separates the deformation and rigid body deformations by using:
• a coordinate system that deforms with the element; i.e., the element coordinates or
• a coordinate system that rigidly rotates with the nodes; i.e., the body coordinates.
Then, it compares the current orientation of the element coordinate system with the initial element
coordinate system, using the rigidly rotated body coordinate system, to determine the deformations.
ˆT ˆ ˆ ˆ ˆ ˆ
d = { δ IJ, θ xJI, θ yI, θ yJ, θ zI, θ zJ } (0-24)
where
δ IJ = length change
ˆ
θ xJI = torsional deformation
ˆ ˆ ˆ ˆ
θ yI, θ yJ, θ zI, θzJ = bending rotational deformations
The superscript ˆ emphasizes that these quantities are defined in the local element coordinate system,
and I and J are the nodes at the ends of the beam.
The beam deformations, defined in (0-19), are the usual small displacement beam deformations (see, for
example, [Przemieniecki 1986]). Indeed, one advantage of the co-rotational formulation is the ease with
which existing small displacement element formulations can be adapted to a large displacement
formulation having small deformations in the element system. Small deformation theories can be easily
accommodated because the definition of the local element coordinate system is independent of rigid body
rotations and hence deformation displacement can be defined directly.
Calculation of Deformations
The elongation of the beam is calculated directly from the original nodal coordinates ( XI, Y I, Z I ) and
the total displacements ( uxI, u yI, u zI ) :
1
δ IJ = -----------o- [ 2 ( X JI u xJI + Y JI u yJI + Z JI u zJI ) + u xJI
2 2
+ u yJI 2
+ u zJI ] (0-25)
l+l
where
X JI = X J – X I (0-26)
The deformation rotations are calculated using the body coordinate components of the original element
coordinate unit vector along the beam axis; i.e., e 10 , as outlined in the previous section. Because the body
coordinate components of initial unit vector e 10 rotate with the node, in the deformed configuration it
Elements 39
CBEAM - Belytschko Beam
indicates the direction of the beam’s axis if no deformations had occurred. Thus comparing the initial unit
vector e 10 with its current orientation e 1 indicates the magnitude of deformation rotations. Forming the
ˆ ˆ
e 1 × e 10 = θ y e 2 + θ z e 3 (0-28)
where
ˆ
θ y is the incremental deformation about the local ŷ axis
ˆ
θ z is the incremental deformation about the local ẑ axis
The calculation is most conveniently performed by transforming the body components of the initial
element vector into the current element coordinate system:
0 0
ê 1x e 1x
ê 1y
0 = [ μ ] T [ λ ] e 1y
0 (0-29)
ê 1z
0 e 1z
0
Substituting the above into (0-28)
e1 e2 e3
ˆ ˆ
e1 × e 10 = det 1 0 0
0
= – ê 1z 0
e 2 + ê 1y e3 = θy e2 + θz e3 (0-30)
0 0 0
ê 1x ê 1y ê 1z
Thus,
θˆy = – ê 1z
0
(0-31)
θˆz = ê 1y
0
(0-32)
40
CBEAM - Belytschko Beam
The torsional deformation rotation is calculated from the vector cross product of initial unit vectors, from
0 0 0
each node of the beam, that were normal to the axis of the beam, i.e., ê 2I and ê 2J ; note that ê 3I and
0
ê 3J could also be used. The result from this vector cross product is then projected onto the current axis
of the beam; i.e.,
e1 e2 e3
ˆ
θ xJI = e 1 ⋅ ( ê 2I
0
× ê 2J
0
) = e 1 det ê x2I
0 0
ê y2I 0
ê z2I
0
= ê y2I 0
ê z2J 0
– ê y2J 0
ê z2I (0-33)
0 0 0
ê x2J ê y2J ê z2J
0 0
Note that the body components of e 2I and e 2J are transformed into the current element coordinate
system before performing the indicated vector products.
• Functional forms relating the overall response of the beam- e.g., moment-curvature relations,
• Direct through-the-thickness integration of the stress.
Currently only the former method, as explained subsequently, is implemented; the direct integration
method is detailed in [Belytschko et al., 1977].
Axial Force
The internal axial force is calculated from the elongation of the beam δ , as given by (0-25), and an
axial stiffness:
ˆ
f xJ = K a δ (0-34)
where:
E is Young’s Modulus
Bending Moments
The bending moments are related to the deformation rotations by
ˆ
m̂ yI K yb 4 + φ y 2 – φ y θyI
= -------------
-
1 + φy 2 – φ 4 + φ ˆ (0-35)
m̂ yJ y y θ yJ
K z 4 + φ z 2 – φ z θˆ zI
b
m̂ zI
= -------------
1 + φ z 2 – φ 4 + φ θˆ
- (0-36)
m̂ zJ z z zJ
where (0-35) is for bending in the x̂ – ẑ plane and (0-36) is for bending in the x̂ – ŷ plane. The bending
constants are given by
b EI yy
K y = ---------
- (0-37)
l0
b EI zz
K z = ---------
- (0-38)
l0
12EI yy 12EI zz
φ y = ---------------- φ z = ---------------2- (0-41)
GA s l 2 GA s l
Hence φ is the shear factor, G the shear modulus, and A s is the effective area in shear.
Torsional Moment
The torsional moment is calculated from the torsional deformation rotation as
ˆ
m̂ xJ = K t θ xJI (0-42)
where
42
CBEAM - Belytschko Beam
GJ-
K t = ------ (0-43)
l0
The above forces are conjugate to the deformation displacements given previously in (0-24); i.e.,
ˆT ˆ ˆ ˆ ˆ ˆ
d = { δ IJ, θ xJI, θ yI, θ yJ, θ zI, θ zJ } (0-45)
where
ˆ T ˆ int
{d} {f} = W (0-46)
ˆT ˆ
f = { f xJ, m̂ xJ, m̂ yI, m̂ yJ, m̂ zI, m̂ zJ } (0-47)
ˆ ˆ m̂ xI = – m̂ xJ
f xI = –f xJ
ˆ m̂ yI + m̂ yJ ˆ ˆ
f zI = – ----------------------
- f zI = – f zJ
l0 (0-48)
ˆ m̂ zI + m̂ zJ ˆ ˆ
f yJ = – ----------------------
- f yI = – f yJ
l0
Updating the Body Coordinate Unit Vectors
The body coordinate unit vectors are updated using the Newmark β -Method [Newmark 1959] with
β = 0 , which is almost identical to the central difference method [Belytschko 1974]. In particular, the
body component unit vectors are updated using the formula
j j
j+1 j dbi Δt 2 d 2 b i
bi = bi + Δt -------- + -------- ---------
- (0-49)
dt 2 dt 2
where the superscripts refer to the time step and the subscripts refer to the three unit vectors comprising
the body coordinate triad. The time derivatives in the above equation are replaced by their equivalent
forms from vector analysis:
j
db
--------i = ω × b i (0-50)
dt
Elements 43
CBEAM - Belytschko Beam
j
d2 bi
---------- = ω × ( ω × b i ) + ( α i × b i ) (0-51)
dt 2
where ω and α are vectors of angular velocity and acceleration, respectively, obtained from the
rotational equations of motion. With the above relations substituted into (0-49), the update formula for
the unit vectors becomes
j+1 j Δt 2
bi = bi + Δt ( ω × b i ) + -------- { [ ω × ( ω × b i ) + ( α i × b i ) ] } (0-52)
2
To obtain the formulation for the updated components of the unit vectors, the body coordinate system is
temporarily considered to be fixed and then the dot product of (0-52) is formed with the unit vector to be
updated. For example, to update the x component of b 3 , the dot product of (0-52), with i = 3 , is
formed with b 1 , which can be simplified to the relation
j+1 j j+1 j Δt 2 j j j
b x3 = b1 ⋅ b y3 = Δtω y + -------- ( ω x ω z + αy ) (0-53)
2
Similarly,
j+1 j j+1 j Δt 2 j j j
b y3 = b2 ⋅ b 3 = Δtω x + -------- ( ω y ω z + αx ) (0-54)
2
j+1 j j+1 j Δt 2 j j j
b x2 = b1 ⋅ b 2 = Δtω z + -------- ( ω x ω y + αz ) (0-55)
2
j+1 j+1
The remaining components b 3 and b 1 are found by using normality and orthogonality, where it
is assumed that the angular velocities ω are small during a time step so that the quadratic terms in the
j+1
update relations can be ignored. Since b 3 is a unit vector, normality provides the relation
j+1
Next, if it is assumed that b x1 ≈ 1 , orthogonality yields (0-57)
44
CBEAM - Belytschko Beam
j+1
The component b x1 is then found by enforcing normality:
The updated components of b 1 and b 3 are defined relative to the body coordinates at time step j . To
complete the update and define the transformation matrix, (0-19), at time stepj + 1 , the updated unit
vectors b 1 and b 3 are transformed to the global coordinate system, using (0-19) with [ λ ] defined at
step j , and their vector cross product is used to form b2
Elements 45
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
• It is incrementally objective (rigid body rotations do not generate strains), allowing for the
treatment of finite strains that occur in many practical applications.
• It is simple, which usually translates into computational efficiency and robustness.
• It is compatible with the brick elements, because the element is based on a degenerated brick
element formulation.
• It includes finite transverse shear strains. The added computations needed to retain this strain
component, compare to those for the assumption of no transverse shear strain, are insignificant.
Geometry
The Hughes-Liu beam element is based on a degeneration of the isoparametric 8-node solid element, an
approach originated by Ahmad et al., [1970]. Recall the solid element isoparametric mapping of the
biunit cube
x ( ξ ,η ,ζ ) = N a ( ξ ,η ,ζ )x a (0-60)
( 1 + ξa ξ ) ( 1 + ηa η ) ( 1 + ζa ζ )
N a ( ξ ,η ,ζ ) = ----------------------------------------------------------------------- (0-61)
8
where x is an arbitrary point in the element, ( ξ ,η ,ζ ) are the parametric coordinates, x a are the global
nodal coordinates of node a , and N a are the element shape functions evaluated at node a , i.e.,
ξ a ,η a ,ζ a are ( ξ ,η ,ζ ) evaluated at node a .
In the beam geometry, ξ determines the location along the axis of the beam and the coordinate pair
( η ,ζ ) defines a point on the cross section. To degenerate the 8-node brick geometry into the 2-node
beam geometry, the four nodes at ξ = – 1 and at ξ = 1 are combined into a single node with three
translational and three rotational degrees of freedom. Orthogonal, inextensible nodal fibers are defined
at each node for treating the rotational degrees of freedom. Figure 0-7 shows a schematic of the bi-unit
cube and the beam element. The mapping of the bi-unit cube into the beam element is separated into
three parts:
x ( ξ ,η ,ζ ) = x ( ξ ) + X ( ξ ,η ,ζ ) = x ( ξ ) + X ζ ( ξ ,ζ ) + X η ( ξ ,η ) (0-62)
46
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
ζ
Bi-unit Cube
ζ
x
η
ξ
η Beam Element
ξ
Nodal Fibers
+
Top Surface zζ
+1
+
Xζ
ζ Xζ
ˆ
Xζ
ˆ
0 X
η
Xζ
-1
Bottom Surface zζ
where x denotes a position vector to a point on the reference axis of the beam, and Xζ and X η are
position vectors at point x on the axis that define the fiber directions through that point. In particular,
x ( ξ ) = N a ( ξ )x a (0-63)
Elements 47
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
X η ( ξ, η ) = N a ( ξ )X ηa ( η ) (0-64)
X ζ ( ξ, ζ ) = N a ( ξ )X ζa ( ζ ) (0-65)
With this description, arbitrary points on the reference line x are interpolated by the one- dimensional
shape function N ( ξ ) operating on the global position of the two beam nodes that define the reference
axis, i.e., x a . Points off the reference axis are further interpolated by using a one-dimensional shape
function along the fiber directions; i.e., Xη a ( η ) and X ζa ( ζ ) where
X ηa ( η ) = z η ( η )X̂ ηa (0-66)
z η ( η ) = N + ( η )z ηa
+ + N ( η )z -
- ηa (0-67)
(1 + η)
N + ( η ) = ----------------- (0-68)
2
( 1 – η)
N - ( η ) = ----------------- (0-69)
2
X ζa ( ζ ) = z ζ ( ζ )X̂ ζa (0-70)
z ζ ( ζ ) = N + ( ζ )z ζa
+ + N ( ζ )z -
- ζa (0-71)
(1 + ζ)
N + ( ζ ) = ----------------- (0-72)
2
(1 – ζ)
N - ( ζ ) = ---------------- (0-73)
2
The Hughes-Liu beam formulation uses four position vectors, in addition to ξ , to locate the reference
+ and x - located on
axis and define the initial fiber directions. Consider the two position vectors x ζa ζa
the top and bottom surfaces, respectively, at node a . Then
1
x ζa = --- [ ( 1 – ζ )x ηa
– + ( 1 + ζ ) ]x +
ζa (0-74)
2
48
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
ˆX = (-------------------------
x ζa+ – x- )
ζa
ζa - (0-75)
+
x ζa – x ζa -
+ = 1
z ζa --- ( 1 – ζ ) ⋅ x ζa
+ – x- (0-76)
2 ζa
- = –1
z ζa --- ( 1 + ζ ) ⋅ x ζa
+ – x- (0-77)
2 ζa
1 – + ( 1 + ζ ) ]x +
x ηa = --- [ ( 1 – ζ )x ηa ηa (0-78)
2
ˆ ( x ηa+ – x- )
ηa
X ηa = --------------------------
+ -
- (0-79)
x ηa – x ηa
+ = 1
z ηa --- ( 1 – η ) ⋅ x ηa
+ – x- (0-80)
2 ηa
- = –1
z ηa --- ( 1 + η ) ⋅ x ηa
+ – x- (0-81)
2 ηa
.
where is the Euclidean norm. The reference surface may be located at the midsurface of the beam or
offset at the outer surfaces. This capability is useful in several practical situations involving contact
surfaces, connection of beam elements to solid elements, and offsetting elements such as for beam
stiffeners in stiffened shells. The reference surfaces are located within the beam element by specifying
the value of the parameters η and ζ , (see lower portion of Figure 0-7). When these parameters take on
the values – 1 or +1 , the reference axis is located on the outer surfaces of the beam. If they are set to
zero, the reference axis is at the center.
The same parametric representation used to describe the geometry of the beam elements is used to
interpolate the beam element displacements; i.e., an isoparametric representation. Again, the
displacements are separated into the reference axis displacements and rotations associated with the
fiber directions:
u ( ξ, η, ζ ) = u ( ξ ) + U ( ξ, η, ζ ) = u ( ξ ) + U ζ ( ξ, ζ ) + U η ( ξ, η ) (0-82)
u ( ξ ) = N a ( ξ )u a (0-83)
U η ( ξ, η ) = N a ( ξ )U ηa ( η ) (0-84)
Elements 49
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
U ζ ( ξ, ζ ) = N a ( ξ )U ζa ( ζ ) (0-85)
U ηa ( η ) = z ηa ( η )Û ηa (0-86)
U ζa ( ζ ) = z ζa ( ζ )Û ζa (0-87)
where u is the displacement of a generic point, u is the displacement of a point on the reference surface,
and U is the ‘fiber displacement’ rotations. The motion of the fibers can be interpreted as either
displacements or rotations as will be discussed.
Hughes and Liu introduced the notation that follows, and the associated schematic shown in Figure 0-8,
to describe the current deformed configuration with respect to the reference configuration:
y = y+Y (0-88)
y = x+u (0-89)
ya = xa + ua (0-90)
Y = X+U (0-91)
Ya = Xa + Ua (0-92)
ˆ ˆ ˆ
Y ηa = X ηa + U ηa (0-93)
Yˆ ζa = Xˆ ζa + U
ˆ
ζa (0-94)
u U
Parallel Construction
X
Reference axis in y
undeformed geometry
x
u Y
x
y
Deformed Configuration
Reference Surface
In the above relations, and in Figure 0-8, the x quantities refer to the reference configuration, the y
quantities refer to the updated (deformed) configuration and the u quantities are the displacements. The
notation consistently uses a superscript bar ( ) to indicate reference surface quantities, a superscript
caret ( ˆ ) to indicate unit vector quantities, lower case letter for translational displacements, and upper
case letters for fiber displacements. Thus to update to the deformed configuration, two vector quantities
are needed: the reference surface displacement u and the associated nodal fiber displacement U . The
nodal fiber displacements are defined in the fiber coordinate system, described in the next subsection.
ΔÛ η1 0 Yˆ η3 – Yˆ η2 Δθ 1
ΔÛ η2 = – Yˆ η3 0 Yˆ η1 Δθ 2 = h η Δθ (0-95)
ΔÛ η3 ˆ ˆ
Y η2 – Y η1 0 Δθ 3
ΔÛ ζ1 0 Yˆ ζ3 – Yˆ ζ2 Δθ 1
ΔÛ ζ2 = – Yˆ ζ3 0 Yˆ ζ1 Δθ 2 = h ζ Δθ (0-96)
ΔÛ ζ3 Yˆ ζ2 – Yˆ ζ1 0 Δθ 3
(0-95) and (0-96) are used to transform the incremental fiber tip displacements to rotational increments
in the equations of motion. The second-order accurate rotational update formulation due to Hughes and
Winget [1980] is used to update the fiber vectors:
Yˆ ηi = R ij ( Δθ )Ŷ ηj
n+1 n
(0-97)
Yˆ ζi = R ij ( Δθ )Ŷ ζi
n+1 n
(0-98)
then
Elements 51
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
ˆn+1 ˆn
ΔÛ ηa = Y ηa – Y ηa (0-99)
ˆn+1 ˆn
ΔÛ ζa = Y ζa – Y ζa (0-100)
where
( 2δ ik + ΔS ik )ΔS jk
R ij ( Δθ ) = δ ij + ------------------------------------------- (0-101)
2D
ΔS ij = e ikj Δθ k (0-102)
1
2D = 2 + --- ( Δθ 12 + Δθ 22 + Δθ32 ) (0-103)
2
y2 – y1
ê 1 = -------------------- (0-104)
y2 – y1
Yˆ η1 + Yˆ η2
e 2′ ---------------------------
= ˆ ˆ
- (0-105)
Y η1 + Y η2
From the vector cross product of these local tangents.
ê 3 = ê 1 × e 2′ (0-106)
ê 2 = ê 3 × ê 1 (0-107)
is defined. This coordinate system rigidly rotates with the deformations of the element.
The transformation of vectors from the global to the local coordinate system can now be defined in terms
of the basis vectors as
52
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
ˆ T
Ax e 1x e 2x e 3x A x
ˆ
A = ˆ
Ay = e 1y e 2y e 3y A y = [ q ] { A } (0-108)
ˆ
Az e 1z e 2z e 3z A z
ˆ
where e ix, e iy, e iz are the global components of the local coordinate unit vectors, A is a vector in the
local coordinates, and A is the same vector in the global coordinate system.
∂Δu i
G ij = ----------- (0-109)
∂y j
where Δu i are the incremental displacements and y j are the deformed coordinates. The incremental
strain and spin tensors are defined as the symmetric and skew-symmetric parts, respectively, of G ij :
1
Δε ij = --- ( G ij + G ji ) (0-110)
2
1
Δω ij = --- ( G ij – G ji ) (0-111)
2
The incremental spin tensor Δω ij is used as an approximation to the rotational contribution of the
Jaumann rate of the stress tensor. Here the Jaumann rate update is approximated as
σ ij = σ ijn + σ ip
n
Δω pj + σ jp
n
Δω pi (0-112)
where the superscripts on the stress tensor refer to the updated ( n + 1 ) and reference ( n )
configurations. This update of the stress tensor is applied before the constitutive evaluation, and the stress
and strain are stored in the global coordinate system.
Elements 53
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
Stress Update
To evaluate the constitutive relation, the stresses and strain increments are rotated from the global to the
local coordinate system using the transformation defined previously in (0-108), viz.
n
σ ijl = q ik σ kn q jn (0-113)
Δε ijl = q ik Δε kn q jn (0-114)
where the superscript l indicates components in the local coordinate system. The stress is
incrementally updated:
1
n+1 n n + ---
2
σ ijl = σ ijl + Δσ ijl (0-115)
n+1
σ ijn + 1 = qki σ kn
l
qnj (0-116)
f aint = Ba σ dυ
T
(0-117)
int
where f a are the internal forces at node a and B a is the strain-displacement matrix in the global
coordinate system associated with the displacements at node a . The B matrix relates six global strain
components to eighteen incremental displacements [three translational displacements per node and the
six incremental fiber tip displacements of (0-99)]. It is convenient to partition the B matrix:
B = [ B 1, B 2 ] (0-118)
Each B a sub matrix is further partitioned into a portion due to strain and spin with the following sub
matrix definitions:
54
CBEAM - DYSHELFORM = 1, Hughes-Liu Beam
B1 0 0 B4 0 0 B7 0 0
0 B2 0 0 B5 0 0 B8 0
0 0 B3 0 0 B6 0 0 B9
Ba = (0-119)
B2 B1 0 B5 B4 0 B8 B7 0
0 B3 B2 0 B6 B5 0 B9 B8
B3 0 B1 B6 0 B4 B9 0 B7
where
∂N a
N a, i = --------- for i = 1,2,3
∂y i
∂ ( N a z ηa )
Bi = ( N aηa ) ,( i – 3 ) = ---------------------- for i = 4,5,6 (0-120)
∂y i – 3
∂ ( N a z ηa )
( N a z ζa ) ,( i – 6 ) = ---------------------- for i = 7,8,9
∂y i – 6
• The derivative of the shape functions are taken with respect to the global coordinates;
• The B matrix is computed on the cross-section located at the mid-point of the axis;
• The resulting B matrix is a 6 × 18 matrix.
f ′ = T t f aint (0-121)
is assembled into the global right hand side internal force vector. T is defined as (also see (0-95)):
I 0
T = 0 hη (0-122)
0 hζ
CQUADA - DYSHELLFORM = 2,
Belytschko-Lin-Tsay Shell
The Belytschko-Lin-Tsay shell element ([Belytschko and Tsay 1981], [Belytschko et al., 1984a]) was
implemented as a computationally efficient alternative to the Hughes-Liu shell element. For a shell
element with five through thickness integration points, the Belytschko-Lin-Tsay shell elements requires
725 mathematical operations compared to 4050 operations for the under integrated Hughes-Liu element.
The selectively reduced integration formulation of the explicit Hughes-Liu element requires 35,350
mathematical operations. Because of its computational efficiency, the Belytschko-Lin-Tsay shell
element is usually the shell element formulation of choice. For this reason, it has become the default shell
element formulation for explicit calculations.
Co-rotational Coordinates
The midsurface of the quadrilateral shell element, or reference surface, is defined by the location of the
element’s four corner nodes. An embedded element coordinate system (see Figure 0-9) that deforms with
the element is defined in terms of these nodal coordinates. Then the procedure for constructing
the co-rotational coordinate system begins by calculating a unit vector normal to the main diagonal of the
element:
s3
ê 3 = --------- (0-123)
s3
2 2 2
s3 = s 31 + s 32 + s 33 (0-124)
s 3 = r 31 × r 42 (0-125)
where the superscript caret ( ˆ ) is used to indicate the local (element) coordinate system.
56
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
s3
yˆ
3
4
eˆ r31
r42 3
eˆ
2
1 s1
xˆ
eˆ
1
r21
2
s 1 = r 21 – ( r 21 ⋅ ê 3 )ê 3 (0-126)
s1
ê 1 = --------- (0-127)
s1
The remaining unit vector is obtained from the vector cross product
ê 2 = ê 3 × ê 1 (0-128)
If the four nodes of the element are coplanar, then the unit vectors ê 1 and ê 2 are tangent to the midplane
of the shell and ê 3 is in the fiber direction. As the element deforms, an angle may develop between the
actual fiber direction and the unit normal ê 3 . The magnitude of this angle may be characterized as
ê 3 ⋅ f – 1 < δ (0-129)
Elements 57
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
where f is the unit vector in the fiber direction and the magnitude of δ depends on the magnitude of the
strains. According to Belytschko et al., for most engineering applications, acceptable values of δ are on
the order of 10-2 and if the condition presented in (0-129) is met, then the difference between the rotation
of the co-rotational coordinates ê and the material rotation should be small.
The global components of this co-rotational triad define a transformation matrix between the global and
local element coordinate systems. This transformation operates on vectors with global components
ˆ ˆ ˆ ˆ
A = ( Ax, Ay, A z ) and element coordinate components A = ( A x, A y, A z ) , and is defined as:
ˆ
Ax e 1x e 2x e 3x Ax
ˆ ˆ
{ A } = A y = e 1y e 2y e 3y ˆ
Ay = [μ]{A} = [ q]T{A} (0-130)
Az e 1z e 2z e 3z ˆ
Az
where e ix, e iy, e iz are the global components of the element coordinate unit vectors. The inverse
transformation is defined by the matrix transpose; i.e.,
ˆ
{ A } = [ μ]T{ A } (0-131)
As in the Hughes-Liu shell element, the displacement of any point in the shell is partitioned into a
midsurface displacement (nodal translations) and a displacement associated with rotations of the
element’s fibers (nodal rotations). The Belytschko-Lin-Tsay shell element uses the Mindlin [1951]
theory of plates and shells to partition the velocity of any point in the shell as:
v = v m – ẑe 3 × θ (0-132)
where v m is the velocity of the mid-surface, θ is the angular velocity vector, and ẑ is the distance along
the fiber direction (thickness) of the shell element. The corresponding co-rotational components of the
velocity strain (rate of deformation) are given by
58
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
ˆ ˆ
ˆ 1 ∂υ i ∂υ j
d ij = --- -------- + -------- (0-133)
2 ∂x̂ j ∂x̂ i
Substitution of (0-132) into the above yields the following velocity-strain relations:
∂v̂ xm ˆ
ˆ ∂θ y
d x = --------- + ẑ -------- (0-134)
∂x̂ ∂x̂
ˆm ˆ
ˆ ∂υ y ∂θ x
dy = ---------
- – ẑ -
------- (0-135)
∂ŷ ∂ŷ
ˆm ˆm ˆ ˆ
ˆ ∂υ x ∂υ y ∂θy ∂θ x
2d xy = ---------- + ---------- + ẑ -------- – -------- (0-136)
∂ŷ ∂x̂ ∂ŷ ∂x̂
ˆm
ˆ ∂υ z ˆ
2d yz = ---------- – θ x (0-137)
∂ŷ
ˆm
ˆ ∂υ z ˆ
2d xz = ---------- + θ y (0-138)
∂x̂
The above velocity-strain relations need to be evaluated at the quadrature points within the shell.
Standard bilinear nodal interpolation is used to define the mid-surface velocity, angular velocity, and the
element’s coordinates (isoparametric representation). These interpolations relations are given by
ν m = N I ( ξ, η )ν I (0-139)
θ m = N I ( ξ, η )θ I (0-140)
x m = N I ( ξ, η )x I (0-141)
where the subscript I is summed over all the nodes of the element and the nodal velocities are
obtained by differentiating the nodal coordinates with respect to time, i.e., υ = x· . The bilinear
I I
shape functions are
1
N 1 = --- ( 1 – ξ ) ( 1 – η ) (0-142)
4
Elements 59
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
1
N 2 = --- ( 1 + ξ ) ( 1 – η ) (0-143)
4
1
N 3 = --- ( 1 + ξ ) ( 1 + η ) (0-144)
4
1
N 4 = --- ( 1 – ξ ) ( 1 + η ) (0-145)
4
The velocity-strains at the center of the element (i.e., at ξ = 0 , and η = 0 ) are obtained by
substitution of the above relations into the previously defined velocity-strain displacement relations,
(0-134) through (0-138). After some algebra, this yields
ˆ ˆ ˆ
d x = B 1I υ xI + ẑB 1I θ yI (0-146)
ˆ ˆ ˆ
d y = B 2I υ yI – ẑB 2I θ xI (0-147)
ˆ ˆ ˆ ˆ ˆ
2d xy = B2I υ xI + B 1I υ yI + ẑ ( B2I θ yI – B1I θ xI ) (0-148)
ˆ ˆ ˆ
2d xz = B 1I υ zI + N I θyI (0-149)
ˆ ˆ ˆ
2d yz = B 2I υ zI – N I θ xI (0-150)
where
∂N
B 1I = ---------I (0-151)
∂x̂
∂N I
B 2I = --------- (0-152)
∂ŷ
The shape function derivatives B aI are also evaluated at the center of the element; i.e., at ξ = 0 , and
η = 0.
ˆR ˆ
f αβ = σαβ dẑ (0-153)
R
m̂ αβ = – ẑσ̂ αβ dẑ (0-154)
where the superscript, R , indicates a resultant force or moment, and the Greek subscripts emphasize the
limited range of the indices for plane stress plasticity.
The above element midplane force and moment resultants are related to the local nodal forces and
moments by invoking the principle of virtual power and integrating with a one-point quadrature. The
relations obtained in this manner are
ˆ ˆR ˆR
f xI = A (B 1I f xx + B 2I f xy ) (0-155)
ˆ ˆR ˆR
f yI = A (B 2I f yy + B 1I f xy ) (0-156)
ˆ ˆR ˆR
f zI = Aκ (B 1I f xz + B 2I f yz ) (0-157)
κˆ R
m̂ xI = A B 2I m̂ yy
R R
+ B 1I m̂ xy – --- f yz (0-158)
4
κˆ R
m̂ yI = – A B 1I m̂ xx
R R
+ B 2I m̂ xy – --- f xz (0-159)
4
m̂ zI = 0 (0-160)
where A is the area of the element, and κ is the shear factor from the Mindlin theory. In the Belytschko-
Lin-Tsay formulation, κ is used as a penalty parameter to enforce the Kirchhoff normality condition as
the shell becomes thin.
The above local nodal forces and moments are then transformed to the global coordinate system using
the transformation relations given previously as (0-130). The global nodal forces and moments are then
appropriately summed over all the nodes and the global equations of motion are solved for the next
increment in nodal accelerations.
Elements 61
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
The hourglass control used by Belytschko et al., extends an earlier derivation by Flanagan and
Belytschko [1981], (see also Kosloff and Frazier [1978], Belytschko and Tsay [1983]). The hourglass
shape vector, τ I , is defined as
τ I = h I – ( hJ x̂ aJ )B aI (0-161)
where
+1
h = –1 (0-162)
+1
–1
is the basis vector that generates the deformation mode that is neglected by one-point quadrature. In
(0-161) and the reminder of this subsection, the Greek subscripts have a range of 2; e.g.,
x aI = ( x̂ 1I, x̂ 2I ) = ( x̂ I, ŷ I ) .
The hourglass shape vector then operates on the generalized displacements, in a manner similar to
(0-146) through (0-150), to produce the generalized hourglass strain rates
ˆ
q· αB = τ I θαI (0-163)
ˆ
q· 3B = τ I υ zI (0-164)
ˆ
q· αM = τ I υ αI (0-165)
where the superscripts B and M denote bending and membrane modes, respectively. The corresponding
hourglass stress rates are then given by
62
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
·B r θ Et 3 A
Q α = ----------------- B βI B βI q· αB (0-166)
192
·B r w κGt 3 A
Q 3 = ---------------------- BβI B βI q· B (0-167)
12 3
·M r m EtA
Q α = --------------- B βI B βI q· αM (0-168)
8
where t is the shell thickness and the parameters, r θ, r w , and r m are generally assigned values between
0.01 and 0.05.
Finally, the hourglass stresses, which are updated from the stress rates in the usual way; i.e.,
·
Q n + 1 = Q n + ΔtQ (0-169)
H B
m̂ αI = τI Qα (0-170)
ˆf H = τ Q B (0-171)
3I I 3
ˆf H = τ Q M (0-172)
αI I α
where the superscript H emphasizes that these are internal force contributions from the hourglass
deformations. These hourglass forces are added directly to the previously determined local internal forces
due to deformations (0-155) through (0-160). These force vectors are orthogonalized with respect to rigid
body motion.
In the hourglass procedure, the in-plane strain field (subscript p ) is decomposed into the one point strain
field plus the stabilization strain field:
· ·0 ·s
εp = εp + εp (0-173)
Elements 63
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
where the stabilization strain field, which is obtained from the assumed strain fields of Pian and Sumihara
[1984], is given in terms of the hourglass velocity field as
·s
ε p = W m q· m + zW b q· b (0-174)
Here, W m and W b play the role of stabilization strain velocity operators for membrane and bending:
p p
f 1( ξ, η ) f 4 ( ξ, η )
p p
Wm = f 2( ξ, η ) f 5 ( ξ, η ) (0-175)
p p
f 3( ξ, η ) f 6 ( ξ, η )
p p
– f 4( ξ, η ) f 1( ξ, η )
W b = – f p( ξ, η ) p
f 2 ( ξ, η ) (0-176)
5
p p
– f 6 ( ξ, η ) f 3 ( ξ, η )
p
where the terms f i ( ξ, η ) i = 1, 2, …, 6 , are rather complicated and the reader is referred to the
reference [Englemann and Whirley, 1991].
To obtain the transverse shear assumed strain field, the procedure given in [Bathe and Dvorkin, 1984] is
used. The transverse shear strain field can again be decomposed into the one point strain field plus the
stabilization field:
· ·0 ·s
εs = εs + ε s (0-177)
·s
ε s = W s q· s (0-178)
f s( ξ, η ) – g s ξ g s η g s ξ g s η
Ws = 1 1 2 3 3 (0-179)
f 2s( ξ, η ) g 4s ξ g 4s η – g 2s ξ g1s η
64
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
In their formulation, the hourglass forces are related to the hourglass velocity field through an
incremental hourglass constitutive equation derived from an additive decomposition of the stress into a
“one-point stress,” plus a “stabilization stress.” The integration of the stabilization stress gives a resultant
constitutive equation relating hourglass forces to hourglass velocities.
The in-plane and transverse stabilization stresses are updated according to:
·
τ ss, n + 1 = τ ss, n + Δtc s C s ε ss (0-180)
where the tangent matrix is the product of a matrix C , which is constant within the shell domain, and a
scalar c that is constant in the plane but may vary through the thickness.
The stabilization stresses can now be used to obtain the hourglass forces:
h
---
2 T
Qm = h
W m τ ps dA d
– ---
2A
h
--- (0-181)
2 T s
Qb = h
W b τ p dA dz
– ---
2A
h---
2 T s
Qs = h
W s τ s dA dz
– ---
2A
In order to include warpage in the formulation it is convenient to define nodal fiber vectors as shown in
Figure 0-11. The geometry is interpolated over the surface of the shell from:
x = x m + ζp = ( x I + ζp I )N I ( ξ, η ) (0-182)
ζh
where: ζ = ------ and ζ is a parametric coordinate which varies between -1 to +1.
2
The in plane strain components are given by:
c
d xx = b xI v̂ xI + ζ ( b xI v̂ xI + b xI p· xI ) (0-183)
c
d yy = b yI v̂ yI + ζ ( b yI v̂ yI + b yI p· yI ) (0-184)
1 c c
d xy = --- bxI v̂ yI + b yI v̂ xI + ζ ( b xI v̂ yI + b xI p· yI + b yI v̂ xI + b yI p· xI ) (0-185)
2
30
29
24 Belystchkno-Tsay
L = 12
Y Displacement (104)
b = 1.1 20
t = .32
twist = 90 degrees 16 Hughes-Liu
E = 29 000 000 Belytschko-Wong-Chiang
ν = .22
12
0
0 5 10 15 19
time (ms)
66
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
Figure 0-10 The Twisted Beam Problem Fails with the Belytschko-Tsay Shell Element
p3
p1
p2
h
c
b xI p ŷ2 – p ŷ4 p ŷ3 – p ŷ1 p ŷ4 – p ŷ2 p ŷ1 – p ŷ3
= (0-186)
bc p x̂2 – p x̂4 p x̂3 – p x̂1 p x̂4 – p x̂2 p x̂1 – p x̂3
yI
For a flat geometry the normal vectors are identical and no coupling can occur. Two methods are used
c
by Belytschko for computing b iI and the reader is referred to his papers for the details. Both methods
have been tested and comparable results were obtained.
γ̂ xz = – N I ( ξ ,η )θ ŷI (0-187)
γ̂ yz = – N I ( ξ ,η )θ x̂I (0-188)
I I K K
θ x̂I = ( e n ⋅ e x̂ )θn + ( e n ⋅ e x̂ )θ n (0-189)
Elements 67
CQUADA - DYSHELLFORM = 2, Belytschko-Lin-Tsay Shell
I I K K
θ ŷI = ( e n ⋅ e ŷ )θ n + ( e n ⋅ e ŷ )θ n (0-190)
I
The rotation θ n comes from the nodal projection
I 1 I I 1-
θ n = --- ( θnI + θnJ ) + ------ ( υ̂ zJ – υ̂ zJ ) (0-191)
2 L IJ
where the subscript n refers to the normal component of side I as seen in Figure 0-12 and L IJ is the
length of side IJ .
yˆ
K
r
eˆ
y
L
K
e en
nk
i
K
I eˆ
X
Figure 0-12 Vector and Edge Definitions for Computing the Transverse Shear
Strain Components
68
CTRIA3 - DYSHELLFORM = 4, C0 Triangular Shell
CTRIA3 - DYSHELLFORM = 4,
C0 Triangular Shell
The C 0 shell element due to Kennedy, Belytschko, and Lin [1986] has been implemented as a
computationally efficient triangular element complement to the Belytschko-Lin-Tsay quadrilateral shell
element ([Belytschko and Tsay 1981], [Belytschko et al., 1984a]). For a shell element with five through-
the-thickness integration points, the element requires 649 mathematical operations (the Belytschko-Lin-
Tsay quadrilateral shell element requires 725 mathematical operations) compared to 1417 operations for
the Marchertas-Belytschko triangular shell [Marchertas and Belytschko 1974] (referred to as the BCIZ
[Bazeley, Cheung, Irons, and Zienkiewicz 1965] triangular shell element).
Triangular shell elements are offered as optional elements primarily for compatibility with local user grid
generation and refinement software. Many computer aided design (CAD) and computer aided
manufacturing (CAM) packages include finite element mesh generators, and most of these mesh
generators use triangular elements in the discretization. Similarly, automatic mesh refinement algorithms
are typically based on triangular element discretization. Also, triangular shell element formulations are
not subject to zero energy modes inherent in quadrilateral element formulations.
The triangular shell element’s origins are based on the work of Belytschko et al., [Belytschko, Stolarski,
and Carpenter 1984b] where the linear performance of the shell was demonstrated. Because the triangular
shell element formulations parallels closely the formulation of the Belytschko-Lin-Tsay quadrilateral
shell element presented in the previous section, the following discussion is limited to items related
specifically to the triangular shell element.
Co-rotational Coordinates
The mid-surface of the triangular shell element, or reference surface, is defined by the location of the
element’s three nodes. An embedded element coordinate system (see Figure 0-13) that deforms with the
element is defined in terms of these nodal coordinates. The procedure for constructing the co-rotational
coordinate system is simpler than the corresponding procedure for the quadrilateral, because the three
nodes of the triangular element are guaranteed coplanar.
zˆ
yˆ
3
ê 3
eˆ 2
1 xˆ
eˆ 1
2
Figure 0-13 Local Element Coordinate System for C 0 Shell Element
Elements 69
CTRIA3 - DYSHELLFORM = 4, C0 Triangular Shell
The local x-axis, x̂ , is directed from node 1 to 2. The element’s normal axis, ẑ , is defined by the vector
cross product of a vector along x̂ with a vector constructed from node 1 to node 3. The local y-axis, ŷ ,
is defined by a unit vector cross product of ê 3 with ê 1 , which are the unit vectors in the ẑ directions,
respectively. As in the case of the quadrilateral element, this triad of co-rotational unit vectors defines a
transformation between the global and local element coordinate systems (see (0-130) and (0-131)).
Velocity-Strain Relations
As in the Belytschko-Lin-Tsay quadrilateral shell element, the displacement of any point in the shell is
partitioned into a mid-surface displacement (nodal translations) and a displacement associated with
rotations of the element’s fibers (nodal rotations). The Kennedy-Belytschko-Lin triangular shell element
also uses the Mindlin [Mindlin 1951] theory of plates and shells to partition the velocity of any point in
the shell (recall (0-132)):
ν = v m – ẑe 3 × θ (0-192)
where ν m is the velocity of the mid-surface, θ is the angular velocity vector, and ẑ is the distance along
the fiber direction (thickness) of the shell element. The corresponding co-rotational components of the
velocity strain (rate of deformation) were given previously in (0-146) through (0-150).
Standard linear nodal interpolation is used to define the midsurface velocity, angular velocity, and the
element’s coordinates (isoparametric representation). These interpolation functions are the area
coordinates used in triangular element formulations. Substitution of the nodally interpolated velocity
fields into the velocity-strain relations (see Belytschko et al., for details), leads to strain rate-velocity
relations of the form
dˆ = Bv̂ (0-193)
where d̂ are the velocity strains (strain rates), the elements of B are derivatives of the nodal
interpolation functions, and the ν̂ are the nodal velocities and angular velocities.
It is convenient to partition the velocity strains and the B matrix into membrane and bending
contributions. The membrane relations are given by
70
CTRIA3 - DYSHELLFORM = 4, C0 Triangular Shell
υ̂ x1
M
υ̂ y1
dˆ x ŷ 3 0 ŷ 3 0 0 0
1 υ̂ x2
dˆ y = ----------
x̂ 2 ŷ 3 0 x̂ 3 – x̂ 2 0 – x̂ 3 0 x̂ 2 (0-194)
υ̂ y2
2dˆ xy x̂ 3 – x̂ 2 –ŷ 3 – x̂ 3 ŷ 3 x̂ 2 0
υ̂ x3
υ̂ y3
or
ˆM
d = B M ν̂ (0-195)
θ̂ x1
ˆ
θ y1
κ̂ x 0 – ŷ 3 0 ŷ 3 0 0 ˆ
– 1 θ x2
κ̂ y = ---------- x̂ 3 – x̂ 2 0 x̂ 3 0 –x̂ 2 0 (0-196)
x̂ 2 ŷ 3 ˆ
θ y2
2κˆ xy ŷ 3 x̂ 3 – x̂ 2 – ŷ 3 – x̂ 3 0 x̂ 2
ˆ
θ x3
ˆ
θ y3
or
ˆ def
κ̂ M = BM θ (0-197)
The local element velocity strains are then obtained by combining the above two relations:
Elements 71
CTRIA3 - DYSHELLFORM = 4, C0 Triangular Shell
M
ˆ ˆ
dx d x κ̂ x
ˆM ˆ
dˆ y dˆ y – ẑ κ̂ y = d – ẑκ (0-198)
2dˆ xy 2dˆ xy 2κˆ xy
ˆ
2d xz 1
= --------------
2dˆ yz 6x̂ 2 ŷ 3
– ŷ 32 ŷ 3 ( 2x̂ 2 + x̂ 3 ) ŷ 32 ŷ 3 ( 3x̂ 2 – x̂ 3 ) 0 x̂ 2 ŷ 3
ŷ 3 ( x̂ 2 –def
2x̂ 2 ) x̂ 22 – xˆ 32 – ŷ 32 ( x̂ 2 + x̂ 3 ) x̂ 3 ( x̂ 3 – 2x̂ 2 ) – 3x̂ 2 ŷ 3 x̂ 2 ( 2x̂ 3 – x̂ 2 )
θ̂x1
ˆ
θy1
ˆ
θx2
θˆ y2
θˆ x3
θˆ y3
(0-199)
or
ˆS ˆ def
d = BS θ (0-200)
All of the above velocity-strain relations have been simplified by using one-point quadrature.
ˆ
In the above relations, the angular velocities θ def are the deformation component of the angular velocity
ˆ
θ obtained by subtracting the portion of the angular velocity due to rigid body rotation; i.e.,
ˆ def ˆ ˆ rig
θ = θ–θ (0-201)
The two components of the rigid body angular velocity are given by
72
CTRIA3 - DYSHELLFORM = 4, C0 Triangular Shell
ˆ rig υ̂ z1 – υ̂ z2
θ y = --------------------- (0-202)
x̂ 2
( υ̂ z3 – υ̂ z1 )x̂ 2 – ( υ̂ z2 – υ̂ z1 )x̂ 3
θ̂ xrig = ---------------------------------------------------------------------- (0-203)
x̂ 2 ŷ 3
The first of the above two relations is obtained by considering the angular velocity of the local x-axis
about the local y-axis. Referring to Figure 0-14, by construction nodes 1 and 2 lie on the local x-axis and
the distance between the nodes is x̂ 2 ; i.e., the x̂ distance from node 2 to the local coordinate origin at
node 1. Thus, the difference in the nodal ẑ velocities divided by the distance between the nodes is an
average measure of the rigid body rotation rate about the local y-axis.
zˆ
yˆ
xˆ
1 2
zˆ
yˆ
3
xˆ
1 2
Figure 0-14 Element Configurations with Node 3 Aligned with Node 1 (left) and Node 3
Aligned with Node 2 (right)
The second relation is conceptually identical, but is implemented in a slightly different manner due to the
arbitrary location of node 3 in the local coordinate system. Consider the two local element configurations
shown in Figure 0-14. For the left-most configuration, where node 3 is the local y-axis, the rigid body
rotation rate about the local x-axis is given by
Elements 73
CTRIA3 - DYSHELLFORM = 4, C0 Triangular Shell
rig υ̂ z3 – υ̂ z1
θ̂ x – left = --------------------- (0-204)
ŷ 3
and for the rightmost configuration the same rotation rate is given by
rig υ̂ z3 – υ̂ z2
θ̂ x – right = --------------------- (0-205)
ŷ 3
Although both of these relations yield the average rigid body rotation rate, the selection of the correct
relation depends on the configuration of the element; i.e., on the location of node 3. Since every element
in the mesh could have a configuration that is different in general from either of the two configurations
shown in Figure 0-14, a more robust relation is needed to determine the average rigid body rotation rate
about the local x-axis. In most typical grids, node 3 will be located somewhere between the two
configurations shown in Figure 0-14. Thus, a linear interpolation between these two rigid body rotation
rates was devised using the distance x̂ 3 as the interpolant:
x̂ 3 x̂ 3
θ x = θ x – left 1 – ----- + θ x – right -----
ˆ rig ˆ rig ˆ rig
(0-206)
x̂ 2 x̂ 2
Substitution of (0-204) and (0-205) into (0-206) and simplifying produces the relations given previously
as (0-203).
ˆR
f αβ = σ̂αβ dẑ (0-207)
αβ dẑ
m̂ αβ ˆ
R = – ẑσ (0-208)
where the superscript R indicates a resultant force or moment and the Greek subscripts emphasize the
limited range of the indices for plane stress plasticity.
The above element midplane force and moment resultant are related to the local nodal forces and
moments by invoking the principle of virtual power and performing a one-point quadrature. The relations
obtained in this manner are
74
CTRIA3 - DYSHELLFORM = 4, C0 Triangular Shell
ˆ
f x1
ˆ
f y1 ˆR
f xx
ˆ
f x2 T R
= AB M ˆf yy (0-209)
ˆ
f y2
ˆf R
ˆ
f x3 xy
ˆ
f y3
m̂ x1
m̂ y1 R
m̂ xx ˆR
m̂ x2 T T f xz
= AB M m̂ yy
R + ABS (0-210)
m̂ y2 ˆR
f yz
m̂ xy
R
m̂ x3
m̂ y3
ˆ ˆ ˆ
The remaining nodal forces, the ẑ component of the force ( f z3, f z2, f z1 ) , are determined by
successively solving the following equilibration equations
ˆ
m̂ x1 + m̂ x2 + m̂ x3 + ŷ 3 f z3 = 0 (0-211)
ˆ ˆ
m̂ y1 + m̂ y2 + m̂ y3 – x̂ 3 f z3 – x̂ 2 f z2 = 0 (0-212)
ˆ ˆ
f z1 + f z2 + ˆf z3 = 0 (0-213)
which represent moment equilibrium about the local x-axis, moment equilibrium about the local y-axis,
and force equilibrium in the local z-direction, respectively.
Elements 75
CTRIA3 - DYSHELLFORM = 3, Marchertas-Belytschko Triangular Shell (BCIZ)
CTRIA3 - DYSHELLFORM = 3,
Marchertas-Belytschko Triangular Shell (BCIZ)
The Marchertas-Belytschko [1974] triangular shell element, or the BCIZ triangular shell element, was
developed in the same time period as the Belytschko beam element [Belytschko, Schwer, and Klein,
1977], see “CBEAM - Belytschko Beam” on page 34, forming the first generation of co-rotational
structural elements developed by Belytschko and co-workers. Although the Marchertas-Belytschko shell
element is relatively expensive (i.e., the C 0 triangular shell element with five through-the-thickness
integration points requires 649 mathematical operations compared to 1,417 operations for the
Marchertas-Belytschko triangular shell), it is maintained in SOL 700 for compatibility with earlier user
models. However, as the user community moves to application of the more efficient shell element
formulations, the use of the Marchertas-Belytschko triangular shell element will decrease.
Element Coordinates
Figure 6-15a shows the element coordinate system, ( x̂, ŷ, ẑ ) originating at Node 1, for the Marchertas-
Belytschko triangular shell. The element coordinate system is associated with a triad of unit vectors
( e 1, e 2, e 3 ) the components of which form a transformation matrix between the global and local
coordinate systems for vector quantities. The nodal or body coordinate system unit vectors b 1, b 2, b 3
are defined at each node and are used to define the rotational deformations in the element, see
“Co-rotational Technique” on page 34.
The unit normal to the shell element e 3 is formed from the vector cross product
e 3 = l 21 × l 31 (0-214)
where l 21 and l 31 are unit vectors originating at Node 1 and pointing towards Nodes 2 and 3,
76
CTRIA3 - DYSHELLFORM = 3, Marchertas-Belytschko Triangular Shell (BCIZ)
Materials 71
Materials
72
Materials
Materials
In addition to most materials available in structures workspace, the following material models suitable
for explicit applications are also available in MD Explicit. Here, we will highlight the theoretical
background of these material models. These are the material models that are most commonly used for
typical structural applications. For a more detailed description of all the available materials in MD
Explicit, please refer to MD R2 Nastran Quick Reference Guide.
1 Elastic
2 Orthotropic Elastic
3 Kinematic/Isotropic Elastic-Plastic
5 Soil and Crushable/Non-crushable Foam
6 Viscoelastic
7 Blatz-Ko Rubber
9 Null Hydrodynamics
10 Isotropic-Elastic-Plastic-Hydrodynamic
12 Isotropic-Elastic-Plastic
13 Elastic-Plastic with Failure Model
14 Soil and Crushable Foam with Failure Model
15 Johnson/Cook Strain and Temperature Sensitive Plasticity
18 Power Law Isotropic Plasticity
19 Strain Rate Dependent Isotropic Plasticity
20 Rigid
22 Composite Damage Model
24 Piecewise Linear Isotropic Plasticity
26 Honeycomb
27 Compressible Mooney-Rivlin Rubber
28 Resultant Plasticity
29 Forced Limited Resultant Fomulation
30 Shape-Memory Superelastic Material
31 Slightly Compressible Rubber Model
32 Laminated Glass Model
34 Fabric
40 Nonlinear Orthotropic
54-55 Composite Damage Model
Materials 73
Materials
In the table below, a list of the available material models and the applicable element types are given.
Some materials include strain rate sensitivity, failure, equations of state, and thermal effects and this
Materials 75
Materials
is also noted. General applicability of the materials to certain kinds of behavior is suggested in the
last column.
Gn General
Cm Composites
Material Number
Strain-Rate Effects
Cr Ceramics
Equation-of-State
Fl Fluids
Thermal Effects
Fm Foam
Thick Shells
Gl Glass
Thin Shells
Hy Hydro-dyn
Mt Metal
Beams
Failure
Bricks
Pl Plastic
Rb Rubber
Material Title Sl Soil/Cone
1 Elastic Y Y Y Y Gn, Fl
2 Orthotropic Elastic (Anisotropic - solids) Y Y Y Cm, Mt
3 Plastic Kinematic/Isotropic Y Y Y Y Y Y Cm, Mt, Pl
5 Soil and Foam Y Fm, Sl
6 Linear Viscoelastic Y Y Y Y Y Y Y Rb
7 Blatz-Ko Rubber Y Y Rb, Polyurethane
9 Null Material Y Y Y Y Fl, Hy
10 Elastic Plastic Hydro (dynamics) Y Y Y Hy, Mt
12 Isotropic Elastic Plastic Y Y Y Y Mt
13 Isotropic Elastic-Plastic with Failure Y Y Mt
14 Soil and Foam with Failure Y Y Fm, Sl
15 Johnson/Cook Plasticity Model Y Y Y Y Y Y Hy, Mt
18 Power Law Plasticity (Isotropic) Y Y Y Y Y Mt, Pl
19 Strain Rate Dependent Rate Plasticity Y Y Y Y Y Mt, Pl
20 Rigid Y Y Y Y
22 Composite Damage Y Y Y Y Cm
24 Piecewise Linear Isotropic Plasticity Y Y Y Y Y Y Mt, Pl
26 Honeycomb Y Y Y Cm, Fm, Sl
27 Mooney-Rivlin Rubber Y Y Rb
28 Resultant Plasticity Y Y Mt
29 Forced Limited Resultant Formaultion Y
30 Shaped Memory Alloy Y Mt
31 Slightly Compressible Rubber Y Rb
32 Laminated Glass (Composite) Y Y Y Cm, Gl
76
Materials
Gn General
Cm Composites
Material Number
Strain-Rate Effects
Cr Ceramics
Equation-of-State
Fl Fluids
Thermal Effects
Fm Foam
Thick Shells
Gl Glass
Thin Shells
Hy Hydro-dyn
Mt Metal
Beams
Failure
Bricks
Pl Plastic
Rb Rubber
Material Title Sl Soil/Cone
34 Fabric Y
40 Nonlinear Orthotropic Y Y Y Cm
54 Composite Damage with Chang Failure Y Y Cm
55 Composite Damage with Tsai-Wu Failure Y Y Cm
57 Low Density Urethane Foam Y Y Y Fm
58 Laminated composite Fabric Y
59 Composite Failure - Plasticity Based Y Y Y Cm, Cr
62 Viscous foam (Crash Dummy) Y Y Fm
63 Isotropic Crushable Foam Y Y Fm
64 Rate Sensitive Power-Law Plasticity Y Y Y Y Mt
66 Linear Elastic Discrete Beam Y Y
67 Nonlinear Elastic Discrete Beam Y Y
68 Nonlinear Plastic Discrete Beam Y Y Y
69 SID Damper Discrete Beam Y Y
70 Hydraulic Gas Damper Discrete Beam Y Y
71 Cable Discrete Beam Y
72 Concrete Damage Y Y Y Y Sl
72R Concrete Damage Release III Y Y Y Y Sl
73 Low Density Viscous Foam Y Y Y Fm
74 Elastic Spring Discrete Beam Y
76 General Viscoelastic (Maxwell model) Y Y Rb
77 Hyperelastic and Ogden Rubber Y Rb
79 Hysteretic Soil (Elasto-Perfectly Plastic) Y Y Sl
80 Ramberg-Osgood Y Y Y Y Y Y Y Y
81 Plastic with Damage (Elasto-Plastic) Y Y Y Y Y Y Mt, Pl
83 Fu-Chang’s Foam Y Y Y Fm
87 Cellular Rubber Y Y Rb
Materials 77
Materials
Gn General
Cm Composites
Material Number
Strain-Rate Effects
Cr Ceramics
Equation-of-State
Fl Fluids
Thermal Effects
Fm Foam
Thick Shells
Gl Glass
Thin Shells
Hy Hydro-dyn
Mt Metal
Beams
Failure
Bricks
Pl Plastic
Rb Rubber
Material Title Sl Soil/Cone
89 Plastic Polymer Y
93 Elastic Six Degrees of Freedom Spring Y
Discrete Beam
94 Inelastic Spring Discrete Beam Y
95 Inelastic Six Degrees of Freedom Spring Y
Discrete Beam
97 General Joint Discrete Beam Y
98 Simplified Johnson-Cook Y Y Y Y
99 Simplified Johnson-Cook Orthotropic Y Y Y Y
Damage
100 Spot weld Y
112 Finite Elastic Strain Plasticity Y
114 Layered Linear Plasticity Y Y
119 General Nonlinear Six Degrees of Freedom Y
Discrete Beam
121 General Nonlinear One Degree of Freedom Y
Discrete Beam
123 Modified Piecewise Linear Plasticity Y Y
126 Modified Honeycomb Y
127 Arruda-Boyce rubber Y
158 Composite Fabric Y Y Y Y Cm
181 Simplified Rubber Y RB
196 General Spring Discrete Beam Y
B01 Seatbelt
S01 Spring Elastic (Linear) Y
S02 Damper Viscous (Linear) Y Y
S03 Spring Elastoplastic (Isotropic) Y
S04 Spring Nonlinear Elastic Y Y
78
Materials
Gn General
Cm Composites
Material Number
Strain-Rate Effects
Cr Ceramics
Equation-of-State
Fl Fluids
Thermal Effects
Fm Foam
Thick Shells
Gl Glass
Thin Shells
Hy Hydro-dyn
Mt Metal
Beams
Failure
Bricks
Pl Plastic
Rb Rubber
Material Title Sl Soil/Cone
S05 Damper Nonlinear Viscous Y Y
S06 Spring General Nonlinear Y
S07 Spring Maxwell (Three Parameter Y Y
Viscoelastic)
S08 Spring Inelastic (Tension or Compression Y
S13 Spring Tri-linear Degrading
S14 Spring Squat Shearwall
S15 Spring Muscle
SW1 Spot Weld (Simple Damage-Failure) Y
SW2 Spot Weld (Resultant-based Failure Y
Criteria)
SW3 Spot Weld (Stress-based Failure) Y
SW4 Spot Weld (Rate Dependent Stress-based Y
Failure)
SW5 Spot Weld (Additional Failure) Y
· n+1⁄2
s ij∇
n+1⁄2
= 2Gε' ij (3-1)
and pressure
p n + 1 = K ln V n + 1 (3-2)
where G and K are the elastic shear and bulk moduli, respectively, and V is the relative volume; i.e.,
the ratio of the current volume to the initial volume.
For standard MD Nastran solution sequence, this would be the same as using MAT1 to define a linear
elastic material.
Materials 79
Materials
S = C ⋅ E = TtClT ⋅ E (3-3)
l 12 m 12 n 12 l1 m1 m1 n1 n1 l1
l 22 m 22 n 22 l2 m2 m2 n2 n2 l2
l 32 m 32 n 32 l3 m3 m3 n3 n3 l3
T = (3-4)
2l 1 l 2 2m 1 m 2 2n 1 n 2 ( l 1 m 2 + l 1 m 1 ) ( m 1 n 2 + m 2 n 1 ) ( n 1 l 2 + n 2 l 1 )
2l 2 l 3 2m 2 m 3 2n 2 n 3 ( l 2 m 3 + l 3 m 2 ) ( m 2 n 3 + m 3 n 2 ) ( n 2 l 3 + n 3 l 2 )
2l 3 l 1 2m 3 m 1 2n 3 n 1 ( l 3 m 1 + l 1 m 3 ) ( m 3 n 1 + m 1 n 3 ) ( n 3 l 1 + n 1 l 3 )
and x' i denotes the material axes. The constitutive matrix C l is defined in terms of the material axes as
υ 21 υ 31
1 - -------
------- – - – -------- 0 0 0
E11 E 22 E 33
υ 12 1 υ 32
– -------- -------- – -------- 0 0 0
E 11 E 22 E 33
υ 13 υ 23 1
– -------- – -------- -------- 0 0 0
C l–1 = E 11 E 22 E33 (3-6)
1-
--------
0 0 0 0 0
G 12
1
0 0 0 0 --------- 0
G 23
1
0 0 0 0 0 ---------
G 31
80
Materials
Since C l is symmetric
υ 12 υ 21
-------
- = -------
- , ect. (3-8)
E 11 E 22
The vector of Green-St. Venant strain components is
Et = E 11 ,E 22 ,E 33 ,E 12 ,E 23 ,E 31 (3-9)
ρ ∂x ∂x
σ ij = ----- --------i- --------j- S k (3-10)
ρ o ∂X k ∂X
After computing S ij , we use Equation 3-10 to obtain the Cauchy stress. This model will predict realistic
behavior for finite displacement and rotations as long as the strains are small.
For standard MD Nastran solution sequences, this would be the same as using MAT9 to define a linear
anisotropic material. For shell elements, you would have used the MAT2 or MAT8 option.
1--- σ y2
φ = ξ ij ξ ij – ------ = 0 (3-11)
2 3
where
ξ ij = S ij α ij (3-12)
σ y = σ 0 + βE p ε eff
p
(3-13)
Materials 81
Materials
2 ·p
α ij∇ = ( 1 – β ) --- Ep ε ij (3-14)
3
Hence,
Et
Yield
Stress
ln ----
l
l0
β = 0 Kinematic Hardening
β = 1 Isotropic Hardening
Strain rate is accounted for using the Cowper and Symonds [Jones 1983] model which scales the yield
stress by a strain rate dependent factor
ε 1⁄p
·
ρ y = 1 + ---- ( σ 0 + βE p ε eff
p
(3-16)
C
·
where p and C are user-defined input constants and ε is the strain rate defined as:
82
Materials
· · ·
ε + ε ij ε ij (3-17)
The current radius of the yield surface, σ y , is the sum of the initial yield strength, σ 0 , plus the growth
βE p ε eff
p
, where E p is the plastic hardening modulus
Et E
E p = -------------- (3-18)
E – Et
p
and ε eff is the effective plastic strain
t
1/2
--3- εij εij
2·p ·p
ε eff
p
= dt (3-19)
0
The plastic strain rate is the difference between the total and elastic (right superscript) strain rates:
·p · ·e
ε ij = ε ij – ε ij (3-20)
In the implementation of this material model, the deviatoric stresses are updated elastically, as described
for model 1, but repeated here for the sake of clarity:
where
*
Let s ij represent the trial elastic deviatoric stress state at n+1
Materials 83
Materials
* * 1 *
s ij = σ ij – --- σ kk σ ij (3-22)
3
and
3
φ = --- ξ ij* ξ ij* – σ y2 = Λ 2 – σ y2 < 0 for elastic or neutral loading (3-24)
2 > 0 for plastic harding
For plastic hardening then:
n+1 pn
Λ – σy pn
ε eff
p
= ε eff - = ε eff
+ ------------------- + Δε eff
p
(3-25)
3G + Ep
Scale back the stress deviators:
p
3GΔε eff
σ ijn + 1 = σ ij* – - ξ ij*
------------------- (3-26)
Λ
and update the center:
( 1 – β )E p Δε eff p
α ijn + 1 = α ijn + ------------------------------------ (3-27)
Λ
σ 33 + λ ( Δε 11 + Δε 22 )
Δε 33 = – ---------------------------------------------------- (3-28)
λ + 2μ
where λ and γ are Lamé’s constants.
84
Materials
When the trial stress is within the yield surface, the strain increment is elastic and the stress update is
completed. Otherwise, for the plastic plane stress case, secant iteration is used to solve Equation (3-26)
for the normal strain increment ( Δε33 ) required to produce a zero normal stress:
3GΔε eff pi ξ
33
σ 33
i = σ 33
* – ---------------------------- (3-29)
Λ
Here, the superscript i indicates the iteration number.
The secant iteration formula for Δε 33 (the superscript p is dropped for clarity) is
Δε 33 i
– Δε 33 i–1
Δε 33
i+1
= Δε 33
i–1
– -------------------------------- σ 33
i–1 (3-30)
σ 33i – σi – 1
33
where the two starting values are obtained from the initial elastic estimate and by assuming a purely
plastic increment; i.e.,
Δε 33
1
= – ( Δε 11 – Δε 22 ) (3-31)
These starting values should bound the actual values of the normal strain increment.
The iteration procedure uses the updated normal stain increment to update first the deviatoric stress and
then the other quantities needed to compute the next estimate of the normal stress in Equation (3-29). The
i
iterations proceed until the normal stress σ 33 is sufficiently small. The convergence criterion requires
convergence of the normal strains:
Δε 33 i
– Δε 33 i–1
----------------------------------- < 10 – 4 (3-32)
Δε 33 i+1
After convergence, the stress update is completed using the relationships given in Equations (3-26) and
(3-27)
For SOL SEQ 106, 129, or 600, this material model is the same as using MATS1, with the hardening rule
(HR) set to either Isotropic or Kinematic. These models do not allow β between 0 and 1.
1
φ s = --- s ij s ij – ( a 0 + a 1 p + a 2 p 2 ) (3-33)
2
where a 0 , a 1 , and a 2 are user-defined constants. Volumetric yielding is determined by a tabulated
curve of pressure versus volumetric strain. Elastic unloading from this curve is assumed to a tensile cutoff
as illustrated in Figure 3-2.
Pressure
Figure 3-2 Volumetric Strain Versus Pressure Curve for Soil and Crushable Foam Model
Implementation of this model is straightforward. One history variable, the maximum volumetric strain in
compression, is stored. If the new compressive volumetric strain exceeds the stored value, loading is
indicated. When the yield condition is violated, the updated trial stresses, s ij* , are scaled back using a
simple radial return algorithm:
a + a p + a p 2 1 ⁄ 2
= -------------------------------------- s ij*
0 1 2
s ijn + 1 (3-34)
1---
s s
2 ij ij
If the hydrostatic tension exceeds the cutoff value, the pressure is set to the cutoff value and the deviatoric
stress tensor is zeroed.
86
Materials
t
∂ε' ij
s ij = 2 φ ( t – τ ) ---------- dτ (3-35)
∂τ
0
where
φ ( t ) = G ∞ + ( G 0 – G ∞ )e – βt (3-36)
is the shear relaxation modulus. A recursion formula is used to compute the new value of the hereditary
integral at time t n + 1 from its value at time t n . Elastic bulk behavior is assumed:
p = K ln V (3-37)
where pressure is integrated incrementally.
1
= G V –1 C ij – V 1 – 2υ δ ij
– ---------------
S ij (3-38)
where G is the shear modulus, V is the relative volume, υ is Poisson’s ratio, and C ij is the right
Cauchy-Green strain:
∂x ∂x
C ij = -------k- -------k- (3-39)
∂X i ∂X j
ρ ∂x i ∂x j
σ ij = ----- --------- -------- S kl (3-40)
ρ 0 ∂X k ∂X l
where ρ 0 and ρ are the initial and current density, respectively. The default value of υ is 0.463.
Material 27 and 31 better represent incompressible materials.
Materials 87
Materials
σ ij = με′ ij (3-41)
1 · · 1·
p = – --- σ ij δ ij ε' ij = ε ij – --- ε v
3 · · 3
s ij = σ ij + pδ ij ε v = ε ij δ ij
· ·
s ij∇ = 2με' ij = 2Gε' ij
(3-42)
s ij∇ = s· ij – s ip Ω pj – s jp Ω pi (3-43)
· ·
*s n + 1
ij = s ijn + s ip Ω pj + s jp Ω pi + 2Gε' ij dt = s ijn + R ij + 2Gε' ij dt
n
s ijR 2GΔε'ij
(3-44)
88
Materials
where the left superscript, *, denotes a trial stress value. The effective trial stress is defined by
1⁄2
s * = --- *s ijn + 1 *s ijn + 1
3
(3-45)
2
y
and if s * exceeds yield stress σ , the von Mises flow rule:
1 σ y2
φ = --- s ij s ij – ------ ≤ 0 (3-46)
2 3
is violated and we scale the trial stresses back to the yield surface; i.e., a radial return
σ
s ijn + 1 = -----*y *s ijn + 1 = m * s ijn + 1 (3-47)
s
The plastic strain increment can be found by subtracting the deviatoric part of the strain increment that
1 - n + 1 Rn
is elastic, ------ (s – s ij ) , from the total deviatoric increment, Δε' ij , i.e.,
2G ij
1
Δε ijp = Δε' ij – ------- ( s ijn + 1 – s ijR n ) (3-48)
2G
Recalling that,
*s n + 1 R
ij – s ij n
Δε' ij = --------------------------
- (3-49)
2G
and substituting Equation (3-49) into Equation (3-48) we obtain,
( *s ijn + 1 – s ijn + 1 )
Δε ijp = ------------------------------------- (3-50)
2G
Substituting Equation (3-47)
s ijn + 1 = m * s ijn + 1
(1 – m) 1–m
Δε ijp = ------------------ *s ijn + 1 = ------------- s ijn + 1 = dλs ijn + 1 (3-51)
2G 2Gm
Materials 89
Materials
1⁄2
Δε p = --- Δε ijp Δε ijp
2
(3-52)
3
Squaring both sides of Equation (3-51) leads to:
1–m 2
Δε ijp Δε ijp = ------------- *s ijn + 1 *s ijn + 1 (3-53)
2G
or from Equations (3-45) and (3-52):
1 – m 22
Δε = ------------- --- s *2
3--- p 2
(3-54)
2 2G 3
Hence,
1–m s∗ – σ y
∴Δε p = ------------- s∗ = ----------------- (3-55)
3G 3G
where we have substituted for m from Equation (3-47)
σy
m = -----
s∗
If isotropic hardening is assumed then:
σ yn + 1 = σ yn + E p Δε p (3-56)
( s∗ – σ yn + 1 ) ( s∗ – σ yn – E p Δε p )
Δε p = ----------------------------
- = -------------------------------------------- (3-57)
3G 3G
Thus,
( 3G + E p )Δε p = ( s∗ – σ yn )
( s∗ – σ yn )
Δε p = ------------------------
- (3-58)
( 3G + E p )
90
Materials
The algorithm for plastic loading can now be outlined in five simple stress. If the effective trial stress
exceeds the yield stress then
1. Solve for the plastic strain increment:
( s∗ – σ yn )
Δε p = ------------------------
-
( 3G + E p )
2. Update the plastic strain:
εp = ε p + Δε p
n+1 n
σ yn + 1 = σ yn + E p Δε p
4. Compute the scale factor using the yield strength at time n + 1:
σ yn + 1
m = -------------
s∗
5. Radial return the deviatoric stresses to the yield surface:
s ijn + 1 = m *s ijn + 1
σ y2
φ = J 2 – ------ (3-59)
3
where the second stress invariant, J 2 , is defined in terms of the deviatoric stress components as
1
J 2 = --- s ij s ij (3-60)
2
and the yield stress, σ y , is a function of the effective plastic strain, ε eff
p
, and the plastic hardening
modulus, Ep :
σ y = σ 0 + Ep ε eff
p
(3-61)
Materials 91
Materials
ε eff dεeff
p p
= (3-62)
0
where:
p 2--- p p
dε eff = dε dε (3-63)
3 ij ij
and the plastic tangent modulus is defined in terms of the input tangent modulus, E t , as:
EE t
E p = -------------- (3-64)
E – Et
Pressure is given by the expression
p n + 1 = K ------------
1 -
–1 (3-65)
V +1
n
where K is the bulk modulus. This is perhaps the most cost effective plasticity model. Only one history
variable, ε eff
p
, is stored with this model.
This model is not recommended for shell elements. In the plane stress implementation, a one-step radial
return approach is used to scale the Cauchy stress tensor to if the state of stress exceeds the yield surface.
This approach to plasticity leads to inaccurate shell thickness updates and stresses after yielding. This is
the only model in SOL 700 for plane stress that does not default to an iterative approach.
For MD Nastran SOL SEQ 106, 129, and 600, this material model is similar to using the MATS1 option
to define an elastic-plastic material.
or
ε eff
p
> ε max
p
(3-67)
92
Materials
s ij = 0 (3-68)
for all time. The failed element can only carry loads in compression.
σ y = ( A + Bε p ) ( 1 + C ln ε∗ ) ( 1 – T∗ m )
n ·
(3-69)
T – T room
T∗ = -------------------------------
T melt – T room
Constants for a variety of materials are provided in Johnson and Cook [1983].
Due to the nonlinearity in the dependence of flow stress on plastic strain, an accurate value of the flow
stress requires iteration for the increment in plastic strain. However, by using a Taylor series expansion
with linearization about the current time, we can solve for σ y with sufficient accuracy to avoid iteration.
ε f = [ D 1 + D 2 exp D 3 s∗ ] [ 1 + D 4 ln ε∗ ] [ 1 + D 5 T∗ ] (3-70)
where Di , i = 1 ,… ,5 , are input constants and σ∗ is the ratio of pressure divided by effective stress:
Materials 93
Materials
p
σ∗ = -------- (3-71)
σ eff
Fracture occurs when the damage parameter,
Δε p
D = εf -
-------- (3-72)
σ y = kε n = k ( ε yp + ε p ) n (3-73)
where ε yp is the elastic strain to yield and ε p is the effective plastic strain (logarithmic).
A parameter, SIGY, in the input governs how the strain to yield is identified. If SIGY is set to zero, the
strain to yield if found by solving for the intersection of the linearly elastic loading equation with the
strain hardening equation:
σ = Eε
(3-74)
σ = lε n
which gives the elastic strain at yield as:
1
------------
= ---
E n–1
ε yp (3-75)
k
If SIGY yield is nonzero and greater than 0.02 then:
1
σy ---
ε yp -----
n
(3-76)
k
Strain rate is accounted for using the Cowper and Symonds model which scales the yield stress with
the factor
·
ε 1/p
1 + ---- (3-77)
C
·
where ε is the strain rate. A fully viscoplastic formulation is optional with this model which incorporates
the Cowper and Symonds formulation within the yield surface. An additional cost is incurred but the
improvement is results can be dramatic.
This material model is a subset of what may be specified through the MATEP option for SOL600.
Materials 95
Materials
where ε p is the effective plastic strain and E p is given in terms of Young’s modulus and the tangent
modulus by
EE t
E p = -------------- (3-80)
E – Et
Both Young's modulus and the tangent modulus may optionally be made functions of strain rate by
specifying a load curve ID giving their values as a function of strain rate. If these load curve ID's are input
as 0, then the constant values specified in the input are used.
Note: All load curves used to define quantities as a function of strain rate must have the same
number of points at the same strain rate values. This requirement is used to allow
vectorized interpolation to enhance the execution speed of this constitutive model.
This model also contains a simple mechanism for modeling material failure. This option is activated by
specifying a load curve ID defining the effective stress at failure as a function of strain rate. For solid
elements, once the effective stress exceeds the failure stress the element is deemed to have failed and is
removed from the solution. For shell elements the entire shell element is deemed to have failed if all
integration points through the thickness have an effective stress that exceeds the failure stress. After
failure the shell element is removed from the solution.
In addition to the above failure criterion, this material model also supports a shell element deletion
criterion based on the maximum stable time step size for the element, Δt max . Generally, Δt max goes
down as the element becomes more distorted. To assure stability of time integration, the global time step
is the minimum of the Δt max values calculated for all elements in the model. Using this option allows
the selective deletion of elements whose time step Δt max has fallen below the specified minimum time
96
Materials
step, Δt crit . Elements which are severely distorted often indicate that material has failed and supports
little load, but these same elements may have very small time steps and therefore control the cost of the
analysis. This option allows these highly distorted elements to be deleted from the calculation, and,
therefore, the analysis can proceed at a larger time step, and, thus, at a reduced cost. Deleted elements do
not carry any load, and are deleted from all applicable slide surface definitions. Clearly, this option must
be judiciously used to obtain accurate results at a minimum cost.
This material model is a subset of what may be specified through the MATEP option for SOL600.
Young's modulus, E , and Poisson's ratio, υ , are used for determining sliding interface parameters if the
rigid body interacts in a contact definition. Realistic values for these constants should be defined since
unrealistic values may contribute to numerical problem in contact.
• , S 12 shear strength
and C 2 are obtained from material strength measurement. α is defined by material shear stress-strain
measurements. In plane stress, the strain is given in terms of the stress as
1
ε 1 = ------ ( σ 1 – υ 1 σ 2 )
E1
1
ε 2 = ------ ( σ 2 – υ 2 σ 1 ) (3-81)
E2
1
2ε 12 = --------- τ 12 + ατ 13
G 12
τ 12
2
------------ + 3--- ατ 12 4
2G 12 4
τ = ----------------------------------
2
(3-82)
S 12
------------ + 3--- αS 4
2G 12 4 12
which is the ratio of the shear stress to the shear strength.
The matrix cracking failure criteria is determined from
σ2 2
F matrix = ------ + τ (3-83)
S2
where failure is assumed whenever F matrix > 1 . If F matrix > 1 , then the material constants E2 ,
G 12 , υ , and υ 2 are set to zero.
98
Materials
σ2 2 C2 2 σ2
F comp = ----------- + ----------- – 1 ------ + τ (3-84)
2S 12 2S 12 C2
where failure is assumed whenever F comp > 1 . If F comp > 1 , then the material constants E2 , υ 1 ,
and υ 2 are set to zero.
σ1 2
F fiber = ------ + τ (3-85)
S1
Failure is assumed whenever F fiber > 1 . If F fiber > 1 , then the constants E 1 , E 2 , G 12 , υ 1 , and
υ 2 are set to zero.
1 σ2
φ = --- s ij s ij – -----y- ≤ 0 (3-86)
2 3
where
σ y = β [ σ 0 + f h ( ε eff
p
)] (3-87)
f h ( ε eff
p
) = E p ( ε eff
p
) (3-88)
1⁄2
3--- s * s * – σy
2 ij ij
Δε eff
p
= ---------------------------------------- (3-89)
3G + Ep
is the shear modulus and E p is the current plastic hardening modulus. The trial deviatoric stress state
s ij* is scaled back:
σy
s ijn + 1 = ---------------------------
1 ⁄
- s ij*
2
(3-90)
3--- s * s *
2 ij ij
For shell elements, the above equations apply, but with the addition of an iterative loop to solve for the
normal strain increment, such that the stress component normal to the mid surface of the shell element
approaches zero.
Three options to account for strain rate effects are possible:
• Strain rate may be accounted for using the Cowper and Symonds model which scales the yield
stress with the factor
ε- 1 ⁄ p
·
β = 1+ --- (3-91)
C
·
where ε is the strain rate.
• For complete generality a load curve, defining β , which scales the yield stress may be input
instead. In this curve the scale factor versus strain rate is defined.
• If different stress versus strain curves can be provided for various strain rates, the option using
the reference to a table definition can be used.
A fully viscoplastic formulation is optional which incorporates the different options above within the
yield surface. An additional cost is incurred over the simple scaling but the improvement is results can
be dramatic.
If a table ID is specified a curve ID is given for each strain rate. Intermediate values are found by
interpolating between curves.
σ
ij
Unloading and
reloading path
0 Strain –ε
ij
Curve extends into negative strain quadrant Unloading is based on the interpolated
since SOL 700 extrapolates using the two Young’s moduli which must provide an
end points. It is important that the extrapolation unloading tangent that exceeds the
does not extend into the negative stress region. loading tangent.
Note that in Figure 3-3, the “yield stress” at a volumetric strain of zero is nonzero. In the
load curve definition, the “time” value is the volumetric strain and the “function” value is the
yield stress.
• LCB, load curve number for sigma-bb versus either relative volume or volumetric strain
(default: LCB = LCA);
• LCC, the load curve number for sigma-cc versus either relative volume or volumetric strain
(default: LCC = LCA);
• LCS, the load curve number for shear stress versus either relative volume or volumetric strain
(default LCS = LCA);
• V f , relative volume at which the honeycomb is fully compacted;
• E aau , elastic modulus in the uncompressed configuration;
• E bbu , elastic modulus in the uncompressed configuration;
Materials 101
Materials
The behavior before compaction is orthotropic where the components of the stress tensor are uncoupled;
i.e., an a component of strain will generate resistance in the local a direction with no coupling to the local
b and c directions. The elastic moduli vary linearly with the relative volume from their initial values to
the fully compacted values:
E aa = E aau + β ( E – E aau )
E bb = E bbu + β ( E – E bbu )
E cc = E ccu + β ( E – E ccu )
G ab = G abu + β ( G – G abu )
G bc = G bcu + β ( G – G bcu )
G ca = G cau + β ( G – G cau )
(3-92)
where
1 – Vmin
β = max min -------------------
-, 1 , 0 (3-93)
1 – Vf
and G is the elastic shear modulus for the fully compacted honeycomb material
E
G = -------------------- (3-94)
2(1 + ν)
V is defined as the ratio of the current volume over the initial volume; typically,
The relative volume
V = 1 at the beginning of a calculation. The relative volume, V min , is the minimum value reached
during the calculation.
102
Materials
The load curves define the magnitude of the average stress as the material changes density (relative
volume). Each curve related to this model must have the same number of points and the same abscissa
values. There are two ways to define these curves: as a function of relative volume V , or as a function
of volumetric strain defined as:
εV = 1 – V (3-95)
In the former, the first value in the curve should correspond to a value of relative volume slightly less
than the fully compacted value. In the latter, the first value in the curve should be less than or equal to
zero corresponding to tension and should increase to full compaction.
Note: When defining the curves, care should be taken that the extrapolated values do not lead to
negative yield stresses.
At the beginning of the stress update we transform each element’s stresses and strain rates into the local
element coordinate system. For the uncompacted material, the trial stress components are updated using
the elastic interpolated moduli according to:
trial
n+1 n
σ aa = σ aa + E aa Δε aa
trial
n+1 n
σ bb = σ bb + E bb Δε bb
trial
n+1 n
σ cc = σ cc + E cc Δε cc
trial
n+1 n
σ ab = σ ab + E ab Δε ab
trial
n+1 n
σ bc = σ bc + E bc Δε bc
trial
n+1 n
σ ca = σ ca + E ca Δε ca
(3-96)
Then we independently check each component of the updated stresses to ensure that they do not exceed
the permissible values determined from the load curves; e.g., if
then
λσ ijn + 1
trial
The parameter λ is either unity or a value taken from the load curve number, LCSR, that defines λ
as a function of strain rate. Strain rate is defined here as the Euclidean norm of the deviatoric strain
rate tensor.
For fully compacted material we assume that the material behavior is elastic-perfectly plastic and updated
the stress components according to:
n+1⁄2
s ijtrial = s ijn + 2GΔε ijdev (3-99)
1
Δε ijdev = Δε ij – --- Δε kk δ ij (3-100)
3
We next check to see if the yield stress for the fully compacted material is exceeded by comparing:
1⁄2
= --- s ijtrial s ijtrial
trial 3
s eff (3-101)
2
the effective trial stress, to the yield stress σ y . If the effective trial stress exceeds the yield stress, we
simply scale back the stress components to the yield surface:
σ y trial
s ijn + 1 = ----------
-s
trial ij
(3-102)
s eff
We can now update the pressure using the elastic bulk modulus, K:
n+1⁄2
p n + 1 = p n – KΔε kk
E
K = -----------------------
3 ( 1 – 2ν ) (3-103)
and obtain the final value for the Cauchy stress:
After completing the stress update, we transform the stresses back to the global configuration.
104
Materials
W ( I 1 ,I 2 ,I 2 ) = A ( I 1 – 3 ) + B ( I 2 – 3 ) + C ---2- – 1 + D ( I 3 – 1 ) 2
1
(3-105)
I
3
where
C = .5∗ A + B (3-106)
A ( 5υ – 2 ) + B ( 11υ – 5 )
D = ---------------------------------------------------------- (3-107)
2 ( 1 – 2υ )
υ = Poisson’s ratio
2 2 2
I1 = λ 1 + λ 2 + λ3
2 2 2 2 2 2
I2 = λ 1 λ 2 + λ 2 λ 3 + λ3 λ 1
2 2 2
I3 = λ1 λ2 λ3
(3-108)
Recommended values for Poisson’s ratio are between .490 and .495 or higher. Lower values may lead to
instabilities. In the derivation of the constants C and D incompressibility is assumed.
elements, lend themselves very cleanly to a resultant formulation. The elements are set by the
SHELLFORM parameter.
Membrane σy
h
---
2
ny = h y
σ dζ = hσ y
– ---
2
Bending σy
h
---
h2
2
my = σ ζ dζ
h--- y
= ----- σ y
– 4
2
In applying this model to shell elements the resultants are updated incrementally using the midplane
strains ε m and curvatures κ:
Δn = ΔtCε m (3-109)
h3
Δm = Δt ------ Cκ (3-110)
12
where the plane stress constitutive matrix is given in terms of Young’s Modulus E and Poisson’s ratio
ν as:
2 – m m + m 2 + 3m 2
m = mxx (3-111)
xx yy yy xy
Defining
2 – n n + n 2 + 3n 2
n = n xx (3-112)
xx yy yy xy
2 – m m + m 2 + 3m 2
m = mxx (3-113)
xx yy yy xy
1 1
mn = m xx n xx – --- m xx n yy – --- n xx m yy + m y n y + 3m xy n xy (3-114)
2 2
the Ilyushin yield function becomes
4 mn 16m-
f ( m, n ) = n + ------------- + --------- ≤ n y2 = h 2 σ y2 (3-115)
h 3 h2
106
Materials
In our implementation we update the resultants elastically and check to see if the yield condition
is violated:
f ( m, n ) > n y2 (3-116)
n y2
α = ----------------- (3-117)
f ( m, n )
We update the yield stress incrementally:
σ yn + 1 = σ yn + E p Δε plastic
eff
(3-118)
where E p is the plastic hardening modulus which in incremental plastic strain is approximated by
f ( m, n ) – n y
Δε plastic
eff = -------------------------------
- (3-119)
h ( 3G + E p )
Kennedy, et. al., report that this model predicts results that may be too stiff; users of this model should
proceed cautiously.
In applying this material model to the Belytschko beam, the flow rule changes to
4m̂ y2 4m̂ z2
f ( m, n ) = ˆf x2 + ---------- + ---------- ≤ n y2 = A 2 σ y2 (3-120)
3I yy 3I zz
have been updated elastically according to Equations (3-112) through (3-114). The yield condition is
checked Equation (3-116), and if it is violated, the resultants are scaled as described above.
This model is frequently applied to beams with nonrectangular cross sections. The accuracy of the results
obtained should be viewed with some healthy suspicion. No workhardening is available with this model.
constraint any form of characteristic may be used including flat or falling curves. Different load curves
and scale factors may be specified at each node and about each of the local s and t axes.
Membrane σy
h
---
2
ny = h y
σ dζ = hσ y
– ---
2
Bending σy
h
---
h2
2
my = σ ζ dζ
h--- y
= ----- σ y
– 4
2
It is not possible to make the plastic moment vary with axial load.
M8
M7
M6
M5
M4
M3
Force
M2
M1
Displacement
For an intermediate value of the end moment, MD Nastran SOL 700 interpolates between the curves to
determine the allowable force.
A co-rotational technique and moment-curvature relations are used to compute the internal forces. The
co-rotational technique will not be treated here as we will focus solely on the internal force update and
computing the tangent stiffness. For this, we use the notation:
E = Young’s modulus
G = Shear modulus
A = Cross sectional area
As = Effective area in shear
ln = Reference length of beam
l n + 1= Current length of beam
I yy = Second moment of inertia about y
I zz = Second moment of inertia about z
J = Polar moment of inertia
ei = ith local base vector in the current configuration
yI = nodal vector in y direction at node I in the current configuration
zI = nodal vector in z direction at node I in the current configuration
Materials 109
Materials
We emphasize that the local y and z base vectors in the reference configuration always coincide with the
corresponding nodal vectors. The nodal vectors in the current configuration are updated using the
Hughes-Winget formula while the base vectors are computed from the current geometry of the element
and the current nodal vectors.
Internal Forces
Elastic Update
In the local system for a beam connected by nodes I and J, the axial force is updated as
where
EA-
K ael = ------ (3-122)
ln
δ = l n + 1 – l n .. (3-123)
where
GJ-
K tel = ------ (3-125)
ln
1
θ t = --- eTl ( y I × y J + z I × z J ) . (3-126)
2
The bending moments are updated as
mel n el
y = my + Ay θy (3-127)
mel n el
z = mz + Az θz (3-128)
where
1 EI **- 4 + φ* 2 – φ *
A *el = --------------- --------- (3-129)
1 + φ* l n 2 – φ 4 + φ
* *
110
Materials
12EI **
φ * = ------------------- (3-130)
GA * l n t n
Plastic Correction
After the elastic update the state of force is checked for yielding as follows. As a preliminary note we
emphasize that whenever yielding does not occur the elastic stiffnesses and forces are taken as the new
stiffnesses and forces.
The yield moments in direction i at node I as functions of plastic rotations are denoted mYiI ( θ iIP ) . This
function is given by the user but also depends on whether a plastic hinge has been created. The theory for
plastic hinges is given in the LS-DYNA Keyword User’s Manual [Hallquist 2003] and is not treated here.
Whenever the elastic moment exceeds the plastic moment, the plastic rotations are updated as
m iIel – mYiI
θ iIP ( n + 1 ) = θ iIP ( n ) + --------------------------------------------------------------- (3-133)
∂mYiI
max 0.001, A iel( II ) + ---------- -
∂θ iIP
and the moment is reduced to the yield moment
el
el A i ( II )
A in( II+ )1 = A i ( II ) 1 – α --------------------------------------------------------------- (3-135)
∂m iIY
max 0.001, Ai ( II ) + ----------el
-
∂θ iIP
where α ≤ 1 is a parameter chosen such that the moment-rotation matrix remains positive definite.
Materials 111
Materials
The yield moment in torsion is given by mYt ( θPt ) and is provided by the user. If the elastic torsional
moment exceeds this value, the plastic torsional rotation is updated as
m tel – mYt
θ tP ( n + 1 ) = θ tP ( n ) + ------------------------------------------------------------
- (3-136)
∂m t Y
max 0.0001, K tel + ----------
∂θ tP
and the moment is reduced to the yield moment
K
el
el
K tn + 1
t
-
= K t 1 – α ------------------------ (3-137)
Y
el ∂m t
K t + ---------P-
∂θ t
Y
el ∂f a
K an + 1 = max 0.05K a ,-------- (3-139)
∂ε
We neglect the influence of change in bending moments when computing this parameter.
Damping
Damping is introduced by adding a viscous term to the internal force on the form
112
Materials
δ
d θt
f v = D ----- (3-140)
dt θ y
θz
K ael
K tel
D = γ (3-141)
Ael
y
A zel
Transformation
The internal force vector in the global system is obtained through the transformation
f gn + 1 = Sf tn + 1 (3-142)
where
– e1 0 – e3 ⁄ l n + 1 –e3 ⁄ l n + 1 e2 ⁄ l n + 1 e2 ⁄ l n + 1
0 –e1 e2 0 e3 0
S = (3-143)
e1 0 e3 ⁄ l n + 1 e3 ⁄ l n + 1 –e2 ⁄ l n + 1 –e2 ⁄ l n + 1
0 e1 0 e2 0 e3
f an + 1
mnt + 1
f ln + 1 = (3-144)
m yn + 1
m zn + 1
Materials 113
Materials
Tangent Stiffness
Derivation
The tangent stiffness is derived from taking the variation of the internal force
δf gn + 1 = K geo δu + K mat δu
where
T
δu = δxTI δωTI δxTJ δωTj . (3-146)
There are two contributions to the tangent stiffness: geometrical and material. The geometrical
contribution is given (approximately) by
1
K geo = R ( f ln + 1 ⊗ I )W – ---------------------- Tf n + 1 L (3-147)
l n + 1 ln + 1 l
where
R1 0 R 3 ⁄ l n + 1 R 3 ⁄ l n + 1 – R2 ⁄ l n + 1 –R 2 ⁄ ( n + 1 )
0 R1 –R2 0 –R3 0
R = (3-148)
–R1 0 –R 3 ⁄ ln + 1 –R3 ⁄ ln + 1 R2 ⁄ ln + 1 R2 ⁄ ln + 1
0 –R1 0 –R2 0 –R3
0 0 –e3 –e3 e2 e2
T = 00 0 0 0 0 (3-150)
0 0 e 3 e3 – e2 –e 2
00 0 0 0 0
and I is the 3 by 3 identity matrix. We use ⊗ as the outer matrix product and define
Ri v = ei × v . (3-152)
K an + 1
K tn + 1
K = (3-154)
A yn + 1
A zn + 1
p = K ( θ – 3αξ S ε L
. (3-156)
t = 2G ( e – ξ S ε L n )
Here K and G are bulk and shear modulii, θ and e are volumetric and shear logarithmic strains and α
and ε L are constant material parameters. There is an option to define the bulk and shear modulii as
functions of the martensite fraction according to
K = KA + ξS ( K S – K A )
(3-157)
G = GA + ξ S ( G S – GK A
Materials 115
Materials
in case the stiffness of the martensite differs from that of the austenite. Furthermore, the unit vector n is
defined as
n = e ⁄ ( e + 10 –12 ) (3-158)
and a loading function is introduced as
F = 2G e + 3αKθ – βξ S (3-159)
where
β = ( 2G + 9α 2 K )ε L . (3-160)
For the evolution of the martensite fraction ξ S in the material, the following rule is adopted
F – R sAS > 0 ·
F -
·
F>0 ξS = –( 1 – ξS ) ------------------
F – R fAS
ξS < 1
F – R sSA < 0 ·
F -
·
F>0 ξ S = ξ S ------------------
F – R fSA
ξS > 0
. (3-161)
Here R sAS , R fAS , RsSA , and R fSA are constant material parameters. The Cauchy stress is finally
obtained as
σ = τ⁄J (3-162)
where J is the Jacobian of the deformation.
F n is known from time t n and the deformation gradient at time t n + 1 , F , is known. We form the left
Cauchy-Green tensor as B = FF T which is diagonalized to obtain the principal values and directions
Λdiag ( λ i ) and Q . The volumetric and principal shear logarithmic strains are given by
116
Materials
θ = log ( J )
e i = log ( λ i ⁄ J 1 ⁄ 3
where
J = λ1 λ2 λ3
n
is the total Jacobian of the deformation. Using Equation (3-159) with ξ S = ξ S , a value F trial of the
loading function can be computed. The discrete counterpart of Equation (3-161) becomes
F trial – R sAS > 0
F trial – βΔξ S – min ( max ( F n ,RsAS ) ,R fAS )
F trial – F n > 0 Δξ S = – ( 1 – ξ S – Δξ S ) ------------------------------------------------------------------------------------------------
n
-
AS
F trial – βΔξ S – R f
ξ Sn < 1
F trial – R sSA > 0
F trial – βΔξ S – min ( max ( F n ,R sSA ) ,R fSA )
F trial – F n > 0 Δξ S = ( ξ S + Δξ S ) ------------------------------------------------------------------------------------------------
n
-
trial – βΔξ – R SA
F S f
ξ Sn < 1
(3-163)
n+1
If none of the two conditions to the left are satisfied, set ξ S = ξ Sn , F n + 1 = F trial , and compute
the stress σ n + 1 using Equations (3-155), (3-156), (3-157), (3-162) and ξ S = ξ Sn . When phase
transformation occurs according to a condition to the left, the corresponding equation to the right is
solved for Δξ S . If the bulk and shear modulii are constan,t this is an easy task. Otherwise, F trial as
well as β depends on this parameter and makes things a bit more tricky. We have that
ES – EA
F trial = F ntrial 1 + ------------------ Δξ S
En
ES – EA
β = β n 1 + ------------------ Δξ S
En
.
where E S and EA are Young’s modulii for martensite and austenite, respectively. The subscript n is
introduced for constant quantities evaluated at time t n . To simplify the upcoming expressions, these
relations are written
Materials 117
Materials
β = β n + ΔβΔξ S
(3-164)
respectively, where we have for simplicity set
Inserting these expressions into Equation (3-163) results in
˜n
f ( Δξ S ) = Δβ ( 1 – ξ Sn )Δξ S2 + ( R fAS – F AS + ( β n – ΔF trial ) ( 1 – ξ Sn ) )Δξ S +
˜n
( 1 – ξ Sn ) ( F AS – F ntrial ) = 0
(3-165)
and
˜n
f ( Δξ S ) = Δβξ Sn Δξ S2 + ( FSA – R fSA + ( β n – ΔF trial )ξ Sn )Δξ S +
˜n
ξ Sn ( F SA – F ntrial ) = 0
˜n
F AS = min ( max ( F n, RsAS ) ,R fAS )
˜n
F SA = min ( max ( F n, RfSA ) ,R sSA )
.
The solutions to these equations are approximated with two Newton iterations starting in the point
Δξ S = 0 .. Now set ξ Sn + 1 = min ( 1 ,max ( 0 ,ξ Sn + Δξ S ) ) and compute σ n + 1 and F n + 1
according to Equations (3-155), (3-156), (3-157), (3-158), (3-160) and ξ S = ξ Sn + 1 .
δp = K ( δθ – 3αδξ S ε L ) + δK ( θ – 3αξ S ε L )
δt = 2G ( δe – δξS ε L n – ξ S ε L δn ) + 2δG ( e – ξ S ε L n )
. (3-166)
The variation of the unit vector in Equation (3-157) can be written
1
δn = -------------------------- ( I – n ⊗ n )δe
e + 10 – 12
118
Materials
where I is the fourth order identity tensor. For the variation of martensite fraction, we introduce the
indicator parameters H AS and H SA that should give information of the probability of phase
transformation occurring in the next stress update. Set initially H AS = H SA = 0 and change them
according to
F trial – R sAS > 0
F trial – F n > 0 H = 1
AS
ξ Sn + Δξ S ≤ 1
F trial – R sSA > 0
F trial – F n > 0 H = 1
SA
ξ Sn + Δξ S ≥ 0
using the quantities computed in the previous stress update. For the variation of the martensite fraction
we take variations of Equations(3-165) and (3-166) with
δξ S = γ ( 2Gn:δe + 3αKδθ )
where
n
( 1 – ξ S )H AS ξ Sn H SA
γ = ------------------------------------------------------------------------------------
- + ----------------------------------------------------------------------
-.
˜n ˜n
R fAS – F AS + ( β n – ΔF ntrial ) ( 1 – ξ Sn ) F SA – R fSA + ( β n – ΔFntrial )ξ Sn
As can be seen, we use the value of γ obtained in the previous stress update since this is easier to
implement and will probably give a good indication of the current value of this parameter.
The variation of the material parameters K and G results in
δK = ( K S – K A )δξ S
δG = ( G S – G A )δξ
and, finally, using the identities
Materials 119
Materials
n:δe = n:δε
δθ = i:δε
δτ = iδp + δt
results in
ξ S ε L dev
δτ = 2G 1 – -------------------------- I + K ( 1 – 9α 2 Kγε L + 3αγ ( K S – K A ) ( θ – 3αξ S ε L ) )i ⊗ i +
– 12
e + 10
2γG ( K S – K A ) ( θ – 3αξ S ε L )i ⊗ n + 6γαK ( G S – G A ) ( e – ξ S ε L )n ⊗ i +
ξS εL
2G ----------------------------------- – 2Gγε L + 2γ ( G S – G A ) ( e – ξ D ε L )
e + 1 – 10 –12
n ⊗ n – 6KGαγε L ( i ⊗ n + n ⊗ i ) }δε
where I dev is the fourth order deviatoric identity tensor. In general this tangent is not symmetric because
of the terms on the second line in the expression above. We simply use a symmetrization of the tangent
stiffness above in the implementation. Furthermore, we transform the tangent to a tangent closer related
to the one that should be used in the MD Nastran SOL 700 implementation,
ξS εL
–1
2G 1 – --------------------------
– 12
I dev + K ( 1 – 9α 2 Kγε L + 3αγ ( K S – K A ) ( θ – 3αξ S ε L ) )i ⊗ i +
e + 10
K S – K A ) ( θ – 3αξ S ε L ) + 3γαK ( G S – G A ) ( e – ξ S ε L ) – 6KGαγε L ) ( i ⊗ n + n ⊗ i ) +
ξS εL n ⊗ n δε
2G -------------------------- – 2Gγε + 2γ ( G – G ) ( e – ξ ε )
e + 10 –12 L S A S L
The strain energy functional, U , is defined in terms of the input constants as:
where the strain invariants can be expressed in terms of the deformation gradient matrix, F ij , and the
Green-St. Venant strain tensor, E ij :
120
Materials
J = F ij
I 1 = E ii . (3-168)
1--- ij
I 2 = δ pq E pi Eq
2
The derivative of U with respect to a component of strain gives the corresponding component of stress
∂U
S ij = --------- (3-169)
∂E ij
where, S ij , is the second Piola-Kirchhoff stress tensor which is transformed into the Cauchy
stress tensor:
ρ ∂x i ∂x j
σ ij = ----- --------- -------- S kl (3-170)
ρ o ∂X k ∂X l
is also included which can be used to reduce the tendency for these material/elements to be crushed when
the no-compression option is invoked.
If the airbag material is to be approximated as an isotropic elastic material, then only one Young’s
modulus and Poisson’s ratio should be defined. The elastic approximation is very efficient because the
local transformations to the material coordinate system may be skipped. If orthotropic constants are
defined, it is very important to consider the orientation of the local material system and use great care in
setting up the finite element mesh.
If the reference configuration of the airbag is taken as the folded configuration, the geometrical accuracy
of the deployed bag will be affected by both the stretching and the compression of elements during the
folding process. Such element distortions are very difficult to avoid in a folded bag. By reading in a
reference configuration, such as the final unstretched configuration of a deployed bag, any distortions in
the initial geometry of the folded bag will have no effect on the final geometry of the inflated bag. This
is because the stresses depend only on the deformation gradient matrix:
∂x i
F ij = -------- (3-171)
∂X j
where the choice of X j may coincide with the folded or unfold configurations. It is this unfolded
configuration which may be specified here. When the reference geometry is used, then the
no-compression option should be active. With the reference geometry, it is possible to shrink the airbag
and then perform the inflation. Although the elements in the shrunken bag are very small, the time step
can be based on the reference geometry so a very reasonable time step size is obtained. The reference
geometry based time step size is optional in the input.
The parameters fabric leakage coefficient, FLC, fabric area coefficient, FAC, and effective leakage area,
ELA, for the fabric in contact with the structure are optional for the Wang-Nefske and hybrid inflation
models. It is possible for the airbag to be constructed of multiple fabrics having different values of
porosity and permeability. The gas leakage through the airbag fabric then requires an accurate
determination of the areas by part ID available for leakage. The leakage area may change over time due
to stretching of the airbag fabric or blockage when the outer surface of the bag is in contact with the
structure. MD Nastran SOL 700 can check the interaction of the bag with the structure and split the areas
into regions that are blocked and unblocked depending on whether the regions are in contact or not,
respectively. Typically, the parameters, FLC and FAC, must be determined experimentally and their
variation with time and pressure are optional inputs that allow for maximum modeling flexibility.
compression. See [Matzenmiller and Schweizerhof, 1990]. This model is only valid for thin
shell elements.
The Chang/Chang criteria is given as follows:
for the tensile fiber mode,
σ aa 2
σ aa > 0 then e c2 = -------- – 1 > 0 failed , (3-173)
Xc < 0 elastic
E a = ν ba = ν ab = 0
for the tensile matrix mode,
σ bb2 σ ab 2 ( Y c – Y t )σ bb
2
e md = ---------- + -------- + ------------------------------ – 1 > 0 failed (3-176)
Yc Yt Sc Yc Yt < 0 elastic
For β = 1 , we get the original criterion of Hashin [1980] in the tensile fiber mode.
For β = 0 , we get the maximum stress criterion which is found to compare better to experiments.
Failure can occur in any of four different ways:
Materials 123
Materials
d3plot
History Variable Description Value Component
1. ef(i) tensile fiber mode 81
2. ec(i) compressive fiber mode 1 - elastic 82
3. em(i) tensile matrix mode 0 - failed 83
4. ed(i) compressive matrix mode 84
5. efail max[ef(ip)] 85
6. dam damage parameter -1 - element intact 86
10-8 - element in crashfront
+1 - element failed
124
Materials
These variables can be plotted in MD Patran as element components 81, 82, ..., 80+ NEIPS. The
following components, defined by the sum of failure indicators over all through-thickness integration
points, are stored as element component 7 instead of the effective plastic strain:
unloading and rate effects are due to Chang, et. al., [1994]. These latter additions have greatly expanded
the usefulness of this model.
Unloading
Curves
Strain Strain
Figure 3-7 Behavior of the Low-density Urethane Foam Model
The model uses tabulated input data for the loading curve where the nominal stresses are defined as a
function of the elongations, ε i , which are defined in terms of the principal stretches, λ i , as:
εi = λ i – 1 (3-177)
The stretch ratios are found by solving for the eigenvalues of the left stretch tensor, V ij , which is
obtained via a polar decomposition of the deformation gradient matrix, F ij :
F ij = R ik U kj = V ik R kj (3-178)
The update of Vij follows the numerically stable approach of Taylor and Flanagan [1989]. After solving
for the principal stretches, the elongations are computed and, if the elongations are compressive, the
corresponding values of the nominal stresses, τ i , are interpolated. If the elongations are tensile, the
nominal stresses are given by
τ i = Eε i (3-179)
τi
σ i = ---------- (3-180)
λiλk
The stresses are then transformed back into the global system for the nodal force calculations.
126
Materials
When hysteretic unloading is used, the reloading will follow the unloading curve if the decay constant,
β , is set to zero. If β is nonzero the decay to the original loading curve is governed by the expression:
1 – e – βt (3-181)
The bulk viscosity, which generates a rate dependent pressure, may cause an unexpected volumetric
response and, consequently, it is optional with this model.
Rate effects are accounted for through linear viscoelasticity by a convolution integral of the form
∂ε kl
t
σ ij = g (t – τ ) --------- (3-182)
0 ijkl ∂τ
where g ijkl ( t – τ ) is the relaxation function. The stress tensor, , σ ijr augments the stresses determined
from the foam, σ ijf ; consequently, the final stress, σ ij , is taken as the summation of the two
contributions:
Since we wish to include only simple rate effects, the relaxation function is represented by one term from
the Prony series:
g ( t ) = α0 + α m e – βt (3-184)
m=1
given by,
g ( t ) = E d e –β1 t (3-185)
This model is effectively a Maxwell fluid which consists of a damper and spring in series. We
characterize this in the input by a Young's modulus, E d , and decay constant, β 1 . The formulation is
performed in the local system of principal stretches where only the principal values of stress are
computed and triaxial coupling is avoided. Consequently, the one-dimensional nature of this foam
material is unaffected by this addition of rate effects. The addition of rate effects necessitates twelve
additional history variables per integration point. The cost and memory overhead of this model comes
primarily from the need to “remember” the local system of principal stretches.
effective strain (compression/tension including shear) is reached. The stress limits are factors used to
limit the stress in the softening part to a given value,
thus, the damage value is slightly modified such that elastoplastic like behavior is achieved with the
threshold stress. As a factor for SLIMxx a number between 0.0 and 1.0 is possible. With a factor of
1.0, the stress remains at a maximum value identical to the strength, which is similar to ideal elastoplastic
SLIMTx is often reasonable; however, for compression
behavior. For tensile failure a small value for
SLIMCx = 1.0 is preferred. This is also valid for the corresponding shear value. If SLIMxx is
smaller than 1.0, then localization can be observed depending on the total behavior of the lay-up. If the
user is intentionally using SLIMxx < 1.0 , it is generally recommended to avoid a drop to zero and set
the value to something in between 0.05 and 0.10.
Then elastoplastic behavior is achieved in the limit which often leads to less numerical problems.
Defaults for SLIMxx = 1.0E-8 .
The crashfront-algorithm is started if and only if a value for TSIZE (time step size, with element
elimination after the actual time step becomes smaller than TSIZE) is input
The damage parameters can be written to the postprocessing database for each integration point as the
first three additional element variables and can be visualized.
Material models with FS=1 or FS=-1 are favorable for complete laminates and fabrics, as all directions
are treated in a similar fashion.
For material model FS=1, an interaction between normal stresses and shear stresses is assumed for the
evolution of damage in the a- and b-directions. For the shear damage is always the maximum value of
the damage from the criterion in a- or b- direction is taken.
For material model FS=-1, it is assumed that the damage evolution is independent of any of the other
stresses. A coupling is present only via the elastic material parameters and the complete structure.
In tensile and compression directions and in a- as well as in b- direction, different failure surfaces can be
assumed. The damage values, however, increase only when the loading direction changes.
In addition, a stress limiter can be used to keep the stress constant via the SLIMS parameter. This value
must be less than or equal to 1.0 and positive, which leads to an elastoplastic behavior for the shear part.
The default is 1.0E-08, assuming almost brittle failure once the strength limit SC is reached.
τ
SC
TAU1
SLIMS*SC
γ
GAMMA1 GMS
Figure 3-8 Stress-strain Diagram for Shear
E1
V2
E
2
E 1t = E 1 ( V –n 1 )
(3-187)
V 2t = V 2 ( abs ( 1 – V ) ) n
where V is the relative volume defined by the ratio of the current to initial volume. Typical values are
(units of N ,mm, s )
E 1 = 0.0036
n 1 = 4.0
V 2 = 0.0015
E 2 = 100.0
n 2 = 0.2
ν = 0.05
·n
σ ijtrial = σ ijn + Eε ij + 1 ⁄ 2 Δt n + 1 ⁄ 2 (3-188)
The magnitudes of the principal values, σ itrial , i = 1 ,3 , are then checked to see if the yield stress,
σ y , is exceeded and if so they are scaled back to the yield surface:
σ itrial
if σ y < σ itrial then σ in + 1 = σ y --------------- (3-189)
σ itrial
130
Materials
After the principal values are scaled, the stress tensor is transformed back into the global system. As seen
in Figure 3-10, the yield stress is a function of the natural logarithm of the relative volume, V ; i.e., the
volumetric strain.
σ ij a
d E
c b
Volumetric Strain – ln V
Figure 3-10 Yield Stress Versus Volumetric Strain Curve for the Crushable Foam
μ( T) -
---------------- = B1 + B2 T + B3 T 2 (3-193)
μ ( 293 )
The relationship between heat capacity (specific heat) and temperature may be characterized by a cubic
polynomial equation as follows:
Cp = G1 + G2 T + G3 T 2 + G4 T 3 (3-194)
A fully viscoplastic formulation is optional. An additional cost is incurred but the improvement in results
can be dramatic.
lookup, if the origin of the load curve is at [0,0], as shown in Figure 3-11b, and the tension and
compression responses are symmetric.
R R
E E
S S
U U
L L
T T
A A
N N
T T
DISPLACEMENT
DISPLACEMENT
(a) (b)
R
E
S
U
L
T
A
N
T
PLASTIC DISPLACEMENT
Figure 3-12 Resultant Forces and Moments Limited by the Yield Definition
Materials 133
Materials
Catastrophic failure, based on force resultants, occurs if the following inequality is satisfied:
Fr 2 Fs 2 Ft 2 Mr 2 Ms 2 Mt 2
---------- + ---------- + ---------- - + ------------ + ------------ – 1 ≥ 0
+ ----------- (3-195)
F rfail F sfail F tfail M rfail M sfail M tfail
ur 2 u s 2 u t 2 θ r 2 θs 2 θ t 2
- + ---------- + ---------- + ---------- + ---------- + ---------- – 1 ≥ 0
--------- (3-196)
u rfail u sfail u tfail θ rfail θ sfail θ tfail
After failure, the discrete element is deleted. If failure is included, either one or both of the criteria may
be used.
C Ap 2
F = SF KA p V p -----t-l + C 2 V p ρ fluid ---------t- – 1 – f ( s + s 0 ) + V p g ( s + s 0 ) } (3-197)
A0 CA0
where K is a user defined constant or a tabulated function of the absolute value of the relative velocity,
V p is the piston's relative velocity, C is the discharge coefficient, A p is the piston area, A 0t is the total
134
Materials
open areas of orifices at time t , ρ fluid is the fluid density, C 1 is the coefficient for the linear term, and
C 2 is the coefficient for the quadratic term.
In the implementation, the orifices are assumed to be circular with partial covering by the orifice
controller. As the piston closes, the closure of the orifice is gradual. This gradual closure is taken into
account to insure a smooth response. If the piston stroke is exceeded, the stiffness value, k , limits further
movement; i.e., if the damper bottoms out in tension or compression, the damper forces are calculated by
replacing the damper by a bottoming out spring and damper, k and c , respectively. The piston stroke
must exceed the initial length of the beam element. The time step calculation is based in part on the
stiffness value of the bottoming out spring. A typical force versus displacement curve at constant relative
Materials 135
Materials
velocity with only the linear velocity term active is shown in Figure 3-14. The factor, SF , which scales
the force defaults to 1.0 and is analogous to the adjusting ring on the damper.
F
O
R
C
E Last orfice
closes
Displacement
As the damper is compressed two actions contribute to the force that develops. First, the gas is
adiabatically compressed into a smaller volume. Secondly, oil is forced through an orifice. A profiled pin
may occupy some of the cross-sectional area of the orifice; thus, the orifice area available for the oil
varies with the stroke. The force is assumed proportional to the square of the velocity and inversely
proportional to the available area. The equation for this element is:
C0 n
F = SCLF ⋅ K h ----- + P 0 --------------- – P a ⋅ A p
V 2
(3-198)
a0 C0 – S
136
Materials
where S is the element deflection and V is the relative velocity across the element.
E ⋅ area
K = ------------------------------------------------------------- (3-201)
( initial length – offset )
The area and offset are defined on either the cross section or element cards in the SOL 700 input. For a
slack cable the offset should be input as a negative length. For an initial tensile force the offset should be
positive. If a load curve is specified, the Young’s modulus will be ignored and the load curve will be used
instead. The points on the load curve are defined as engineering stress versus engineering strain; i.e., the
change in length over the initial length. The unloading behavior follows the loading.
∂ε kl
0 gijkl ( t – τ ) --------
t
σ ijr = - dτ (3-202)
∂τ
where g ijkl ( t – τ ) is the relaxation function. The stress tensor, σ ijr , augments the stresses
determined from the foam, σ ijf ; consequently, the final stress, σ ij , is taken as the summation of the
two contributions:
Since we wish to include only simple rate effects, the relaxation function is represented by up to six terms
of the Prony series:
g ( t ) = α0 + α m e –βt (3-204)
m=1
This model is effectively a Maxwell fluid which consists of a dampers and springs in series. The
formulation is performed in the local system of principal stretches where only the principal values of
stress are computed and triaxial coupling is avoided. Consequently, the one-dimensional nature of this
foam material is unaffected by this addition of rate effects. The addition of rate effects necessitates 42
additional history variables per integration point. The cost and memory overhead of this model comes
primarily from the need to “remember” the local system of principal stretches and the evaluation of the
viscous stress components
· · ΔL· -
F = F 0 + Kf ( ΔL ) 1 + C1 ⋅ ΔL + C2 ⋅ sgn ( ΔL ) ln max 1. , -----------
DLE
+DΔL· + g ( ΔL )h ( ΔL· ) (3-206)
where C1 and C2 are damping coefficients for nonlinear behavior, DLE is a factor to scale time
units, and g ( ΔL ) . is an optional load curve defining a scale factor versus deflection for load curve ID,
h ( ΔL ⁄ dt ) .
In these equations, ΔL is the change in length
ΔL = currentlength – initiallength .
Failure can occur in either compression or tension based on displacement values of CDF and TDF,
respectively. After failure no forces are carried. Compressive failure does not apply if the spring is
initially zero length.
The cross sectional area is defined on the beam property card for the discrete beam elements. The
square root of this area is used as the contact thickness offset if these elements are included in the
contact treatment.
∂ε
t
σ ij = g (t – τ ) --------kl- dτ (3-207)
0 ijkl ∂τ
where g ijkl ( t – τ ) is the relaxation function.
If we wish to include only simple rate effects for the deviatoric stresses, the relaxation function is
represented by six terms from the Prony series:
We characterize this in the input by shear modulii, G i , and decay constants, β i . An arbitrary number of
terms, up to 6, may be used when applying the viscoelastic model.
Note: This curve defines stress versus time where time is defined on a logarithmic scale. For best
results, the points defined in the load curve should be equally spaced on the logarithmic
scale. Furthermore, the load curve should be smooth and defined in the positive quadrant.
If nonphysical values are determined by least squares fit, SOL 700 will terminate with an
error message after the initialization phase is completed. If the ramp time for loading is
included, then the relaxation which occurs during the loading phase is taken into account.
This effect may or may not be important.
For volumetric relaxation, the relaxation function is also represented by the Prony series in terms of
bulk modulii:
N
–βk t
k( t) = Km e m (3-209)
m=1
W H ( J ) , is included in the strain energy functional which is function of the relative volume, ( J ) ,
[Ogden, 1984]:
W ( J 1 ,J 2 ,J ) = C pq ( J 1 – 3 ) p ( J 2 – 3 ) q + W H ( J )
p ,q = 0
J 1 = I 1 I 3– 1 ⁄ 2
J 2 = I 2 I 3– 1 ⁄ 2
(3-210)
In order to prevent volumetric work from contributing to the hydrostatic work the first and second
invariants are modified as shown. This procedure is described in more detail by Sussman and Bathe
[1987]. For the Ogden model the energy equation is given as:
3 n
μj 1
W∗ = ----
-(λ – 1 ) + --- K ( J – 1 ) 2
*α 1
(3-211)
αj i 2
i = 1j = 1
where the asterisk ( * ) indicates that the volumetric effects have be eliminated from the principal
stretches, λ j* . See Ogden [1984] for more details.
Rate effects are taken into account through linear viscoelasticity by a convolution integral of the form:
t ∂ε kl
σ ij = 0 g ijkl ( t – τ ) --------- dτ
∂τ
(3-212)
∂Ekl
S ij G ijkl ( t – τ ) ---------- dτ
t
(3-213)
0 ∂τ
where g ijkl ( t – τ ) and G ijkl ( t – τ ) are the relaxation functions for the different stress measures. This
stress is added to the stress tensor determined from the strain energy functional.
If we wish to include only simple rate effects, the relaxation function is represented by six terms from the
Prony series:
g ( t ) = α0 + α m e – βt (3-214)
m=1
given by,
Materials 141
Materials
Gi e
–βi t
g(t) = (3-215)
j=1
This model is effectively a Maxwell fluid which consists of a dampers and springs in series. We
characterize this in the input by shear moduli, G i , and decay constants, β i . The viscoelastic behavior is
optional and an arbitrary number of terms may be used.
The constants ( a 0 , a 1 , a 2 ) govern the pressure sensitivity of the yield stress. Only the ratios between
these values are important - the absolute stress values are taken from the stressstrain curve.
The stress strain pairs ( ( γ1 ,τ1 ) , … ( γ5 ,τ5 ) ) define a shear stress versus shear strain curve. The first
point on the curve is assumed by default to be (0,0) and does not need to be entered. The slope of the
curve must decrease with increasing γ . Not all five points need be to be defined. This curve applies at
the reference pressure; at other pressures, the curve variesaccording to a 0 , a 1 , and a 2 as in the soil
and crushable foam model, Material 5.
The elastic moduli G and K are pressure sensitive.
G = G0 ( p – p 0 ) b
K = K0 ( p – p0 ) b
where G 0 and K 0 are the input values, p is the current pressure, p o the cut-off or referencepressure
(must be zero or negative). If p attempts to fall below p o (i.e., more tensile) the shear stresses are set to
zero and the pressure is set to p o . Thus, the material has no stiffness orstrength in tension. The pressure
in compression is calculated as follows:
142
Materials
1
------------
p = [ –K0 ln ( V ) ] b
1 –
where V is the relative volume; i.e., the ratio between the original and current volume.
γ τ τ r if γ ≥ 0
---- = ---- + α ----
γy τy τy
---γ- = ----
τ τ r if γ < 0
– α ----
γy τy τ y (3-216)
where γ is the shear strain and τ is the shear stress. The model approaches perfect plasticity as the stress
exponent r → ∞ . These equations must be augmented to correctly model unloading and reloading
·
material behavior. The first load reversal is detected by γγ < 0 . After the first reversal, the stress-strain
relationship is modified to
( γ – γ0 ) ( τ – τ0 ) ( τ – τ 0 ) rif γ ≥ 0
- + α -----------------
------------------ = ----------------- -
2γ y 2τ y 2τ y
( γ – γ0 ) ( τ – τ0 ) ( τ – τ 0 ) rif γ < 0
- – α -----------------
------------------ = ----------------- -
2γ y 2τ y 2τ
y (3-217)
where γ 0 and τ 0 represent the values of strain and stress at the point of load reversal. Subsequent load
p = K εv (3-218)
given by a load curve ID, and the initial yield stress, SIGY, multiplied by the Cowper-Symonds rate term
as follows:
·p
p ·p ε eff 1 ⁄ p
σ y ( ε eff ,ε eff ) = σ ys ( ε eff
p
) + SIGY ⋅ -
------- (3-220)
C
where the plastic strain rate is used. If SIGY=0, the following equation is used instead where the static
stress, σ ys ( ε eff
p
) , must be defined by a load curve:
·p
p ·p ε eff 1 ⁄ p
σ y ( ε eff ,ε eff ) = σ ys 1 + -------- (3-221)
C
This latter equation is always used if the viscoplastic option is off.
For complete generality a load curve (LCSR) to scale the yield stress may be input instead. In this curve
the scale factor versus strain rate is defined.
144
Materials
The constitutive properties for the damaged material are obtained from the undamaged material
properties. The amount of damage evolved is represented by the constant, ω , which varies from zero if
no damage has occurred to unity for complete rupture. For uniaxial loading, the nominal stress in the
damaged material is given by:
P
σ nominal = --- (3-222)
A
where P is the applied load and A is the surface area. The true stress is given by:
P
σ true = --------------------- (3-223)
A – A loss
where A loss is the void area. The damage variable can then be defined:
A loss 0≤ω≤1
ω = -----------
A (3-224)
In this model damage is defined in terms of plastic strain after the failure strain is exceeded:
ε eff
p
– ε failure
p
ω = ----------------------------------------
- if ε failure
p ≤ ε eff
p ≤ εp
rupture (3-225)
ε rapture
p
– ε failure
p
After exceeding the failure strain softening begins and continues until the rupture strain is reached.
By default, deletion of the element occurs when all integration points in the shell have failed.
Materials 145
Materials
Note in Figure 3-18 that the origin of the curve is at (0,0). It is permissible to input the failure strain, fs ,
as zero for this option. The nonlinear damage curve is useful for controlling the softening behavior after
the failure strain is reached.
σ yield
Nominal stress
after failure
p
0
ε eff
Failure Begins ω = 1
ω = 0
Rupture
1
Damage
p
ε eff – fs
Failure
E· ( t ) = E· L ( t ) + E· N ( t ) (3-227)
E· N is an expression for the past history of E N . A postulated constitutive equation may be written as:
∞
σ( t) = [ E tN ( τ ) ,S ( t ) ] dτ (3-228)
τ=0
∞
E tN ( τ ) = E N ( t – τ ) (3-229)
It is assumed that the material remembers only its immediate past, i.e., a neighborhood about τ = 0.
Therefore, an expansion of E tN ( τ ) in a Taylor series about τ = 0 yields:
∂ E tN
E tN ( τ ) = EN( 0 ) + - ( 0 ) dt
--------- (3-231)
∂t
Hence, the postulated constitutive equation becomes:
σ ( t ) = σ∗ ( E N ( t ), E· N ( t ), S ( t ) ) (3-232)
Materials 147
Materials
∂ E tN ·
where we have replaced ---------- by E N , and σ∗ is a function of its arguments.
∂t
For a special case,
σ ( t ) = σ∗ ( E· N ( t ), S ( t ) ) (3-233)
we may write
E· tN = f ( S ( t ), s ( t ) ) (3-234)
which states that the nonlinear strain rate is the function of stress and a state variable which represents
the history of loading. Therefore, the proposed kinetic equation for foam materials is:
σ tr ( σ S ) 2n0
E· N = -------- D 0 exp –c 0 ---------------
- (3-235)
σ ( σ 2)
where D 0, c 0 and n0 are material constants, and S is the overall state variable. If either D 0 = 0 or
c 0 → ∞ then the nonlinear strain rate vanishes.
S· ij = [ c 1 ( a ij R – c 2 S ij ) P + c 3 W n1 ( W· N ) n2 I ij ] R (3-236)
E· N - n3
R = 1 + c4 ---------- – 1 (3-237)
c5
· N)
P = tr ( σ E (3-238)
W = tr ( σ dE ) (3-239)
σ = ( σ ij σ ij ) 1 / 2
E· = ( E· ij E· ij ) 1 / 2 (3-240)
·
E N = ( E· ij E· ij ) 1 / 2
N N
148
Materials
In the implementation by Fu Chang the model was simplified such that the input constants a ij and the
state variables S ij are scalars.
W ( J 1 ,J 2 ,J ) = C pq ( J 1 – 3 ) p ( J 2 – 3 ) q + W H
p, q = 0
–1 ⁄ 3
J1 = I1 I3 (3-241)
–2 ⁄ 3
J2 = I2 I3
In order to prevent volumetric work from contributing to the hydrostatic work the first and second
invariants are modified as shown. This procedure is described in more detail by Sussman and
Bathe [1987].
Materials 149
Materials
The effects of confined air pressure in its overall response characteristics are included by augmenting the
stress state within the element by the air pressure.
where σ ijsk is the bulk skeletal stress and σ air is the air pressure computed from the equation:
p0 γ
σ air = – -------------------- (3-243)
1+γ–φ
where p 0 is the initial foam pressure usually taken as the atmospheric pressure and γ defines the
volumetric strain
γ = V – 1 + γ0 (3-244)
where V is the relative volume of the voids and γ 0 is the initial volumetric strain which is typically zero.
The rubber skeletal material is assumed to be incompressible.
Rate effects are taken into account through linear viscoelasticity by a convolution integral of the form:
t ∂ε kl
σ ij = 0 gijkl ( t – τ ) --------
∂τ
- dt (3-245)
or in terms of the second Piola-Kirchhoff stress, S ij , and Green's strain tensor, , Eij
t ∂ E kl
S ij = 0 G ijkl ( t – τ ) ---------- dt
∂τ
(3-246)
where g ijkl ( t – τ ) and G ijkl ( t – τ ) are the relaxation functions for the different stress measures. This
stress is added to the stress tensor determined from the strain energy functional.
Since we wish to include only simple rate effects, the relaxation function is represented by one term from
the Prony series:
g ( t ) = a0 + αm + e – βt (3-247)
m=1
given by
g ( t ) = E d e –β1 t . (3-248)
150
Materials
This model is effectively a Maxwell fluid which consists of a damper and spring in series. We
characterize this in the input by a shear modulus, G , and decay constant, β 1 .
F Y = F y ( Δ L plastic ) (3-249)
F T = F n + K ⋅ Δ L· ⋅ Δ t (3-250)
Y if F T > F Y
F = F
FT if F T ≤ F Y
(3-251)
The final force, which includes rate effects and damping, is given by:
· · Δ L· - ·
Fn + 1 = F ⋅ 1 + C 1 ⋅ Δ L + C 2 ⋅ sgn ( Δ L ) ln max 1. , ----------- h ( Δ L· )
+ D Δ L + g ( Δ L )(3-252)
DLE
where
It is advisable to include all spot welds which can be arbitrarily placed within the structure, which provide
the slave nodes, and spot welded materials, which define the master segments, within a single type 7 tied
interface. As a constraint method, multiple type 7 interfaces are treated independently which can lead to
significant problems if such interfaces share common nodal points. The offset option, “o 7”, should not
be used with spot welds.
Materials 153
Materials
The weld material is modeled with isotropic hardening plasticity coupled to two failure models. The first
model specifies a failure strain which fails each integration point in the spot weld independently. The
second model fails the entire weld if the resultants are outside of the failure surface defined by:
N rr 2 N rs 2 N rt 2 M rr 2 M ss 2
--------- T rr 2
- + ---------- + --------- + ----------- + ----------- – 1 --------
- = 0 (3-253)
N rr N rs N rt M rr M ss T rr
F F F F F F
where the numerators in the equation are the resultants calculated in the local coordinates of the cross
section, and the denominators are the values specified in the input. If the user defined parameter, NF,
which the number of force vectors stored for filtering, is nonzero the resultants are filtered before failure
is checked. The default value is set to zero which is generally recommended unless oscillatory resultant
forces are observed in the time history databases. Even though these welds should not oscillate
significantly, this option was added for consistency with the other spot weld options. NF affects the
storage since it is necessary to store the resultant forces as history variables.
If the failure strain is set to zero, the failure strain model is not used. In a similar manner, when the value
of a resultant at failure is set to zero, the corresponding term in the failure surface is ignored. For example,
if only N rrF is nonzero, the failure surface is reduced to N rr = N rrF . None, either, or both of the
failure models may be active depending on the specified input values.
The inertias of the spot welds are scaled during the first time step so that their stable time step size is Δt.
A strong compressive load on the spot weld at a later time may reduce the length of the spot weld so that
stable time step size drops below Δ t . If the value of Δ t is zero, mass scaling is not performed, and the
spot welds will probably limit the time step size. Under most circumstances, the inertias of the spot welds
are small enough that scaling them will have a negligible effect on the structural response and the use of
this option is encouraged.
Spotweld force history data is written into the SWFORC ASCII file. In this database the resultant
moments are not available, but they are in the binary time history database.
The constitutive properties for the damaged material are obtained from the undamaged material
properties. The amount of damage evolved is represented by the constant, ω , which varies from zero if
no damage has occurred to unity for complete rupture. For uniaxial loading, the nominal stress in the
damaged material is given by
P
σ nominal = --- (3-254)
A
where P is the applied load and A is the surface area. The true stress is given by:
P
σ true = --------------------- (3-255)
A – A loss
154
Materials
where A loss is the void area. The damage variable can then be defined:
A loss
ω = ----------- 0≤ω≤1
A (3-256)
In this model damage is defined in terms of plastic strain after the failure strain is exceeded:
p p
ε eff – ε failure
- if ε pfailure ≤ ε peff ≤ ε prupture
ω = --------------------------------------- (3-257)
p p
ε rupture – ε failure
After exceeding the failure strain softening begins and continues until the rupture strain is reached.
ur 2 us 2 ut 2 θt 2 θ s 2 θt 2
----------- + ----------- + ----------- + ----------- + ----------- + ----------- – 1. ≥ 0
u rtfail u stfail u ttfail θ rtfail θ stfail θ ttfail
u r 2 u s 2 ut 2 θt 2 θs 2 θt 2
- + ------------ + ------------ + ------------ + ------------ + ------------ – 1. ≥ 0
-----------
u rcfail u scfail u tcfail θrcfail θ scfail θ tcfail
(3-258)
Materials 155
Materials
After failure the discrete element is deleted. If failure is included either the tension failure or the
compression failure or both may be used.
Unloading Curve
R
R
E
E
S
S
U
U
L
L
T
T
A
A
N
N
T
T
Loading-unloading Curve
DISPLACEMENT DISPLACEMENT
Unload = 0 Unload = 1
R R
E E
S S
U U
L L
Unloading Curve
T T
A A
Unloading Curve N N
T T
DISPLACEMENT
DISPLACEMENT
OFFSET x Umin
Umin
Unload = 2 Unload = 3
where
and G is the elastic shear modulus for the fully compacted honeycomb material
E
G = --------------------- (3-261)
2(1 + υ)
The relative volume, ν , is defined as the ratio of the current volume over the initial volume, and
typically, V = 1 at the beginning of a calculation.
The load curves define the magnitude of the stress as the material undergoes deformation. The first value
in the curve should be less than or equal to zero corresponding to tension and increase to full compaction.
Care should be taken when defining the curves so the extrapolated values do not lead to negative
yield stresses.
At the beginning of the stress update we transform each element’s stresses and strain rates into the local
element coordinate system. For the uncompacted material, the trial stress components are updated using
the elastic interpolated modulii according to:
σ aa
n + 1 trial = σ n + E Δε σ ab = σ ab
n + 2 G Δε
n+1 trial
aa aa aa ab ab
σ bb = σ bb
n + E Δε σ bc = σ bc
n + 2 G Δε
n+1 trial n+1 trial
bb bb bc bc
σ cc
n+1 trial
= σ cc
n + E Δε
cc cc σ ca
n + 1 trial = σ n + 2 G Δε
ca ca ca
(3-262)
We then independently check each component of the updated stresses to ensure that they do not exceed
the permissible values determined from the load curves, e.g., if
σ ijn + 1 > λσ ij ( ε ij )
trial
(3-263)
then
λσ ijn + 1
trial
σ ijn + 1 = σ ij ( ε ij ) ----------------------
- (3-264)
σ in + 1
trial
The components of σ ij ( ε ij ) are defined by load curves. The parameter λ is either unity or a value taken
from the load curve number, LCSR, that defines λ as a function of strain-rate. Strain-rate is defined here
as the Euclidean norm of the deviatoric strain-rate tensor.
For fully compacted material we assume that the material behavior is elastic-perfectly plastic and updated
the stress components according to:
Materials 157
Materials
n+1⁄2
s ijtrial = s ijn + 2 G Δε ijdev (3-265)
1
Δε ijdev = Δε ij – --- Δε kk δ ij (3-266)
3
We now check to see if the yield stress for the fully compacted material is exceeded by comparing
1⁄2
= --- s ijtrial s ijtrial
trial 3
s eff (3-267)
2
the effective trial stress to the yield stress, σ y . If the effective trial stress exceeds the yield stress, we
simply scale back the stress components to the yield surface
σ y trial
s ijn + 1 = ----------
-s
trial ij
(3-268)
s eff
We can now update the pressure using the elastic bulk modulus, K
n+1⁄2
p n + 1 = p n – K Δε kk
E (3-269)
K = ------------------------
3 ( 1 – 2υ )
and obtain the final value for the Cauchy stress
After completing the stress update we transform the stresses back to the global configuration.
158
Materials
In Figure 3-23, note that the "yield stress" at a strain of zero is nonzero. In the load curve definition, the
"time" value is the directional strain and the "function" value is the yield stress.
Curve extends into negative strain quadrant Unloading is based on the interpolated
since SOL 700 will extrapolate using the two Young’s moduli which must provide an
end points. It is important that the unloading tangent that exceeds the loading
extrapolation does not extend into the tangent.
negative stress region.
The hydrostatic work term is expressed in terms of the bulk modulus, K , and J , as:
K
W H ( J ) = ---- ( H – 1 ) 2 (3-272)
2
Materials 159
Materials
Rate effects are taken into account through linear viscoelasticity by a convolution integral of the form:
t ∂ε kl
σ ij = 0 gijkl ( t – τ ) --------
∂τ
- dτ (3-273)
or in terms of the second Piola-Kirchhoff stress, S ij , and Green's strain tensor, Eij ,
t ∂ E kl
S ij = 0 Gijkl ( t – τ ) ---------
∂τ
- dτ (3-274)
where g ijkl ( t – τ ) and G ijkl ( t – τ ) are the relaxation functions for the different stress measures. This
stress is added to the stress tensor determined from the strain energy functional.
If we wish to include only simple rate effects, the relaxation function is represented by six terms from the
Prony series:
g ( t ) = α0 + α m e –βt (3-275)
m=1
given by,
Gi e
–βi t
g(t) = (3-276)
i=1
This model is effectively a Maxwell fluid which consists of a dampers and springs in series. We
characterize this in the input by shear moduli, G i , and decay constants, βi . The viscoelastic behavior is
optional and an arbitrary number of terms may be used.
When viscoelastic terms are not included, this model is similar to the use of the Arruda Boyce model in
solution 600, defined in the MATHE option. When viscoelasticity is included, the formulation of these
two models are different.
∂ε kl
0 gijkl ( t – τ ) --------
t
σ ij = - dτ (3-277)
∂τ
160
Materials
where g ijkl t – τ is the relaxation function for different stress measures. This stress is added to the stress
tensor determined from the strain energy functional. Since we wish to include only simple rate effects,
the relaxation function is represented by six terms from the Prony series:
g(t) = G m e –βm t
m=1
We characterize this in the input by the shear moduli, G i , and the decay constants, β i . An arbitrary
number of terms, not exceeding 6, may be used when applying the viscoelastic model. The composite
failure is not directly affected by the presence of the viscous stress tensor.
τ ij = q ik q jl τ kl
E
(3-279)
The q ij are the components of the orthogonal tensor containing the eigenvectors of the principal basis.
The Cauchy stress is then given by
σ ij = J –1 τ ij (3-280)
Materials 161
Materials
∂ iiE
TKE
C iijj = λj -------- – 2τ iiE δ ij
∂λj
TKE
λj τ ii – λ i2 τ jjE
2 E
i ≠ j ,( λ i ≠ λj ) (no sum)
C ijij = -----------------------------
-
λ 12 – λ j2
TKE
λi ∂τ iiE ∂τ iiE i ≠ j ,λ i = λ j
C ijij = ---- --------- – ---------
2 ∂λ i ∂λ j
(3-281)
These components are transformed to the standard basis according to
TKE TKE
C ijkl = qip q jg q kr q ls C pqrs (3-282)
and finally the constitutive tensor relating the rate of deformation to the Truesdell rate of Cauchy stress
is obtained through
TC TK
C ijkl = J –1 C ijkl (3-283)
3
1
τ iE = f ( λ i ) + K ( J – 1 ) – ---
3 f ( λk ) (3-284)
k=1
where f is a load curve determined from uniaxial data (possible at different strain rates). Furthermore,
K is the bulk modulus and J is the relative volume change of the material. This stress cannot be deduced
from a strain energy function unless f ( λ ) = E ln λ for some constitutive parameter E . A
consequence of this is that when using the formulas in the previous section the resulting tangent stiffness
matrix is not necessarily symmetric. We remedy this by symmetrizing the formulas according to
162
Materials
2
--- λ i f' ( λi )
λ ∂τ ∂τ ∂τ
E E E
ii if i = j
= --- λ j --------- + λi --------- = KJ +
--------
- 1 ii jj 3
j ∂λ j symn 2 ∂λ j ∂λ i –1
- ( λ i f·( λi ) + λ j f' ( λ j ) ) otherwise
--
6
(3-285)
Two Remarks
The function f introduced in the previous section depends not only on the stretches but for some choices
of input also on the strain rate. Strain rate effects complicate things for an implicit analyst and here one
also has to take into account whether the material is in tension/compression or in a loading/unloading
stage. We believe that it is of little importance to take into account the strain rate effects when deriving
the tangent stiffness matrix and therefore this influence has been disregarded.
For the fully integrated brick element, we have used the approach in material model 77 to account for the
constant pressure when deriving the tangent stiffness matrix. Experiments have shown that this is crucial
to obtain a decent implicit performance for nearly incompressible materials.
An elastic-plastic stress σ d is added to model the frequency independent damping properties of rubber.
This stress is deviatoric and determined by the shear modulus G and the yield stress σ Y . This part of
the stress is updated incrementally as
n+1
σ̃ d = σ dn + 2 GI dev Δε (3-286)
where Δε . is the strain increment. The trial stress is then radially scaled (if necessary) to the yield surface
according to
σY
min 1 ,--------
n+1
σ dn + 1 = σ̃ d
σ eff
n+1
where σ eff is the effective von Mises stress for the trial stress σ̃ d
The elastic tangent stiffness contribution is given by
C d = 2 GI dev (3-287)
and if yield has occurred in the last time step the elastic-plastic tangent is used
Materials 163
Materials
3G
C d = 2 GI dev – ------2- σ d ⊗ σ d (3-288)
σY
F = F 0 + K Δ L + D Δ L· (3-289)
but if the load curve ID is specified, the force is then given by:
Δ L·
F = F 0 + Kf ( Δ L ) 1 + C 1 ⋅ Δ L· + C 2 ⋅ sgn ( Δ L· ) ln max 1. ,------------ + D Δ L· + g ( Δ L ) h ( Δ L· )
DLE
In these equations, .L is the change in length
F Y = F y ( Δ L plastic ) (3-290)
F T = F n + K Δ L· Δ t (3-291)
Y if F T > F Y
F = F (3-292)
FT if F T ≤ F Y
The final force, which includes rate effects and damping, is given by:
Δ L·
F n + 1 = F ⋅ 1 + C 1 ⋅ Δ L· + C 2 ⋅ sgn ( Δ L· ) ln max 1. ,------------ + D Δ L· + g ( Δ L ) h ( Δ L· )
DLE
164
Materials
Unless the origin of the curve starts at (0,0), the negative part of the curve is used when the spring force
is negative where the negative of the plastic displacement is used to interpolate, F y . The positive part of
the curve is used whenever the force is positive.
The cross-sectional area is defined on the section card for the discrete beam elements. See
*SECTION_BEAM. The square root of this area is used as the contract thickness offset if these elements
are included in the contact treatment.
Boundary Conditions 157
Boundary Conditions
158
Boundary Conditions
Boundary Conditions
The motion of part or all of a model can be prescribed by Boundary and Initial conditions.
Single-Point Constraints
A single-point constraint is used to prescribe the motion of a translational or rotational degree of freedom.
The constraint is effective throughout the analysis and is used to specify boundary conditions or planes
of symmetry.
A single-point constraint is defined by an SPCn entry. The SPC entry defines the constraints on one grid
point, while the SPC1 defines the constraints to be applied to a set of grid points. Several sets of SPC
entries can be defined in the Bulk Data Section, but only those selected in the Case Control Section using
the SPC = n command are incorporated in the analysis.
Single-point constraints can also be defined using the GRIDt entry. These constraints are present for the
entire analysis and do not need to be selected in Case Control. This is valid only for SPC and SPC1.
Since Sol 700 is an explicit code, there is no matrix decomposition. Therefore, the problems of singular
matrices that occur with some implicit codes do not exist. All, or part of the Lagrangian mesh can be
entirely unconstrained and can undergo rigid body motion. SOL 700 correctly calculates the motion of
the mesh. Similarly, the redundant degrees of freedom, such as the in-plane rotation of shell elements, do
not need to be constrained since they do not affect the solution. The only constraints that are needed are
those representing the boundary conditions of the model and those necessary for any planes of symmetry.
• SPC
• SPC1
• SPCADD
• SPCD (at present, only velocity is available)
• SPCD2
Multi-Point Constraints
MPCs are special element types which define a rigorous behavior between several specified nodes. The
following MPC types which are supported for MD Nastran SOL 700:
• MPC
• RBAR
• RBE2
• RBE2A
• RBE2D
• RBE2F
Boundary Conditions 159
Boundary Conditions
• RBE3
• RBE3D
• RBJOINT
• REJSTIFF
• RCONN
References:
“MPC” in the MD Nastran Quick Reference Guide.
• Rigid - Deformable contact between a deformable body and a rigid body, for two- or
three-dimensional cases.
Contact problems involve a variety of different geometric and kinematic situations. Some contact
problems involve small relative sliding between the contacting surfaces, while others involve large
sliding or penetration followed by perforation and failure. Some contact problems involve contact over
large areas, while others involve contact between discrete points. The general Contact Body approach in
SOL 600 is also adopted by MD Nastran Explicit Nonlinear (SOL 700) to model contact and can be used
to handle most contact problem definitions.
Contact Bodies
There are two types of contact bodies in MD Nastran Explicit Nonlinear – deformable and rigid.
Deformable bodies are simply a collection of finite elements as shown below.
Rigid Walls
A rigid wall is a plane through which specified slave grid points cannot penetrate. The rigid wall provides
a convenient way of defining rigid targets in impact analyses.
Any number of rigid walls can be specified using WALL entries. The orientation of each wall is defined
by the coordinates of a point on the wall and a vector that is perpendicular to the wall and points towards
the model.
At each time step, a check is made to determine whether the slave grid points have penetrated the wall.
These slave points are defined using a Case Control SET entry, and there can be any number of them.
Since a check is made for every slave point at each time step, you should specify only those points as
slave points that are expected to contact the wall in order to ensure the most efficient solution.
If a slave point is found to have penetrated the wall, it is moved back towards the wall so that its
momentum is conserved. If the slave point subsequently moves away from the wall, it is allowed to do so.
Slave points cannot have any other constraint. They can, however, be part of other contact.
Contact Detection
During the incremental procedure, each potential contact node is first checked to see whether it is near a
contact segment. The contact segments are either edges of other 2- D deformable bodies, faces of 3-D
deformable bodies, or segments from rigid bodies. By default, each node could contact any other segment
including segments on the body that it belongs to. This allows a body to contact itself. To simplify the
computation, it is possible to use the BCTABLE entry to indicate that a particular body will or will not
contact another body. This is often used to indicate that a body will not contact itself. During the iteration
process, the motion of the node is checked to see whether it has penetrated a surface by determining
whether it has crossed a segment.
For more detailed discussion on the Contact, refer to Chapter 5 of this guide and see Chapter 12 in the
SOL 600 User’s Guide.
162
Lagrangian Loading
Lagrangian Loading
This section covers the different ways that the analysis model can be loaded. The facilities available are:
• Concentrated Loads and Moments
• Pressure Loads
• Initial Conditions
Concentrated loads and moments can be applied to any grid point using the DAREA, FORCE,
FORCE2, entries in combination with a TLOADn entry.
The types of concentrated load that can be applied are discussed in the following section.
Pressure Loads
Pressure loads are applied to the faces of solid elements and to shell elements. Pressure loads are defined
using the PLOAD or PLOAD4 entry in combination with a TLOADn entry.
TLOAD2 also references a set of PLOAD and/or PLOAD4 entries. Each entry selects the face of the
element to be loaded by its grid points and defines the scale factor to be applied to the curve of pressure
versus time. The actual pressure acting on the element p el is given as follows:
p el ( t ) = Ap ( t )
G4
G3 G3
G1 G1
G2 G2
The RFORCE entry defines enforced motion due to a centrifugal acceleration field. This motion affects
all structural elements present in the problem. The GRAV defines an enforced motion due to a
gravitational acceleration field. This motion affects all Lagrangian elements.
Grid points with enforced motion cannot be:
• Attached to a rigid body.
• A slave point for a rigid wall.
To specify the motion of a rigid body, the enforced motion of the rigid-body center of gravity must be
defined. To do so, set the TYPE field of the TLOAD1 and TLOAD2 entries to 12. The G field on the
DAREA, FORCE entry references the property number of the rigid body.
Initial Conditions
The initial velocity of grid points can be defined using TIC, TIC3, and TICD entries. This allows the
initial state of the model to be set prior to running the analysis. It is important to recognize the difference
164
Lagrangian Loading
between initial velocities and enforced velocities. Enforced velocities specify the motion of grid points
throughout the transient analysis. Initial velocities, on the other hand, specify the velocity of grid points
at the beginning of the analysis. Thereafter, the velocities are determined by the calculation.
Where TIC and TIC3 set only the initial grid-point velocity, the TICD entry can be used to set the initial
value of any valid grid point variable. It can also refer to a local coordinate system by including the CID1
and/or CID2 entry in the list.
Boundary Conditions 165
Boundary and Loading Conditions - Theoretical Background
∂b N t t ds
1
(4-1)
a Gaussian quadrature rule is used. To locate any point of the surface under consideration, a position
vector, γ , is defined:
r = f 1 ( ξ ,η ) i 1 + f 2 ( ξ ,η ) i 2 + f 3 ( ξ ,η ) i 3 (4-2)
where:
f i ( ξ ,η ) = φj xi
j
(4-3)
j=1
and i 1 , i 2 , and i 3 are unit vectors in the x 1 , x 2 , and x 3 directions (see 4-2).
Nodal quantities are interpolated over the four-node linear surface by the functions:
1
φ i = --- ( 1 + ξξ i ) ( 1 + ηη i ) (4-4)
4
so that the differential surface area ds may be written in terms of the curvilinear coordinates as:
ds = J d ξ d η
where J is the surface Jacobian defined by:
∂r ∂r
J = ------ × ------ = ( EG – F 2 ) 1 ⁄ 2 (4-5)
∂ξ ∂η
166
Boundary and Loading Conditions - Theoretical Background
in which:
∂r ∂r
E = ------ ⋅ ------
∂ξ ∂ξ
∂r ∂r
F = ------ ⋅ ------
∂ξ ∂η
∂r ∂r
G = ------ ⋅ ------
∂η ∂η (4-6)
∂r ∂r
n = J –1 ------ × ------ (4-7)
∂ξ ∂η
and the global components of the traction vector can now be written:
t i = n i φj p j (4-8)
j=1
3 2
x3
ξ
4
x2
i3
1
i2 rξ,η
i1 x1
1 1 t
–1 –1 N t J ( dξ ) dη (4-9)
One such integral is computed for each surface segment on which a pressure loading acts. Note that the
Jacobians cancel when (4-7) and (4-6) are put into (4-9). Equation 4-9 is evaluated with one-point
integration analogous to that employed in the volume integrals. The area of an element side is
approximated by 4 J where J = J ( 0, 0 ) .
Displacement Constraints
Translational and rotational boundary constraints are imposed either globally or locally by setting the
constrained acceleration components to zero. If nodal single point constraints are employed, the
constraints are imposed in a local system. The user defines the local system by specifying a vector u l in
the direction of the local x-axis, x l , and a local in-plane vector v l . After normalizing u l , the local x l ,
y l , and z l axes are given by:
ul
x l = --------- (4-10)
ul
xl × vl
z l = ------------------- (4-11)
xl × vl
yl = zl × xl (4-12)
A transformation matrix q is constructed to transform the acceleration components to the local system:
x lt
q = y lt (4-13)
z lt
168
Boundary and Loading Conditions - Theoretical Background
·
and the nodal translational and rotational acceleration vectors a I and ω I , for node I are transformed to
the local system:
a I l = qaI (4-14)
· ·
ωIl = q ω I (4-15)
and the constrained components are zeroed. The modified vectors are then transformed back to the
global system:
t
a I = q aIl (4-16)
· t·
ωI = q ωIl (4-17)
f ebody = v m
ρ N t Na base dυ = m e abase (4-18)
where a base is the base acceleration and m e is the element (lumped) mass matrix.
Contact 169
Contact
170
Contact
Contact
This section will summarize those parts of the contact methods applicable to SOL 700, and will describe
the theory of the algorithms used.
To activate contact in SOL 700, the Case Control entry BCONTACT must be given. There are three
methods available to define contact:
1. All elements in the model in one contact definition, using default settings
Case Control: BCONTACT=ALL
This will result in automatic contact detection between all elements in the model. For SOL 700,
it is advisable to use this method.
2. User defined Contact Bodies, using default or non-default settings
Case Control: BCONTACT = n
Bulk Data: BCBODY, BSURF, or BCBOX or BCPROP or BCMATL, and BCTABLE
It is possible to define contact bodies (BCBODY), and specify which contact bodies need to be
checked for contact (BCTABLE). This method using BCBODY definitions provides extreme
flexibility and is compatible with the Implicit Nonlinear solution (SOL 600).
A contact body is defined by the Bulk Data entry BCBODY, which references a set of elements
(BSURF), a set of elements inside a box (BCBOX), or certain property IDs (BCPROP) or with
certain material IDs (BCMATL).
Often used definitions related to contact methods are:
Single Surface Contact: This refers to any contact definition where no master is defined.
Master Slave Contact: This refers to any contact definition where a master is defined.
• 2-D contact, since SOL 700 only applies to 3-D (DIM on BCBODY)
• A rigid BCBODY defined by patches or geometric entities (BEHAV=RIGID on BCBODY)
- Rigid BCBODY as a symmetry plane (ISTYP on BCBODY)
- Motion/Load controlled rigid BCBODY (CONTROL on BCBODY)
(See note below on rigid body modeling in SOL 700)
• Body smoothing (IDSPL on BCBODY)
• User-defined distance below which a node is (ERROR on BCTABLE)
considered touching
• Separation Force (FNTOL on BCTABLE)
• Interference closure (CINTERF on BCTABLE)
Contact 171
Contact
With these capabilities, a faceted rigid body, similar to a BCBODY with BEHAV=RIGID can be easily
modeled. The BEHAV=RIGID logic will be implemented in the next release of SOL 700.
Penalty Methods
The treatment of sliding and impact along interfaces is an important capability in SOL 700.
Internally, the interfaces are defined in three dimensions by listing in arbitrary order all triangular and
quadrilateral segments that comprise each side of the interface. One side of the interface is designated as
the slave side, and the other is designated as the master side. Nodes lying in those surfaces are referred
to as slave and master nodes, respectively. When slave nodes penetrate the master, normal interface
springs are placed between the penetrated nodes and the master surface. See Slave Search in this chapter
for details on the contact force calculations.
∂ b1 ∂ b2
1 2
b b
∂ B2
1
∂B 2
B0
1
B
0
Preliminaries
Consider the time-dependent motion of two bodies occupying regions B 1 and B 2 in their undeformed
configuration at time zero. Assume that the intersection
B1∩ B2 = φ (5-1)
1 2
is satisfied. Let ∂ B 1 and ∂ B 2 denote the boundaries of B and B , respectively. At some later time,
these bodies occupy regions b 1 and b 2 bounded by ∂ b 1 and ∂ b 2 as shown in Figure 5-1. Because
the deformed configurations cannot penetrate,
( b1 – ∂ b1 )∩ b2 = φ (5-2)
Slave Search
The slave search is common to all interface algorithms implemented in SOL 700. This search finds for
each slave node its nearest point on the master surface. Lines drawn from a slave node to its nearest point
will be perpendicular to the master surface, unless the point lies along the intersection of two master
segments, where a segment is defined to be a 3- or 4-node element of a surface.
Consider a slave node, n s , sliding on a piecewise smooth master surface and assume that a search of the
master surface has located the master node, m s , lying nearest to n s . Figure 5-2 depicts a portion of a
master surface with nodes m s and n s labeled. If m s and n s do not coincide, n s can usually be shown
to lie in a segment s s via the following tests:
( ci × s ) ⋅ ( ci × ci + 1 ) > 0
( ci × s ) ⋅ ( s × ci + 1 ) > 0
(5-3)
where vector c i and c i + 1 are along edges of s 1 and point outward from m s . Vector s is the projection
of the vector beginning at m s , ending at n s , and denoted by g , onto the plane being examined (see
Figure 5-3).
s = g – ( g ⋅ m)m (5-4)
Contact 175
Slave Search
Figure 5-2 Four Master Segments can Harbor Slave Node n s given that m s is the Nearest
Master Node
ci × ci + 1
m = ------------------------ (5-5)
ci × ci + 1
Since the sliding constraints keep n s close but not necessarily on the master surface and since n s may
lie near or even on the intersection of two master segments, the inequalities of Equation (5-3) may be
inconclusive; i.e., they may fail to be satisfied or more than one may give positive results. When this
occurs n s is assumed to lie along the intersection which yields the maximum value for the quantity
g ⋅ ci
----------- i = 1, 2, 3, 4, … (5-6)
ci
When the contact surface is made up of badly shaped elements, the segment apparently identified as
containing the slave node actually may not, as shown in Figure 5-4.
Assume that a master segment has been located for slave node n s and that n s is not identified as lying
on the intersection of two master segments. Then the identification of the contact point, defined as the
point on the master segment which is nearest to n s , becomes nontrivial. Each master surface segment
s 1 , is given the parametric representation:
r = f 1 ( ξ ,η ) i 1 + f 2 ( ξ ,η ) i 2 + f 3 ( ξ ,η ) i 3 (5-7)
176
Slave Search
where
f i ( ξ, η ) – φj xi
j
(5-8)
j=1
When the nearest node fails to contain the segment that harbors the slave node, segments numbered 1-8
are searched in the order shown in Figure 5-4.
1
2 8
6 7
3
5
4
∂r
∂ r- ------
----- × ≠0 (5-9)
∂ξ ∂η
Thus, r represents a master segment that has a unique normal whose direction depends continuously on
the points of s 1 .
Let t be a position vector drawn to slave node n s and assume that the master surface segment s 1 has
been identified with n s . The contact point coordinates ( ξ c, η c ) on s 1 must satisfy
∂ r-
----- ( ξ , η ) ⋅ [ t – r ( ξ c, η c ) ] = 0 (5-10)
∂ξ c c
∂r
------ ( ξ , η ) ⋅ [ t – r ( ξ c, η c ) ] = 0 (5-11)
∂η c c
The physical problem is illustrated in Figure 5-5, which shows n s lying above the master surface.
Equations (5-10) and (5-11) are readily solved for ξ c and η c . One way to accomplish this is to solve
Contact 177
Slave Search
Equation (5-10) for ξ c in terms of η c , and substitute the results into (5-11). This yields a cubic equation
in η c .
η
3 2
ns ∂r
------
∂η
∂r-
----- ξ
x3 t ∂ξ
4
1
r
x2
x1
Figure 5-5 Location of Contact Point when n s lies above Master Segment
The equations are solved numerically. When two nodes of a bilinear quadrilateral are collapsed into a
single node for a triangle, the Jacobian of the minimization problem is singular at the collapsed node.
Fortunately, there is an analytical solution for triangular segments since three points define a plane.
Newton-Raphson iteration is a natural choice for solving these simple nonlinear equations. The method
diverges with distorted elements unless the initial guess is accurate. An expanded search procedure as
discussed in Improvements to the Contact Searching section is used.
Three iterations with a least-squares projection are used to generate an initial guess:
ξ 0 = 0, η 0 = 0,
r
[ r ,ξ r ,η ] Δξ = ,ξ { r ( ξ i, η i ) –t },
r ,ξ
r ,η Δη r ,η
ξ i + 1 = ξ i + 1 + Δξ, η i + 1 = η i + Δη
(5-12)
followed by the Newton-Raphson iterations which are limited to ten iterations, but which usually
converges in four or less.
178
Slave Search
Δξ r ,ξ
[ H ] = – { r ( ξ i, η i ) –
Δη r ,η
r ,χ 0 r ⋅ r,
[ H ] [ r ,ξ r ,η ] + (5-13)
r ,η r ⋅ r ,ξη 0
ξ i + 1 = ξ i + Δξ
η i + 1 = η i + Δη
In concave regions, a slave node may have isoparametric coordinates that lie outside of the [ – 1 ,+1 ]
range for all of the master segments, yet still have penetrated the surface. A simple strategy is used for
handling this case, but it can fail. The contact segment for each node is saved every time step.
If the slave node contact point defined in terms of the isoparametric coordinates of the segment, is just
outside of the segment, and the node penetrated the isoparametric surface, and no other segment
associated with the nearest neighbor satisfies the inequality test, then the contact point is assumed to
occur on the edge of the segment.
In effect, the definition of the master segments is extended so that they overlap by a small amount. In the
hydrocode literature, this approach is similar to the slide line extensions used in two dimensions. This
simple procedure works well for most cases, but it can fail in situations involving sharp concave corners.
Contact 179
Contact Force Calculation
l = n i ( [ t – r ( ξ c, η c ) ] < 0 ) (5-14)
where
n i = n i ( ξ c, η c ) (5-15)
is normal to the master segment at the contact point. The amount of penetration is equal to the value of l.
If slave node ns has penetrated through master segment s i , we add an interface force vector f s :
f mi = φ 2 ( ξ c, η c ) f s if l < 0 (5-17)
dt as
m slave m master 1
k i = f si ⋅ --------------------------------------- ⋅ ------2- (5-18)
m slave + m master dt
With SOFT=1, it is mostly not needed to scale the contact stiffness, even if materials with very different
stiffness properties come into contact.
When SOFT=0 on the BCTABLE entry, the stiffness factor k i for master segment s i is given in terms of
the bulk modulus K i , the volume V i , and the face area A i of the element that contains s i as
f si K i A i2
k i = ----------------- (5-19)
Vi
for brick elements and
180
Contact Force Calculation
f si K i A i
k i = ------------------------------------------------- (5-20)
max ( shell diagonal )
for shell elements. When too much penetration is observed, the contact stiffness can be increased by the
interface stiffness scale factor, f si . This scale factor is 0.1 by default, and can be defined by FACT on the
BCTABLE entry. Larger values may cause instabilities unless the time step size is scaled back in the time
step calculation.
Contact 181
Improvements to the Contact Searching
slave node
Figure 5-6 Failure to find the Contact Segment can be caused by Poor Aspect Ratios in the
Finite Element Mesh
To circumvent the problem caused by bad aspect ratios, an expanded searching procedure is used in
which we attempt to locate the nearest segment rather than the nearest nodal point.
1 2 3 4 5
Normal vector
at node 3
The nearest contact segment to a given node, k , is defined to be the first segment encountered when
moving in a direction normal to the surface away from k . A major deficiency with the nearest node
search is depicted in Figure 5-7 where the nearest nodes are not even members of the nearest contact
182
Improvements to the Contact Searching
segment. Obviously, this would not be a problem for a more uniform mesh. To overcome this problem
we have adopted segment based searching in both surface to surface and single surface contact.
Contact 183
Bucket Sorting
Bucket Sorting
Bucket sorting is now used extensively in the SOL 700 contact algorithms.
The reasons for eliminating slave node tracking by incremental searching is illustrated in Figure 5-8
where surfaces are shown which cause the incremental searches to fail. With bucket sorting incremental
searches may still be used but for reliability they are used after contact is achieved. As contact is lost, the
bucket sorting for the affected nodal points must resume.
In a direct search of a set of N nodes to determine the nearest node, the number of distance comparisons
N – 1 . Since this comparison needs to be made for each node, the total number of
required is
comparisons is N ( N – 1 ) , with each of these comparisons requiring a distance calculation
1 2 = ( xi – xj ) 2 + ( yi – yj ) 2 + ( zi – z j ) 2 (5-21)
that uses eight mathematical operations. The cumulative effect of these mathematical operations for
N ( N – 1 ) compares can dominate the solution cost at less than 100 elements.
The idea behind a bucket sort is to perform some grouping of the nodes so that the sort operation need
only calculate the distance of the nodes in the nearest groups. With this partitioning the nearest node will
either reside in the same bucket or in one of the two adjoining buckets. The number of distance
calculations is now given by
3------
N-
–1 (5-22)
α
where α is the number of buckets. The total number of distance comparisons for the entire one-
dimensional surface is
N ------- – 1
3N
(5-23)
α
Thus, if the number of buckets is greater than 3, then the bucket sort will require fewer distance
comparisons than a direct sort. It is easy to show that the corresponding number of distance comparisons
for two-dimensional and three-dimensional bucket sorts are given by
where b and c are the number of partitions along the additional dimension.
184
Bucket Sorting
Incremental searching may fail on surfaces that are not simply connected. The contact algorithm in SOL
700 avoids incremental searching for nodal points that are not in contact and all these cases are
considered (see Figure 5-8).
tied interface
not yet supported
The cost of the grouping operations, needed to form the buckets, is nearly linear with the number of nodes
N . For typical SOL 700 applications, the bucket sort is 100 to 1000 times faster than the corresponding
direct sort. However, the sort is still an expensive part of the contact algorithm, so that, to further
minimize this cost, the sort is performed every ten or fifteen cycles and the nearest three nodes are stored.
This can be specified by BSORT on the BCTABLE entry. Typically, three to five percent of the
calculational costs will be absorbed in the bucket sorting when most surface segments are included in the
contact definition.
Contact 185
Bucket Sorting in Single Surface Contact
where the coordinate pairs ( x min, x max ) , ( y min, y max ) , and ( z min, z max ) span the entire contact
surface. In this procedure, we loop over the segments rather than the nodal points. For each segment we
use a nested DO LOOP to loop through a subset of buckets from IMIN to IMAX , JMIN to
JMAX , and KMIN to KMAX where:
and PXk , PYk , PZk are the bucket pointers for the kth node. Figure 5-9 shows a segment passing
through a volume that has been partitioned into buckets. The orthogonal distance of each slave node
contained in the box from the segment is determined. The box is subdivided into sixty buckets.
We check the orthogonal distance of all nodes in the bucket subset from the segment. As each segment
is processed, the minimum distance to a segment is determined for every node in the surface and the two
nearest segments are stored. Therefore the required storage allocation is still deterministic. This would
not be the case if we stored for each segment a list of nodes that could possibly contact the segment.
We have now determined for each node, k , in the contact surface the two nearest segments for contact.
Having located these segments we permanently store the node on these segments which is nearest to node
k . When checking for interpenetrating nodes we check the segments surrounding the node including the
nearest segment since during the steps between bucket searches it is likely that the nearest segment may
change. It is possible to bypass nodes that are already in contact and save some computer time; however,
if multiple contacts per node are admissible then bypassing the search may lead to unacceptable errors.
Contact 187
Accounting For the Shell Thickness
Thickness changes in the contact are accounted for “if and only if” the shell thickness change option is
flagged on the PARAM* DYCONTHKCHG. Each cycle, as the shell elements are processed, the nodal
thicknesses are stored for use in the contact algorithms. The interface stiffness may change with thickness
depending on the input options used.
To account for the nodal thickness, the maximum shell thickness of any shell connected to the node is
taken as the nodal thickness and is updated every cycle. The projection of the node is done normal to the
contact surface:
188
Initial Contact Penetrations
brick
shell
Inner penetration if
edge is too close
Figure 5-13 Undetected Penetration due to rounding the Edge of the Shell Element
To avoid problems with initial penetrations, the following recommendations should be considered:
• Adequately offset adjacent surfaces to account for part thickness during the mesh
generation phase.
• Use consistently refined meshes on adjacent parts which have significant curvatures.
• Be very careful when defining thickness on shell and beam section definitions - especially for
rigid bodies.
Contact 189
Initial Contact Penetrations
• Scale back part thickness if necessary. Scaling a 1.5mm thickness to .75mm should not cause
problems but scaling to .075mm might. Alternatively, define a smaller contact thickness by part
ID. Warning: if the part is too thin contact failure will probably occur
• Use spot welds instead of merged nodes to allow the shell mid surfaces to be offset.
190
Contact Energy Calculation
nsn nmn 1
n + ---
2
Δ Fi × Δ dist islave + Δ Fi × Δ dist imaster
n+1 n slave master
E contact = E contact +
i=1 i=1
Where nsn is the number of slave nodes, nmn is the number of master nodes, Δ F islave is the
interface force between the ith slave node and the contact segment Δ F imaster is the interface force
between the ith master node and the contact segment, Δ dist islave is the incremental distance the ith
slave node has moved during the current time step, and Δ dist imaster is the incremental distance the ith
master node has moved during the current time step. In the absence of friction the slave and master side
energies should be close in magnitude but opposite in sign. The sum, E contact , should equal the stored
energy. Large negative contact energy is usually caused by undetected penetrations. Contact energies are
reported in the GLSTAT file. In the presence of friction and damping discussed below the interface energy
can take on a substantial positive value especially if there is, in the case of friction, substantial sliding.
Contact 191
Friction
Friction
Friction in SOL 700 is based on a Coulomb formulation. Let f * be the trial force, f n the normal force,
n
k the interface stiffness, μ the coefficient of friction, and f the frictional force at time n. The frictional
algorithm, outlined below, uses the equivalent of an elastic plastic spring. The steps are as follows:
1. Compute the yield force, Fy :
Fy = μ fn (5-28)
Δ e = r n + 1 ( ξ cn + 1, η cn + 1 ) – r n + 1 ( ξ cn, η cn ) (5-29)
f∗ = f n – k Δ e (5-30)
4. Check the yield condition:
f n + 1 = f∗ if f∗ ≤ F y (5-31)
F y f∗
f n + 1 = ---------- if f∗ > F y (5-32)
f∗
An exponential interpolation function smooths the transition between the static, μ s , and dynamic, μ d ,
coefficients of friction where ν is the relative velocity between the slave node and the master segment:
μ = μ d + ( μ s – μ d ) e –c ν (5-33)
where
ΔeΔT
ν = -------------- (5-34)
• Inflator models
Airbag Definition
The airbags are defined by using the AIRBAG entry in SOL 700. Airbags can be automatically inflated
by defining a gas flow rate in Eulerian domain using full gas dynamics method. A second method, the
conventional uniform pressure method, is also available but it is not as accurate as the full gas dynamics.
All related airbag input definitions such as inflator and porosity models, environmental parameters, are
grouped in enties that can be directly input in the AIRBAG entry. Every entity has certain parameters
associated with it that has to be input immediately following the name of the entity. The order of which
entity is entered first in the AIRBAG card is immaterial as long as the associated parameters are defined
right after the name of the entity. The following entity groups are available:
“CFD” Entries for this entity describe the properties of the Eulerian domain. Only one
"CFD" section can be defined. If this section is not defined, the uniform pressure
method is used.
“ENVIRONM” Entries for this entity describe the properties of the environmental conditions for the
airbag. Only one “ENVIRONM” section can be defined. If not defined, the values
for “INITIAL” will be used.
“INITIAL” Entries for this entity describe the properties of the environmental conditions for the
airbag. Only one “INITIAL” section can be defined and is required.
“INFLATOR” Entries for this entity describe the properties of an inflator that is attached to the
airbag. More than one inflator may be defined.
“CGINFLTR” Entries for this entity describe the properties of a cold gas inflator that is attached to
the airbag. More than one inflator may be defined.
“SMALHOLE” Entries for this entity describe the properties of a small hole in the airbag. More than
one SMALHOLE may be defined. A small hole should be used when the size of the
hole is of the same order as the size of the elements of the Euler mesh.
“LARGHOLE” Entries for this entity describe the properties of a large hole in the airbag. More than
one LARGHOLE may be defined. A large hole should be used when the size of the
hole is of the larger than the size of the elements of the Euler mesh.
“PERMEAB” Entries for this entity describe the properties of the permeability of the airbag fabric.
More than one PERMEAB may be defined.
“CONVECT” Entries for this entity describe the properties of the loss of energy of the gas in the
airbag by means of convection through the airbag surface. More than one CONVECT
may be defined.
Airbag and Occupant Safety 197
Airbag Definition
“RADIATE” Entries for this entity describe the properties of the loss of energy of the gas in the
airbag by means of radiation through the airbag surface. More than one RADIATE
may be defined. When this option is used, the Stephan-Boltzmann constant must be
defined by PARAM, SBOLTZ.
“GAS” Entries for this entity describe the properties of gases. More than 1 GAS may be
defined and the gases defined in one AIRBAG entry can be referenced by other
AIRBAG entries.
There are other entries in SOL 700 to define airbag properties. These are:
INFLATR1 Standard inflator defined by mass flow rate and static temperature of a
single inflowing gas.
INFLHYB1 Hybrid inflator defined by mass flow rate and static temperature of multiple
inflowing gasses.
For both the uniform pressure model and the full gas dynamics (CFD) method, the inflator location and
area are defined by means of a subsurface created by a BSURF, BCPROP, BCMATL, or BCSEG
parameters. These parameters are referenced by BFID field on the INFLATOR entity. It can only
reference shell elements that belong to the airbag surface, as defined by BFID. The characteristics of the
inflator are specified on an INFLATOR entity on AIRBAG card. This entry references tables for the mass
flow rate and the temperature of the inflowing gas.
A model can be defined containing both (CFD) and uniform pressure model for the airbag. These two
options can be defined with identical inflator characteristics. When the CFD entity and its associated
parameters are omitted from the AIRBAG card, the uniform pressure method will be used. In case the
CFD entity is present, the airbag will use the Euler method from the start of the simulation. In case the
value of SWITCH is nonzero, the airbag will switch from an Eulerian representation to a Uniform
Pressure formulation.
c p m· T i = c v m· T t + c v m· T· t
p· V = m· RT t + mRT· t
and, finally, we obtain the desired mass flow rate:
c v p· V
·
m = -------------
-
c p RT t
Porosity in Airbags
Porosity is defined as the flow of gas through the airbag surface. There are two ways to model this:
1. Holes: The airbag surface contains a discrete hole.
2. Permeability: The airbag surface is made from material that is not completely sealed.
Airbag and Occupant Safety 199
Airbag Definition
The same porosity models are available for both the uniform pressure airbag model as the Eulerian
coupled (CFD) airbag model. The porous flow can be either to and from the environment or into and from
another uniform pressure model.
Holes
Flow through holes as defined on the SMALHOLE entries is based on the theory of one-dimensional gas
flow through a small orifice. LARGHOLE entries define flow through a hole with the velocity method.
The velocity method can only be active for Eulerian airbags. When the SMALHOLE is used on the
AIRBAG card, the theory of one-dimensional gas flow through a small orifice is applied.
Velocity Method
The transport of mass through the porous area is based on the velocity of the gas in the Eulerian
elements,relative to the moving of coupling surface (airbag fabric).
Eulerian Element
SX int
Face of the coupling
surface that intersects
v the Eulerian element
Coupling Surface
The volume of the Eulerian materian transported through the faces of the coupling surface that intersect
an Eulerian element is equal to
V trans = – d t ⋅ α ⋅ ( v ⋅ A )
where
V trans = transported volume during one time step ( V trans > 0 for the outflow; V trans < 0 for
the inflow).
dt = time step.
α = porosity coefficient.
v = velocity vector of the gas in the Eulerian mesh
A = area of the face of the coupling surface that intersects the Eulerian element A is equal to
the area of the face that lies inside the Eulerian element.
The transport mass through the porous area is equal to the density of the gas times the
transported volume.
200
Airbag Definition
Pressure Method
The transport of mass through the porous area is based on the pressure difference between the gas in the
Eulerian element and the outside pressure. The outside pressure is the pressure as specified on the
ENVIRONM section in the AIRBAG entry.
Eulerian Element
SX int
Face of the coupling
surface that intersects
v the Eulerian element
Coupling Surface
The volume of the Eulerian material transported through the faces of the coupling surface that intersect
an Eulerian element is equal to:
γ+1
2 p ργ pexh 2 / γ p exh -----------
V trans = dt ⋅ α ⋅ ( A ⋅ A ) ⋅ ------------ --------- – --------- γ
γ–1 p p
where
V trans = transported volume during one ture step ( V trans > 0 for outflow; V trans < 0 for inflow).
dt = time step.
α = porosite coefficient.
v = velocity vector of the gas in the Eulerian mesh.
A = area of the face of the coupling surface that intersects the Eularian element A is equal to
the area of the face that lies inside the Eulerian element.
p= = pressure of the gas in the Eulerian element.
ρ = density of the gas in the Eulerian element.
γ = adiabatic exponent = C p ⁄ Cv .
p exh = pressure of the face.
The pressure at the face is approximated by the one-dimensional isentropic expansion of the gas to the
critical pressure or the environmental pressure ascending to
Airbag and Occupant Safety 201
Airbag Definition
p > pc
p = env
p env < p c
r
2 -----------
p c p ⋅ ----------- r – 1
γ+1
In case the outside pressure is greater than the pressure of the gas, inflow through the coupling surface
occurs. This porosity model can only be used for ideal gases; i. e., materials modeled with the gamma
law equation of state (EOSGAM).
Permeability
Permeability is defined as the velocity of gas through a surface area depending on the pressure
differenceover that area.
On the PERMEAB entity in AIRBAG card, permeability can be specified by either a coefficient or a
pressure dependent table:
1. Coefficient: Massflow = coeff * pressure_difference
Gas Velocity
δ ( massflow) )
coeff = ---------------------------------
δ ( pressdiff )
Coeff
press_diff
2. Table
Gas Velocity
press_diff
The velocity of the gas flow can never exceed the sonic speed:
202
Airbag Definition
where γ is the gas constant of in- or outflowing gas and T crit is the critical temperature.
The critical temperature can be calculated as follows:
T crit 2 -
--------- – ---------------
T gas ( γ + 1 )
∂ xi
F ij = --------
∂ Xj
where the choice of X j may coincide with the folded or unfold configurations. It is this unfolded
configuration which may be specified here.
Note that a reference geometry which is smaller than the initial airbag geometry will not induce initial
tensile stresses.
If a liner is included and the LNRC parameter set to 1 in MATD034, compression is disabled in the liner
until the reference geometry is reached; i.e., the fabric element becomes tensile.
where h ( t ) is the time-dependent heat transfer coefficient, A is the (sub)surface area for heat
transfer, T is the temperature inside the airbag, and T env is the environment temperature.
2. Radiation:
where e is the gas emissivity, A is the (sub)surface area for heat transfer, T is the temperature
inside the airbag, and T env is the environment temperature.
204
Seatbelts
Seatbelts
Belt elements are single degree of freedom elements connecting two nodes. When the strain in an element
is positive (i.e. the current length is greater then the unstretched length), a tension force is calculated from
the material characteristics and is applied along the current axis of the element to oppose further
stretching. The unstretched length of the belt is taken as the initial distance between the two nodes
defining the position of the element plus the initial slack length.
Seatbelt shell elements must be used with caution. The seatbelt shells distribute the loading on the surface
of the dummy more realistically than the two node belt elements. For the seatbelt shells to work with
sliprings and retractors it is necessary to use a logically regular mesh of quadrilateral elements.
Slipring
Retractor
Top View:
RN5 SN5
The ordering of the nodes and elements are important for seatbelt shells.
Airbag and Occupant Safety 205
Seatbelts
SEATBELT_PRETENSIONER
Pretensioners allow modeling of five types of active devices which tighten the belt during the initial
stages of a crash. Types 1 and 5 represent a pyrotechnic device which spins the spool of a retractor,
causing the belt to be reeled in. The user defines a pull-in versus time curve which applies once the
pretensioner activates. Types 2 and 3 represent preloaded springs or torsion bars which move the buckle
when released. The pretensioner is associated with any type of spring element including rotational. Note
that when the preloaded spring, locking spring, and any restraints on the motion of the associated nodes
are defined in the normal way; the action of the pretensioner is merely to cancel the force in one spring
until (or after) it fires. With the second type, the force in the spring element is canceled out until the
pretensioner is activated. In this case, the spring in question is normally a stiff, linear spring which acts
as a locking mechanism, preventing motion of the seat belt buckle relative to the vehicle. A preloaded
spring is defined in parallel with the locking spring. This type avoids the problem of the buckle being free
to ‘drift’ before the pretensioner is activated. Type 4, a force type, is described below.
To activate the pretensioner, the following sequence of events must occur:
1. Any one of up to four sensors must be triggered.
2. Then a user-defined time delay occurs.
3. Then the pretensioner acts.
Type 1 pretensioner is intended to simulate a pyrotechnic retractor. Each retractor has a loading (and
optional unloading) curve that describes the force on the belt element as a function of the amount of belt
that has been pulled out of the retractor since the retractor locked. The type 1 pretensioner acts as a shift
of this retractor load curve. An example will make this clear. Suppose at a particular time that 5mm of
belt material has left the retractor. The retractor responds with a force corresponding to 5mm pull-out on
it's loading curve. But suppose this retractor has a type 1 pretensioner defined, and, at this instant of time,
the pretensioner specifies a pull-in of 20mm. The retractor then responds with a force that corresponds
to (5mm + 20mm) on it's loading curve. This results in a much larger force. The effect can be that belt
material will be pulled in, but there is no guarantee. The benefit of this implementation is that the force
vs. pull-in load curve for the retractor is followed and no unrealistic forces are generated. Still, it may be
difficult to produce realistic models using this option, so two new types of pretensioners have been added.
These are available in MD Nastran r2 and later versions.
Types 2 and 3 are simple triggers for activating or deactivating springs, which then pull on the buckle.
No changes have been made to these, and they are not discussed here.
The type 4 pretensioner takes a force vs. time curve (see Figure 6-4). Each time step, the retractor
computes the desired force without regard to the pretensioner. If the resulting force is less than that
specified by the pretensioner load curve, then the pretensioner value is used instead. As time goes on,
the pretensioner load curve should drop below the forces generated by the retractor, and the pretensioner
is then essentially inactive. This provides for good control of the actual forces, so no unrealistic values
are generated. The actual direction and amount of belt movement is unspecified, and depends on the
206
Seatbelts
other forces being exerted on the belt. This is suitable when the force the pretensioner exerts over time
is known.
Force
Retractor Pull-out Force
Defined Force
vs.
Time Curve
Time
Figure 6-4 Force versus Time Pretensioner. At the intersection, the retractor locks.
The type 5 pretensioner is essentially the same as the old type 1 pretensioner, but with the addition of a
force limiting value. The pull-in is given as a function of time, and the belt is drawn into the retractor
exactly as desired. However, if at any point the forces generated in the belt exceed the pretensioner force
limit, then the pretensioner is deactivated and the retractor takes over. In order to prevent a large
discontinuity in the force at this point, the loading curve for the retractor is shifted (in the abscissa) by
the amount required to put the current (pull-out, force) on the load curve. For example, suppose the
current force is 1000, and the current pull-out is -10 (10mm of belt has been pulled in by the
pretensioner). If the retractor would normally generate a force of 1000 after 25mm of belt had been pulled
OUT, then the load curve is shifted to the left by 3, and remains that way for the duration of the
calculation. So that at the current pull-in of 10, it generates the force normally associated with a pull out
of 25. If the belt reaches a pull out of 5, the force is as if it were pulled out 40 (5 + the shift of 35), and
so on. This option is included for those who liked the general behavior of the old type 1 pretensioner, but
has the added feature of the force limit to prevent unrealistic behavior.
The type 6 pretensioner is a variation of the type 4 pretensioner, with features of the type 5 pretensioner.
A force vs. time curve is input and the pretensioner force is computed each cycle. The retractor linked to
this pretensioner should specify a positive value for PULL, which is the distance the belt pulls out before
it locks. As the pretensioner pulls the belt into the retractor, the amount of pull-in is tracked. As the
pretensioner force decreases and drops below the belt tension, belt will begin to move back out of the
retractor. Once PULL amount of belt has moved out of the retractor (relative to the maximum pull in
encountered), the retractor will lock. At this time, the pretensioner is disabled, and the retractor force
curve is shifted to match the current belt tension. This shifting is done just like the type 5 pretensioner. It
Airbag and Occupant Safety 207
Seatbelts
is important that a positive value of PULL be specified to prevent premature retractor locking which
could occur due to small outward belt movements generated by noise in the simulation.
SEATBELT_RETRACTOR
The unloading curve should start at zero tension and increase monotonically (i.e., no segments of
negative or zero slope).
Retractors allow belt material to be paid out into a belt element. Retractors operate in one of two regimes:
unlocked when the belt material is paid out, or reeled in under constant tension and locked when a user
defined force-pullout relationship applies.
The retractor is initially unlocked, and the following sequence of events must occur for it to
become locked:
1. Any one of up to four sensors must be triggered. (The sensors are described below.)
2. Then a user-defined time delay occurs.
3. Then a user-defined length of belt must be paid out (optional).
4. Then the retractor locks and once locked, it remains locked.
In the unlocked regime, the retractor attempts to apply a constant tension to the belt. This feature allows
an initial tightening of the belt and takes up any slack whenever it occurs. The tension value is taken from
the first point on the force-pullout load curve. The maximum rate of pull out or pull in is given by 0.01 ×
fed length per time step. Because of this, the constant tension value is not always achieved.
In the locked regime, a user-defined curve describes the relationship between the force in the attached
element and the amount of belt material paid out. If the tension in the belt subsequently relaxes, a
different user-defined curve applies for unloading. The unloading curve is followed until the minimum
tension is reached.
The curves are defined in terms of initial length of belt. For example, if a belt is marked at 10mm intervals
and then wound onto a retractor, and the force required to make each mark emerge from the (locked)
retractor is recorded, the curves used for input would be as follows:
Pyrotechnic pretensions may be defined which cause the retractor to pull in the belt at a
predetermined rate. This overrides the retractor force-pullout relationship from the moment when the
pretensioner activates.
If desired, belt elements may be defined which are initially inside the retractor. These will emerge as belt
material is paid out, and may return into the retractor if sufficient material is reeled in during unloading.
208
Seatbelts
Elements e2, e3 and e4 are initially inside the retractor, which is paying out material into element e1.
When the retractor has fed L crit into e1, where
Element e2 emerges with an unstretched length of 1.1∞minimum length ; the unstretched length of
element e1 is reduced by the same amount. The force and strain in e1 are unchanged; in e2, they are set
equal to those in e1. The retractor now pays out material into e2.
If no elements are inside the retractor, e2 can continue to extend as more material is fed into it.
As the retractor pulls in the belt (for example, during initial tightening), if the unstretched length of the
mouth element becomes less than the minimum length, the element is taken into the retractor.
To define a retractor, the user enters the retractor node, the ‘mouth’ element (into which belt material will
be fed), e1 in Figure 6-5, up to 4 sensors which can trigger unlocking, a time delay, a payout delay
(optional), load and unload curve numbers, and the fed length. The retractor node is typically part of the
vehicle structure; belt elements should not be connected to this node directly, but any other feature can
be attached including rigid bodies. The mouth element should have a node coincident with the retractor
but should not be inside the retractor. The fed length would typically be set either to a typical element
initial length, for the distance between painted marks on a real belt for comparisons with high speed film.
The fed length should be at least three times the minimum length.
If there are elements initially inside the retractor (e2, e3 and e4 in the Figure 6-5) they should not be
referred to on the retractor input, but the retractor should be identified on the element input for these
elements. Their nodes should all be coincident with the retractor node and should not be restrained or
constrained. Initial slack will automatically be set to 1.1 × minimum length for these elements; this
overrides any user-defined value.
Airbag and Occupant Safety 209
Seatbelts
Weblockers can be included within the retractor representation simply by entering a ‘locking up’
characteristic in the force pullout curve (see Figure 6-6. The final section can be very steep (but must have
a finite slope).
Element 1
Before
Element 1
Element 2
Element 3
Element 2
Element 4
After
Element 3
Element 4
with weblockers
without weblockers
FORCE
PULLOUT
Figure 6-6 Retractor Force Pull Characteristics
SEATBELT_SENSOR
Sensors are used to trigger locking of retractors and activate pretensioners. Four types of sensors are
available which trigger according to the following criteria:
Type 1 – When the magnitude of x-, y-, or z- acceleration of a given node has remained
above a given level continuously for a given time, the sensor triggers. This does
not work with nodes on rigid bodies.
Type 2 – When the rate of belt payout from a given retractor has remained above a given
level continuously for a given time, the sensor triggers.
Type 3 – The sensor triggers at a given time.
Type 4 – The sensor triggers when the distance between two nodes exceeds a given
maximum or becomes less than a given minimum. This type of sensor is intended
for use with an explicit mass/spring representation of the sensor mechanism.
By default, the sensors are inactive during dynamic relaxation. This allows initial tightening of the belt
and positioning of the occupant on the seat without locking the retractor or firing any pretensioners.
However, a flag can be set in the sensor input to make the sensors active during the dynamic relaxation
phase.
SEATBELT_SLIPRING
Sliprings allow continuous sliding of a belt through a sharp change of angle. Two elements (1 & 2 in
Figure 6-7) meet at the slipring. Node B in the belt material remains attached to the slipring node, but belt
material (in the form of unstretched length) is passed from element 1 to element 2 to achieve slip. The
amount of slip at each timestep is calculated from the ratio of forces in elements 1 and 2. The ratio of
forces is determined by the relative angle between elements 1 and 2 and the coefficient of friction, μ.
Airbag and Occupant Safety 211
Seatbelts
The tension in the belts are taken as T 1 and T 2 , where T 2 is on the high tension side and T1 is the force
on the low tension side. Thus, if T 2 is sufficiently close to T 1 , no slip occurs; otherwise, slip is just
sufficient to reduce the ratio T 2 ⁄ T 1 to e μΘ . No slip occurs if both elements are slack. The out-of-
balance force at node B is reacted on the slipring node; the motion of node B follows that of slipring node.
If, due to slip through the slipring, the unstretched length of an element becomes less than the minimum
length (as entered on the belt material card), the belt is remeshed locally: the short element passes through
the slipring and reappears on the other side (see Figure 6-7). The new unstretched length of e1 is
1.1 × minimum length . Force and strain in e2 and e3 are unchanged; force and strain in e1 are now
equal to those in e2. Subsequent slip will pass material from e3 to e1. This process can continue with
several elements passing in turn through the slipring.
To define a slipring, the user identifies the two belt elements which meet at the slipring, the friction
coefficient, and the slipring node. The two elements must have a common node coincident with the
slipring node. No attempt should be made to restrain or constrain the common node for its motion will
automatically be constrained to follow the slipring node. Typically, the slipring node is part of the vehicle
body structure and, therefore, belt elements should not be connected to this node directly, but any other
feature can be attached, including rigid bodies.
Slipring
B
Element 2
Element 1
Element 1 Element 3
Element 2
Element 3
Before After
Figure 6-7 Elements Passing Through Slipring
212
Occupant Dummy Models
Simulation
214
Perform the Simulation
Common subcase
Common subcase defines the master output requests for your analysis. Output requested in the common
case will become the default for all subcases. The software will write all of these requests and conditions
to the Case Control section. In addition, you can define cases which let you perform multiple analyses
with different load and/or constraint sets.
Create Subcase
The subcase dialog box lets you select the loads and constraints to apply to your analysis.You can apply
boundary conditions as both the common level or in subcases. If your analysis requires multiple load or
constraint sets, you must create subcases.
MPCs: pick a constraint set to define constraint equations.
Simulation 215
Perform the Simulation
Output Requests
Use the Output Requests dialog box to identify the types of output that you want from the analysis. You
can select to output any combination of the results shown on the following form:
Solution Parameters
Here you can specify Solution Parameters such as whether or not to include Large Displacements,
Follower Forces, etc.
.
216
Perform the Simulation
Subcase Parameters
Here you specify Subcase Parameters such as Ending Time for the simulation, Number of Time Steps.
Simulation 217
Perform the Simulation
Output File
The purpose of this section is to introduce the output file generated by a typical run. After an input file is
submitted for execution, several output files may be generated.
218
Results
Results
If you submitted your analysis directly from SimXpert, the results will be attached upon job completion.
If your analysis ran externally to SimXpert but you have the model in SimXpert, select Attach Results
from the File menu and navigate to your results file. If your analysis ran externally to SimXpert and you
do not have the model in SimXpert, select Import Results from the File menu and navigate to your
results file. This will read in model information as well as results from the results file.
Output data is also stored in Result Cases. If you run your model with several different loading conditions
or through several different analysis types, SimXpert will keep the output data from each analysis in a
different Result Case.
Postprocessing can be divided into two main categories: Chart and State Plot. State Plot postprocessing
can be further divided into the following types of plots:
1. Deformation
Can be displayed in any available render style. You can turn the display of the undeformed shape
on or off as desired.
2. Fringe
Color code your model based on result value
3. Vector
Show vectors representing direction and magnitude of result value.
All plot types can be animated. You can animate a single result case by applying a progressive scale factor
to the deformation result or you can animate over a selection of result sets. All plot types can be displayed
together in the same window if desired.
State Plots
The State Plot properties form is the main control for postprocessing. It is from here you can control
which result set is being displayed, how your model is displayed and select your desired postprocessing
options.
You can use dynamic rotation to manipulate your model while results are displayed and also during
animation.
• How to Select the Data used for postprocessing
The state plot properties form is accessed by clicking on the Results toolbox and selecting State Plot.
To choose the data used in the display, in the Results Cases list click the desired Result Case (A in figure),
then click the Results Type to use (B), and, if desired, you can make selections for Derivation and Target
Entities. You can limit the Results Cases listed using the Filter text box. If you are animating multiple
Results Cases, simply select all desired cases in the Results Cases list.
• Deformed Shape Plots
The following are some of the options control the appearance of your deformed shape plot:
Simulation 219
Results
1. Deformed display scaling - True: applies the related scale factor to the actual computed values for
displacement To see a plot of your actual displacement set the scale factor to 1. Relative: applies
the scale factor to display the maximum displacement as a percentage of your maximum model
dimension. This is usually an exaggerated plot for ease in visualization.
2. Deformed Shape - controls rendering for Deformed shape display.
3. Undeformed Shape - can display or remove the undeformed model, as well as control how it is
Contour Style View Options
• Fringe Plots
The Fringe tab allows access to data transforms. These options are very important to understand since
they control how the data is converted from pure discrete numbers to a visual representation. Improper
selection of data transforms can lead to erroneous interpretation of the results.
You can specify which domain in the model to use for result averaging in order to obtain an accurate
representation of the results. The following domains are available:
1. Property - results will not be averaged across property boundaries
2. Material - results will not be averaged across material boundaries
3. All Entities - results will be averaged at all common nodes
4. Target Entities - results will be averaged only between the elements selected as the target entities
5. Element Type - results will not be averaged across boundaries between different types of elements
6. None - no averaging of results between any elements. Fringe values are based on individual
element results only.
When the averaging domain is set to anything other than All Entities the resulting graphics may not be
as smooth but this is a more accurate representation of the results when discontinuities exist in the
structure.
You can compare the difference between using an averaging domain of None to one of the other averaged
domains to assess your mesh quality. If there is a large difference in the maximum result these between
the two fringe plots, especially at locations that do not have sharp corners or breaks in the model, the
mesh may need to be refined in that area.
The Result averaging method controls how SimXpert converts the results from pure data at element
centroids, corners, and nodes to the actual continuous graphical representation. The following methods
are available.
1. Derive/Average - calculates the selected result value first then applies a simple average of all the
contributing nodes.
2. Average/Derive - averages the contributions of the common nodes then derives the result
3. Difference - computes the minimum and maximum results for the elements sharing a common
node. The difference between the maximum and minimum contributor to each node is plotted.
The fringe plot of this difference is an indicator of mesh quality: result differences between
neighboring elements should not be large. If large discontinuities are found the mesh should be
refined in that area. Nodal results will have zero max-difference.
220
Results
Vector attributes - allows you to choose whether to scale vector lengths are adjusted, and how the vectors
are color coded.
• Animation
Animation attributes- controls number of frames, delay, and type of scale factor used to generate the
animation (Sinusoid, Linear, etc.)
Chart Plots
SimXpert can create XY plots of results. Control over the contents of a chart is provided by the chart
properties form. The chart properties form is accessed by clicking on the Results toolbox and selecting
Chart. It allows you to control the Results Cases and Results Type shown on the XY data plots. Choose
the Results Cases and the Results Type from the appropriate list boxes. You can limit the Results Cases
you see in the list box by using the Filter button.The Chart Properties form controls whether an XY plot
is displayed, and what type of chart to display.
Example 221
Example
ET = 29.E3 psi
Steps
Following are the steps to complete the impact model.
1. Launch SimXpert
Select MD Explicit as the Workspace
14. Specify the time step for d3plot and time hstory output
Job Parameters -> PARAM
Enter DYDTOUT for N1
Enter 0.0001 for V1
Enter STEPFCTL for N1
Enter 0.1 for V1
Click Create
Click Exit
18. Submit the exported input file tapered_beam.bdf for analysis by MD Nastran.
Open MD Nastran
Select the input file: tapered_beam.bdf
Click Open
Click Run
19. Attach the Analysis Results (after the spawned MD Nastran job is completed)
File -> Attach Results
Select Results
Click the File Path icon
Select the result file: tapered_beam.dytr.d3plot
Click Open
Click OK