Вы находитесь на странице: 1из 337

Modeling, Integrated Design & Analysis Software

Benchmark Series








Contents











































Benchmark Series
midas NFX









Contents



Benchmark Manual Contents

1. Introduction 1-1
2. Linear static analysis 2-1
2.1 Twisted beam under tip load / 2-1
2.2 Curved cantilevered beam under tip load / 2-3
2.3 Cantilevered beam under various tip loads / 2-6
2.4 Pinched hemispherical shell with hole / 2-11
2.5 Elliptic membrane under uniform outward pressure / 2-13
2.6 Plane truss under concentrated load / 2-15
2.7 Skew plate under uniform pressure / 2-16
2.8 Thick plate under uniform pressure / 2-18
2.9 Cantilever with variable thickness / 2-20
2.10 Scordellis-Lo barrel vault / 2-22
2.11 Z-section cantilever / 2-24
2.12 Thick cylinder subjected to internal pressure load / 2-26
2.13 Hemisphere under point loads / 2-29
2.14 Uniformly distributed load on a circular plate / 2-31
2.15 Thick/Thin square plates / 2-33
2.16 Cylindrical shell patch test / 2-36
2.17 Membrane with hot-spot / 2-38
2.18 Solid cylinder/taper/sphere temperature loading / 2-40
2.19 Laminated strip under three-point bending / 2-43
2.20 Laminated thick cylinder / 2-45
2.21 Three-layer sandwich shell under normal pressure / 2-48
2.22 Axisymmetric shell under pressure / 2-50
References / 2-51
3. Free vibration analysis 3-1
3.1 Pin-ended cross / 3-1
3.2 Pin-ended double cross / 3-3
3.3 Free square frame / 3-5
3.4 Cantilever with off-center point masses / 3-7







Contents



3.5 Deep simply-supported beam / 3-9
3.6 Circular ring / 3-11
3.7 Thin square cantilever plate / 3-13
3.8 Thin square cantilever plate-mesh distortion / 3-17
3.9 Simply-supported thick square plate / 3-19
3.10 Simply-supported thin square plate / 3-22
3.11 Free thin square plate / 3-25
3.12 Simply-supported thin annular plate / 3-28
3.13 Simply-supported thick annular plate / 3-31
3.14 Simply-supported thick annular plate (axisymmetric) / 3-34
3.15 Clamped thin rhombic plate / 3-36
3.16 Clamped thick rhombic plate / 3-38
3.17 Cantilevered square membrane / 3-40
3.18 Cantilevered tapered membrane / 3-42
3.19 Free annular membrane / 3-44
3.20 Cantilever beam mesh distortion / 3-46
3.21 Free cylinder axisymmetric vibration / 3-47
3.22 Thick hollow sphere uniform radial vibration / 3-49
3.23 Cylindrical shell with rigid diaphragm / 3-51
References / 3-54
4. Linear buckling analysis 4-1
4.1 Column buckling / 4-1
4.2 Three member frame / 4-3
4.3 Uniaxially compressed clamped square plate / 4-5
4.4 Rectangular plate under concentrated center loads / 4-7
4.5 Axially compressed cylinder / 4-9
4.6 L-bracket plate under in-plane load / 4-11
References / 4-13
5. Steady-state heat transfer analysis 5-1
5.1 One-dimensional heat transfer with heat source / 5-1
5.2 Two-dimensional heat transfer with convection / 5-3
5.3 Two-dimensional heat transfer in bi-material / 5-5
5.4 One-dimensional heat transfer with radiation / 5-7
5.5 Radiation shape factor calculation / 5-9







Contents



5.6 Perpendicular surfaces in radiant balance / 5-11
References / 5-12
6. Transient heat transfer 6-1
6.1 One-dimensional transient heat transfer - I / 6-1
6.2 One-dimensional transient heat transfer - II / 6-3
6.3 Transient heat transfer with convection / 6-7
6.4 Transient heat transfer with heat generation / 6-9
6.5 Axisymmetric transient heat transfer with convection / 6-11
References / 6-12
7. Linear dynamic analysis 7-1
7.1 Deep simply supported beam / 7-1
7.2 Simply supported thin square plate / 7-4
7.3 Simply supported thick square plate / 7-7
7.4 Laminated strip under three-point bending / 7-11
7.5 Damping options in transient analyses / 7-13
7.6 Response spectrum of a simply supported beam / 7-15
7.7 Linear dynamic analyses of a rod / 7-18
7.8 Double cantilever under multiple base excitations / 7-21
7.9 Cantilever subjected to earthquake motion / 7-24
7.10 Modal frequency response with residual modes / 7-29
7.11 Steady state dynamics for two-dimensional elements / 7-31
7.12 Cantilever with a tip mass / 7-33
References / 7-35
8. Prestressed analysis 8-1
8.1 Eigenvalue of a beam prestressed by axial force / 8-1
8.2 Vibration of a cable under tension / 8-3
8.3 Vibration of a rotating plate / 8-4
References / 8-6
9. Geometric nonlinearity 9-1
9.1 Snap-through problem for a simple truss element / 9-1
9.2 Bifurcation problem for a simple truss element / 9-4
9.3 Snap-back problem for a simple truss element / 9-6
9.4 Rigid-body rotation / 9-9







Contents



9.5 Straight cantilever / 9-11
9.6 Curved cantilever / 9-18
9.7 Z-shaped 3D cantilever / 9-21
9.8 Torsional buckling of cantilever / 9-23
9.9 Pear-shaped cylinder under end shortening / 9-25
9.10 Curved elastic cantilever under transverse end load / 9-27
9.11 Buckling of a flat plate subjected to in-plane shear / 9-29
9.12 Hinged spherical shell under pressure loading / 9-31
9.13 Pinched hemispherical shell / 9-33
9.14 Reticulated space trusses / 9-35
9.15 Hinged panel subjected to a point load / 9-37
9.16 Circular plate under uniform surface pressure / 9-39
9.17 Lee's frame buckling problem / 9-41
9.18 Pull-out of an open cylindrical shell / 9-43
9.19 Slit annular plate under line force / 9-45
9.20 Pinched cylinder with rigid diaphragms / 9-47
9.21 Composite cylindrical shell under uniform load / 9-49
9.22 Semi-cylindrical laminated shell / 9-51
References / 9-53
10. Material nonlinearity 10-1
10.1 2D plane strain plasticity / 10-1
10.2 2D plane stress plasticity / 10-5
10.3 3D plasticity / 10-9
10.4 Kinematic hardening plasticity / 10-13
10.5 Two-bar assembly plasticity / 10-16
10.6 Rigid punch plasticity / 10-20
10.7 Thermal ratcheting of uniform beam / 10-22
10.8 An extended straight truss / 10-25
10.9 Square plate under uniformly distributed load / 10-26
10.10 Uniformly loaded circular plate / 10-28
10.11 Two coaxial tubes / 10-30
10.12 Residual stress problem / 10-32
10.13 Nonlinear equation solution tests / 10-34
10.14 Stiffened cylindrical panel / 10-37







Contents



10.15 Necking of a circular bar / 10-39
10.16 Pressurized rubber disc / 10-41
10.17 Inflation of a spherical rubber balloon / 10-43
10.18 Inflation of an axisymmetric ellipsoidal balloon / 10-45
10.19 Contact between a rigid body and a hyperelastic body / 10-47
References / 10-49
11. Contact anal ysi s 11-1
11.1 Contact patch test / 11-1
11.2 Rigid punch contact / 11-3
11.3 Hertzian contact / 11-5
11.4 Sliding wedge with linear springs / 11-7
11.5 Cantilever beam loaded against a rigid curvilinear surface / 11-8
11.6 Sliding and rolling of a ring on a rigid surface / 11-9
11.7 Two contacting rings / 11-11
11.8 Crushing of a pipe / 11-13
References / 11-15
12. Explicit dynamic analysis 12-1
12.1 Impact load on a rod by a mass at a constant velocity / 10-1
12.2 Flexure of a deep beam / 10-4
12.3 Simply supported thick square plate / 10-6
12.4 Laminated strip under three-point bending / 10-8
12.5 Clamped cylindrical panel under impulsive load / 10-10
12.6 Impact of a rod on rigid wall / 12-11
12.7 Frictional contact between pad and rotating drum / 12-13
12.8 Pre-tensioned belt on a rigid drum / 12-16
12.9 Clamped shallow spherical cap / 12-18
12.10 Hemispherical Load / 12-20
12.11 Twisted Cantilever Beam / 12-22
12.12 Billet Upset / 12-23
12.13 Actuated beam connected by revolute joints / 12-26
References / 12-28



1
midas NFX Benchmark Series 1-1
1. Introduction
This is the benchmark series for midas NFX to demonstrate the performance and capabilities of the midas
NFX solver performing wide range of finite element analyses ranging from linear static to explicit dynamic
analyses. We have selected well-known problems that have gained wide-spread acceptance as standards for
testing the performance of finite element programs. These include problems from the suite of NAFEMS
(National Agency for Finite Element Methods and Standards, National Engineering Laboratory, Glasgow, U.K.)
benchmarks and problems included in several technical publications and books that are widely used to validate
the performance of finite element solvers. Summary of the benchmark problems organized in terms of their
references is given in Table 1.1

Table 1. 1 Summary of benchmark problems organized with respect to references
Reference type
Analysis
NAFEMS Journal papers Books and others Total
Linear static analysis 12 5 5 22
Free vibration analysis 22 1 23
Linear buckling analysis 2 4 6
Steady-state heat transfer 2 4 6
Transient heat transfer 2 3 5
Linear dynamic analysis 4 3 5 12
Prestressed analysis 1 2 3
Geometry nonlinearity 14 2 6 22
Material nonlinearity 10 1 8 19
Contact analysis 7 1 8
Explicit dynamic analysis 2 6 5 13
Total 75 21 43 139

Performance of the midas NFX solver applied to linear static, free vibration and linear buckling problems is
demonstrated in chapters 2, 3 and 4, respectively. Collection of benchmarks problems in steady-state and
transient heat transfer analyses are included in chapters 5 and 6. Performance of the solver in handling linear
dynamics and prestressed problems are exhibited in chapters 7 and 8. Benchmarks for advanced nonlinear
problems including geometric, material and boundary nonlinearities are included in chapters 9 through 11.
Finally, results of nonlinear explicit dynamics benchmark tests are summarized in chapter 12.

Due to the inherent limitation of individual finite element model, a given set of elements tend to yield superior
performance for certain set of problems while the results for other types of problems may be poor. The basic
philosophy behind midas NFX benchmark series is to transparently provide both the performance of midas NFX
solver and some of its limitations. Therefore, for each problem, especially for linear problem, we have
thoroughly tested as many types of element models as possible with different base geometries and formulations.

Each problem section is organized as three parts; section header, problem description and result summary. In
the section header, information regarding the reference, type of elements tested and midas NFX files used to
solve the problem are included. It is followed by a brief description of the benchmark problem. The results are
provided in tables and figures. Many problems have tables organized in terms of the element types used to obtain

-
Introduction
1-2
the results. Table 1.2 summarizes the element groups and element types that are mentioned in the tables.

Table 1.2 Element groups and element types included in each group
Element Group Element type
Bar and rod BAR-2, ROD-2
Membrane, shell, plane-strain TRIA-3, TRIA-6, QUAD-4, QUAD-8
Surface TRIAS-3, TRIAS-6, QUADS-4, QUADS-8
Layered shell TRIAL-3, TRIAL-6, QUADL-4 QUADL-8
Continuum solid TETRA-4, TETRA-10, PYRAM-5, PYRAM-13, PENTA-6, PENTA-15,
HEXA-8, HEXA-20
Layered solid PENTAL-6, PENTAL-15, HEXAL-8, HEXAL-20
Axisymmetric solid TRIAX-3, TRIAX-6, QUADX-4, QUADX-8

For each problem, analyses can be carried out inside the midas NFX GUI environment using the binary input
file (.nfx) mentioned in the problem header.



1
midas NFX Benchmark Series 2-1
2.1 Twisted beam under tip load

REFERENCE MacNeal et al. [2-1]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic01.nfx





Figure 2.1.1 Twisted beam model

Material data Youngs modulus
Poissons ratio
E = 29 MPa
v = 0.22
Section property Thickness t = 0.32 m

Table 2.1.1 Displacements
Y
u and
Z
u at point A obtained using shell elements
*
obtained using shell element formulations with 6-dof per node. Shell edge is considered to be folded if the angle between
adjacent shells is greater than 10.
X
Z
Y
Out-of-plane shear
F
Y
= 1 N
In-plane shear
F
Z
= -1 N
A
1.1 12
Units : m

Load
A
Z
u [m]
in-plane shear
Z
F
A
Y
u [m]
out-of-plane shear
Y
F
Reference 5.42410
-3
1.75410
-3

Element type Number of elements
TRIA-3 2(122)
6.09210
-3

5.31210
-3
*
1.67410
-3

1.70910
-3
*
QUAD-4 212 5.39210
-3
1.73310
-3

TRIA-6 2(122) 5.40810
-3
1.75410
-3

QUAD-8 212 5.41510
-3
1.75510
-3

Figure 2.1.1 shows a twisted beam structure with two tip loads. Fixed boundary condition is assigned to root of
the beam. A unit load is applied at the tip in two different directions. Displacements at point A are determined.
Solid model has single layer in the thickness direction.


Linear static analysis
2-2
Table 2.1.2 Displacements
Y
u and
Z
u at point A obtained using solid elements

Table 2.1.3 Displacements
Y
u and
Z
u at point A obtained using layered solid elements

[ Note ]
If there is no node at point A in solid and layered solid element, the averaged value is calculated from the nearest two nodes.


Load
A
Z
u [m]
in-plane shear
Z
F
A
Y
u [m]
out-of-plane shear
Y
F
Reference
5.42410
-3
1.75410
-3

Element type Number of elements
TETRA-4 144 0.38710
-3
0.27110
-3

PENTA-6 481 2.34610
-3
0.74610
-3

HEXA-8 2121 5.42110
-3
1.74310
-3

TETRA-10 144 5.44410
-3
1.77110
-3

PENTA-15 481 5.41110
-3
1.75010
-3

HEXA-20 2121 5.42610
-3
1.75410
-3


Load
A
Z
u [m]
in-plane shear
Z
F
A
Y
u [m]
out-of-plane shear
Y
F
Reference
5.42410
-3
1.75410
-3

Element type Number of elements
PENTAL-6 481 5.35210
-3
1.47510
-3

HEXAL-8 2121 5.43310
-3
1.74510
-3

PENTAL-15 481 5.41110
-3
1.75010
-3

HEXAL-20 2121 5.41310
-3
1.75010
-3



3
midas NFX Benchmark Series 2-3
2.2 Curved cantilevered beam under tip load

REFERENCE MacNeal et al. [2-1]
KEYWORDS bar elements, shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic02.nfx





Figure 2.2.1 Curved cantilever beam model

Material data

Youngs modulus
Poissons ratio
E = 1.010
7
psi
v = 0.25
Section property Thickness t = 0.1 in

Table 2.2.1 Displacements
Y
u and
Z
u at point A obtained using bar elements



X
Z
Y
Out-of-plane
shear
F
Z
= 1 lbf
In-plane shear
F
Y
= -1 lbf
8.24
8.64
A
Units : in

Load
A
Y
u [in]
in-plane shear
Y
F
A
Z
u [in]
out-of-plane shear
Z
F
Reference -8.73410
-2
5.02210
-1

Element type Number of elements
BAR-2 6 -8.73510
-2
4.96810
-1

Figure 2.2.1 shows a curved beam structure under tip loads in two different directions. Fixed boundary
condition is assigned to the root of the beam. A unit load is applied at the free end in two different directions.
Displacements at point A are determined. Solid model has single layer in the thickness direction.


Linear static analysis
2-4
Table 2.2.2 Displacements
Y
u and
Z
u at point A obtained using shell elements
*
obtained using shell element formulations with 6-dof per node.

Table 2.2.3 Displacements
Y
u and
Z
u at point A obtained using solid elements
*
obtained using enhanced higher order tetrahedral formulation.



Load
A
Y
u [in]
in-plane shear
Y
F
A
Z
u [in]
out-of-plane shear
Z
F
Reference -8.73410
-2
5.02210
-1

Element type Number of elements
TRIA-3 1(62)
-0.22210
-2

7.05910
-2
*
4.34710
-1

4.98410
-1
*
QUAD-4 16 -8.54310
-2
4.73910
-1

TRIA-6 1(62) -8.75610
-2
4.72710
-1

QUAD-8 16 -8.85110
-2
4.81310
-1


Load
A
Y
u [in]
in-plane shear
Y
F
A
Z
u [in]
out-of-plane shear
Z
F
Reference -8.73410
-2
5.02210
-1

Element type Number of elements
TETRA-4 76 -0.23610
-2
0.04410
-1

PENTA-6 1(62)1 -0.22110
-2
0.34710
-1

HEXA-8 161 -8.53510
-2
4.41510
-1

TETRA-10 76
-8.78310
-2

-9.043310
-2
*
4.59710
-1

4.73710
-1
*
PENTA-15 1(62)1 -8.75410
-2
4.68310
-1

HEXA-20 161 -8.85010
-2
4.76610
-1



5
midas NFX Benchmark Series 2-5
Table 2.2.4 Displacements
Y
u and
Z
u at point A obtained using layered solid elements

[ Note ]
If there is no node at point A, the averaged value is calculated from the nearest two or four nodes.


Load
A
Y
u [in]
in-plane shear
Y
F
A
Z
u [in]
out-of-plane shear
Z
F
Reference -8.73410
-2
5.02210
-1

Element type Number of elements
PENTAL-6 1(62)1 -0.22110
-2
3.20910
-1

HEXAL-8 161 -8.53410
-2
4.72710
-1

PENTAL-15 1(62)1 -8.75410
-2
4.68310
-1

HEXAL-20 161 -8.81210
-2
4.67410
-1



Linear static analysis
2-6
2.3 Cantilevered beam under various tip loads

REFERENCE MacNeal et al. [2-1]
KEYWORDS bar elements, shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic03.nfx







Figure 2.3.1 Cantilever beam model

Material data
Youngs modulus
Poissons ratio
E = 1.010
7
psi
v = 0.3
Section property Thickness t = 0.1 in

Table 2.3.1 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using bar elements

Mesh A
Mesh B
45
o
45
o
Mesh C
45
o
A
X
Y
Z
X
Y
Z
X
Y
Z
B
C
A
B
C
A
B
C

Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

BAR-2 6 3.00010
-5
4.32110
-1
1.08110
-1
3.41010
-2

Figure 2.3.1 shows a straight beam under tip loads in three different directions and twisting moment. Fixed
boundary condition is assigned to the root of the beam. Unit forces and moment are applied at the free end.
Displacements and twist angle at point A are determined by using three kinds of meshes (mesh A, B and C)
using shell, solid and layered solid elements. Same problem was also solved using bar elements. Solid model
has single layer of elements in the thickness direction. Twist angle at point A is determined from relative
displacement of point B and C.


7
midas NFX Benchmark Series 2-7
Table 2.3.2 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using shell elements, mesh A
*
obtained using shell element formulations with 6-dof per node.

Table 2.3.3 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using shell elements, mesh B
*
obtained using shell element formulations with 6-dof per node.



Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

TRIA-3 1(62)
3.00010
-5

4.19910
-1

0.03410
-1

0.81110
-1
*
2.97010
-2


QUAD-4 16 3.00010
-5

4.23910
-1

4.23910
-1
*
1.07410
-1

1.06810
-1
*
3.01910
-2

3.51210
-2
*
TRIA-6 1(62) 3.00010
-5
4.28310
-1
1.07410
-1
3.04710
-2

QUAD-8 16 3.00010
-5
4.29810
-1
1.08110
-1
3.03210
-2


Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

TRIA-3 1(62)
3.00010
-5

4.17710
-1

0.01610
-1

0.83510
-1
*
3.13710
-2


QUAD-4 16
3.00010
-5

4.17210
-1

4.18610
-1
*
0.24010
-1

1.06310
-1
*
3.02010
-2

3.51310
-2
*
TRIA-6 1(62) 3.00010
-5
4.28010
-1
1.05010
-1
3.06210
-2

QUAD-8 16 3.00010
-5
4.30210
-1
1.05710
-1
3.01610
-2



Linear static analysis
2-8
Table 2.3.4 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using shell elements, mesh C
*
obtained using shell element formulations with 6-dof per node.

Table 2.3.5 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using solid elements, mesh A
*
obtained using enhanced higher order tetrahedral formulation.







Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

TRIA-3 1(62)
3.00010
-5


4.20110
-1


0.02410
-1

0.70010
-1
*


3.50810
-2


QUAD-4 16
3.00010
-5


4.23810
-1

4.23610
-1
*
0.86010
-1

1.06410
-1
*
3.02210
-2

3.51510
-2
*
TRIA-6 1(62) 3.00010
-5
4.29010
-1
1.05810
-1
3.04910
-2

QUAD-8 16 3.00010
-5
4.30210
-1
1.08010
-1
3.02910
-2


Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

TETRA-4 222 3.00010
-5
0.03110
-1
0.02310
-1
0.02910
-2

PENTA-6 1(62)1 3.00010
-5
0.51010
-1
0.03410
-1
0.07710
-2

HEXA-8 161 3.00010
-5
4.24910
-1
1.07210
-1
2.92010
-2

TETRA-10 222
3.00010
-5


3.97310
-1

4.08510
-1
*
1.02110
-1

1.05010
-1
*
2.88310
-2

2.96210
-2
*
PENTA-15 1(62)1 3.00010
-5
4.30110
-1
1.07410
-1
2.96010
-2

HEXA-20 161 3.00010
-5
4.32210
-1
1.08110
-1
2.93510
-2



9
midas NFX Benchmark Series 2-9
Table 2.3.6 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using solid elements, mesh B
*
obtained using enhanced higher order tetrahedral formulation.

Table 2.3.7 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using solid elements, mesh C
*
obtained using enhanced higher order tetrahedral formulation.







Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

TETRA-4 222 3.00010
-5
0.02510
-1
0.01510
-1
0.03010
-2

PENTA-6 1(62)1 3.00010
-5
0.10110
-1
0.01610
-1
0.10210
-2

HEXA-8 161 3.00010
-5
0.49010
-1
0.23110
-1
2.90910
-2

TETRA-10 222
3.00010
-5


3.89610
-1

4.01710
-1
*
0.99910
-1

1.03010
-1
*
2.88010
-2

2.96110
-2
*
PENTA-15 1(62)1 3.00010
-5
4.27010
-1
1.04910
-1
2.95610
-2

HEXA-20 161 3.00010
-5
4.29810
-1
1.05710
-1
2.98510
-2


Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

TETRA-4 222 3.00010
-5
0.03010
-1
0.02210
-1
0.03510
-2

PENTA-6 1(62)1 3.00010
-5
0.14910
-1
0.02310
-1
0.15910
-2

HEXA-8 161 3.00010
-5
2.78610
-1
0.84910
-1
2.93110
-2

TETRA-10 222
3.00010
-5


4.05310
-1

4.17710
-1
*
1.03310
-1

1.06510
-1
*
2.88810
-2

2.97310
-2
*
PENTA-15 1(62)1 3.00010
-5
4.24110
-1
1.05810
-1
2.96610
-2

HEXA-20 161 3.00010
-5
4.31910
-1
1.08010
-1
2.94810
-2



Linear static analysis
2-10
Table 2.3.8 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using layered solid elements, mesh A

Table 2.3.9 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using layered solid elements, mesh B

Table 2.3.10 Displacements , ,
X Y Z
u u u and rotation
X
u at point A obtained using layered solid elements, mesh C

[ Note ]
If there is no node at point A, B or C, the averaged value is calculated from the nearest two or four nodes.


Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

PENTAL-6 1(62)1 3.00010
-5
4.23210
-1
0.03410
-1
1.99810
-2

HEXAL-8 161 3.00010
-5
4.24910
-1
1.07210
-1
3.02010
-2

PENTAL-15 1(62)1 3.00010
-5
4.30110
-1
1.07410
-1
2.96010
-2

HEXAL-20 161 3.00010
-5
4.28910
-1
1.07510
-1
2.90210
-2


Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

PENTAL-6 1(62)1 3.00010
-5
4.05810
-1
0.01610
-1
1.97510
-2

HEXAL-8 161 3.00010
-5
4.21410
-1
0.23010
-1
2.76310
-2

PENTAL-15 1(62)1 3.00010
-5
4.27110
-1
1.04910
-1
2.95610
-2

HEXAL-20 161 3.00010
-5
4.09810
-1
0.97810
-1
2.90210
-2


Load
A
X
u [in]
extension
X
F
A
Y
u [in]
out-of-plane
Y
F
A
Z
u [in]
in-plane
Z
F
A
X
u [rad]
twist
X
M
Reference 3.00010
-5
4.32110
-1
1.08110
-1
3.40810
-2

Element
type
Number of
elements

PENTAL-6 1(62)1 3.00010
-5
4.01710
-1
0.02310
-1
2.11910
-2

HEXAL-8 161 3.00010
-5
4.26510
-1
0.84910
-1
2.67210
-2

PENTAL-15 1(62)1 3.00010
-5
4.24210
-1
1.05810
-1
2.96610
-2

HEXAL-20 161 3.00010
-5
4.23610
-1
1.06010
-1
2.90210
-2



11
midas NFX Benchmark Series 2-11
2.4 Pinched hemispherical shell with hole

REFERENCE MacNeal et al. [2-1], Simo et al. [2-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic04.nfx





Figure 2.4.1 Pinched hemisphere shell model with hole

Material data
Youngs modulus
Poissons ratio
E = 6.82510
7
psi
v = 0.3
Section property Thickness t = 0.04 in


Z
18
o
X
F = 1
Y
F = 1
A
B
Figure 2.4.1 shows one quadrant of pinched hemispherical shell with inward and outward forces at point B and
point A. The hemisphere has 18 hole at the top, and the quadrant of the hemisphere is modeled utilizing
symmetric boundary conditions. Displacement at point A is determined. Solid model has single layer in the
thickness direction.


Linear static analysis
2-12
Table 2.4.1 Displacement
X
u at point A obtained using shell elements

Table 2.4.2 Displacement
X
u at point A obtained using solid elements

Table 2.4.3 Displacement
X
u at point A obtained using layered solid elements
*
obtained using higher order layered hexahedral elements with reduced integration

[ Note ]
If the mesh is not perfectly symmetric, averaged displacement from point A and B is used.
If there is no node at point A and B, the averaged value is calculated from the nearest two nodes.


A
X
u [in]
Reference 9.410
-2

Ref. 2-1
, 9.310
-2

Ref. 2-2

Number of elements per side 4 8 12
Element
type
TRIA-3 9.47910
-2
9.37510
-2
9.31710
-2

QUAD-4 9.70010
-2
9.44510
-2
9.37210
-2

TRIA-6 0.93210
-2
4.03910
-2
7.16710
-2

QUAD-8 8.07310
-2
9.29610
-2
9.33510
-2


A
X
u [in]
Reference 9.410
-2

Ref. 1
, 9.310
-2

Ref. 2

Number of elements per side 4 8 12
Element
type
PENTA-6 0.00410
-2
0.01510
-2
0.03310
-2

HEXA-8 1.01710
-2
7.41110
-2
8.87810
-2

PENTA-15 0.52510
-2
3.61110
-2
6.91410
-2

HEXA-20 6.79710
-2
9.25510
-2
9.35510
-2


A
X
u [in]
Reference 9.410
-2

Ref. 1
, 9.310
-2

Ref. 2

Number of elements per side 4 8 12
Element
type
PENTAL-6 0.37810
-2
2.86010
-2
6.21610
-2

HEXAL-8 9.23510
-2
9.30610
-2
9.31910
-2

PENTAL-15 0.52510
-2
3.60910
-2
6.90810
-2

HEXAL-20
0.20010
-2

6.44010
-2
*
2.37810
-2

9.26510
-2
*
5.89710
-2

9.36610
-2
*


13
midas NFX Benchmark Series 2-13
2.5 Elliptic membrane under uniform outward pressure

REFERENCE NAFEMS [2-3]
KEYWORDS membrane elements, solid elements
MODEL FILENAME Linearstatic05.nfx





Figure 2.5.1 Elliptic membrane model

Material data
Youngs modulus
Poissons ratio
E = 210 GPa
v = 0.3
Section property Thickness t = 0.1 m







X
Y
A
B
C
2 2
3.25 3.25
1
1
2.75
2.75
Units : m
t = 0.1
Figure 2.5.1 shows the elliptic membrane problem. The membrane is subjected to uniform outward normal
pressure of 10MPa at the outer edge BC. A quarter of the elliptic membrane is discretized with symmetric
boundary conditions using coarse and fine meshes of membrane and solid elements. Y-normal stress at point A
is obtained and compared with the reference solution given in the standard NAFEMS benchmarks.


Linear static analysis
2-14
Table 2.5.1 Stress
YY
o at point A obtained using shell elements
*
obtained using shell element formulations with 6-dof per node.

Table 2.5.2 Stress
YY
o at point A obtained using solid elements





A
YY
o [MPa]
Reference 92.7
Number of elements per side 32 64
Element
type
TRIA-3
52.9

65.4*
72.9

84.1*
QUAD-4
60.9

67.0*
80.2

83.3*
TRIA-6 87.7 93.0
QUAD-8 85.4 91.7

A
YY
o [MPa]
Reference 92.7
Number of elements per side 32 64
Element
type
PENTA-6 53.3 73.3
HEXA-8 60.9 80.3
PENTA-15 89.0 93.2
HEXA-20 86.7 91.8


15
midas NFX Benchmark Series 2-15
2.6 Plane truss under concentrated load

REFERENCE McCormac [2-4]
KEYWORDS rod elements
MODEL FILENAME Linearstatic06.nfx





Figure 2.6.1 Plane truss model

Material data Youngs modulus E = 3.010
4
psi
Section property Area A = 1.0 in
2
(element - a)
A = 2.0 in
2
(element - b)
A = 1.5 in
2
(element - c)
A = 3.0 in
2
(element - d)
A = 4.0 in
2
(element - e)

Table 2.6.1 Displacement
Z
u at point A obtained using rod elements




20
kips
20
kips
20
kips
A
B C
a
a a
a
b b
c c d d
d
e e
180 180 180 180
180
X
Z
Units : in

A
Z
u [in]
Reference -2.63
ROD-2 13 elements -2.63
Figure 2.6.1 shows a two-dimensional truss structure with different section properties. The truss structures in
loaded vertically at points A, B and C. The maximum vertical deflection which occurs at point A is determined
utilizing rod elements.


Linear static analysis
2-16
2.7 Skew plate under uniform pressure

REFERENCE NAFEMS [2-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic07.nfx







Figure 2.7.1 Skew plate model

Material data Youngs modulus
Poissons ratio
E = 210 GPa

v = 0.3
Section property Thickness t = 0.01 m

Table 2.7.1 Maximum principal stress at bottom surface
1 P
o obtained using shell elements

X
Y
A B
C
D
E
1
1
30
o
150
o
150
o
30
o
Units : m
t = 0.01
simply-supported on all edges
AB, BC, CD and DA

1
E
P
o [MPa]
Reference 0.802
Number of elements per side 2 4 8
Element
type
TRIA-3 0.804 0.783 0.804
QUAD-4 0.666 0.799 0.799
TRIA-6 1.145 0.898 0.835
QUAD-8 0.668 0.771 0.763
A skewed plate model with obtuse angles of 150 degrees (see Figure 2.7.1) subjected to uniform pressure of
700 Pa is evaluated. Maximum principal stress on the lower surface at the center of the plate (point E) is
obtained using shell, solid and layered solid elements. The reference value is taken from the standard
NAFEMS benchmarks.


17
midas NFX Benchmark Series 2-17
Table 2.7.2 Maximum principal stress at bottom surface
1 P
o obtained using solid elements
*
obtained from surface elements attached to lower face of solid mesh
Table 2.7.3 Maximum principal stress at bottom surface
1 P
o obtained using layered solid elements
*
obtained using higher order layered hexahedral elements with reduced integration




1
E
P
o [MPa]
Reference 0.802
Number of elements per side 2 4 8
Element
type
PENTA-6
0.329
0.327
*

0.087
0.845
*

0.251
0.251
*

HEXA-8
0.336
0.364
*

0.675
0.747
*

0.736
0.838
*

PENTA-15
0.570
0.570
*

0.769
0.773
*

0.792
0.793
*

HEXA-20
0.533
0.527
*

0.736
0.719
*

0.723
0.722
*


1
E
P
o [MPa]
Reference 0.802
Number of elements per side 2 4 8
Element
type
PENTAL-6 0.340 0.647 0.744
HEXAL-8 0.336 0.674 0.731
PENTAL-15 0.570 0.769 0.792
HEXAL-20
0.168
0.533
*

0.385
0.736
*

0.602
0.723
*



Linear static analysis
2-18
2.8 Thick plate under uniform pressure

REFERENCE NAFEMS [2-3]
KEYWORDS solid elements, layered solid elements
MODEL FILENAME Linearstatic08.nfx





Figure 2.8.1 Thick plate model

Material data
Youngs modulus
Poissons ratio
E = 210 GPa

v = 0.3









X
Z
Y
0.6
X
Y
1.25 2
1
1.75
1
2
2
2
= + |
.
|

\
|
y
x
1
25 . 3 25 . 3
2 2
= |
.
|

\
|
+ |
.
|

\
| y x
Units : m
z displacements fixed along mid-plane
Figure 2.8.1 shows a thick plate clamped on the middle surface, subjected to uniform pressure of 1MPa applied
on the top surface. Utilizing symmetry, a quarter of the plate is modeled using solid and layered solid elements.
Normal stress (
D
YY
o ) at point D is evaluated. The reference solution is taken from the NAFEMS standard
benchmarks.


19
midas NFX Benchmark Series 2-19
Table 2.8.1 Stress
D
YY
o at point D obtained using solid elements
*
obtained from surface elements attached to upper face of solid mesh

Table 2.8.2 Stress
D
YY
o at point D obtained using layered solid elements



D
YY
o (MPa)
Reference 5.38
Number of elements per side 32 64
Element
type
TETRA-4
1.889
2.358
*

1.970
2.924
*

PENTA-6
4.878
4.501
*

5.751
5.392
*

HEXA-8
5.321
6.544
*

5.578
5.786
*

TETRA-10
6.596
5.878
*

5.943
5.250
*

PENTA-15
6.131
5.605
*

6.039
5.302
*

HEXA-20
5.421
5.252
*

5.662
4.945
*


D
YY
o (MPa)
Reference 5.38
Number of elements per side 32 64
Element
type
PENTAL-6 4.878 5.751
HEXAL-8 5.321 5.577
PENTAL-15 6.131 6.593
HEXAL-20 5.422 6.321


Linear static analysis
2-20
2.9 Cantilever with variable thickness

REFERENCE Young et al. [2-5]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic09.nfx






Figure 2.9.1 Cantilever model with variable thickness

Material data Youngs modulus
Poissons ratio
E = 1.010
3
ksi

v = 0.3


Table 2.9.1 Displacement
A
Z
u subjected to end moment, and
A
Y
u subjected to end tension at point A obtained using shell
elements

20
3
F
Y
= 50 Kips/in
M
X
= 3 in-Kips/in
1
100
A
Y
X
Z
Units : in

Load
A
Z
u [in]
End moment
A
Y
u [in]
End tension
Reference 20.0 2.7465
Element type Number of elements
TRIA-3 (52) 19.00 2.6924
QUAD-4 5 19.98 2.7252
TRIA-6 (52) 19.28 2.7096
QUAD-8 5 20.07 2.7364
Cantilever model with thickness that varies linearly along the length is shown in Figure 2.9.1. The cantilever is
subjected to either end moment or tension. The root of the cantilever is fixed. For the end moment load, the
vertical displacement at point A is determined. For the tension load, the longitudinal displacement at point A is
determined. The cantilever is modeled using shell with variable thickness, solid and layered solid elements.


21
midas NFX Benchmark Series 2-21
Table 2.9.2 Displacement
A
Z
u subjected to end moment, and
A
Y
u subjected to end tension at point A obtained using solid
elements

Table 2.9.3 Displacement
A
Z
u subjected to end moment, and
A
Y
u subjected to end tension at point A obtained using
layered solid elements

[ Note ]
If there is no node at point A, the averaged value is calculated from the nearest two or four nodes.




Load
A
Z
u [in]
End moment
A
Y
u [in]
End tension
Reference 20.0 2.7465
Element type Number of elements
TETRA-4 60 0.27 2.6930
PYRAM-5 30 0.50 2.6931
PENTA-6 (52)1 6.09 2.7031
HEXA-8 51 19.95 2.7026
TETRA-10 60 19.67 2.7293
PYRAM-13 30 18.99 2.7287
PENTA-15 (52)1 19.79 2.7351
HEXA-20 51 19.97 2.7356

Load
A
Z
u [in]
End moment
A
Y
u [in]
End tension
Reference 20.0 2.7465
Element type Number of elements
PENTAL-6 (52)1 19.42 2.7043
HEXAL-8 51 19.95 2.7017
PENTAL-15 (52)1 19.79 2.7351
HEXAL-20 51 20.05 2.7287


Linear static analysis
2-22
2.10 Scordellis-Lo barrel vault

REFERENCE MacNeal et al. [2-1], Simo et al. [2-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic10.nfx





Figure 2.10.1 Scordellis-Lo barrel vault model

Material data Youngs modulus
Poissons ratio
E = 4.3210
8
lbf/ft
2
v = 0

Table 2.10.1 Vertical displacement
Z
u at point A obtained using shell elements
*
obtained using shell element formulations with 6-dof per node.
X
Y
t = 0.25
40
o
50
Weight density () =
360 lbf/ft
3
R = 25
Units : ft
Z
X
Y
t = 0.25
40
o
50
Weight density () =
360 lbf/ft
3
R = 25
Z
A

Z
u [ft]
Reference -0.3024
Number of elements per side 4 6 8
Element
type
TRIA-3
-0.2064
-0.2832*
-0.2437
-0.2907*
-0.2615
-0.2963*
QUAD-4 -0.3189 -0.3078 -0.3057
TRIA-6 -0.2793 -0.2957 -0.2995
QUAD-8 -0.3002 -0.3006 -0.3007
Figure 2.10.1 shows the Scordelis-Lo roof problem, in which a short cylindrical shell section is subjected to
gravity forces of 360 lbf/ft
3
. Utilizing symmetric boundary conditions, a quarter of the cylinder is modeled
with shell, solid and layered solid elements. The vertical displacement at the mid-point of the free edge at point
A is determined with meshes with different refinement level. The reference value of -0.3024 mentioned in [2-1]
and [2-2] is taken for comparison.


23
midas NFX Benchmark Series 2-23
Table 2.10.2 Vertical displacement
Z
u at point A obtained using solid elements

Table 2.10.3 Vertical displacement
Z
u

at point A obtained using layered solid elements



Z
u [ft]
Reference -0.3024
Number of elements per side 4 6 8
Element
type
PENTA-6 -0.0163 -0.0228 -0.0291
HEXA-8 -0.3146 -0.3087 -0.3071
PENTA-15 -0.2567 -0.2893 -0.2978
HEXA-20 -0.3039 -0.3037 -0.3038

Z
u [ft]
Reference -0.3024
Number of elements per side 4 6 8
Element
type
PENTAL-6 -0.1134 -0.1964 -0.2422
HEXAL-8 -0.3228 -0.3100 -0.3077
PENTAL-15 -0.2567 -0.2893 -0.2978
HEXAL-20 -0.2475 -0.2900 -0.2988


Linear static analysis
2-24
2.11 Z-section cantilever

REFERENCE NAFEMS [2-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic11.nfx






Figure 2.11.1 Z-section cantilever model (all units in meters)

Material data Youngs modulus
Poissons ratio
E = 210 GPa
v = 0.3
Section property Thickness t = 0.1 m


X
Z
Y
50
1
2
10
1
2.5
Units : m
S
S
Figure 2.11.1 shows a Z-section cantilever clamped on one end. Torque moment of 1.2 MN m is applied on the
other end by uniformly distributed edge shears, S = 0.6 MN at each flange. The cantilever model is discretized
using shell, solid and layered solid elements. Axial stress,
A
x
o is evaluated at point A and compared with the
reference solution. The solution from the standard NAFEMS benchmarks is taken as the reference.


25
midas NFX Benchmark Series 2-25
Table 2.11.1 Stress
A
X
o at point A obtained using shell elements
*
obtained using shell element formulations with 6-dof per node.

Table 2.11.2 Stress
A
X
o at point A obtained using solid elements
*
obtained by averaging stresses at top and bottom nodes

Table 2.11.3 Stress
A
X
o at point A obtained using shell elements tied to layered solid elements by contact condition


A
X
o [MPa]
Reference -108
Element type Number of elements
TRIA-3 48
-30.80
-76.65*
QUAD-4 24
-110.23
-115.43*
TRIA-6 48 -107.76
QUAD-8 24 -110.09

A
X
o [MPa]
Reference -108
Element type Number of elements
PENTA-6 80 -3.047*
HEXA-8 40 -107.357*
PENTA-15 80 -104.509*
HEXA-20 40 -107.094*

A
X
o [MPa]
Reference -108
Element type Number of elements
QUAD-4 (PENTAL-6) 16 (16) -110.11
QUAD-4 (HEXAL-8) 16 (8) -112.96


Linear static analysis
2-26
2.12 Thick cylinder subjected to internal pressure load

REFERENCE Young et al. [2-5]
KEYWORDS plane strain elements, solid elements, axisymmetric elements
MODEL FILENAME Linearstatic12_axisymmetric.nfx
Linearstatic12_planestrain.nfx
Linearstatic12_solid.nfx






Figure 2.12.1 Thick cylinder model

Material data Youngs modulus
Poissons ratio
E = 220 GPa
v = 0.0
Section property Major radius
Minor radius
a = 0.2 m
b = 0.1m



b
a

(infinite)
b
a
Plane strain
problem
Axisymmetric
problem
Thick infinite cylinder subjected to internal of 100 MPa applied to the inner surface of the cylinder is shown in
Figure 2.12.1. The model is formulated as a plane problem and an axisymmetric problem. The circumferential
stresses at the inner surfaces as well as the radial displacements at the inner and the outer surfaces are
evaluated. The reference solution is taken from [2-5].



27
midas NFX Benchmark Series 2-27
Table 2.12.1 Radial displacement
R
u at inner and outer surfaces and maximum circumferential stress
max
u
o obtained
using plane strain elements

Table 2.12.2 Radial displacement
R
u at inner and outer surfaces and maximum circumferential stress
max
u
o obtained
using axisymmetric elements


R
u [m]
max
u
o [MPa]
Inner surface Outer surface
Reference 7.93610
-5
6.34910
-5
166.667
Element
type
Number of
elements



TRIA-3
90 7.85910
-5
6.2910
-5
163.608
360 7.91710
-5
6.3410
-5
165.876
QUAD-4
45 7.91110
-5
6.3410
-5
162.943
180 7.93010
-5
6.3510
-5
165.662
TRIA-6
91 7.90910
-5
6.33310
-5
165.200
360 7.92910
-5
6.34510
-5
166.738
QUAD-8
45 7.90210
-5
6.33610
-5
166.063
180 7.93110
-5
6.34510
-5
166.340

R
u [m]
max
u
o [MPa]
Inner surface Outer surface
Reference 7.93610
-5
6.34910
-5
166.667
Element
type
Number of
elements



TRIAX-3
10 7.91310
-5
6.33810
-5
164.949
20 7.93010
-5
6.34610
-5
166.185
QUADX-4
5 7.93710
-5
6.34910
-5
144.444
10 7.93710
-5
6.34910
-5
154.545
TRIAX-6
10 7.93710
-5
6.34910
-5
165.270
20 7.93710
-5
6.34910
-5
166.274
QUADX-8
5 7.93610
-5
6.34910
-5
164.714
10 7.93710
-5
6.34910
-5
166.100


Linear static analysis
2-28
Table 2.12.3 Radial displacement
R
u at inner and outer surfaces and maximum circumferential stress
max
u
o obtained
using solid elements




R
u [m]
max
u
o [MPa]
Inner surface Outer surface
Reference 7.93610
-5
6.34910
-5
166.667
Element
type
Number of
elements



HEXA-8
10 7.93010
-5
6.34610
-5
165.668
80 7.93510
-5
6.34910
-5
166.396
PENTA-6
20 7.89210
-5
6.33910
-5
154.201
160 7.92310
-5
6.34710
-5
160.433
HEXA-20
5 7.90210
-5
6.33610
-5
166.068
40 7.92810
-5
6.34610
-5
166.424
PENTA-15
10 7.90310
-5
6.33110
-5
165.300
80 7.92810
-5
6.34510
-5
166.253


29
midas NFX Benchmark Series 2-29
2.13 Hemisphere under point loads

REFERENCE NAFEMS [2-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic13.nfx





Figure 2.13.1 Hemisphere quadrant model









X
Z
Y
A
B
R = 10
2 kN
2 kN
100
2 2 2
= + + z y x
Units : m
Material data Youngs modulus
Poissons ratio
E = 68.25 GPa
v = 0.3
Section property Thickness t = 0.04 m
Figure 2.13.1 shows a quadrant of the hemispheric shell with inward and outward forces at point A and point
B. Utilizing symmetric conditions, the quadrant model is discretized using shell, solid and layered solid
elements. Single element in the thickness direction is used for meshes with solid and layered solid elements.
Radial displacement at point A is determined. The reference solution is taken from the standard NAFEMS
benchmarks.


Linear static analysis
2-30
Table 2.13.1 Displacement
X
u at point A obtained using shell elements

Table 2.13.2 Displacement
X
u at point A obtained using solid elements

Table 2.13.3 Displacement
X
u at point A obtained using layered solid elements
*
obtained using higher order layered hexahedral elements with reduced integration

[ Note ]
If the mesh is not perfectly symmetric, averaged displacement from point A and B is used.
If there is no node at point A and B, the averaged value is calculated from the nearest two nodes.



A
X
u [m]
Reference 1.85010
-1

Number of elements per side 4 8
Element type
TRIA-3 1.85610
-1
1.85210
-1

QUAD-4 1.08910
-1
1.83210
-1

TRIA-6 0.13210
-2
0.70310
-2

QUAD-8 1.48010
-1
1.83210
-1


A
X
u [m]
Reference 1.85010
-1

Number of elements per side 4 8
Element type
PENTA-6 9.42210
-5
3.53310
-4

HEXA-8 9.95510
-3
1.19210
-1

PENTA-15 7.71110
-3
0.62110
-1

HEXA-20 1.23310
-1


1.81810
-1


A
X
u [m]
Reference 1.85010
-1

Number of elements per side 4 8
Element type
PENTAL-6 0.16510
-1
1.00410
-1

HEXAL-8 1.01510
-1
1.78510
-1

PENTAL-15 7.70410
-3
0.62110
-1

HEXAL-20
2.43510
-3

1.20310
-1
*
0.31210
-1

1.82310
-1
*


31
midas NFX Benchmark Series 2-31
2.14 Uniformly distributed load on a circular plate

REFERENCE Batoz et al. [2-6]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic14.nfx





Figure 2.14.1 Circular plate model

Material data Youngs modulus
Poissons ratio
E = 210 GPa
v = 0.3
Section property Thickness t = 5 mm

Table 2.14.1 Displacement
A
Z
u at point A obtained using shell elements




X
Y
Z
A
2
Units : m

A
Z
u [m]
Reference -6.510
-3

Element type Number of elements
TRIA-3 53 -6.37910
-3

QUAD-4 38 -6.51610
-3

TRIA-6 53 -6.52810
-3

QUAD-8 38 -6.41310
-3

Figure 2.14.1 shows a quarter of a circular plate. The circular plate is clamped around its perimeter and is
subjected to uniform pressure load of 1MPa. Utilizing symmetric boundary conditions, a quarter of the plate is
discretized using shell, solid and layered solid elements. Vertical displacement at the center (point A) is
determined.


Linear static analysis
2-32
Table 2.14.2 Displacement
A
Z
u at point A obtained using solid elements

Table 2.14.3 Displacement
A
Z
u at point A obtained using layered solid elements
*
obtained using higher order layered hexahedral elements with reduced integration


A
Z
u [m]
Reference -6.510
-3

Element type Number of elements
PENTA-6 53 -4.82210
-5

HEXA-8 38 -5.63210
-3

PENTA-15 53 -6.02510
-3

HEXA-20 38 -6.22810
-3


A
Z
u [m]
Reference -6.510
-3

Element type Number of elements
PENTAL-6 53 -5.05010
-3

HEXAL-8 38 -6.45310
-3

PENTAL-15 53 -6.02510
-3

HEXAL-20 38
-6.00010
-3

-6.22810
-3
*


33
midas NFX Benchmark Series 2-33
2.15 Thick/Thin square plates

REFERENCE Zienkiewicz et al. [2-7]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Linearstatic15a.nfx
Linearstatic15b.nfx






Figure 2.15.1 Square plate with clamped edges

Material data Youngs modulus E = 29 KPa (Model A)
E = 29 GPa (Model B)
Poissons ratio v = 0.3

Section property Model A
Model B
Thickness t = 0.1 m
Thickness t = 0.001 m

A
1
1
Uniform Pressure
P = 1 Pa
X
Y
Units : m
Figure 2.15.1 shows a square plate with side length of 1m with clamped edges. Uniform pressure of 1 Pa is
applied. Two configurations are considered. Model A has thickness of 0.1m and Model B has thickness of
0.001m. Utilizing symmetric conditions, a quarter of the plate is discretized using shell, solid and layered solid
elements. Maximum vertical displacement at the center of the plate (point A) is determined for both models.


Linear static analysis
2-34
Table 2.15.1 Displacement
Z
u

obtained at point A using shell elements, t=0.1 m (Model A)

Table 2.15.2 Displacement
Z
u

obtained at point A using solid elements, t=0.1 m (Model A)

Table 2.15.3 Displacement
Z
u obtained at point A using layered solid elements, t=0.1 m (Model A)



A
Z
u [m]
Reference 5.64510
-4

Finite element mesh 22 44 88
Element type
TRIA-3 5.54810
-4
5.67510
-4
5.67010
-4

QUAD-4 6.14510
-4
5.81510
-4
5.70410
-4

TRIA-6 6.46310
-4
5.88210
-4
5.72210
-4

QUAD-8 6.37010
-4
6.10710
-4
5.86910
-4


A
Z
u [m]
Reference 5.64510
-4

Finite element mesh 221 441 881
Element type
PENTA-6 2.75210
-4
4.35810
-4
5.16910
-4

HEXA-8 5.35510
-4
5.48810
-4
5.51610
-4

PENTA-15 4.96610
-4
5.23410
-4
5.35810
-4

HEXA-20 5.01010
-4
5.23310
-4
5.35510
-4


A
Z
u [m]
Reference 5.64510
-4

Finite element mesh 221 441 881
Element type
PENTAL-6 4.07610
-4
5.28610
-4
5.47410
-4

HEXAL-8 5.35510
-4
5.48810
-4
5.51610
-4

PENTAL-15 4.96310
-4
5.23010
-4
5.34910
-4

HEXAL-20 4.72310
-4
5.20910
-4
5.34410
-4



35
midas NFX Benchmark Series 2-35
Table 2.15.4 Displacement
Z
u obtained at point A using shell elements, t=0.001 m (Model B)

Table 2.15.5 Displacement
Z
u obtained at point A using solid elements, t=0.001 m (Model B)

Table 2.15.6 Displacement
Z
u obtained at point A using layered solid elements, t=0.001 m (Model B)
*
obtained using higher order layered hexahedral elements with reduced integration




A
Z
u [m]
Reference 4.76310
-4

Finite element mesh 22 44 88
Element type
TRIA-3 4.53310
-4
4.72510
-4
4.75910
-4

QUAD-4 4.56210
-4
4.71410
-4
4.76510
-4

TRIA-6 4.58710
-4
4.81710
-4
4.79210
-4

QUAD-8 4.76910
-4
4.76710
-4
4.76510
-4


A
Z
u [m]
Reference 4.76310
-4

Finite element mesh 221 441 881
Element type
PENTA-6 0.00510
-5
0.02110
-5
0.08010
-5

HEXA-8 4.66710
-4
4.73510
-4
4.75710
-4

PENTA-15 0.30710
-5
1.51710
-4
4.03510
-4

HEXA-20 1.21310
-4
2.86310
-4
4.60610
-4


A
Z
u [m]
Reference 4.76310
-4

Finite element mesh 221 441 881
Element type
PENTAL-6 0.02110
-5
0.57610
-5
0.92910
-4

HEXAL-8 4.66710
-4
4.73510
-4
4.75710
-4

PENTAL-15 0.30710
-5
1.51710
-4
4.03510
-4

HEXAL-20
0.17110
-5

1.21310
-4
*
1.66210
-4

2.86310
-4
*
4.49710
-4

4.60610
-4
*


Linear static analysis
2-36
2.16 Cylindrical shell patch test

REFERENCE NAFEMS [2-3]
KEYWORDS shell elements, layered solid elements
MODEL FILENAME Linearstatic16.nfx





Figure 2.16.1 Cylindrical shell patch test
Material data Youngs modulus
Poissons ratio
E = 210 GPa
v = 0.3
Section property Thickness t = 0.01 m

Table 2.16.1 Tangential stress
uu
o at point E obtained using shell elements for load cases 1 and 2
A
B
C
D
0.5
r = 1.0
30
o
z
A
B
C
D
E
0.3
0.5
30
o
20
o
t=0.01
15
o
Units : m

E
uu
o [MPa]
Reference 60
Load Case
Edge moment
(Case 1)
Outward pressure
(Case 2)
Element type Number of elements
TRIA-3 (42) 43.61 45.37
QUAD-4 4 52.36 68.21
TRIA-6 (42) 35.75 60.09
QUAD-8 4 51.82 56.64

Figure 2.16.1 shows the geometry of model. Two loading conditions are considered. In case 1, the shell is
subjected to uniform moment of 1 kN m/m applied on the edge DC. In case 2, the shell is under uniform
outward normal pressure of 0.6 MPa. Outer surface tangential stress at point E is determined. Comparison is
made with the reference value of 60MPa.


37
midas NFX Benchmark Series 2-37
Table 2.16.2 Tangential stress
uu
o at point E obtained using layered solid elements for load cases 1 and 2
*
obtained using higher order layered hexahedral elements with reduced integration



E
uu
o [MPa]
Reference 60
Load Case
Edge moment
(Case 1)
Outward pressure
(Case 2)
Element type Number of elements
PENTAL-6 (42) 43.12 67.13
HEXAL-8 4 52.62 76.37
PENTAL-15 (42) 12.89 61.05
HEXAL-20 4
14.06
49.35*
58.99
60.66*



Linear static analysis
2-38
2.17 Membrane with hot-spot

REFERENCE NAFEMS [2-3]
KEYWORDS shell elements
MODEL FILENAME Linearstatic17.nfx





Figure 2.17.1 Membrane with hot spot at center

Material data Youngs modulus
Poissons ratio
E = 100.0 GPa
v = 0.3
Section property Thickness t = 1.0 mm
Thermal property Thermal expansion
coefficient
= 0.001


A
B
C
X
Y
D
Hot-Spot
2
20
20
r

Units : mm
Figure 2.17.1 shows membrane problem with hot-spot in the center of the square plate. A quarter (ABCD) is
modeled utilizing symmetric boundary conditions. Inside the circular hot-spot (r < 1 mm), thermal strain oT of
0.001 is applied. Y-normal stress at point D (1mm away from center point, outside hot spot) is determined and
compared with the reference solution of 50MPa given in the standard NAFEMS benchmarks


39
midas NFX Benchmark Series 2-39
Table 2.17.1 Stress
Y
o at point D obtained using shell elements



D
Y
o [MPa]
Reference 50.00
Element type Number of elements
TRIA-3 56 54.30
QUAD-4 28 46.12
TRIA-6 56 54.40
QUAD-8 28 44.27


Linear static analysis
2-40
2.18 Solid cylinder/taper/sphere temperature loading

REFERENCE NAFEMS [2-3]
KEYWORDS solid elements, layered solid elements, axisymmetric elements
MODEL FILENAME Linearstatic18_Axisymmetric.nfx
Linearstatic18_Solid.nfx






Figure 2.18.1 Solid cylinder/taper/sphere model

Material data Youngs modulus
Poissons ratio
Thermal expansion coefficient
E = 210 GPa
v = 0.3
o = 2.310
-4
/


r
Z
A B
D
C
E
H I
G
F
1 0.4
0.345
0.345
0.400
0.700
0.2929 0.7071
45
o
1
1.4
D
F
G
I
H
E
C
A
B
Y X
Z
Units : m
Figure 2.18.1 shows the cylinder model. It is subjected to thermal load due to temperature change that is linear
in the radial and axial directions, given by
2 2
T x y z A = + +
. Symmetric conditions are applied at the x-z and
y-z planes. Top and bottom plane are constrained in the z-direction. The z-normal stress at point A is
determined using a coarse mesh and a fine mesh. The reference solution is taken from the standard NAFEMS
benchmarks.


41
midas NFX Benchmark Series 2-41
Table 2.18.1 Stress
z
o at point A obtained using solid elements, coarse mesh

Table 2.18.2 Stress
z
o at point A obtained using solid elements, fine mesh

Table 2.18.3 Stress
z
o at point A obtained using axisymmetric elements, coarse mesh







A
z
o [MPa]
Reference -105
Element type Number of elements
TETRA-4 167 -72.531
PENTA-6 40 -82.438
HEXA-8 20 -89.997
TETRA-10 167 -97.913
PENTA-15 40 -101.341
HEXA-20 20 -93.271

A
z
o

[MPa]
Reference -105
Element type Number of elements
TETRA-4 320 -94.550
PENTA-6 804 -93.966
HEXA-8 160 -98.651
TETRA-10 320 -104.568
PENTA-15 804 -104.205
HEXA-20 160 -99.665

A
z
o

[MPa]
Reference -105
Element type Number of elements
TRIAX-3 10 -54.844
QUADX-4 5 -94.503
TRIAX-6 10 -101.335
QUADX-8 5 -93.965


Linear static analysis
2-42
Table 2.18.4 Stress
z
o at point A obtained using axisymmetric elements, fine mesh




A
z
o

[MPa]
Reference -105
Element type Number of elements
TRIAX-3 40 -79.275
QUADX-4 20 -100.415
TRIAX-6 40 -104.205
QUADX-8 20 -99.948


43
midas NFX Benchmark Series 2-43
2.19 Laminated strip under three-point bending

REFERENCE NAFEMS [2-8]
KEYWORDS layered shell elements, layered solid elements, stacked layered solid elements
MODEL FILENAME Linearstatic19.nfx







Figure 2.19.1 Laminated strip model

Material data Youngs modulus
Shear modulus
Poissons ratios
E
1
=100 GPa

,
E
2
= E
3
=5 GPa
G
12
=3 GPa, G
13
= G
23
=2 GPa
v
12
= v
13
=0.4, v
23
=0.3


X
Y
0
o
fiber direction 10
X
Z
10 10 15 15
1
10 N/mm
A B
E
C,E
D
0.4
0.1
0.1
0.1
0.1
0.1
0.1
E
D
C
90
o
0
o
90
o
0
o
0
o
90
o
0
o
Units : mm
Figure 2.19.1 shows laminated strip with [0/90/0/90/0/90/0] layup under three-point bending. Layered shell
and layered solid elements are employed to obtain the linear static response of the laminate. For the model with
layered solid elements, the entire laminate is modeled using either single element in the thickness direction or
using multiple layered solid elements stacked with each element representing individual ply. The vertical
displacement and the in-plane bending stress are computed at point E and the inter-laminar shear stress is
computed at point D. Comparison is made with the reference solution given in the NAFEMS publication for
composite benchmarks.


Linear static analysis
2-44
Table 2.19.1 Bending stress
11
o at E, interlaminar shear stress
31
o

at D and Z deflection
Z
u

at E obtained using
layered shell elements
*
obtained using higher order layered quadrilateral elements with reduced integration

Table 2.19.2 Bending stress
11
o

and Z deflection
Z
u

at point E obtained using layered solid elements with single
element in the thickness direction

Table 2.19.3 Bending stress
11
o

and Z deflection
Z
u

at point E obtained using layered solid elements stacked in the
thickness direction



E
11
o

[MPa]
31
D
o

[MPa]
E
Z
u

[mm]
Reference 683.9 -4.1 -1.06
Element type
Number of
elements

TRIAL-3 40 668.8 -4.05 -1.06
QUADL-4 210 679.3 -4.07 -1.06
TRIAL-6 40 699.7 -4.02 -1.06
QUADL-8 210
683.6
682.1*
-4.08
-4.05*
-1.09
-1.06*

E
11
o

[MPa]

E
Z
u

[mm]
Reference 683.9 -1.06
Element type
Number of
elements

PENTAL-6 401 625.0 -1.05
HEXAL-8 2101 625.1 -1.05
PENTAL-15 401 682.3 -1.05
HEXAL-20 2101 680.4 -1.05

E
11
o

[MPa]

E
Z
u

[mm]
Reference 683.9 -1.06
Element type
Number of
elements

PENTAL-6 407 633.6 -1.05
HEXAL-8 2107 633.8 -1.05
PENTAL-15 407 699.9 -1.06
HEXAL-20 2107 695.9 -1.06


45
midas NFX Benchmark Series 2-45
2.20 Laminated thick cylinder

REFERENCE NAFEMS [2-8]
KEYWORDS layered shell elements, layered solid elements, stacked layered solid elements
MODEL FILENAME Linearstatic20.nfx








Figure 2.20.1 Laminated thick cylinder model, all dimensions in mm

Material data

Inner cylinder
(Isotropic)
Outer cylinder
(Orthotropic)

E
1
=210 GPa, v =0.3 , o=210
-5
/

E
1
=130 GPa

,
E
2
= E
3
=5 GPa
G
12
=10 GPa, G
13
= G
23
=5 GPa
v
12
= v
13
= v
23
=0.25


200
23
25
27
2
1
Orthotropic
material orientation
Y
X
A
B
C
D
Y
Z
Z = 0 Z = 100
Internal
pressure
P=200 MPa
Units : mm
Laminated thick cylinder shown in Figure 2.20.1 subjected to both internal pressure load and thermal load is
evaluated. The cylinder model has two different layers the inner layer is isotropic and the outer layer is
orthotropic with stiffness greater in the circumferential direction. Symmetric boundary conditions are applied
on surfaces represented by AB and CD and at Z=0. Two cases are considered herein. In case 1, the cylinder is
subjected to internal pressure of 200 MPa. In case 2, the cylinder is subjected to both internal pressure load and
temperature rise of 130. Hoop stress at Z=0(average value of all boundary nodes) are computed and
compared with the reference values provided in the NAFEMS publication for composite benchmarks.


Linear static analysis
2-46
Table 2.20.1 Hoop stress
uu
o

in inner cylinder at r=23, r=25 and outer cylinder at r=25, r=27 subjected to internal
pressure (case 1) obtained using layered shell elements

Table 2.20.2 Hoop stress
uu
o

in inner cylinder at r=23, r=25 and outer cylinder at r=25, r=27 subjected to internal
pressure (case 1) obtained using layered solid elements with single element in the thickness direction

Table 2.20.3 Hoop stress
uu
o

in inner cylinder at r=23, r=25 and outer cylinder at r=25, r=27 subjected to internal
pressure (case 1) obtained using layered solid elements stacked in the thickness direction




i
uu
o

at r=23
[MPa]
i
uu
o

at r=25
[MPa]
o
uu
o

at r=25
[MPa]
o
uu
o

at r=27
[MPa]
Reference 1565.3 1429.7 874.7 759.1
Element
type
Number of
elements

TRIAL-3 128 1427.0 1417.6 880.4 873.2
QUADL-4 88 1411.3 1413.5 873.7 873.7

i
uu
o

at r=23
[MPa]
i
uu
o

at r=25
[MPa]
o
uu
o

at r=25
[MPa]
o
uu
o

at r=27
[MPa]
Reference 1565.3 1429.7 874.7 759.1
Element
type
Number of
elements

PENTAL-6 1281 1634.4 1359.0 841.0 691.6
HEXAL-8 881 1645.4 1362.8 838.0 678.4
PENTAL-15 321 1590.9 1349.5 880.7 786.6
HEXAL-20 441 1590.0 1349.9 881.0 787.5

i
uu
o

at r=23
[MPa]
i
uu
o

at r=25
[MPa]
o
uu
o

at r=25
[MPa]
o
uu
o

at r=27
[MPa]
Reference 1565.3 1429.7 874.7 759.1
Element
type
Number of
elements

PENTAL-6
1282
1284
1584.3
1580.2
1392.0
1415.8
878.9
879.7
737.9
749.6
HEXAL-8
882
884
1581.6
1577.3
1390.1
1413.3
877.6
878.2
743.9
753.7
PENTAL-15
322
324
1568.6
1567.1
1431.1
1429.4
874.6
874.5
759.1
758.4
HEXAL-20
442
444
1567.4
1565.9
1431.9
1430.2
875.1
874.9
759.7
759.0


47
midas NFX Benchmark Series 2-47
Table 2.20.4 Hoop stress
uu
o

in inner cylinder at r=23, r=25 and outer cylinder at r=25, r=27 subjected to internal
pressure and uniform temperature change (case 2) obtained using layered shell elements

Table 2.20.5 Hoop stress
uu
o

in inner cylinder at r=23, r=25 and outer cylinder at r=25, r=27 subjected to internal
pressure and uniform temperature change (case 2) obtained using layered solid elements with single element
in the thickness direction

Table 2.20.6 Hoop stress
uu
o

in inner cylinder at r=23, r=25 and outer cylinder at r=25, r=27 subjected to internal
pressure and uniform temperature change (case 2) obtained using layered solid elements stacked in the
thickness direction




i
uu
o

at r=23
[MPa]
i
uu
o

at r=25
[MPa]
o
uu
o

at r=25
[MPa]
o
uu
o

at r=27
[MPa]
Reference 1381.0 1259.6 1056.0 936.1
Element
type
Number of
elements

TRIAL-3 128 1253.8 1239.5 1058.4 1048.1
QUADL-4 88 1238.2 1236.5 1053.8 1053.7

i
uu
o

at r=23
[MPa]
i
uu
o

at r=25
[MPa]
o
uu
o

at r=25
[MPa]
o
uu
o

at r=27
[MPa]
Reference 1381.0 1259.6 1056.0 936.1
Element
type
Number of
elements

PENTAL-6 1281 1469.2 1191.1 1023.0 870.4
HEXAL-8 881 1476.9 1193.9 1020.9 861.0
PENTAL-15 321 1417.5 1149.6 1063.5 974.3
HEXAL-20 441 1416.5 1149.9 1063.7 975.2

i
uu
o

at r=23
[MPa]
i
uu
o

at r=25
[MPa]
o
uu
o

at r=25
[MPa]
o
uu
o

at r=27
[MPa]
Reference 1381.0 1259.6 1056.0 936.1
Element
type
Number of
elements

PENTAL-6
1282
1284
1396.9
1395.2
1224.6
1247.7
1062.1
1062.5
909.6
923.5
HEXAL-8
882
884
1393.6
1391.6
1222.0
1244.5
1060.4
1060.5
917.5
928.8
PENTAL-15
322
324
1384.5
1383.1
1261.1
1259.6
1056.1
1055.9
935.4
934.7
HEXA-20
442
444
1383.2
1381.9
1261.8
1260.4
1056.5
1056.3
936.1
935.4


Linear static analysis
2-48
2.21 Three-layer sandwich shell under normal pressure

REFERENCE NAFEMS [2-8]
KEYWORDS layered shell elements
MODEL FILENAME Linearstatic21.nfx







Figure 2.21.1 Three-layer sandwich model

Material data

Face sheets


Core

E
X
=10.010
6
psi, E
Y
=4.010
6
psi,
G
XY
=1.87510
6
psi, v
XY
=0.3

E
X
= E
Y
=10 psi,
G
XY
=10 psi, G
XZ
=3.010
4
psi, G
YZ
=1.210
4
psi
v
XY
=0.0


10
10
0.750
0.028
0.028
simply-supported on all edges
Core
X
Y
A B
Uniform normal
pressure P=100 psi
Face sheets C D
E
Units : in
A three-layer sandwich shell with soft middle core as shown in Figure 2.21.1, is evaluated with layered shell
elements. The sandwich panel is simply-supported on all four edges and is under uniform normal pressure of
100 psi. Utilizing symmetry, a quarter of the sandwich shell of which the corner points are ABCD, is modeled
using layered shell elements. Vertical displacement, in-plane normal stresses are extracted at the center of the
full model (point C), and in-plane shear stress is extracted at point E, which is the center of the quarter model.
Both points are located at the bottom of the sandwich shell. The results are compared with the reference values
provided in the NAFEMS publication for composite benchmarks.


49
midas NFX Benchmark Series 2-49
Table 2.21.1 Displacement
Z
u , inplane normal stresses
22
,
11
o o at point C and inplane shear stress
12
o at point E
obtained using layered shell elements



Z
u [in.]
11
C
o [psi]
22
C
o [psi]
12
E
o [psi]
Reference -0.123 34449 13350 -5067.5
Element
type
Number of
elements

TRIAL-3 128 -0.122 35492 13864 -5254.7
QUADL-4 88 -0.122 35721 13956 -5293.1
TRIAL-6 32 -0.123 37036 14437 -5268.6
QUADL-8 44 -0.128 36091 14278 -4824.1


Linear static analysis
2-50
2.22 Axisymmetric shell under pressure

REFERENCE NAFEMS [2-3]
KEYWORDS axisymmetric elements
MODEL FILENAME Linearstatic22.nfx




Figure 2.22.1 Axisymmetric shell model

Material data Youngs modulus
Poissons ratio
E = 210 GPa
v = 0.3

Table 2.22.1 Hoop stress
uu
o at point D obtained using axisymmetric elements
Units : m
r
Y
0.5
0.25
E
A
D
0.25
0.0625
36
o
t=0.01

D
uu
o

[MPa]
Reference 94.55
Element type Number of elements
TRIAX-3
(4x21)x1
(4x42)x2
55.61
71.93
QUADX-4
21x1
42x2
66.59
78.41
TRIAX-6
(4x21)x1
(4x42)x2
88.78
90.05
QUADX-8
21x1
42x2
88.84
90.08
Figure 2.22.1 shows the axisymmetric shell structure subjected to uniform normal pressure. Points A and E are
constrained in vertical and radial directions respectively. Hoop stress on outer surface at point D is determined
using a coarse mesh and a fine mesh. The reference solution is taken from the standard NAFEMS benchmarks.


51
midas NFX Benchmark Series 2-51
References

[2-1] R.H. MacNeal and R.L. Harder, A Proposed Standard Set of Problems to Test Finite Element Accuracy,
Finite Element Analysis and Design, Vol. 1, pp. 3-20, 1985
[2-2] J.C. Simo, D.D. Fox and M.S. Rifai, On a Stress Resultant Geometrically Exact Shell Model. Part II:
The Linear Theory, Computer Methods in Applied Mechanics and Engineering, Vol. 73, pp. 53-92, 1989
[2-3] NAFEMS, The Standard NAFEMS Benchmarks, Rev. 3, NAFEMS, Glasgow, 1990
[2-4] J.C. McCormac, Structural Analysis, International Textbook Company, Scranton, PA, 1965
[2-5] W.C. Young and R.G. Budynas, Roarks Formulas for Stress and Strain, 7
th
Edition, McGraw-Hill, New
York, 2002
[2-6] J. Batoz and I. Katili, On a Simple Triangular Reissner/Mindlin Plate Element Based on Incompatible
Modes and Discrete Constraints, International Journal for Numerical Methods in Engineering, Vol. 35,
pp. 1603-1632, 1992
[2-7] O. Zienkiewicz and Z. XU, Linked Interpolation for Reisner-Midlin Plate Elements: Part I-A simple
quadrilateral, International Journal for Numerical Methods in Engineering, Vol. 36, pp. 3043-3056,
1993
[2-8] NAFEMS, Composite Benchmarks, Ref . R0031, Issue 2, NAFEMS, Glasgow, 2001



1
midas NFX Benchmark Series 3-1
3.1 Pin-ended cross

REFERENCE NAFEMS [3-1]
KEYWORDS bar elements
MODEL FILENAME FreeVib01.nfx





Figure 3.1.1 Pin-ended cross model
Y
X
5 5
5
5
0.125
0.125
Units : m
Figure 3.1.1 shows the pin-ended cross model. The model is modeled with 4 bar elements per arm. Free-
vibration analysis is carried out. The natural frequencies of the lowest modes, of which the vibration mode
shapes are depicted in Figure 3.1.2 are compared with the reference values given in the NAFEMS benchmarks.


Free vibration analysis
3-2

Figure 3.1.2 Vibration mode shapes


Material data
Youngs modulus
Density
E = 200 GPa
= 8000 kg/m
3

Section property Square cross-section 0.125 m 0.125m

Table 3.1.1 Natural frequencies in Hz obtained using bar elements



Mode 1 Mode 2&3 Mode 4
Mode 5 Mode 6&7 Mode 8
Mode Number 1 2,3 4 5 6,7 8
Reference 11.336 17.709 17.709 45.345 57.390 57.390
Element
type
Number of
elements

BAR-2 4 per arm 11.338 17.689 17.717 45.483 57.371 57.690


3
midas NFX Benchmark Series 3-3
3.2 Pin-ended double cross

REFERENCE NAFEMS [3-2]
KEYWORDS bar elements
MODEL FILENAME FreeVib02.nfx







Figure 3.2.1 Pin-ended double cross model
Y
X
5 5
5
5
0.125
0.125
Units : m
All arms equal length
Figure 3.2.1 shows the pin-ended double cross model. The model is modeled with 4 bar elements per arm.
Free-vibration analysis is carried out. The natural frequencies of the lowest modes, of which the vibration
mode shapes are depicted in Figure 3.2.2 are compared with the reference values given in the NAFEMS
benchmarks.



Free vibration analysis
3-4

Figure 3.2.2 Vibration mode shapes


Material data
Youngs modulus
Density
E = 200 GPa
= 8000 kg/m
3

Section property Square cross-section 0.125 m 0.125m

Table 3.2.1 Natural frequencies in Hz obtained using bar elements




Mode 1 Mode 2&3 Mode 4,5,6,7&8
Mode 9 Mode 10&11 Mode 12,13,14,15&16
Mode Number 1 2,3 4,5,6,7,8 9 10,11
12,13,14,
15,16
Reference 11.336 17.709 17.709 45.345 57.390 57.390
Element
type
Number of
elements

BAR-2 4 per arm 11.330 17.658 17.687 45.386 57.152 57.469


5
midas NFX Benchmark Series 3-5
3.3 Free square frame

REFERENCE NAFEMS [3-1]
KEYWORDS bar elements
MODEL FILENAME FreeVib03.nfx





Figure 3.3.1 Free square frame model

Y
X
10
10
0.25
0.25
Units : m
Figure 3.3.1 shows a square frame model with no boundary conditions. The square frame is modeled with 4
bar elements per arm. Frequencies of the lowest vibration modes excluding rigid-body modes are compared
with the reference solutions given in the NAFEMS benchmarks. The vibration mode shapes are depicted in
Figure 3.3.2.


Free vibration analysis
3-6

Figure 3.3.2 Vibration mode shapes

Material data
Youngs modulus
Density
E = 200 GPa
= 8000 kg/m
3

Section property Square cross-section 0.25 m 0.25m

Table 3.3.1 Natural frequencies in Hz obtained using bar elements




Mode Number 4 5 6,7 8 9 10,11
Reference 3.261 5.668 11.136 12.849 24.570 28.695
Element
type
Number of
elements

BAR-2 4 per arm 3.262 5.666 11.135 12.802 24.629 28.721


7
midas NFX Benchmark Series 3-7
3.4 Cantilever with off-center point masses

REFERENCE NAFEMS [3-2]
KEYWORDS bar elements
MODEL FILENAME FreeVib04.nfx





Figure 3.4.1 Cantilever with off-center point masses


Figure 3.4.2 Vibration mode shapes
Y
X
10
Units : m
2
2
0.5
M
1
M
2
M
1
= 10000 kg (along X, Y, Z)
M
2
= 1000 kg (along X, Y, Z)
Mode 4 Mode 5 Mode 6
Mode 4 Mode 5 Mode 6
Y
X
X
Z
Y
X
Z
Y
Y
X
Y
X
X
Z
Y
Figure 3.4.1 shows a cantilever model with two off-center point masses. This problem is characterized by the
torsional-flexural coupling. The entire cantilever is discretized by 5 bar elements. The natural frequencies of
the lowest modes, of which the vibration mode shapes are depicted in Figure 3.4.2 are compared with the
reference values given in the NAFEMS benchmarks.




Free vibration analysis
3-8
Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Circular cross-section R = 0.25m

Table 3.4.1 Natural frequencies in Hz obtained using bar elements



Mode Number 1 2 3 4 5 6
Reference 1.723 1.727 7.413 9.972 18.155 26.957
Element
type
Number of
elements

BAR-2 5 per arm 1.722 1.726 7.410 9.948 18.053 26.716


9
midas NFX Benchmark Series 3-9
3.5 Deep simply-supported beam

REFERENCE NAFEMS [3-2]
KEYWORDS bar elements, solid elements
MODEL FILENAME FreeVib05.nfx





Figure 3.5.1 Deep simply-supported beam


Figure 3.5.2 Vibration mode shapes

10
Units : m
2
2
X
Z
Y
x = y = z = R
X
= 0 at A, y = z = 0 at B
A B
Mode 1&2 Mode 3 Mode 4
Mode 5&6 Mode 7 Mode 8&9
Y
X
Y
X
Y
X
FLEXURAL TORSIONAL EXTENSIONAL
Y
X
Y
X
Y
X
FLEXURAL TORSIONAL FLEXURAL
Figure 3.5.1 shows the deep simply-supported beam problem. These tests are adopted from the NAFEMS
benchmark problems FV5 (bar) and 51 (solid). The reference frequencies are taken from the respective
problems in NAFEMS.


Free vibration analysis
3-10
Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Square cross-section 2.0 m 2.0m

Table 3.5.1 Natural frequencies in Hz obtained using bar elements

Table 3.5.2 Natural frequencies in Hz obtained using solid elements




Mode Number 1,2 3 4 5,6 7 8,9
Reference 42.649 77.542 125.00 148.31 233.10 284.55
Element
type
Number of
elements

BAR-2 5 42.675 77.841 125.52 150.43 241.24 300.11
Mode Number 1 2 3 4 5
Reference 38.200 85.210 152.23 245.53 297.05
Element
type
Number of
elements

TETRA-4 282 46.381 95.982 179.77 292.81 325.68
PYRAM-5 180 40.572 88.060 163.38 272.39 305.46
PENTA-6 60 39.317 85.659 158.54 264.38 298.68
HEXA-8 10x1x3 38.277 83.952 157.57 264.92 298.33
TETRA-10 29 38.461 88.167 159.58 264.95 306.90
PYRAM-13 180 38.217 86.979 155.52 257.18 300.85
PENTA-15 10 37.950 86.354 154.08 248.98 300.77
HEXA-20 5x1 38.166 86.816 155.95 256.33 302.76


11
midas NFX Benchmark Series 3-11
3.6 Circular ring

REFERENCE NAFEMS [3-1]
KEYWORDS bar elements
MODEL FILENAME FreeVib06.nfx





Figure 3.6.1 circular ring model

Units : m
Z
X
Y
1
0.1
An unconstrained circular ring model is shown in Figure 3.6.1. Bar elements are used to obtain the lowest
vibration modes excluding rigid body modes. The frequencies obtained using free vibration analysis are
compared with those from NAFEMS benchmarks.


Free vibration analysis
3-12

Figure 3.6.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Circular cross-section R = 0.05 m

Table 3.6.1 Natural frequencies in Hz obtained using bar elements






Mode 7&8 Mode 9&10 Mode 11&12
Mode 13&14 Mode 15 Mode 16
OUT OF PLANE IN PLANE OUT OF PLANE
IN PLANE OUT OF PLANE IN PLANE
Mode Number 7,8 9,10 11,12 13,14 15 16
Reference 51.849 53.382 148.77 150.99 286.98 289.51
Element
type
Number of
elements

BAR-2 20 52.212 53.776 148.92 151.26 285.34 285.34


13
midas NFX Benchmark Series 3-13
3.7 Thin square cantilever plate

REFERENCE NAFEMS [3-1]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib07A.nfx
FreeVib07B.nfx





Figure 3.7.1 Thin square cantilever plate model

Y
X
10
Units : m
t = 0.05
10 C
L
Z
x = y = R
z
= 0 at all nodes,
z = R
y
= R
x
= 0 along y-axis
x = y = R
z
= 0 at all nodes,
z = R
y
= R
x
= 0 along y-axis, R
x
= 0 along y = 5m
case A
case B
Figure 3.7.1 shows a thin square cantilever plate model. Half of the model is discretized using shell, solid and
layered solid elements. Depending on the applied boundary conditions, free vibration analyses of two sub-
cases are conducted; Case A considers only the symmetric modes and Case B considers only the anti-
symmetric vibration modes.


Free vibration analysis
3-14

Figure 3.7.2 Vibration mode shapes (case A)


Figure 3.7.3 Vibration mode shapes (case B)

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Thickness t = 0.05 m










Mode 1 Mode 2 Mode 3
Mode 4 Mode 5 Mode 6
C
L
C
L
C
L
C
L
C
L
C
L
Mode 1 Mode 2 Mode 3
Mode 4 Mode 5 Mode 6
C
L
C
L
C
L
C
L
C
L
C
L


15
midas NFX Benchmark Series 3-15
Table 3.7.1 Natural frequencies in Hz obtained using shell elements; symmetric vibration modes (case A)

Table 3.7.2 Natural frequencies in Hz obtained using solid elements; symmetric vibration modes (case A)

Table 3.7.3 Natural frequencies in Hz obtained using layered solid elements; symmetric vibration modes (case A)
*
obtained using high-order layered hexahedral element with reduced integration









Mode Number 1 2 3 4 5 6
Reference 0.421 2.582 3.306 6.555 7.381 11.402
Element
type
Number of
elements

TRIA-3 64 0.419 2.639 3.368 6.951 7.974 13.032
QUAD-4 8x4 0.418 2.615 3.337 6.752 7.905 12.681
TRIA-6 16 0.418 2.637 3.290 6.748 7.809 12.636
QUAD-8 4x2 0.419 2.569 3.281 6.566 7.497 12.129
Mode Number 1 2 3 4 5 6
Reference 0.421 2.582 3.306 6.555 7.381 11.402
Element
type
Number of
elements

HEXA-8 8x4x1 0.419 2.656 3.353 6.747 8.251 12.654
PENTA-15 16 0.435 2.909 4.043 9.401 10.253 21.691
HEXA-20 4x2x1 0.431 2.664 3.426 7.617 7.885 14.343
Mode Number 1 2 3 4 5 6
Reference 0.421 2.582 3.306 6.555 7.381 11.402
Element
type
Number of
elements

PENTAL-6 64 0.422 2.723 4.458 8.295 10.887 18.295
HEXAL-8 8x4x1 0.419 2.656 3.353 6.747 8.251 12.654
PENTAL-15 16 0.435 2.909 4.043 9.401 10.253 21.691
HEXAL-20* 4x2x1 0.431 2.664 3.426 7.616 7.885 14.339


Free vibration analysis
3-16
Table 3.7.4 Natural frequencies in Hz obtained using shell elements; anti-symmetric vibration modes (case B)

Table 3.7.5 Natural frequencies in Hz obtained using solid elements; anti-symmetric vibration modes (case B)

Table 3.7.6 Natural frequencies in Hz obtained using layered solid elements; anti-symmetric vibration modes (case B)
*
obtained using high-order layered hexahedral element with reduced integration



Mode Number 1 2 3 4 5 6
Reference 1.029 3.753 7.730 8.561 N/A N/A
Element
type
Number of
elements

TRIA-3 64 1.035 3.864 8.288 9.414 12.337 18.293
QUAD-4 8x4 1.026 3.807 8.220 9.194 11.974 18.004
TRIA-6 16 1.042 3.804 7.901 8.930 11.335 16.581
QUAD-8 4x2 1.024 3.730 7.625 8.606 11.167 16.807
Mode Number 1 2 3 4 5 6
Reference 1.029 3.753 7.730 8.561 N/A N/A
Element
type
Number of
elements

HEXA-8 8x4x1 1.029 3.840 8.356 9.416 11.977 18.248
PENTA-15 16 1.380 5.759 11.481 15.275 25.009 28.983
HEXA-20 4x2x1 1.043 3.830 8.237 9.102 14.733 17.537
Mode Number 1 2 3 4 5 6
Reference 1.029 3.753 7.730 8.561 N/A N/A
Element
type
Number of
elements

PENTAL-6 64 1.330 5.637 11.855 13.450 25.963 29.350
HEXAL-8 8x4x1 1.029 3.840 8.356 9.416 11.977 18.248
PENTAL-15 16 1.380 5.759 11.481 15.275 25.009 28.983
HEXAL-20* 4x2x1 1.043 3.830 8.237 9.102 14.732 17.537


17
midas NFX Benchmark Series 3-17
3.8 Thin square cantilever platemesh distortion

REFERENCE NAFEMS [3-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib08.nfx





Figure 3.8.1 Free thin square plate model

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Thickness t = 0.05 m

Table 3.8.1 Natural frequencies in Hz obtained using regular and distorted shell element meshes
*
obtained using shell element formulations with 6-dof per node.


Y
X
10
10
Test 1 Test 2 Test 3 Test 4
Units : m
Cantilevered
Mode Number 1 2 3 4 5 6
Reference 0.421 1.029 2.582 3.306 3.753 6.555
Element
type
Mesh
QUAD-4
Regular
0.418
0.417*
1.044
1.009*
2.756
2.666*
3.505
3.465*
4.199
3.829*
7.483
6.870*
Distorted
0.418
0.417*
1.043
1.007*
2.789
2.679*
3.539
3.485*
4.192
3.857*
7.515
6.965*
QUAD-8
Regular 0.421 1.015 2.723 3.442 3.895 5.667
Distorted 0.422 1.022 2.701 3.413 3.897 5.278
Figure 3.8.1 shows the cantilever plate problem. To assess the mesh distortion sensitivity of the modal
responses, the cantilever plate problem is solved using several distorted shell and layered solid meshes.


Free vibration analysis
3-18
Table 3.8.2 Natural frequencies in Hz obtained using regular and distorted solid element meshes

Table 3.8.3 Natural frequencies in Hz obtained using regular and distorted layered solid element meshes
*
obtained using high-order layered hexahedral element with reduced integration


Mode Number 1 2 3 4 5 6
Reference 0.421 1.029 2.582 3.306 3.753 6.555
Element
type
Mesh
HEXA-8 Regular 0.422 1.042 2.935 3.583 4.184 7.369
HEXA-20
Regular 0.443 1.064 2.827 3.604 4.108 7.664
Distorted 0.443 1.093 2.888 3.883 4.323 7.149
Mode Number 1 2 3 4 5 6
Reference 0.421 1.029 2.582 3.306 3.753 6.555
Element
type
Mesh
HEXAL-8
Regular 0.422 1.042 2.935 3.583 4.184 7.369
Distorted 0.421 1.042 2.928 3.614 4.172 7.370
HEXAL-20*
Regular 0.443 1.064 2.827 3.603 4.108 7.658
Distorted 0.443 1.089 2.852 4.015 4.333 7.244


19
midas NFX Benchmark Series 3-19
3.9 Simply-supported thick square plate

REFERENCE NAFEMS [3-1]
KEYWORDS shell elements
MODEL FILENAME FreeVib09.nfx







Figure 3.9.1 Simply-supported thick square plate model

Y
X
10
Units : m
t = 1
10
Z
x = y = R
z
= 0 at all nodes,
z = R
y
= R
x
= 0 along y-axis
x = y = R
z
= 0 at all nodes,
z = 0 along all 4 edges
case A
case B
A simply-supported thick square plate problem is depicted in Figure 3.9.1. The problem is discretized using
shell elements. Two types of simply-supported boundary conditions (Case A and Case B) are employed. For
both cases, all nodes are constrained in the translational degrees of freedom in X and Y directions and Z
rotational degrees of freedom. Also, Z-directional translational degrees of freedom along 4 edges are
constrained for both cases. In addition, X-directional rotation is constrained along edges parallel to X-axis and
Y-directional rotation is constrained along edges parallel to Y-axis for case A only.


Free vibration analysis
3-20

Figure 3.9.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Thickness t = 1 m



















Mode 1 Mode 2&3 Mode 4
Mode 5&6 Mode 7&8 Mode 9&10


21
midas NFX Benchmark Series 3-21
Table 3.9.1 Natural frequencies in Hz
*
obtained using high-order quadrilateral shell element with reduced integration


Mode Number 1 2,3 4 5,6 7,8 9,10
Reference 45.897 109.44 167.89 204.51 256.50 336.62
Element
type
Number of
elements
BC
type

TRIA-3 128
A 47.386 117.76 188.58 233.57 305.04 394.14
B 46.177 115.96 184.37
231.46,
231.85
299.64 391.45
QUAD-4 8x8
A 46.493 114.46 176.33 226.93 280.64
373.36
383.73
B 45.128 112.60 172.20
225.22,
225.45
276.10
367.21
382.57
TRIA-6 32
A 45.029 106.46 159.49
198.39,
202.95
258.52 324.69
B 43.847 105.30 157.59
197.82,
198.88
252.74 321.18
QUAD-8 4x4
A
45.021
45.936*
105.38
110.41*
162.37
170.38*
189.62
212.81*
248.99
269.96*
250.89
344.77*
B
44.307
44.134*
104.46
107.85*
160.58
164.19*
188.68,
188.79
210.07*,
210.58*
247.09
260.32*
248.22
342.80*


Free vibration analysis
3-22
3.10 Simply-supported thin square plate

REFERENCE NAFEMS [3-1]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib10.nfx





Figure 3.10.1 Simply-supported thin square plate model

Y
X
10
Units : m
t = 0.05
10
Z
x = y = R
z
= 0 at all nodes,
R
x
= 0 along edges x = 0 & x = 10 m,
R
y
= 0 along edges y = 0 & y = 10 m
Figure 3.10.1 shows a thin plate model. The simply-supported thin square plate is evaluated using shell, solid
and layered solid elements. Modal analysis is carried out to obtain the lowest natural frequencies.


23
midas NFX Benchmark Series 3-23

Figure 3.10.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
= 0.3
= 8000 kg/m
3

Section property Thickness t = 0.05 m

Table 3.10.1 Natural frequencies in Hz obtained using shell elements













Mode 1 Mode 2&3 Mode 4
Mode 5&6 Mode 7&8
Mode Number 1 2,3 4 5,6 7,8
Reference 2.377 5.942 9.507 11.884 15.449
Element
type
Number of
elements

TRIA-3 128 2.461 6.459 10.901 13.986 19.200
QUAD-4 8x8 2.414 6.295 10.171 13.750 17.671
TRIA-6 32 2.365 5.960 9.247 11.675, 12.671 16.145
QUAD-8 4x4 2.384 6.055 9.980 12.753 14.132


Free vibration analysis
3-24
Table 3.10.2 Natural frequencies in Hz obtained using solid elements
*
obtained using lumped mass matrix

Table 3.10.3 Natural frequencies in Hz obtained using layered solid elements
*
obtained using lumped mass matrix
**
obtained using high-order layered hexahedral element with reduced integration


Mode Number 1 2,3 4 5,6 7,8
Reference 2.377 5.942 9.507 11.884 15.449
Element
type
Number of
elements

HEXA-8 8x8 2.414 6.318 10.131 13.956 17.597
PENTA-15 32 2.593 7.550 15.990 22.664, 36.058 55.554
HEXA-20 4x4
2.454
2.450*
7.108
6.909*
16.247
13.503*
16.247, 16.611
13.788*
25.967
17.565*
Mode Number 1 2,3 4 5,6 7,8
Reference 2.377 5.942 9.507 11.884 15.449
Element
type
Number of
elements

PENTAL-6* 128 2.440 6.294 10.665 13.235 18.736
HEXAL-8 8x8
2.414
2.353*
6.318
5.924*
10.131
9.140*
13.956
12.275*
17.597
14.893*
PENTAL-15 32 2.593 7.550 15.990 22.664, 36.058 55.554
HEXAL-20** 4x4 2.454 7.108 16.247 16.247, 16.611 25.967


25
midas NFX Benchmark Series 3-25
3.11 Free thin square plate

REFERENCE NAFEMS [3-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib11.nfx





Figure 3.11.1 Free thin square model

Y
X
10
Units : m
t = 0.05
10
Z
x = y = R
z
= 0 at all nodes
Figure 3.11.1 shows a thin plate model. The free vibration analysis is carried out of the unconstrained thin
square plate using shell, solid and layered solid elements. The lowest natural frequencies excluding rigid body
motions are obtained and compared with the those from the NAFEMS benchmarks.



Free vibration analysis
3-26

Figure 3.11.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
= 0.3
= 8000 kg/m
3

Section property Thickness t = 0.05 m

Table 3.11.1 Natural frequencies in Hz obtained using shell elements












Mode 4 Mode 5 Mode 6
Mode 7&8 Mode 9 Mode 10
Mode Number 4 5 6 7,8 9,10
Reference 1.622 2.360 2.922 4.233 7.416
Element
type
Number of
elements

TRIA-3 128 1.621 2.392 2.984 4.294 7.851
QUAD-4 8x8 1.630 2.391 2.984 4.271 7.839
TRIA-6 32 1.623 2.353 2.899 4.211 7.318
QUAD-8 4x4 1.620 2.363 2.929 4.189 7.247


27
midas NFX Benchmark Series 3-27
Table 3.11.2 Natural frequencies in Hz obtained using solid elements

Table 3.11.3 Natural frequencies in Hz obtained using layered solid elements
**
obtained using high-order layered hexahedral element with reduced integration

Mode Number 4 5 6 7,8 9,10
Reference 1.622 2.360 2.922 4.233 7.416
Element
type
Number of
elements

HEXA-8 8x8 1.633 2.403 3.007 4.279 7.942, 8.031
PENTA-15 32 1.643 2.427 3.008 4.734 7.841
HEXA-20 4x4 1.631 2.398 2.951 4.241 7.658
Mode Number 4 5 6 7,8 9,10
Reference 1.622 2.360 2.922 4.233 7.416
Element
type
Number of
elements

PENTAL-6 128 1.643 2.405 3.010 4.436 8.048
HEXAL-8 8x8 1.633 2.403 3.007 4.279 7.942, 8.031
PENTAL-15 32 1.643 2.427 3.008 4.734 7.841
HEXAL-20** 4x4 1.631 2.398 2.951 4.241 7.658


Free vibration analysis
3-28
3.12 Simply-supported thin annular plate

REFERENCE NAFEMS [3-1]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib12.nfx





Figure 3.12.1 Simply-supported annular plate model

Units : m
Y
1.8
Z
X
6
t = 0.06
x = y = R
z
= 0 at all nodes,
z = R
x
= 0 around circumference
A simply-supported annular plate model is shown in Figure 3.12.1. The thin annular plate is simply supported
on the outer boundary. Free vibration analysis is carried out to obtain the lowest vibration frequencies using
shell, solid and layered solid elements.


29
midas NFX Benchmark Series 3-29

Figure 3.12.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
= 0.3
= 8000 kg/m
3

Section property Thickness t = 0.06 m

Table 3.12.1 Natural frequencies in Hz obtained using shell elements










Mode 1 Mode 2&3 Mode 4&5
Mode 6 Mode 7&8 Mode 9&10
Mode Number 1 2,3 4,5 6 7,8 9
Reference 1.870 5.137 9.673 14.850 15.573 18.382
Element
type
Number of
elements

TRIA-3 320 1.897 5.267 10.139 15.642 16.690 19.494
QUAD-4 160 1.876 5.264 10.035 15.416 16.351 19.251
TRIA-6 96 1.875 5.181 9.624 14.870 15.358 18.541
QUAD-8 48 1.873 5.151 9.726 14.918 15.786 18.532


Free vibration analysis
3-30
Table 3.12.2 Natural frequencies in Hz obtained using solid elements

Table 3.12.3 Natural frequencies in Hz obtained using layered solid elements
*
obtained using high-order layered hexahedral element with reduced integration

Mode Number 1 2,3 4,5 6 7,8 9,10
Reference 1.870 5.137 9.673 14.850 15.573 18.382
Element
type
Number of
elements

HEXA-8 160 1.894 5.294 10.213 15.830 16.895
19.640
19.641
PENTA-15 96 2.016 5.615 11.445 17.188 20.386 23.439
HEXA-20 48 1.876 5.125 9.781 14.950 16.338 18.390
Mode Number 1 2,3 4,5 6 7,8 9,10
Reference 1.870 5.137 9.673 14.850 15.573 18.382
Element
type
Number of
elements

PENTAL-6 320 2.040 5.748 11.285 17.772 19.754 25.289
HEXAL-8 160 1.876 5.374 10.370 15.603 17.036 19.561
PENTAL-15 96 2.016 5.615 11.445 17.188 20.386 23.439
HEXAL-20 48
1.889
1.877*
5.254
5.121*
10.288
9.795*
15.716
14.941*
18.368
16.319*
19.498
18.341*


31
midas NFX Benchmark Series 3-31
3.13 Simply-supported thick annular plate

REFERENCE NAFEMS [3-1]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib13.nfx





Figure 3.13.1 Simply-supported annular plate model

Units : m
Y
1.8
Z
X
6
t = 0.6
x = y = R
z
= 0 at all nodes,
z = R
x
= 0 around circumference
A simply-supported annular plate model is shown in Figure 3.13.1. The thick annular plate is simply supported
on the outer boundary. Free vibration analysis is carried out to obtain the lowest vibration frequencies using
shell, solid and layered solid elements.



Free vibration analysis
3-32

Figure 3.13.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
= 0.3
= 8000 kg/m
3

Section property Thickness t = 0.6 m

Table 3.13.1 Natural frequencies in Hz obtained using shell elements










Mode 1 Mode 2&3 Mode 4&5
Mode 6 Mode 7&8 Mode 9
Mode Number 1 2,3 4,5 6 7,8 9,10
Reference 18.58 48.92 92.59 140.15 N/A 166.36
Element
type
Number of
elements

TRIA-3 320 18.83 50.78 96.89 147.36 155.69 178.71
QUAD-4 160 18.62 50.35 95.57 145.42 151.95 174.44
TRIA-6 96 18.53 49.01 91.27 138.05 142.56 165.72
QUAD-8 48 18.52 49.19 90.67 135.95 141.00 163.40


33
midas NFX Benchmark Series 3-33
Table 3.13.2 Natural frequencies in Hz obtained using solid elements

Table 3.13.3 Natural frequencies in Hz obtained using layered solid elements

Mode Number 1 2,3 4,5 6 7,8 9,10
Reference 18.58 48.92 92.59 140.15 N/A 166.36
Element
type
Number of
elements

PENTA-6 320 20.85 63.75 116.40 169.09 184.80 219.28
HEXA-8 160 18.81 50.72 96.77 149.70 154.92 179.21
PENTA-15 96 18.78 50.09 94.99 143.53 152.52 174.53
HEXA-20 48 18.64 49.18 93.34 141.85 147.71 168.74
Mode Number 1 2,3 4,5 6 7,8 9,10
Reference 18.58 48.92 92.59 140.15 N/A 166.36
Element
type
Number of
elements

PENTAL-6 320 18.88 51.45 98.47 149.95 159.68 183.79
HEXAL-8 160 18.64 50.99 96.92 147.77 154.78 178.42
PENTAL-15 96 18.78 50.09 95.00 143.55 152.56 174.56
HEXAL-20 48 18.68 49.77 94.01 143.52 149.34 171.92


Free vibration analysis
3-34
3.14 Simply-supported thick annular plate (axisymmetric)

REFERENCE NAFEMS [3-1]
KEYWORDS axisymmetric elements, solid elements
MODEL FILENAME FreeVib14_Axisymmetric.nfx
FreeVib14_Solid.nfx






Figure 3.14.1 Simply-supported axisymmetric annular plate model


Figure 3.14.2 Vibration mode shapes


1.8
Units : m
4.2
R
Z
C
L
0.6
Mode 1 Mode 2 Mode 3
Mode 4 Mode 5
To obtain the axisymmetric vibration modes of the simply-supported thick annular plate problem. The annular
plate model is modeled as axisymmetric problem using axisymmetric elements and solid elements with
boundary conditions to enforce axisymmetry. The lowest vibration frequencies obtained are compared with the
reference values provided in the NAFEMS benchmarks.


35
midas NFX Benchmark Series 3-35
Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Thickness t = 0.6m

Table 3.14.1 Natural frequencies in Hz using axisymmetirc elements

Table 3.14.2 Natural frequencies in Hz using solid elements



Mode Number 1 2 3 4 5
Reference 18.583 140.15 224.16 358.29 629.19
Element
type
Number of
elements

QUADX-4 5 18.569 147.47 223.64 441.20 695.50
TRIAX-3 10 20.564 224.71 291.81 685.63 731.61
QUADX-8 5 18.582 140.56 224.18 374.05 686.04
TRIAX-6 10 18.595 141.60 224.16 382.95 688.47
Mode Number 1 2 3 4 5
Reference 18.583 140.15 224.16 358.29 629.19
Element
type
Number of
elements

TETRA-4 565 20.910 159.43 224.52 413.64 688.22
PENTA-6 120 18.951 141.08 224.55 366.24 648.70
HEXA-8 15x1x4 18.600 139.20 224.22 362.81 645.35
TETRA-10 59 18.893 139.87 224.60 368.30 664.11
PENTA-15 10 18.658 135.18 224.06 344.42 598.98
HEXA-20 5x1x1 18.389 133.28 223.28 338.86 590.50


Free vibration analysis
3-36
3.15 Clamped thin rhombic plate

REFERENCE NAFEMS [3-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib15.nfx





Figure 3.15.1 Clamped thin rhombic plate model


Figure 3.15.2 Vibration mode shapes

Y
X
45
o
Z
10
10
Units : m
t = 0.05
x = y = R
z
= 0 at all nodes,
z = R
x
= R
y
= 0 along all edges
Mode 1 Mode 2 Mode 3
Mode 4 Mode 5 Mode 6
Figure 3.15.1 shows the thin rhombic plate problem. All sides of the plate are clamped. The lowest vibration
frequencies are obtained using shell, solid and layered solid elements. The results obtained are compared with
the reference values given in the NAFEMS benchmarks.


37
midas NFX Benchmark Series 3-37
Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Thickness t = 0.05m

Table 3.15.1 Natural frequencies in Hz obtained using shell elements

Table 3.15.2 Natural frequencies in Hz obtained using solid elements

Table 3.15.3 Natural frequencies in Hz obtained using layered solid elements
*
obtained using high-order layered hexahedral element with reduced integration



Mode Number 1 2 3 4 5 6
Reference 7.938 12.835 17.941 19.133 24.009 27.922
Element
type
Number of
elements

TRIA-3 288 8.047 13.337 18.925 19.780 25.729 29.923
QUAD-4 12x12 8.049 13.554 19.422 19.764 26.853 30.712
TRIA-6 72 7.644 12.298 17.017 17.906 22.605 26.057
QUAD-8 6x6 7.924 13.024 18.689 19.160 25.871 28.955
Mode Number 1 2 3 4 5 6
Reference 7.938 12.835 17.941 19.133 24.009 27.922
Element
type
Number of
elements

HEXA-8 12x12 8.107 13.818 19.822 20.040 27.539 31.830
PENTA-15 72 9.766 17.571 26.246 28.561 44.132 44.716
HEXA-20 6x6 9.205 16.008 22.479 26.542 34.882 42.779
Mode Number 1 2 3 4 5 6
Reference 7.938 12.835 17.941 19.133 24.009 27.922
Element
type
Number of
elements

PENTAL-6 288 9.018 15.223 23.438 23.484 34.079 36.314
HEXAL-8 12x12 8.107 13.818 19.822 20.041 27.541 31.831
PENTAL-15 72 9.766 17.571 26.246 28.561 44.132 44.716
HEXAL-20* 6x6 9.205 16.008 22.479 26.542 34.882 42.779


Free vibration analysis
3-38
3.16 Clamped thick rhombic plate

REFERENCE NAFEMS [3-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib16.nfx





Figure 3.16.1 Clamped thick rhombic plate model


Figure 3.16.2 Vibration mode shapes

Y
X
45
o
Z
10
10
Units : m
t = 1
x = y = R
z
= 0 at all nodes,
z = R
x
= R
y
= 0 along all edges
Mode 1 Mode 2 Mode 3
Mode 4 Mode 5 Mode 6
Figure 3.16.1 shows the thick rhombic plate problem. All sides of the plate are clamped. The lowest vibration
frequencies are obtained using shell, solid and layered solid elements. The results obtained are compared with
the reference values given in the NAFEMS benchmarks.



39
midas NFX Benchmark Series 3-39
Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Thickness t = 1.0 m

Table 3.16.1 Natural frequencies in Hz obtained using shell elements

Table 3.16.2 Natural frequencies in Hz obtained using solid elements


Table 3.16.3 Natural frequencies in Hz obtained using layered solid elements
*
obtained using high-order layered hexahedral element with reduced integration


Mode Number 1 2 3 4 5 6
Reference 133.95 201.41 265.81 282.74 334.45 N/A
Element
type
Number of
elements

TRIA-3 288 136.14 210.85 283.59 293.57 364.96 408.85
QUAD-4 12x12 135.80 213.17 287.49 291.66 372.78 414.89
TRIA-6 72 129.66 194.92 257.41 270.74 324.59 365.20
QUAD-8 6x6 114.50 171.54 225.55 226.73 289.76 308.23
Mode Number 1 2 3 4 5 6
Reference 133.95 201.41 265.81 282.74 334.45 N/A
Element
type
Number of
elements

PENTA-6 288 144.21 227.10 307.01 315.39 397.74 446.48
HEXA-8 12x12 140.56 225.25 306.63 307.35 402.39 446.76
PENTA-15 72 140.47 214.20 284.72 298.77 361.28 406.50
HEXA-20 6x6 140.49 214.41 286.69 299.47 367.19 409.34
Mode Number 1 2 3 4 5 6
Reference 133.95 201.41 265.81 282.74 334.45 N/A
Element
type
Number of
elements

PENTAL-6 288 141.24 221.56 300.51 311.23 389.90 438.75
HEXAL-8 12x12 140.76 225.80 308.41 308.09 348.19 406.18
PENTAL-15 72 140.54 214.30 284.86 298.92 361.47 406.72
HEXAL-20 6x6
141.83
140.56*
218.68
214.51*
295.59
286.84*
302.38
299.62*
383.36
367.40*
417.55
409.55*


Free vibration analysis
3-40
3.17 Cantilevered square membrane

REFERENCE NAFEMS [3-1]
KEYWORDS membrane elements
MODEL FILENAME FreeVib17.nfx





Figure 3.17.1 Cantilevered square membrane model

Y
X
10
Units : m
10
Z
z = 0 at all nodes
x = y = 0 along y-axis
Figure 3.17.1 shows the cantilevered square membrane problem. Membrane elements are employed to obtain
the lowest vibration frequencies. The results are compared with the reference values provided in the NAFEMS
benchmarks.


41
midas NFX Benchmark Series 3-41

Figure 3.17.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3


Table 3.17.1 Natural frequencies in Hz obtained using membrane elements


Mode 1 Mode 2 Mode 3
Mode 4 Mode 5 Mode 6
Mode Number 1 2 3 4 5 6
Reference 52.404 125.69 140.78 222.54 241.41 255.74
Element
type
Number of
elements

TRIA-3 128 53.848 126.35 145.43 237.64 250.51 267.11
QUAD-4 8x8 52.726 126.06 142.76 226.95 247.22 259.43
TRIA-6 32 52.646 125.87 141.76 225.67 243.90 257.64
QUAD-8 4x4 52.603 125.85 141.41 224.45 243.00 256.67


Free vibration analysis
3-42
3.18 Cantilevered tapered membrane

REFERENCE NAFEMS [3-2]
KEYWORDS membrane elements
MODEL FILENAME FreeVib18.nfx





Figure 3.18.1 Cantilevered tapered membrane model


Figure 3.18.2 Vibration mode shapes
10
2.5
Y
X
Z
2.5
Units : m
z = 0 at all nodes
x = y = 0 along y-axis
Mode 1 Mode 2 Mode 3
Mode 4 Mode 5 Mode 6
Figure 3.18.1 shows the cantilevered tapered membrane problem. Membrane elements are employed to obtain
the lowest vibration frequencies. The results are compared with the reference values provided in the NAFEMS
benchmarks.



43
midas NFX Benchmark Series 3-43
Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3


Table 3.18.1 Natural frequencies in Hz obtained using membrane elements
*
obtained using shell element formulations with 6-dof per node.



Mode Number 1 2 3 4 5 6
Reference 44.623 130.03 162.70 246.05 379.90 391.44
Element
type
Number of
elements

TRIA-3 256
45.643
44.725*
134.53
130.60*
162.92
162.89*
258.37
247.94*
393.48
384.41*
404.95
393.13*
QUAD-4 16x8
44.647
44.628*
131.04
130.20*
162.80
162.72*
250.33
246.82*
391.54
382.03*
393.10
392.84*
TRIA-6 64 44.645 130.19 162.72 247.05 383.47 391.58
QUAD-8 8x4 44.632 130.11 162.71 246.48 381.58 391.53


Free vibration analysis
3-44
3.19 Free annular membrane

REFERENCE NAFEMS [3-1]
KEYWORDS membrane elements
MODEL FILENAME FreeVib19.nfx





Figure 3.19.1 Free annular membrane model

Units : m
Y
1.8
Z
X
6
z = 0 at all nodes
Figure 3.19.1 shows the unconstrained annular membrane problem. Membrane elements are employed to
obtain the lowest vibration frequencies excluding rigid body modes. The results are compared with the
reference values provided in the NAFEMS benchmarks.



45
midas NFX Benchmark Series 3-45

Figure 3.19.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Thickness t = 0.6 m

Table 3.19.1 Natural frequencies in Hz obtained using membrane elements


Mode 4&5 Mode 6 Mode 7&8
Mode 9&10 Mode 11&12 Mode 13&14
Mode Number 4,5 6 7,8 9,10 11,12 13,14
Reference 129.24 226.17 234.74 264.66 336.61 376.79
Element
type
Number of
elements

TRIA-3 96 135.43 227.63 235.19 278.91 343.71 400.79
QUAD-4 3x16 128.64 225.22 234.87 270.75
339.87,
339.88
389.29
TRIA-6 320 129.08 227.36 236.07 269.48 340.43 384.01
QUAD-8 5x32 125.97 224.18 232.94 263.88 335.60 377.57


Free vibration analysis
3-46
3.20 Cantilever beam mesh distortion

REFERENCE NAFEMS [3-1]
KEYWORDS bar elements
MODEL FILENAME FreeVib20.nfx





Figure 3.20.1 Cantilever beam model

Material data
Youngs modulus
Density
E = 200 GPa
= 8000 kg/m
3

Section property Square cross-section 0.125 m 0.125m

Table 3.20.1 Natural frequencies in Hz obtained using bar elements


10
Units : m
A
0.125
0.125
X
Y
a a a b b b b a
x = y = R
z
= 0 at A
Mode Number 1 2 3 4 5 6
Reference 1.010 6.327 17.716 34.717 57.390 85.730
Element
type
Element
length ratios

BAR
a = b 1.010 6.323 17.698 34.694 57.470 86.230
a = 10b 1.010 6.327 17.796 34.873 60.626 101.694
a = 100b 1.010 6.330 17.824 35.080 64.766 104.671
A cantilever beam problem is shown in Figure 3.20.1. Free vibration analyses are carried out with several
ratios of a:b to see the effect of 1-dimensional mesh distortion on the accuracies of the results.


47
midas NFX Benchmark Series 3-47
3.21 Free cylinder axisymmetric vibration

REFERENCE NAFEMS [3-2]
KEYWORDS axisymmetric elements, solid elements
MODEL FILENAME FreeVib21_Axisymmetric.nfxa
FreeVib21_Solid.nfx





Figure 3.21.1 Free cylinder axisymmetric model


Figure 3.21.2 Vibration mode shapes

1.8
Units : m
10
R
L
C
0.4
Z
Mode 2 Mode 3 Mode 4
Mode 5 Mode 6
L
C
L
C
L
C
L
C
L
C
Figure 3.21.1 shows axisymmetric model of the free cylinder problem. Both axisymmetric elements and solid
elements are employed to obtain the free vibration frequencies. Lowest modes not including the rigid body
modes are compared with the reference solutions given in the NAFEMS benchmarks.


Free vibration analysis
3-48
Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3

Section property Thickness t = 0.4m

Table 3.21.1 Natural frequencies in Hz obtained using axisymmetric elements

Table 3.21.2 Natural frequencies in Hz obtained using solid elements


Mode Number 2 3 4 5 6
Reference 243.53 377.41 394.11 397.72 405.28
Element
type
Number of
elements

TRIAX-3 96 244.07 380.87 396.99 408.72 430.38
QUADX-4 16x3 243.96 378.10 394.34 397.87 406.74
TRIAX-6 16 243.50 377.41 394.26 397.90 406.41
QUADX-8 8x1 243.50 377.46 394.30 397.97 406.44
Mode Number 2 3 4 5 6
Reference 243.53 377.41 394.11 397.72 405.28
Element
type
Number of
elements

PENTA-6 96 244.07 380.82 396.68 407.51 430.48
HEXA-8 16x3 243.97 378.33 394.67 398.24 407.06
PENTA-15 16 243.50 377.41 394.26 397.90 406.41
HEXA-20 8x1 243.50 377.33 394.15 397.70 404.74


49
midas NFX Benchmark Series 3-49
3.22 Thick hollow sphere uniform radial vibration

REFERENCE NAFEMS [3-2]
KEYWORDS axisymmetric elements
MODEL FILENAME FreeVib22.nfx






Figure 3.22.1 Thick hollow cylinder model

1.8
Units : m
4.2
R
Z
C
L
O

Z
R
z displacement = 0 at all nodes,
Nodes at same R are constrained to have same r displacement
Figure 3.22.1 shows the axisymmetric model of a hollow thick sphere problem. Utilizing constraint equations,
radial displacements are made identical at nodes with identical radius, thus allowing uniform radial vibrations
only. The lowest uniform radial vibration frequencies obtained with axisymmetric elements are compared with
the respective reference solutions given in the NAFEMS benchmarks.


Free vibration analysis
3-50

Figure 3.22.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 200 GPa
v = 0.3
= 8000 kg/m
3


Table 3.22.1 Natural frequencies in Hz obtained using axisymmetric elements



Mode 1 Mode 2 Mode 3
Mode 4 Mode 5
R
r
R
r
R
r
R
r
R
r
Mode Number 1 2 3 4 5
Reference 369.91 838.03 1451.2 2117.0 2795.8
Element
type
Number of
elements

TRIAX-3 20 371.10 843.28 1473.72 2185.24 2947.42
QUADX-4 10x1 370.09 838.79 1469.97 2188.21 2970.57
TRIAX-6 10 370.83 839.17 1453.25 1937.09 2128.77
QUADX-8 5x1 370.02 838.09 1453.03 1942.56 2131.74


51
midas NFX Benchmark Series 3-51
3.23 Cylindrical shell with rigid diaphragm

REFERENCE Soedel [3-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME FreeVib23.nfx





Figure 3.23.1 Cylindrical shell model with rigid diaphragm

Units : mm
R = 100
200
Y
Z
X
Rigid Diaphragm
Rigid Diaphragm
Figure 3.23.1 shows the cylindrical shell problem with rigid diaphragm. One quarter model is used to obtain
the natural frequencies with symmetric boundary conditions. Free vibration analysis is carried out using shell,
solid and layered solid elements.


Free vibration analysis
3-52

Figure 3.23.2 Vibration mode shapes

Material data
Youngs modulus
Poissons ratio
Density
E = 20.610
4
N/mm
2

= 0.3
= 7.8510
-9
N/g/mm
3
Section property Thickness t = 2 mm

Table 3.23.1 Natural frequencies in Hz obtained using shell elements
*
obtained using lumped mass matrix






Mode 1 Mode 2 Mode 3
Mode 4 Mode 5
Mode Number 1 2 3 4 5
Reference 1342.5 1464.8 1725.9 1892.4 2493.1
Element
type
Number of
elements

TRIA-3* 400 1388.8 1553.7 1739.5 2006.6 2608.4
QUAD-4* 10x20 1343.5 1470.0 1728.8 1910.8 2535.3
TRIA-6 100 1391.1 1563.5 1738.1 2035.9 2681.0
QUAD-8 5x10 1341.6 1471.3 1724.4 1921.1 2569.8


53
midas NFX Benchmark Series 3-53
Table 3.23.2 Natural frequencies in Hz obtained using solid elements
*
obtained using lumped mass matrix

Table 3.23.3 Natural frequencies in Hz obtained using layered solid elements
*
obtained using lumped mass matrix
**
obtained using high-order layered hexahedral element with reduced integration




















Mode Number 1 2 3 4 5
Reference 1342.5 1464.8 1725.9 1892.4 2493.1
Element
type
Number of
elements

HEXA-8* 10x20x1 1355.7 1515.9 1730.8 2021.5 2757.2
HEXA-20 5x10x1 1348.3 1481.7 1727.1 1939.6 2611.7
Mode Number 1 2 3 4 5
Reference 1342.5 1464.8 1725.9 1892.4 2493.1
Element
type
Number of
elements

HEXAL-8* 10x20x1 1346.5 1481.9 1729.3 1940.9 2599.2
HEXAL-20** 5x10x1 1349.2 1483.0 1728.4 1941.8 2615.3


Free vibration analysis
3-54
References

[3-1] NAFEMS, Selected Benchmarks for Natural Frequency Analysis, Issue 2, NAFEMS, Glasgow, 1987
[3-2] NAFEMS, The Standard NAFEMS Benchmarks, Rev. 3, NAFEMS, Glasgow, 1990
[3-3] W. Soedel, Vibrations of Shells and Plates, 2
nd
Edition, Marcel Dekker, New York, 1993




1
midas NFX Benchmark Series 4-1
4.1 Column buckling

REFERENCE Gere et al. [4-1]
KEYWORDS bar elements, shell elements, solid elements
MODEL FILENAME Buckling01_CASE1.nfx
Buckling01_CASE2.nfx
Buckling01_CASE3.nfx
Buckling01_CASE4.nfx






Figure 4.1.1 Column model


Figure 4.1.2 Buckling mode shape
Y
X
15
Units : m
Top : Roller
Bottom : Pin
Top : Free
Bottom : Fixed
Top : Laterally guided
Bottom : Fixed
Top : Roller
Bottom : Fixed
1 tonf 1 tonf 1 tonf 1 tonf
1
0.25
Case 1 Case 2 Case 3 Case 4
Figure 4.1.1 shows the straight column model. Linear buckling analysis is carried out to determine the
buckling modes and the corresponding critical loads of a column subjected to a vertical load. Four kinds of
boundary condition are considered. The lowest buckling modes are sketched in Figure 4.1.2.

Linear buckling analysis
4-2
Material data Youngs modulus 10000 tonf/m
2

Section property
Rectangular cross-
section
0.25 m 1.0m

Table 4.1.1 Critical loads in tonf obtained using bar elements

Table 4.1.2 Critical loads in tonf obtained using shell elements

Table 4.1.3 Critical loads in tonf obtained using solid elements




Case number 1 2 3 4
Reference 0.5712 0.1428 2.2846 1.1684
Element
type
Number of
elements

BAR-2 15 0.5714 0.1428 2.2891 1.1695
Case number 1 2 3 4
Reference 0.5712 0.1428 2.2846 1.1684
Element
type
Number of
elements

TRIA-3 30 0.5741 0.1430 2.3323 1.1812
QUAD-4 15 0.5736 0.1429 2.3245 1.1790
TRIA-6 30 0.5698 0.1427 2.2627 1.1621
QUAD-8 15 0.5705 0.1427 2.2749 1.1656
Case number 1 2 3 4
Reference 0.5712 0.1428 2.2846 1.1684
Element
type
Number of
elements

HEXA-8 15 0.5747 0.1430 2.3421 1.1841
TETRA-10 180 0.5724 0.1429 2.3053 1.1743
PENTA-15 30 0.5712 0.1428 2.2859 1.1687
HEXA-20 15 0.5710 0.1428 2.2827 1.1678


3
midas NFX Benchmark Series 4-3
4.2 Three member frame

REFERENCE Timoshenko et al. [4-2]
KEYWORDS bar elements
MODEL FILENAME Buckling02.nfx





Figure 4.2.1 Three-member frame model


Figure 4.2.2 Buckling mode shape

Y
X
100
Units : in
100
1 lbf 1 lbf

Figure 4.2.1 shows a plane frame structure with two vertical point forces. The buckling load is determined for
each case in which different number of elements per member are considered. In Figure 4.2.2, the typical
buckling mode is depicted.

Linear buckling analysis
4-4
Material data Youngs modulus E = 1 10
6
psi
Section property
Area
Moment of inertia
A = 1.0 in
2
I = 1.0 in
4


Table 4.2.1 Critical loads in lbf obtained using bar elements



Reference 737.9
Element type Number of elements
BAR-2
2 per member 739.8
4 per member 737.6
8 per member 737.5


5
midas NFX Benchmark Series 4-5
4.3 Uniaxially compressed clamped square plate

REFERENCE Chajes [4-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Buckling03.nfx





Figure 4.3.1 Clamped square plate model


Figure 4.3.2 Buckling mode shape

Material data
Youngs modulus
Poissons ratio
E = 11.064 10
5
psi

= 0.3
Section property Thickness t = 0.01 in





Y
X
1
1
Units : in.
P = 1.0 lbf/in.
P
t = 0.01
Modelled portion
Figure 4.3.1 shows a square plate subjected to compressive load. The linear buckling analysis is carried out to
determine the critical load with clamped boundary condition. One quarter model is used with symmetric
boundary condition.

Linear buckling analysis
4-6
Table 4.3.1 Critical distributed load in lbf. obtained using shell elements

Table 4.3.2 Critical distributed load in lbf. obtained using solid elements

Table 4.3.3 Critical distributed load in lbf. obtained using layered solid elements



Reference 100.7
Element type Number of elements
TRIA-3 32 111.7
QUAD-4 16 107.3
TRIA-6 32 97.2
QUAD-8 16 100.6
Reference 100.7
Element type Number of elements
HEXA-8 16 98.0
PENTA-15 32 127.8
HEXA-20 16 111.8
Reference 100.7
Element type Number of elements
HEXAL-8 16 98.0
PENTAL-15 32 127.8
HEXAL-20 16 113.6


7
midas NFX Benchmark Series 4-7
4.4 Rectangular plate under concentrated center loads

REFERENCE Timoshenko et al. [4-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Buckling04.nfx





Figure 4.4.1 Simply supported rectangular plate model

Material data
Youngs modulus
Poissons ratio
E = 200 GPa

= 0.3
Section property Thickness t = 0.01 m

Table 4.4.1 Critical load in kN obtained using shell elements


Y
X
Z
A
B C
D
E
P = 1 kN
2
1
Units : m
t = 0.01
Reference 330
Element type Number of elements
TRIA-3 144 337
QUAD-4 72 327
TRIA-6 36 311
QUAD-8 18 320
Figure 4.4.1 shows a rectangular plate under point force. All sides are simply supported, and the edge AD is
constrained in Y direction. A concentrated force is applied at the center of BC (point E). The linear buckling
analysis is carried out to obtain the lowest critical load.

Linear buckling analysis
4-8
Table 4.4.2 Critical load in kN obtained using solid elements

Table 4.4.3 Critical load in kN obtained using layered solid elements




Reference 330
Element type Number of elements
HEXA-8 72 298
PENTA-15 36 557
HEXA-20 18 330
Reference 330
Element type Number of elements
PENTAL-6 144 457
HEXAL-8 72 298
PENTAL-15 36 559
HEXAL-20 18 348


9
midas NFX Benchmark Series 4-9
4.5 Axially compressed cylinder

REFERENCE Simo et al. [4-4]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Buckling05.nfx






Figure 4.5.1 Axially compressed cylinder model


Figure 4.5.2 Buckling mode shape

Material data
Youngs modulus
Poissons ratio
E = 567 Pa

= 0.3
Section property Thickness t = 0.247 m
P = 0.209 N/m
Clamped
R = 100
L = 35.95
Units : m
Figure 4.5.1 shows a cylindrical shell model under distributed compressive load. Both ends of the cylinder are
clamped in transverse direction. One eighth portion of the cylinder is modeled with symmetric boundary
condition. The linear buckling analysis is carried out to determine the lowest critical load factor. The typical
mode shape is sketched in Figure 4.5.2.

Linear buckling analysis
4-10
Table 4.5.1 Critical load factor obtained using shell elements

Table 4.5.2 Critical load factor obtained using solid elements

Table 4.5.3 Critical load factor obtained using layered solid elements


Reference 1.0833
Element type Number of elements
TRIA-3 840 1.1094
QUAD-4 420 1.1370
TRIA-6 840 1.0104
QUAD-8 420 1.0198
Reference 1.0833
Element type Number of elements
HEXA-8 420 1.0631
HEXA-20 420 0.9951
Reference 1.0833
Element type Number of elements
HEXAL-8 420 1.0714
HEXAL-20 420 0.9933


11
midas NFX Benchmark Series 4-11
4.6 L-bracket plate under in-plane load

REFERENCE Simo et al. [4-4] and Argyris et al. [4-5]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME Buckling06.nfx





Figure 4.6.1 L-bracket plate model


Figure 4.6.2 Buckling mode shapes

Material data
Youngs modulus
Poissons ratio
E = 71.240 GPa

= 0.3
Section property Thickness t = 0.6 mm







240
240
30
Units : mm
t = 0.6
X
Y
P = 1 N
Figure 4.6.1 shows a clamped L-shape plate under in-plane bending load. The linear buckling analysis
performed to obtain the critical load. The critical load induces the lateral buckling, of which the mode shape is
depicted in Figure 4.6.2.

Linear buckling analysis
4-12
Table 4.6.1 Critical load in N obtained using shell elements

Table 4.6.2 Critical load in N obtained using solid elements

Table 4.6.3 Critical load in N obtained using layered solid elements
*
obtained using high-order layered hexahedral element with reduced integration












Reference 1.137 (Simo et al.), 1.155 (Argyris et al.)
Element type Number of elements
TRIA-3 136 1.189
QUAD-4 68 1.199
TRIA-6 34 1.202
QUAD-8 17 1.189
Reference 1.137 (Simo et al.), 1.155 (Argyris et al.)
Element type Number of elements
HEXA-8 68 1.198
PENTA-15 34 1.237
HEXA-20 17 1.188
Reference 1.137 (Simo et al.), 1.155 (Argyris et al.)
Element type Number of elements
PENTAL-6 136 1.373
HEXAL-8 68 1.198
PENTAL-15 34 1.236
HEXAL-20 17
1.262
1.188*


13
midas NFX Benchmark Series 4-13
References

[4-1] J.M. Gere and S.P. Timoshenko, Mechanics of Materials, 2
nd
Edition, Thomson Brooks/Cole, California,
New York, 1984
[4-2] S.P. Timoshenko and J.M. Gere, Theory of Elastic Stability, 2
nd
Edition, McGraw-Hill, New York, 1961
[4-3] A. Chajes, Principles of Structural Stability Theory, Prentice-Hall, Englewood Cliffs, N.J,1974
[4-4] J.C. Simo, D.D. Fox and M.S. Rifai, On a Stress Resultant Geometrically Exact Shell Model. Part III:
The Computational Aspects of the Nonlinear Theory, Computer Methods in Applied Mechanics and
Engineering, Vol. 79, pp. 21-70, 1990
[4-5] J.H. Argyris, H. Balmer, J.St. Doltsinis, P.C. Dunne, M. Haase, M. Hasse, M. Kleiber, G.A. Malejannakis,
H.P. Mlejenek, M. Muller and D.W. Scharpf, Finite Element Method The natural approach, Computer
Methods in Applied Mechanics and Engineering, Vol. 17/18, pp. 1-106, 1979




1
midas NFX Benchmark Series 5-1
5.1 One-dimensional heat transfer with heat source

REFERENCE Lewis et al. [5-1]
KEYWORDS bar elements, shell elements, solid elements, layered solid elements
MODEL FILENAME SteadyStateHeat01.nfx





Figure 5.1.1 column model

Material data Conductivity k = 10+0.1T W/mC
Section property
Rectangular cross-
section
0.1 m 0.1 m


Table 5.1.1 Temperature T at node B obtained using bar elements


A
2
Units : m
T = 5
o
C
X
Element 1 Element 2 B
Q = 36000 J/m
3
hr

B
T [ C]
Reference 5.948
Element
type
Number of
elements

BAR-2 2 5.948
Figure 5.1.1 shows one-dimensional steady-state heat transfer problem with temperature dependent
conductivity. The one-dimensional problem is discretized with 2 in the length direction with internal heat
generation of 36000 J/m
3
hr in element 2. Point A has fixed temperature of 5 C. Temperature at mid point B is
determined.

Steady-state heat transfer analysis
5-2
Table 5.1.2 Temperature T at node B obtained using shell elements


Table 5.1.3 Temperature T at node B obtained using solid elements

Table 5.1.4 Temperature at node B obtained using layered solid elements



Reference
B
T [ C]
5.948
Element
type
Number of
elements

TRIA-3 (2x1) 5.948
QUAD-4 2 5.948
TRIA-6 (2x1) 5.948
QUAD-8 2 5.948
Reference
B
T [ C]
5.948
Element
type
Number of
elements

PENTA-6 (2x2) 5.948
HEXA-8 2 5.948
PENTA-15 (2x2) 5.948
HEXA-20 2 5.948
Reference
B
T [ C]
5.948
Element
type
Number of
elements

PENTAL-6 (2x2) 5.948
HEXAL-8 2 5.948
PENTAL-15 (2x2) 5.948
HEXAL-20 2 5.948


3
midas NFX Benchmark Series 5-3
5.2 Two-dimensional heat transfer with convection

REFERENCE NAFEMS. [5-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME SteadyStateHeat02.nfx





Figure 5.2.1 Rectangular plate model

Material data
Conductivity
Convection coefficient
k = 52 J/mhr
o
C
h = 750.0 W/m
2

o
C









A B
C D
Insulator
Zero flux
Convection to
ambient temperature
of 0
Prescribed
temperature
of 100
E
0.6
0.2
1
Uniform
thickness
Units : m
Figure 5.2.1 shows a two-dimensional heat transfer problem. Temperature of 100
o
C is prescribed on edge AB.
On edges BC and CD convection boundary conditions are applied with ambient temperature at 0
o
C. Edge DA
is insulated. Steady-state heat transfer analysis is carried out and the resulting temperature at point E is
determined.

Steady-state heat transfer analysis
5-4
Table 5.2.1 Temperature T at node E obtained using shell elements

Table 5.2.2 Temperature T at node E obtained using solid elements

Table 5.2.3 Temperature T at node E obtained using layered solid elements


Reference
E
T [ C]
18.3
Element type Number of elements
TRIA-3 6x(10x2) 18.9
QUAD-4 6x10 18.9
TRIA-6 3x(5x2) 18.0
QUAD-8 3x5 18.0
Reference
E
T [ C]
18.3
Element type Number of elements
PENTA-6 6x(10x2)x1 18.9
HEXA-8 6x10x1 18.9
PENTA-15 3x(5x2)x1 18.0
HEXA-20 3x5x1 17.9
Reference
E
T [ C]
18.3
Element type Number of elements
PENTAL-6 6x(10x2)x1 18.9
HEXAL-8 6x10x1 18.9
PENTAL-15 3x(5x2)x1 18.0
HEXAL-20 3x5x1 17.9


5
midas NFX Benchmark Series 5-5
5.3 Two-dimensional heat transfer in bi-material

REFERENCE Holman [5-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME SteadyStateHeat03.nfx






Figure 5.3.1 Bi-material embedded in high-conductivity material

Material data
Conductivity
k = 2.0 W/m C (Material A)

k = 0.3 W/m C (Material B)

Convection coefficient h = 25 W/m
2
C








6
1.5
1
5
Convection to ambient T=30
400
400
400
Material A
Material B Units : cm
A B
Figure 5.3.1 shows bi-material embedded in a high-thermal-conductivity material maintained at 400 C. The
upper surface is exposed to convection environment at 30 C. Temperature at points A and B are determined
and compared with reference the reference solution given in [5-3].

Steady-state heat transfer analysis
5-6
Table 5.3.1 Temperatures T at node A and B obtained using shell elements

Table 5.3.2 Temperatures T at node A and B obtained using solid elements

Table 5.3.3 Temperatures T at node A and B obtained using layered solid elements




A
T [ C]
B
T [ C]
Reference 254.96 247.64
Element type Number of elements
TRIA-3 12x(10x2) 250.97 246.51
QUAD-4 12x10 250.07 246.49
TRIA-6 12x(10x2) 249.86 246.35
QUAD-8 12x10 249.72 246.38

A
T [ C]
B
T [ C]
Reference 254.96 247.64
Element type Number of elements
PENTA-6 12x(10x2)x1 250.97 246.51
HEXA-8 12x10x1 250.07 246.49
PENTA-15 12x(10x2)x1 249.72 246.38
HEXA-20 12x10x1 249.86 246.35

A
T [ C]
B
T [ C]
Reference 254.96 247.64
Element type Number of elements
PENTAL-6 12x(10x2)x1 250.97 246.51
HEXAL-8 12x10x1 250.07 246.49
PENTAL-15 12x(10x2)x1 249.72 246.38
HEXAL-20 12x10x1 249.86 246.35


7
midas NFX Benchmark Series 5-7
5.4 One-dimensional heat transfer with radiation

REFERENCE NAFEMS [5-2]
KEYWORDS bar elements, shell elements, solid elements, layered solid elements
MODEL FILENAME SteadyStateHeat04.nfx







Figure 5.4.1 Column model

Material data
Conductivity
Emissivity
Stefan-Boltzmann constant
k = 55.6 W/mC

= 0.98
= 5.67x10
-8
Wm
2
K
-4

Table 5.4.1 Temperature T at point B obtained using bar elements





A
0.1
Units : m
B
T = 1000 K
Radiation to
Ambient temperature
Of 300 K
X
Uniform cross section

B
T [K]
Reference 927
Element type Number of elements
BAR-2 10 926.4
Figure 5.4.1 shows a one-dimensional bar. Temperature is maintained at 1000 K at one end (point A). The other
end at point B is exposed to radiation to ambient temperature of 300 K. 10 elements in the length direction are
used to model the bar. Temperature at point B obtained using bar, shell, solid and layered solid elements are
compared with the reference solution given in [5-2].

Steady-state heat transfer analysis
5-8
Table 5.4.2 Temperature T at point B obtained using shell elements

Table 5.4.3 Temperature T at point B obtained using solid elements

Table 5.4.4 Temperature T at point B obtained using layered solid elements



B
T [K]
Reference 927
Element type Number of elements
TRIA-3 (10x2) 926.4
QUAD-4 10 926.4
TRIA-6 (10x2) 924.6
QUAD-8 10 924.6

B
T [K]
Reference 927
Element type Number of elements
TETRA-4 87 926.4
PYRAM-5 60 926.4
PENTA-6 (10x2)x1 926.4
HEXA-8 10x1 926.4
TETRA-10 87 926.4
PYRAM-13 60 926.4
PENTA-15 (10x2)x1 926.4
HEXA-20 10x1 926.4

B
T [K]
Reference 927
Element type Number of elements
PENTAL-6 (10x2)x1 926.4
HEXAL-8 10x1 926.4
PENTAL-15 (10x2)x1 926.4
HEXAL-20 10x1 926.4


9
midas NFX Benchmark Series 5-9
5.5 Radiation shape factor calculation

REFERENCE Holman [5-3]
KEYWORDS Radiation exchange elements
MODEL FILENAME SteadyStateHeat05.nfx





Figure 5.5.1 Radiation shape factor between two rectangular surfaces

Table 5.5.1 Radiation shape factors
1 2
F

-- case A: parallel rectangles obtained using two radiation exchange elements


Y
X
D
X
Z
Y
Case A Case B
Plate 1
Plate 2
Plate 1
Plate 2
Parallel rectangles Perpendicular rectangles with common edge
X/D Y/D
1 2
F

Reference
0.1 0.1 0.00316 0.00316
1 0.1 0.02492 0.02492
5 0.1 0.04357 0.04360
1 1 0.19975 0.19982
5 1 0.35892 0.35917
5 5 0.68949 0.69024
Radiation shape factors are computed between two parallel rectangles (case A), and two perpendicular
rectangles with common edge (case B). The results were obtained for several geometric configurations and
compared with analytical solutions [5-3].

Steady-state heat transfer analysis
5-10
Table 5.5.2 Radiation shape factor
1 2
F

-- case B: perpendicular rectangles with common edge obtained using two
radiation exchange elements



Z/X Y/X
1 2
F

Reference
0.1 0.1 0.28375 0.28189
1 0.1 0.43433 0.43251
5 0.1 0.45326 0.43915
1 1 0.20004 0.20004
5 1 0.24690 0.24690
5 5 0.08723 0.08810


11
midas NFX Benchmark Series 5-11
5.6 Perpendicular surfaces in radiant balance

REFERENCE Holman [5-3]
KEYWORDS shell elements, Radiation exchange elements
MODEL FILENAME SteadyStateHeat06.nfx





Figure 5.6.1 Surface model in radiant balance

Material data Conductivity k = 1000 W/mC


Table 5.6.1 Temperature T at surface 2 in K obtained using shell elements
*
average temperature of nodes on plate 2.

Insulated
Ambient temperature
at 300 K
T
1
=1000 K
0.5
0.5
0.5
Units : m
Plate 1
Plate 2

1
= 0.6

2
T [K]
Reference 599.4
Element type Number of elements
CTRIA-3 4 599.8*
CQUAD-4 2 599.4
CTRIA-6 4 599.9*
CQUAD-8 2 599.4
Figure 5.6.1 shows two rectangles placed perpendicular to one another with a common edge. One surface has
fixed temperature of 1000 K, while the other surface is insulated and in radiant balance with the ambient.
Temperature of the insulated surface 2 is determined and compared with the reference. Conductivity is set high
to obtain near-uniform temperature distribution on surface 2.

Steady-state heat transfer analysis
5-12
References

[5-1] R.W. Lewis, K. Morgan, H.R. Thomas and K.N. Seetharamu, The Finite Element Method in Heat
Transfer Analysis, Wiley, West Sussex, 1996
[5-2] NAFEMS, The Standard NAFEMS Benchmarks, Rev. 3, NAFEMS, Glasgow, 1990
[5-3] J.P. Holman, Heat Transfer, 9
th
Edition, McGraw-Hill, New York, 2002





1
midas NFX Benchmark Series 6-1
6.1 One-dimensional transient heat transfer - I

REFERENCE NAFEMS [6-1]
KEYWORDS bar elements, shell elements, solid elements, layered solid elements
MODEL FILENAME TransientHeat01.nfx






Figure 6.1.1 One-dimensional transient heat transfer problem

Material data
Conductivity
Specific heat
Density
k = 35.0 W/mC
C= 440.5 J/kgC
=7200 kg/m
3

Table 6.1.1 Temperature at point C obtained using bar element








A
0.1
Units : m
T = 0
o
C
X
B
T = 100sin(pt/40)
o
C
C
0.02

C
T [ C]
Reference 36.60
Element type Number of elements
BAR-2 10 35.51
Figure 6.1.1 shows one-dimensional transient heat transfer problem with conduction. Temperature of 0 C is
assigned to point A. Time variant temperature of 100sin( / 40) t is set to point B. The initial temperature of 0
C is applied at all of the nodes. Transient heat transfer analysis is carried out with fixed time step of 1 second.
Temperature at point C at time t=32 sec is obtained using various finite elements. Solution from the NAFEMS
benchmarks is taken as a reference for comparison.
Transient heat transfer analysis

6-2
Table 6.1.2 Temperature at point C obtained using shell elements
*
averaged temperature from nodes.

Table 6.1.3 Temperature at point C obtained using solid elements
*
averaged temperature from nodes.

Table 6.1.4 Temperature at point C obtained using layered solid elements
*
averaged temperature from nodes.


C
T [ C]
Reference 36.60
Element type Number of elements
TRIA-3 1x(10x2) 35.51
QUAD-4 1x10 35.51
TRIA-6 1x(10x2) 36.09*
QUAD-8 1x10 36.09

C
T [ C]
Reference 36.60
Element type Number of elements
TETRA-4 1x1x(10x2) 35.51*
PYRAM-5 1x1x10 35.86
PENTA-6 1x1x(10x2) 35.51
HEXA-8 1x1x10 35.51
TETRA-10 1x1x(10x2) 36.09*
PYRAM-13 1x1x10 36.09*
PENTA-15 1x1x(10x2) 36.09*
HEXA-20 1x1x10 36.09

C
T [ C]
Reference 36.60
Element type Number of elements
PENTAL-6 1x1x(10x2) 35.51
HEXAL-8 1x1x10 35.51
PENTAL-15 1x1x(10x2) 36.09*
HEXAL-20 1x1x10 36.09


3
midas NFX Benchmark Series 6-3
6.2 One-dimensional transient heat transfer - II

REFERENCE J. Barlow et al. [6-2]
KEYWORDS bar elements, shell elements, solid elements, layered solid elements
MODEL FILENAME TransientHeat02.nfx






Figure 6.2.1 One-dimensional transient heat transfer problem

Material data
Conductivity
Specific heat
Density
k = 1.0 W/mC
C= 985 J/kgC
=2300 kg/m
3

Table 6.2.1 Temperature at point C obtained using bar element







A
0.1
Units : m
T = 0
o
C
X
B
T = 100sin(pt/40)
o
C
C
0.01

C
T [ C]
Reference 9.62
Element type Number of elements
BAR-2
10 3.46
20 7.11
Figure 6.1.2 represents one-dimensional transient heat transfer problem with conduction. Point A has fixed
temperature of 0 C while the temperature at point B varies with time, given by 100sin( / 40) t . The entire
model is at 0 C initially. Transient heat transfer analysis is carried out with fixed time step of 1 second.
Temperature at point C at time t=58 sec is obtained using various finite elements. Due to the high level of
temperature gradient near point C, high order elements perform significantly better than low order elements.

Transient heat transfer analysis

6-4
Table 6.2.2 Temperature at point C obtained using shell elements
*
averaged temperature from nodes.


C
T [ C]
Reference 9.62
Element type Number of elements
TRIA-3
1x(10x2) 3.46
1x(20x2) 7.11
QUAD-4
1x10 3.46
1x20 7.11
TRIA-6
1x(10x2) 9.07*
1x(20x2) 9.03*
QUAD-8
1x10 9.36
1x20 9.14


5
midas NFX Benchmark Series 6-5
Table 6.2.3 Temperature at point C obtained using solid elements
*
averaged temperature from nodes.


C
T [ C]
Reference 9.62
Element type Number of elements
TETRA-4
1x1x(10x2) 3.46*
1x1x(20x2) 7.05*
PYRAM-5
1x1x10 6.94
1x1x20 7.71
PENTA-6
1x1x(10x2) 3.46
1x1x(20x2) 7.11
HEXA-8
1x1x10 3.46
1x1x20 7.11
TETRA-10
1x1x(10x2) 9.26*
1x1x(20x2) 8.99*
PYRAM-13
1x1x10 7.76*
1x1x20 9.15*
PENTA-15
1x1x(10x2) 9.00*
1x1x(20x2) 8.99*
HEXA-20
1x1x10 9.36
1x1x20 9.14
Transient heat transfer analysis

6-6
Table 6.2.4 Temperature at point C obtained using layered solid elements
*
averaged temperature from nodes.


C
T [ C]
Reference 9.62
Element type Number of elements
PENTAL-6
1x1x(10x2) 3.46
1x1x(20x2) 7.11
HEXAL-8
1x1x10 3.46
1x1x20 7.11
PENTAL-15
1x1x(10x2) 9.00*
1x1x(20x2) 8.99*
HEXAL-20
1x1x10 9.36
1x1x20 9.14


7
midas NFX Benchmark Series 6-7
6.3 Transient heat transfer with convection

REFERENCE Holman [6-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME TransientHeat03.nfx






Figure 6.3.1 Two-dimensional transient heat transfer problem

Material data
Conductivity
Specific heat
Density
k = 3.0 W/mC
C= 800 J/kgC
=1600 kg/m
3








2
0.5
1
Convection to ambient T=50
300
300
Units : cm
A
B
C
D
E
Figure 6.3.1 shows two-dimensional transient heat transfer model whose sides are maintained at temperature of
300 C. The bottom edge of the model is insulated, and the top edge is subjected to heat exchange by
convection. The ambient temperature for convection is 50 C and convective heat transfer coefficient
200W/m
2
C. The entire model is at 300 C initially. Temperature at point E after 12 seconds is obtained using
transient heat transfer analysis with time step of 2 second.
Transient heat transfer analysis

6-8
Table 6.3.1 Temperature at point E obtained using shell elements

Table 6.3.2 Temperature at point E obtained using solid elements

Table 6.3.3 Temperature at point E obtained using layered solid elements


E
T [ C]
Reference 243.32
Element type Number of elements
TRIA-3 20x(10x2) 239.71
QUAD-4 20x10 239.56
TRIA-6 20x(10x2) 239.46
QUAD-8 20x10 239.46

E
T [ C]
Reference 243.32
Element type Number of elements
TETRA-4 20x(10x24)x1 239.56
PYRA-5 20x(10x6)x1 239.56
PENTA-6 20x(10x2)x1 239.71
HEXA-8 20x10x1 239.56
TETRA-10 20x(10x24)x1 240.63
PYRA-13 20x(10x6)x1 239.46
PENTA-15 20x(10x2)x1 239.46
HEXA-20 20x10x1 239.46

E
T [ C]
Reference 243.32
Element type Number of elements
PENTAL-6 20x(10x2)x1 239.71
HEXAL-8 20x10x1 239.56
PENTAL-15 20x(10x2)x1 239.46
HEXAL-20 20x10x1 239.46


9
midas NFX Benchmark Series 6-9
6.4 Transient heat transfer with heat generation

REFERENCE Holman [6-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME TransientHeat04.nfx






Figure 6.4.1 Two-dimensional transient heat transfer problem

Material data
Conductivity
Specific heat
Density
k = 19 W/mC
C= 460 J/kgC
=7800 kg/m
3

Table 6.4.1 Temperature at point E obtained using shell elements

5
2
5
Convection to ambient T=20
Units : mm
A
B
C
D
E
Convection to ambient T=120
Heat Generation
50 MW/m
3

E
T [ C]
Reference 190.70
Element type Number of elements
TRIA-3 5x(5x2) 190.37
QUAD-4 5x5 190.39
TRIA-6 5x(5x2) 190.32
QUAD-8 5x5 190.32
Figure 6.4.1 represents a two-dimensional plane with 50MW/m
3
internal heat generation. The two side edges
are subjected to convection boundary conditions. The edges AC and BD have convection coefficients of
400W/m
2
and 500 W/m
2
respectively. The ambient temperatures for the two edges are 120 C and 20 C,
respectively. The plane is initially at a uniform temperature of 100 C. Temperature at point E after 9 seconds
is obtained by transient heat transfer analysis.
Transient heat transfer analysis

6-10
Table 6.4.2 Temperature at point E obtained using solid elements
*
averaged temperature from nodes.

Table 6.4.3 Temperature at point E obtained using layered solid elements



E
T [ C]
Reference 190.70
Element type Number of elements
PENTA-6 5x(5x2)x1 190.37
HEXA-8 5x5x1 190.39
TETRA-4 5x(5x24)x1 190.37*
PYRA-5 5x(5x6)x1 190.38
PENTA-15 5x(5x2)x1 190.32
HEXA-20 5x5x1 190.32
TETRA-10 5x(5x24)x1 190.32
PYRA-13 5x(5x6)x1 190.32

E
T [ C]
Reference 190.70
Element type Number of elements
PENTAL-6 5x(5x2)x1 190.37
HEXAL-8 5x5x1 190.39
PENTAL-15 5x(5x2)x1 190.32
HEXAL-20 5x5x1 190.32


11
midas NFX Benchmark Series 6-11
6.5 Axisymmetric transient heat transfer with convection

REFERENCE Holman [6-3]
KEYWORDS axisymmetric elements
MODEL FILENAME TransientHeat05.nfx





Figure 6.5.1 Axisymmetric transient heat transfer problem

Material data
Conductivity
Specific heat
Density
k = 35 W/mC
C= 460 J/kgC
=7800 kg/m
3



Table 6.5.1 Temperature at point A obtained using axisymmetric elements

X
Z
Y
A
5
Units : cm
Convection
to ambient T=100

A
T [ C]
Reference 150.0
Element type Number of elements
TRIAX-3 196 151.06
QUADX-4 94 151.06
TRIAX-6 196 151.06
QUADX-8 94 151.06
Figure 6.5.1 shows a 5.0 cm diameter hemisphere with initial temperature 450C. The hemisphere is subjected
to convection with ambient temperature of 100C and convection coefficient of 10 W/m
2
C. Axisymmetric
heat transfer elements are employed to evaluate the transient response of the model. Temperature obtained at
point A at time t=5819 sec is compared with the reference value.
Transient heat transfer analysis

6-12
References

[6-1] NAFEMS, The Standard NAFEMS Benchmarks, Rev. 3, NAFEMS, Glasgow, 1990
[6-2] J. Barlow, G. A. O. Davis, Selected FE Benchmarks in Structural and Thermal Analysis, NAFEMS,
Glasgow, 1987
[6-3] J. P. Holman, Heat Transfer, 9
th
Edition, McGraw-Hill, 2002





1
midas NFX Benchmark Series 7-1
7.1 Deep simply supported beam

REFERENCE NAFEMS [7-1]
KEYWORDS bar elements, solid elements
MODEL FILENAME LinearDynamic01_direct.nfx
LinearDynamic01_modal.nfx
LinearDynamic01_ direct_random.nfx
LinearDynamic01_ modal_random.nfx








X
Z
10
Units : m
2
2
F
0
=10
6
N/m
A


Figure 7.1.1 Simply supported deep beam model

Material data
Youngs modulus
Poissons ratio
Density
Mass proportional damping
Stiffness proportional damping
Modal damping
E = 200 GPa
v = 0.3
p = 8000 kg/m
3
o = 5.36 sec
-1

| = 7.46 10
-5
sec
c = 0.02
Section property Rectangular cross-section 2m 2m
Forcing functions
(1) Harmonic
0
sin 2 F F t f e e t = =
(2) Periodic
0
(sin sin3 ) 2 20 F F t t f f Hz e e e t = = =
(3) Transient
0
0 F F t = >
(4) Random PSD=(10
6
N/m)
2
/Hz
*
Rayleigh damping coefficients, and are chosen to give 0.02 damping in the dominant first mode.
A dynamic system consisting of a simply supported beam is shown in Figure 7.1.1. Four types of time-
variation of the load are selected; (1) harmonic loads containing single frequency (2) periodic forcing function
composed of two harmonic loadings (3) transient step load and (4) random forcing with uniform power
spectral density (PSD). All four types of loads are applied in the transverse direction of the beam and are
uniformly distributed spatially. For the first 3 load types, the peak displacement and stress are obtained either
in the frequency domain or the time domain. For the random forcing, peak displacement and stress PSD are
obtained in the frequency domain. The results are obtained at mid-span using both modal superposition and
direct methods. The responses obtained using various elements corresponding to each load type are
summarized in Tables 7.1.1~7.1.4. The results from the NAFEMS benchmark problems are taken as reference.

Linear dynamic analysis
7-2
Table 7.1.1 Peak responses and frequency of beam subjected to harmonic loads

Table 7.1.2 Peak responses of beam subjected to periodic forcing function

Table 7.1.3 Peak responses of beam subjected to transient step load
Peak
A
Z
u

[mm] Peak
A
o

[MPa] Frequency [Hz]
Reference 13.45 241.9 42.65
Element type
Number of
elements
Direct Modal Direct Modal Direct Modal
BAR 10 13.51 13.51 244.4 244.3 42.60 42.60
HEXA-8 10 13.09 13.09 236.5 236.5 43.40 43.40
PENTA-6 104 12.22 12.23 235.2 235.5 44.75 44.75
HEXA-20 5 13.17 13.17 236.3 236.3 42.95 42.95
PENTA-15 54 13.24 13.23 237.1 237.0 42.85 42.85
PYRAM-13 56 13.12 13.12 235.9 235.9 43.00 43.00
Peak
A
Z
u

[mm] Peak
A
o

[MPa]
Reference 0.951 17.1
Element type
Number of
elements
Direct Modal Direct Modal
BAR 10 0.954 0.954 17.4 17.4
HEXA-8 10 0.961 0.962 17.5 17.5
PENTA-6 104 0.964 0.965 18.6 18.6
HEXA-20 5 0.943 0.944 17.0 17.0
PENTA-15 54 0.944 0.944 17.0 17.0
PYRAM-13 56 0.943 0.944 17.0 17.0
Peak

A
Z
u [mm] Peak time [sec] Peak
A
o

[MPa] Static
A
Z
u [mm]
Reference 1.043 0.0117 18.76 0.538
Element type
Number of
elements
Direct Modal Direct Modal Direct Modal Direct Modal
BAR 10 1.044 1.043 0.0117 0.0116 18.52 18.53 0.537 0.537
HEXA-8 10 1.012 1.011 0.0116 0.0116 18.03 18.05 0.521 0.521
PENTA-6 104 0.946 0.945 0.0112 0.0113 17.99 17.97 0.487 0.487
HEXA-20 5 1.019 1.017 0.0117 0.0116 18.07 17.98 0.525 0.525
PENTA-15 54 1.023 1.022 0.0117 0.0116 18.11 18.04 0.527 0.527
PYRAM-13 56 1.014 1.013 0.0117 0.0116 18.03 17.92 0.522 0.522


3
midas NFX Benchmark Series 7-3
Table 7.1.4 Peak responses and frequency of beam subjected to white noise power spectral density



Peak
A
Z
u

PSD

[mm
2
/Hz]
Peak
A
o PSD

[(N/mm
2
)
2
/Hz]
Frequency [Hz]
Reference 180.90 58516 42.65
Element type
Number of
elements
Direct Modal Direct Modal Direct Modal
BAR 20 181.99 181.86 58855 58817 42.60 42.61

Linear dynamic analysis
7-4
7.2 Simply supported thin square plate

REFERENCE NAFEMS [7-1]
KEYWORDS shell elements
MODEL FILENAME LinearDynamic02_direct.nfx
LinearDynamic02_modal.nfx
LinearDynamic02_direct_random.nfx
LinearDynamic02_modal_random.nfx









z
Y
X
10.0
1
0
.
0
x = y = Rz = 0 at all nodes
Rx = Ry = 0 along 4
edges
Units: m
F
0
=100 N/m
2
A

Figure 7.2.1 Simply supported thin square plate model

A dynamic system consisting of a thin square cantilever plate is shown in Figure 7.2.1. Four types of time-
variation of the load are selected; (1) harmonic loads containing single frequency (2) periodic forcing function
composed of two harmonic loadings (3) transient step load and (4) random forcing with uniform power
spectral density (PSD). All three types of loads are uniformly distributed spatially. All four types of loads are
applied in the transverse direction of the beam and are uniformly distributed spatially. For the first 3 load types,
the peak displacement and stress are obtained either in the frequency domain or the time domain. For the
random forcing, peak displacement and stress PSD are obtained in the frequency domain. The results are
obtained at center point using both modal superposition and direct methods. The responses obtained using
various elements corresponding to each load type are summarized in Tables 7.2.1~7.2.4. The results from the
NAFEMS benchmark problems are taken as reference.


5
midas NFX Benchmark Series 7-5
Material data
Youngs modulus
Poissons ratio
Density
Mass proportional damping
Stiffness proportional damping
Modal damping
E = 200 GPa
v = 0.3
p = 8000 kg/m
3
o = 0.299 sec
-1

| = 1.339 10
-3
sec
c = 0.02
Section property Thickness t = 0.05 m
Forcing functions
(1) Harmonic
0
sin 2 F F t f e e t = =
(2) Periodic
0
(sin sin3 ) 2 1.2 F F t t f f Hz e e e t = = =
(3) Transient
0
0 F F t = >
(4) Random PSD=10
4
Pa
2
/Hz
*
Rayleigh damping coefficients, and are chosen to give 0.02 damping in the dominant first mode.

Table 7.2.1 Peak responses and frequency of thin plate subjected to harmonic loads

Table 7.2.2 Peak responses of thin plate subjected to periodic forcing function


Peak
A
Z
u [mm] Peak
A
o [MPa] Peak frequency [Hz]
Reference 45.42 30.03 2.377
Element
type
Number of
elements
Direct Modal Direct Modal Direct Modal
QUAD-4 88 45.12 45.15 31.68 31.70 2.415 2.415
TRIA-3 128 43.59 43.63 30.38 30.41 2.455 2.455
QUAD-8 44 45.36 45.38 35.05 35.07 2.385 2.385
TRIA-6 32 47.08 47.11 34.93 34.95 2.340 2.340
Peak
A
Z
u [mm] Peak
A
o [MPa]
Reference 2.863 2.018
Element
type
Number of
elements
Direct Modal Direct Modal
QUAD-4 88 2.913 2.914 2.073 2.075
TRIA-3 128 2.883 2.884 2.037 2.039
QUAD-8 44 2.884 2.885 2.275 2.278
TRIA-6 32 2.930 2.932 2.214 2.217

Linear dynamic analysis
7-6
Table 7.2.3 Peak responses of thin plate subjected to transient step load

Table 7.2.4 Peak responses and frequency of thin plate subjected to subjected to white noise power spectral density



Peak

A
Z
u [mm] Peak Time [sec] Peak
A
o [MPa] Static
A
Z
u [mm]
Reference 3.523 0.210 2.484 1.817
Element
type
Number of
elements
Direct Modal Direct Modal Direct Modal Direct Modal
QUAD-4 88 3.474 3.487 0.210 0.212 2.382 2.435 1.770 1.770
TRIA-3 128 3.355 3.368 0.206 0.206 2.282 2.325 1.709 1.709
QUAD-8 44 3.472 3.472 0.214 0.216 2.563 2.628 1.774 1.774
TRIA-6 32 3.598 3.594 0.218 0.220 2.548 2.597 1.842 1.842

Peak
A
Z
u

PSD
[mm
2
/Hz]
Peak
A
o

PSD
[(N/mm
2
)
2
/Hz]
Frequency [Hz]
Reference 2063.2 1025.44 2.377
Element
type
Number of
elements
Direct Modal Direct Modal Direct Modal
QUAD-4 88 1946.5 2040.6 959.57 1005.9 2.415 2.414
QUADL-4 88 1946.5 2040.6 959.57 1005.9 2.415 2.414
HEXA-8 88 1946.3 2039.4 921.04 965.09 2.415 2.414
HEXAL-8 88 1946.3 2039.4 921.04 965.09 2.415 2.414


7
midas NFX Benchmark Series 7-7
7.3 Simply supported thick square plate

REFERENCE NAFEMS [7-1]
KEYWORDS shell elements, solid elements
MODEL FILENAME LinearDynamic03_direct.nfx
LinearDynamic03_modal.nfx
LinearDynamic03_direct_random.nfx
LinearDynamic03_modal_random.nfx









z
Y
X
10.0
1
0
.
0
x = y = Rz = 0 at all nodes
Rx = Ry = 0 along 4
edges
Units: m
F
0
=10
6
N/m
2
A

Figure 7.3.1 Simply supported thick square plate model

A dynamic system consisting of a thick square cantilever plate is shown in Figure 7.3.1. Four types of time-
variation of the load are selected; (1) harmonic loads containing single frequency (2) periodic forcing function
composed of two harmonic loadings (3) transient step load and (4) random forcing with uniform power
spectral density (PSD). All three types of loads are uniformly distributed spatially. All four types of loads are
applied in the transverse direction of the beam and are uniformly distributed spatially. For the first 3 load types,
the peak displacement and stress are obtained either in the frequency domain or the time domain. For the
random forcing, peak displacement and stress PSD are obtained in the frequency domain. The results are
obtained at center point using both modal superposition and direct methods. The responses obtained using
various elements corresponding to each load type are summarized in Tables 7.3.1~7.3.4. The results from the
NAFEMS benchmark problems are taken as reference.

Linear dynamic analysis
7-8
Material data
Youngs modulus
Poissons ratio
Density
Mass proportional damping
Stiffness proportional damping
Modal damping
E = 200 GPa
v = 0.3
p = 8000 kg/m
3
o = 5.772 sec
-1

| = 6.929 10
-5
sec
c = 0.02
Section property Thickness t = 1 m
Forcing functions
(1) Harmonic
0
sin 2 F F t f e e t = =
(2) Periodic
0
(sin sin3 ) 2 20 F F t t f f Hz e e e t = = =
(3) Transient
0
0 F F t = >
(4) Random PSD=10
12
Pa
2
/Hz
*
Rayleigh damping coefficients, and are chosen to give 0.02 damping in the dominant first mode.

Table 7.3.1 Peak responses and frequency of thick plate subjected to harmonic loads


Peak
A
Z
u [mm] Peak
A
o [MPa] Peak Frequency [Hz]
Reference 58.33 800.8 45.90
Element type
Number of
elements
Direct Modal Direct Modal Direct Modal
QUAD-4 88 60.00 60.00 809.7 809.7 46.50 46.50
TRIA-3 128 57.77 57.80 773.7 774.1 47.35 47.35
QUAD-8 44 62.74 62.75 878.2 878.4 45.00 45.00
TRIA-6 32 62.76 62.77 892.4 892.6 45.05 45.05
HEXA-8 96 59.15 59.16 777.2 777.3 46.80 46.80
HEXA-20 48 58.57 58.57 873.1 873.1 46.10 46.10
PENTA-15 64 56.77 56.78 759.4 759.6 47.00 47.00


9
midas NFX Benchmark Series 7-9
Table 7.3.2 Peak responses of thick plate subjected to periodic forcing function


Table 7.3.3 Peak responses of thick plate subjected to transient step load




Peak
A
Z
u [mm] Peak
A
o [MPa]
Reference 4.929 67.67
Element type
Number of
elements
Direct Modal Direct Modal
QUAD-4 88 5.300 5.303 72.53 72.59
TRIA-3 128 5.428 5.431 73.54 73.60
QUAD-8 44 5.061 5.063 72.24 72.30
TRIA-6 32 5.067 5.069 73.67 73.73
HEXA-8 96 5.341 5.344 70.96 71.01
HEXA-20 48 5.053 5.056 76.64 76.71
PENTA-15 64 5.213 5.216 70.79 70.85
Peak

A
Z
u [mm] Peak Time [sec] Peak
A
o

[MPa] Static
A
Z
u [mm]
Reference 4.524 0.0108 62.11 2.333
Element
type
Number of
elements
Direct Modal Direct Modal Direct Modal Direct Modal
QUAD-4 88 4.565 4.545 0.0107 0.0105 59.03 59.58 2.339 2.339
TRIA-3 128 4.398 4.380 0.0106 0.0105 56.11 56.07 2.253 2.253
QUAD-8 44 4.827 4.845 0.0110 0.0109 66.84 67.99 2.422 2.422
TRIA-6 32 4.807 4.819 0.0109 0.0107 66.75 68.36 2.434 2.434
HEXA-8 96 4.505 4.489 0.0108 0.0109 56.66 57.05 2.310 2.310
HEXA-20 48 4.467 4.458 0.0107 0.0106 64.25 65.38 2.283 2.283
PENTA-15 64 4.323 4.307 0.0106 0.0104 54.98 55.68 2.214 2.214

Linear dynamic analysis
7-10
Table 7.3.4 Peak responses and frequency of thick plate subjected to subjected to white noise power spectral density



Peak
A
Z
u PSD

[mm
2
/Hz]
Peak
A
o PSD

[(N/mm
2
)
2
/Hz]
Frequency [Hz]
Reference 3401.81 641200 45.90
Element type
Number of
elements
Direct Modal Direct Modal Direct Modal
QUAD-4 88 3600.01 3614.12 655540 655815 46.50 46.49
QUADL-4 88 3600.01 3614.12 655540 655815 46.50 46.49
HEXA-8 88 3498.51 3498.09 604028 604040 46.80 46.82
HEXAL-8 88 3498.51 3498.09 604028 604040 46.80 46.82


11
midas NFX Benchmark Series 7-11
7.4 Laminated strip under three-point bending

REFERENCE NAFEMS [7-2]
KEYWORDS layered shell elements, layered solid elements
MODEL FILENAME LinearDynamic04.nfx






Figure 7.4.1 Laminated strip model

Material data
Youngs modulus
Shear modulus
Poissons ratio
Density
Mass proportional damping
Stiffness proportional damping
E
1
=100 GPa

,
E
2
= E
3
=5 GPa
G
12
=3 GPa, G
13
= G
23
=2 GPa
v
12
= v
13
=0.4, v
23
=0.3
p = 0.1 kg/mm
3
o = 8.4 sec
-1

| = 7.0 10
-4
sec


X
Y
0
o
fiber direction 10
X
Z
10 10 15 15
1
10 N/mm
A B
E
C,E
D
0.4
0.1
0.1
0.1
0.1
0.1
0.1
E
D
C
90
o
0
o
90
o
0
o
0
o
90
o
0
o
Units : mm
Figure 7.4.1 shows laminated strip with [0/90/0/90/0/90/0] layup under three-point bending. Linear transient
analyses with both modal and direct methods are carried out using layered shell and solid elements to obtain
the steady-state response of the laminate. During the analysis, the inertial effects are damped out to obtain
steady state responses. The resulting vertical displacement and the inplane bending stress are obtained at point
E. Interlaminar shear stress is recovered at point D for layered shell elements.

Linear dynamic analysis
7-12
Table 7.4.1 Bending stress
11
o at E, interlaminar shear stress
31
o

at D and Z deflection
Z
u

at E
*
obtained using layered quadrilateral elements with reduced integration


E
11
o

[MPa]
D
31
o

[MPa]
E
Z
u

[mm]
Reference 683.9 -4.1 -1.06
Element type
Number of
elements
Direct Modal Direct Modal Direct Modal
QUADL-4 102 679.3 679.3 -4.07 -4.07 -1.06 -1.06
TRIAL-3 10(22) 683.6 654.9 -4.05 -4.05 -1.06 -1.06
QUADL-8 10(22)
683.6
682.1
*

683.6
682.1
*

-4.08
-4.05
*

-4.08
-4.05
*

-1.09
-1.06
*

-1.09
-1.06
*

TRIAL-6 10(22) 703.8 703.8 -4.02 -4.02 -1.06 -1.06
HEXAL-8 102 625.1 625.1 -4.96 -4.96 -1.05 -1.05
PENTAL-6 102 626.0 626.0 -4.93 -4.93 -1.05 -1.05
HEXAL-20 102 680.4 680.4 -4.92 -4.92 -1.05 -1.05
PENTAL-15 10(22) 683.1 683.1 -4.91 -4.91 -1.05 -1.05


13
midas NFX Benchmark Series 7-13
7.5 Damping options in transient analyses

REFERENCE Thomson [7-3]
KEYWORDS bar elements, shell elements, solid elements
MODEL FILENAME LinearDynamic05.nfx










Figure 7.5.1 Massless rod with various damping options

Material data
Youngs modulus
Concentrated mass
E = 69.0 10
9
N/m
2

m = 277.700 10
3
kg
Section property Rectangular cross-section 0.02m 0.0325m





m
A: Cross-section area
E: Youngs modulus
0.254
A
t = 0.02
Beam
Cross-section
d = 0.03225
Units: m
Figure 7.5.1 shows a massless rod constrained to move in the horizontal direction only. One end of the model
is constrained and a concentrated mass is attached on the other end. Effects of various damping options
including damper, modal damping and Rayleigh damping are investigated. Logarithmic decrement over n
cycles given by:
1 ( )
ln
( )
u t
n u t nT
o
| |
=
|
+
\ .

is evaluated using linear transient analyses employing both direct time integration method and modal
superposition method. Closed form solution is taken for reference. Bar, plane and solid elements are used to
model the problem. Transient responses are identical with different elements.



Linear dynamic analysis
7-14
Table 7.5.1 Logarithmic decrement with various damping options in transient analyses with direct time integration

Table 7.5.2 Logarithmic decrement with various damping options in transient analyses with modal superposition method



Damping Options
Reference Direct Time Integration
Damper
Value [kg/s]
Coefficient of Stiffness Proportional
Damping [sec]
0 0.00159248 0.126 0.127
279.011 0 0.126 0.127
279.011 0.00159248 0.252 0.251
Damping Options
Reference Modal Superposition
Modal
Damping
Coefficient of Stiffness Proportional
Damping [sec]
0 0.00159248 0.126 0.125
0.02 0 0.126 0.125
0.02 0.00159248 0.252 0.251


15
midas NFX Benchmark Series 7-15
7.6 Response spectrum of a simply supported beam

REFERENCE Biggs [7-4]
KEYWORDS bar elements, shell elements, solid elements, layered solid elements
MODEL FILENAME LinearDynamic06.nfx





A B X
Z
Y
6096
Units : mm
x=y=z=Rx=0 at A, y=z=0 at B
37.0
355.6
Y
Z

Figure 7.6.1 Simply-supported beam (10 elements)

Material data
Youngs modulus
Density
E = 206.8 GPa
p = 1.0473 10
5
kg/m
3

Section property Rectangular cross-section 37.0 mm 355.6 mm


Figure 7.6.1 shows a simply-supported beam problem with rectangular cross section. Response spectrum
defined in Table 7.6.1 is applied in the vertical direction of the beam at its ends. Response spectrum analyses
are carried out with various types of elements. Damping is assumed to be zero and lumped mass is utilized.
Maximum displacement, bending moment and stress at mid-span are obtained using square root of the sum of
squares (SRSS) summation method.

Linear dynamic analysis
7-16
Table 7.6.1 Response spectra definition(unit : m)

Table 7.6.2 Response spectrum analysis results obtained using bar elements

Table 7.6.3 Response spectrum analysis results obtained using shell elements


Frequency [Hz] 5.000 6.000 6.098 7.000 8.000
Period [sec] 0.2000 0.1667 0.1640 0.1429 0.1250
Type
(scale factor)
Acceleration(0.5) 39.258 32.716 32.190 28.042 28.521
Velocity(1.0) 0.6248 0.4339 0.4201 0.3188 0.2837
Displacement(1.0) 0.0199 0.0115 0.0110 0.0072 0.0056
Result at mid-span
Displacement
[mm]
Stress
[MPa]
Moment
10
5
[Nm]
Reference 14.2 140.4 1.095
Element type Spectra type
BAR-2
Displacement 14.2 138.4 1.079
Velocity 14.1 138.1 1.077
Acceleration 14.1 138.1 1.077
Result at mid-span Displacement [mm] Stress [MPa]
Reference 14.2 140.4
Element type Spectra type
QUAD-4
Displacement 13.8 132.4
Velocity 13.7 132.1
Acceleration 13.7 132.0
TRIA-3
Displacement 13.9 134.6
Velocity 13.9 134.2
Acceleration 13.9 134.2
QUAD-8
Displacement 14.2 138.4
Velocity 14.1 138.1
Acceleration 14.1 138.1
TRIA-6
Displacement 14.2 138.8
Velocity 14.2 138.6
Acceleration 14.2 138.6


17
midas NFX Benchmark Series 7-17
Table 7.6.4 Response spectrum analysis results obtained using solid elements

Table 7.6.5 Response spectrum analysis results obtained using layered solid elements


Result at mid-span Displacement [mm] Stress [MPa]
Reference 14.2 140.4
Element type Spectra type
HEXA-8
Displacement 13.7 132.3
Velocity 13.7 132.0
Acceleration 13.7 131.9
HEXA-20
Displacement 14.1 138.4
Velocity 14.1 138.1
Acceleration 14.1 138.0
PENTA-15
Displacement 14.1 137.6
Velocity 14.1 137.3
Acceleration 14.1 137.3
Result at mid-span Displacement [mm] Stress [MPa]
Reference 14.2 140.4
Element type Spectra type
HEXAL-8
Displacement 13.7 132.3
Velocity 13.7 132.0
Acceleration 13.7 131.9
PENTAL-6
Displacement 13.9 133.5
Velocity 13.8 133.1
Acceleration 13.8 133.1
HEXAL-20
Displacement 14.1 137.5
Velocity 14.1 137.2
Acceleration 14.1 137.2
PENTAL-15
Displacement 14.1 137.6
Velocity 14.1 137.3
Acceleration 14.1 137.3

Linear dynamic analysis
7-18
7.7 Linear dynamic analyses of a rod

REFERENCE Chopra [7-5]
KEYWORDS rod elements, membrane elements, solid elements, layered solid elements
MODEL FILENAME LinearDynamic07.nfx





A D
30
Units : m
X=y=z=Rx=0 at A, y=z=0 at B/C/D
B C
g
x
F
X
Z
Y

Figure 7.7.1 One-dimensional rod model

Material data
Youngs modulus
Density
E = 5 Pa
p = 1/90 kg/m
3

Section property Cross-section Area A = 2.0 m
2

Analysis condition
Modal transient with tip load
Modal transient with base acceleration
Modal frequency with tip load
Response spectrum
F = 10 N, 10% damping
x
g
=1.0 m/sec
2
, 10% damping

F = 5 N, 10% damping
Displacement spectra, , 2% damping

2.0
P
s
e
u
d
o

D
i
s
p
l
a
c
e
m
e
n
t

(
m
)
Frequency
1.0
0.0 . =
0.1 . =
0.01

Figure 7.7.2 displacement response spectra
One-dimensional rod model is discretized with three truss elements as depicted in Figure 7.7.1. The problem is
modeled with not only rod but membrane, solid and layered solid elements also. Responses of the rod model
are evaluated using modal methods including free vibration, transient, frequency and response spectrum.
Comparison is made with closed form solutions.


19
midas NFX Benchmark Series 7-19
Table 7.7.1 Normal mode results

Table 7.7.2 Displacement and acceleration at point D using modal transient analysis with tip load

Table 7.7.3 Total displacement at point D using modal transient analysis with base acceleration


Mode Number 1 2 3
Eigenvalue 1.2058 9.0000 16.794
Frequency [Hz] 0.1748 0.4775 0.6522
Eigenvector
A 0.000 0.000 0.000
B 0.866 1.732 0.866
C 1.500 0.000 -1.500
D 1.732 -1.732 1.732
Participation factor 0.718 0.192 0.0516
Generalized mass 1.000 1.000 1.000
Effective mass 0.5158 0.0370 0.00266
Result type Displacement [m] Acceleration [m/sec
2
]
Time step [sec] 0.1 0.2 0.3 0.1 0.2 0.3
Reference 0.4387 1.686 3.598 81.42 66.71 48.06
Element type
ROD-2 0.4387 1.686 3.598 81.42 66.71 48.06
QUAD-4 0.4387 1.686 3.598 81.42 66.71 48.06
HEXA-8 0.4387 1.686 3.598 81.44 66.71 48.06
HEXAL-8 0.4387 1.686 3.598 81.44 66.71 48.06
Result type Displacement [m]
Time step [sec] 0.1 0.2 0.3
Reference 0.244 10
-4
1.965 10
-4
6.692 10
-4

Element type
ROD-2 0.244 10
-4
1.965 10
-4
6.692 10
-4

QUAD-4 0.244 10
-4
1.965 10
-4
6.692 10
-4

HEXA-8 0.244 10
-4
1.965 10
-4
6.692 10
-4

HEXAL-8 0.244 10
-4
1.965 10
-4
6.692 10
-4


Linear dynamic analysis
7-20
Table 7.7.4 Stress of element 1 and reaction force at point A using modal frequency analysis with tip load

Table 7.7.5 Peak displacement at point D using response spectrum analysis with 2% modal damping ratio



Result type Stress [Pa] Reaction force [N]
Forcing frequency [Hz] 0.01 0.175 0.477 0.01 0.175 0.477
Reference 2.51 15.50 3.988 5.019 31.00 7.977
Element type
ROD-2 2.51 15.49 3.993 5.019 30.97 7.987
QUAD-4 2.51 15.49 3.993 5.019 30.97 7.987
HEXA-8 2.51 15.49 3.993 5.019 30.97 7.987
HEXAL-8 2.51 15.49 3.993 5.019 30.97 7.987
Result type Displacement [m]
Combination method ABS SRSS TENP NRL CQC
Reference 2.902 2.248 2.248 2.767 2.246
Element type
ROD-2 2.902 2.248 2.248 2.767 2.246
QUAD-4 2.902 2.248 2.248 2.767 2.246
HEXA-8 2.902 2.248 2.248 2.767 2.246
HEXAL-8 2.902 2.248 2.248 2.767 2.246


21
midas NFX Benchmark Series 7-21
7.8 Double cantilever under multiple base excitations

REFERENCE Kiureghian et al. [7-6]
KEYWORDS bar elements, shell elements, solid elements
MODEL FILENAME LinearDynamic08.nfx









Figure 7.8.1 Double cantilever subjected to base excitations


Material data
Youngs modulus
Density
E = 3.0 10
7
psi
p = 7.8 10
4
lbf s
2
/in
4

Section property Rectangular cross-section 2 in 1 in

( )
g
u t t ( )
g
u t
x
y
t = 2
Beam
Cross-section
d = 1
Units: in
236.22
Figure 7.8.1 shows a double cantilever beam model subjected to excitation by base motions. The ground
motion is applied on both ends in the form of prescribed acceleration as provided in Figure 7.8.2. On the left
end, the acceleration is applied with a time shift of 0.25 seconds. Various types of finite elements including
beam, shell and solid elements are employed. Linear transient response of the vertical deflection at the mid-
point of the beam model is obtained by direct time integration and by using modal superposition technique.
Figure 7.8.3 compares the results obtained by the two methods using 20 beam elements. Numerical damping
effect included in the direct time integration method accounts for the difference between the responses
obtained using direct integration method and mode superposition method. Minimum deflection over the 2
second period is summarized in Table 7.8.1 for different elements and methods used in the analyses. Difference
in minimum vertical deflection at mid-point is negligibly small.

Linear dynamic analysis
7-22

Figure 7.8.2 Acceleration record of base motion


Figure 7.8.3 Vertical displacement at mid-point of beam obtained using bar elements

-100
-50
0
50
100
150
0 0.5 1 1.5 2
B
a
s
e

A
c
c
e
l
e
r
a
t
i
o
n

[
i
n
/
s
e
c
2
]
Time [sec]
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0 0.5 1 1.5 2
V
e
r
t
i
c
a
l

D
i
s
p
l
a
c
e
m
e
n
t

[
i
n
]
Time [sec]
BAR (Direct)
BAR (Modal)


23
midas NFX Benchmark Series 7-23
Table 7.8.1 Minimum vertical displacement at mid-point of the beam(unit: in)


Method Direct transient Modal superposition
Bar Element
Type
BAR-2 -0.527 -0.528
Shell Element
Type
QUAD-4 -0.522 -0.530
TRIA-3 -0.522 -0.530
QUAD-8 -0.527 -0.528
TRIA-6 -0.525 -0.530
Solid Element
Type
HEXA-8 -0.522 -0.530
HEXA-20 -0.527 -0.528
PENTA-15 -0.526 -0.530
PYRAM-13 -0.525 -0.530
TETRA-10 -0.524 -0.530

Linear dynamic analysis
7-24
7.9 Cantilever subjected to earthquake motion

REFERENCE Hilber et al. [7-7] and Hurty et al. [7-8]
KEYWORDS bar elements
MODEL FILENAME LinearDynamic09.nfx








7620
Units : mm
25.4
50.8
g
x
X
Y
X
Z
Y

Figure 7.9.1 vertical cantilever beam

Material data
Youngs modulus
Density
E = 206.8 GPa
p = 7780 kg/m
3

Section property Rectangular cross-section 50.8mm 25.4mm


Figure 7.9.1 shows a vertical cantilever beam model. Seismic analyses are carried out in which the anchor
point of the beam model is subjected to prescribed motion. The time history of acceleration of the El Centro N-
S as shown in Figure 7.9.2 is applied in the horizontal direction. Transient response of the structure is obtained
using direct and modal superposition methods. Maximum displacement and velocity at top location is
summarized in Table 7.9.1. Response spectrum analyses are also conducted based on the response spectra
shown in Figures 7.9.3 and 7.9.4 using the ABS and SRSS combination methods. Also the effects of using
baseline correction are investigated. Baseline correction method adjusts the acceleration record to inhibit the
structure from drifting away. Figures 7.9.5 and 7.9.6 show the tip and base displacement with and without
baseline correction.


25
midas NFX Benchmark Series 7-25

Figure 7.9.2 El Centro N-S acceleration history



Figure 7.9.3 Displacement spectra for the period range 0.03-10 sec



-0.3
-0.2
-0.1
0.0
0.1
0.2
0.3
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
N
o
r
m
a
l
i
z
e
d

a
c
c
e
l
e
r
a
t
i
o
n
Time [sec]
1.0E-03
1.0E-02
1.0E-01
1.0E+00
1.0E+01
1.00E-02 1.00E-01 1.00E+00 1.00E+01
D
i
s
p
l
a
c
e
m
e
n
t

[
i
n
]
Period [sec]

Linear dynamic analysis
7-26
Table 7.9.1 Maximum displacement and velocity at the top of the column provided by transient analysis


Figure 7.9.4 Velocity spectra for the period range 0.03-10 sec


1.0E-01
1.0E+00
1.0E+01
1.0E+02
1.00E-02 1.00E-01 1.00E+00 1.00E+01
V
e
l
o
c
i
t
y

[
i
n
/
s
e
c
]
Period [sec]
Result type Displacement [mm] Velocity [m/sec]
Reference 59.2 0.508
Analysis type
Number of
elements

Direct transient
10 58.9 0.439
20 58.9 0.438
50 58.9 0.438
Modal transient
10 59.2 0.512
20 59.1 0.515
50 59.1 0.516


27
midas NFX Benchmark Series 7-27
Table 7.9.2 Maximum displacement and velocity at the top of the column provided by response spectrum analysis


Figure 7.9.5 Absolute displacement of the cantilevers tip with and without baseline correction


-250
-200
-150
-100
-50
0
50
100
150
200
0 1 2 3 4 5 6 7 8 9 10
T
o
t
a
l

d
i
s
p
l
a
c
e
m
e
n
t

[
m
m
]
Time [sec]
corrected
original
Result type Displacement [mm] Velocity [m/sec]
Reference 59.2 0.508
Number of
elements
Spectrum type Combination method
10
Displacement
ABS 67.2 0.639
SRSS 57.0 0.392
Velocity
ABS 70.8 0.640
SRSS 61.0 0.395
20
Displacement
ABS 67.2 0.638
SRSS 57.0 0.392
Velocity
ABS 70.8 0.639
SRSS 61.0 0.395
50
Displacement
ABS 67.2 0.638
SRSS 57.0 0.392
Velocity
ABS 70.8 0.639
SRSS 61.0 0.395

Linear dynamic analysis
7-28

Figure 7.9.6 Base displacement with and without baseline correction



-200
-150
-100
-50
0
50
100
150
0 1 2 3 4 5 6 7 8 9 10
D
i
s
p
l
a
c
e
m
e
n
t

[
m
m
]
Time [sec]
corrected
original


29
midas NFX Benchmark Series 7-29
7.10 Modal frequency response with residual modes

REFERENCE Dickens et al. [7-9]
KEYWORDS discrete elements
MODEL FILENAME LinearDynamic10.nfx






Figure 7.10.1 Spring-mass system

Material data
Lumped mass
Spring stiffness
Modal damping
m = 1.0 kg
k = 10000 N/m
c = 0.02

Figure 7.10.2 Displacement amplitude response for DOF 3
1 2 3 4
k k k
m m m 0.5m
F
k k
u
3
u
4
u
1
u
2
1.E-06
1.E-05
1.E-04
1.E-03
1.E-02
3 45
D
i
s
p
l
a
c
e
m
e
n
t

r
e
s
p
o
n
s
e

Frequency [Hz]
All modes
with residual modes
without residual modes
In modal superposition methods, midas NFX incorporates residual modes to enhance the accuracy of the
responses otherwise limited by the number of eigenmodes used. Figure 7.10.1 shows a four DOF spring-mass
system with an excitation force at DOF 3 to investigate the effect of adopting residual modes in modal
frequency response. For the analysis single eigenmode is used with and without additional residual modes to
obtain the frequency response of the system. Comparison is made with the frequency response obtained with
all four eigenmodes used as base vectors. The modal damping is 2% for all modes and residual vectors.

Linear dynamic analysis
7-30

Figure 7.10.3 Acceleration amplitude response for DOF 1

Table 7.10.1 Displacement and percentage error at 3 Hz

Table 7.10.2 Acceleration and percentage error at 3 Hz


1.E-02
1.E-01
1.E+00
1.E+01
1.E+02
3 45
A
c
c
e
l
e
r
a
t
i
o
n

r
e
s
p
o
n
s
e
Frequency [Hz]
All modes
with residual modes
without residual modes
DOF 1 DOF 2 DOF 3 DOF 4
Reference 4.52E-05 8.89E-05 1.29E-04 6.53E-05
All Modes
% error
4.52E-05
0.0%
8.89E-05
0.0%
1.29E-04
0.0%
6.53E-05
0.0%
with residual modes
% error
4.54E-05
0.3%
8.89E-05
0.0%
1.29E-04
0.0%
6.51E-05
-0.3%
without residual modes
% error
6.60E-05
31.5%
1.05E-04
15.5%
1.01E-04
-27.5%
5.65E-05
-15.5%
DOF 1 DOF 2 DOF 3 DOF 4
Reference 1.61E-02 3.16E-02 4.60E-02 2.32E-02
All Modes
% error
1.61E-02
0.0%
3.16E-02
0.0%
4.60E-02
0.0%
2.32E-02
0.0%
with residual modes
% error
1.61E-02
0.3%
3.16E-02
0.0%
4.60E-02
0.0%
2.31E-02
-0.3%
without residual modes
% error
2.35E-02
31.5%
3.74E-02
15.5%
3.61E-02
-27.5%
2.01E-02
-15.5%


31
midas NFX Benchmark Series 7-31
7.11 Steady state dynamics for two-dimensional elements

REFERENCE Thomson [7-3]
KEYWORDS membrane elements
MODEL FILENAME LinearDynamic11.nfx







Figure 7.11.1 2D steady state dynamics model

Material data
Youngs modulus
Poissons ratio
Density
Mass proportional damping
Stiffness proportional damping
E = 2.0 10
7
Pa
v = 0.0
p = 8000 kg/m
3
o = 5.36 sec
-1
| = 7.46 10
-5
sec




L = 1.0 m
A B
C D
L = 1.0 m
A B
D C
F = 30000 N/m
Figure 7.11.1 shows a two-dimensional square model with constraint applied on edge AD and harmonic line
force on edge BC. Harmonic response analyses are performed to present the consistency between direct and
modal methods. The first peak displacement and normal stress are determined for a various two-dimensional
element types. Coupled mass is applied for high order elements and lumped mass is used for low order
elements. Rayleigh damping is considered.

Linear dynamic analysis
7-32
Table 7.11.1 Peak displacements and stress at resonant frequency

*
Reference values are obtained using modal frequency analyses employing low order hybrid quadrilateral elements.





Peak displacement
[mm]

Peak stress
[N/mm
2
]

Frequency
[Hz]
Reference 16.94 0.478 12.16
Element type
Number of
elements
Direct Modal Direct Modal Direct Modal
QUAD-4 22 16.94 16.94 0.478 0.478 12.16 12.16
TRIA-3 2(22) 17.56 17.56 0.476 0.467 12.07 12.07
QUAD-8 22 16.54 16.45 0.518 0.541 12.32 12.48
TRIA-6 2(22) 17.26 16.46 0.518 0.545 12.44 12.48


33
midas NFX Benchmark Series 7-33
7.12 Cantilever with a tip mass

REFERENCE Paz [7-10]
KEYWORDS bar elements
MODEL FILENAME LinearDynamic12.nfx







Figure 7.12.1 A tower model subjected to sinusoidal force


Model data
Effective spring
constant
Concentrated mass
k=100,000 lbf/in

m=100 lb sec
2
/in
Equivalent material data
Youngs Modulus
Poissons ratio
E=2.0 10
6
psi
v=0.3
Section property Square cross-section A = 1 in
2






m
K
5
( ) sin
10
o
o
F t F t
F lbf
e =
=
( ) u t
Units: in
L =100 in
A tower system as shown in Figure 7.12.1 with effective spring constant of k and concentrated mass m is
modeled using bar elements with equivalent properties. The tower is subjected to sinusoidal force applied at its
top. Transient response of the system is obtained using linear transient analyses employing both direct and
modal methods. Displacement, velocity and acceleration at t=0.1, 0.2 and 0.3 seconds are compared with the
closed form exact solution. Damping is not taken into consideration.

Linear dynamic analysis
7-34
Table 7.12.1 Horizontal displacement at t=0.1, 0.2, and 0.3seconds

Table 7.12.2 Horizontal velocity at t=0.1, 0.2, and 0.3seconds

Table 7.12.3 Horizontal acceleration at t=0.1, 0.2, and 0.3seconds


Time 0.1 0.2 0.3
Reference 1.6076 -3.1865 4.7420
Method
Direct Transient 1.6079 -3.1843 4.6982
Modal Superposition 1.6082 -3.1848 4.6992
Time 0.1 0.2 0.3
Reference 2.9379 -11.6917 26.0822
Method
Direct Transient 3.3498 -13.3339 29.7521
Modal Superposition 3.3439 -13.3084 29.6941
Time 0.1 0.2 0.3
Reference -1466.51 2907.53 -4298.04
Method
Direct Transient -1447.58 2867.97 -4231.91
Modal Superposition -1449.45 2870.58 -4235.68


35
midas NFX Benchmark Series 7-35
References

[7-1] NAFEMS, Selected Benchmarks for Forced Vibration, Ref . R0016, NAFEMS, Glasgow, 1993
[7-2] NAFEMS, Composite Benchmarks, Ref . R0031, Issue 2, NAFEMS, Glasgow, 2001
[7-3] W.T. Thomson, Theory of Vibration with Application, 4
th
Edition, Prentice-Hall, Englewood Cliffs, N.J.,
1993
[7-4] J.M. Biggs, Introduction to Structural Dynamics, McGraw-Hill, Inc., New York, 1964
[7-5] A.K. Chopra, Dynamics of Structures: Theory and Applications to Earthquake Engineering, Prentice-Hall,
Englewood Cliffs, N.J., 1995
[7-6] A. Der Kiureghian and A. Neuenhofer, Response Spectrum Method for Multi-Support Seismic
Excitations, Earthquake Engineering and Structural Dynamics, Vol. 21, pp. 713-740, 1992
[7-7] H.M. Hilber, T.J.R. Hughes and R.L. Taylor, Improved Numerical Dissipation of Time Integration
Algorithms in Structural Dynamics, Earchquake Engineering and Structural Dynamics, Vol. 5, pp. 283-
292, 1977
[7-8] W.C. Hurty and M.F. Rubinstein, Dynamics of Structures, Prentice-Hall, Englewood Cliffs, N.J., 1964
[7-9] J.M. Dickens, J.M. Nakagawa and M.J. Wittbrodt, A Critique of Mode Acceleration and Modal
Truncation Augmentation Methods for Modal Response Analysis, Computers & Structures, Vol. 62, pp.
985-998, 1997
[7-10] M. Paz, Structural Dynamics: Theory and Computation, 4
th
Edition, Chapman & Hall, International
Thomson Publishing, 1997


1
midas NFX Benchmark Series 8-1
8.1 Eigenvalue of a beam prestressed by axial force

REFERENCE Timoshenko [8-1]
KEYWORDS bar elements, shell elements, solid elements, layered solid elements
MODEL FILENAME PreStress01.nfx






Figure 8.1.1 Prestressed column model

Material data
Youngs modulus
Poissons ratio
E = 1.87 10
11
psi

= 0.0
Section property Circular cross-section R =2.54mm

Table 8.1.1 Natural frequencies in Hz obtained using bar elements


Z
X
127
Units : mm
5.08
4448 N
Mode number 1 2 3
Reference 883.00 2717.1 5709.6
Element type Number of elements
BAR-2 10 882.26 2706.2 5649.8
Figure 8.1.1 shows a straight column model with simply supported boundary conditions. Free vibration
analysis is carried out to determine the natural frequencies of the column prestressed by an axial force. The
column has a circular cross section. Shell and solid models which have equivalent width and thickness are also
used for comparison.

Prestressed analysis
8-2
Table 8.1.2 Natural frequencies in Hz obtained using shell elements

Table 8.1.3 Natural frequencies in Hz obtained using solid elements

Table 8.1.4 Natural frequencies in Hz obtained using layered solid elements




Mode number 1 2 3
Reference 883.00 2717.1 5709.6
Element type Number of elements
TRIA-3 40 883.37 2730.0 5786.3
QUAD-4 10x1 882.77 2729.5 5797.9
TRIA-6 40 883.01 2703.6 5626.0
QUAD-8 10x1 883.18 2706.9 5641.8
Mode number 1 2 3
Reference 883.00 2717.1 5709.6
Element type Number of elements
HEXA-8 10x1x1 882.15 2723.4 5771.7
PENTA-15 40x1 882.94 2712.7 5683.2
HEXA-20 10x1x1 882.26 2724.1 5773.1
Mode number 1 2 3
Reference 883.00 2717.1 5709.6
Element type Number of elements
PENTAL-6 40x1 882.82 2724.9 5764.9
HEXAL-8 10x1x1 882.26 2724.1 5773.1
PENTAL-15 40x1 882.94 2712.7 5683.2
HEXAL-20 10x1x1 883.05 2715.9 5701.7


3
midas NFX Benchmark Series 8-3
8.2 Vibration of a cable under tension

REFERENCE Thomson [8-2]
KEYWORDS rod elements
MODEL FILENAME PreStress02.nfx





Figure 8.2.1 Prestressed cable model

Material data Youngs modulus E = 206.84 GPa
Section property Area A = 1.979 mm
2

Table 8.2.1 Natural frequencies in Hz obtained using rod elements


Z
X
2.54
Units : m
2224 N
Mode number 1 2 3 4
Reference 74.70 149.0 224.0 299.0
Element type Number of elements
ROD-2 13 74.53 148.0 219.3 287.4
Figure 8.2.1 shows a cable modeled using 13 rod elements. Tensile force is first applied to the cable in a static
step. In the next step, free vibration analysis is carried out to determine the natural frequencies of the
prestressed structure.

Prestressed analysis
8-4
8.3 Vibration of a rotating plate

REFERENCE Hibbit [8-3]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME PreStress03.nfx





Figure 8.3.1 Rotating plate model

Material data
Youngs modulus
Poissons ratio
E = 2.17 10
11
psi

= 0.0
Section property Thickness t = 3.0 mm



Units : mm
Y
Z
X
150
328
= 25Hz, 75Hz
Figure 8.3.1 shows a flat plate built into a rigid wheel spinning about its axis. The free vibration analysis is
carried out to determine the natural frequency under the prestressed state by centrifugal force. The root of the
plate is assumed to be clamped. Two different rotating speeds are considered.


5
midas NFX Benchmark Series 8-5
Table 8.3.1 Natural frequencies in Hz obtained using shell elements

Table 8.3.2 Natural frequencies in Hz obtained using solid elements

Table 8.3.3 Natural frequencies in Hz obtained using layered solid elements



Mode number 1
Rotating speed [Hz] 25 75
Reference 41.74 104.27
Element type Number of elements
TRIA-3 24 41.50 103.17
QUAD-4 6x1 41.24 102.47
TRIA-6 24 41.50 103.17
QUAD-8 6x1 41.53 103.28
Mode number 1
Rotating speed [Hz] 25 75
Reference 41.74 104.27
Element type Number of elements
HEXA-8 6x1x1 41.34 102.96
TETRA-10 129 41.82 104.30
PYRAM-13 36 41.72 104.09
PENTA-15 24x1 41.72 104.02
HEXA-20 6x1x1 41.41 103.03
Mode number 1
Rotating speed [Hz] 25 75
Reference 41.74 104.27
Element type Number of elements
PENAL-6 24x1 41.62 103.71
HEXAL-8 6x1x1 41.34 102.96
PENTAL-15 24x1 41.72 104.02
HEXAL-20 6x1x1 41.57 103.76

Prestressed analysis
8-6
References

[8-1] S.P. Timoshenko, Vibration Problems in Engineering, 2
nd
Edition , D. Van Nostrand Company, New York,
1937
[8-2] W.T. Thomson, Theory of Vibration with Applications, 4
th
Edition, Prentica-Hall, Englewood Cliffs, N.J,
1993
[8-3] H.D. Hibbit, Some Follower Forces and Load Stiffness, International Journal for Numerical Methods
in Engineering, Vol. 14, pp. 937-941,1979






1
midas NFX Benchmark Series 9-1
9.1 Snap-through problem for a simple truss element

REFERENCE NAFEMS [9-1]
KEYWORDS rod elements
MODEL FILENAME GeometricNonlinearStatic01.nfx





Figure 9.1.1 Simple truss model for snap-through process

Material data
Youngs modulus
Poissons ratio
E = 500 10
3
N/mm
2
v = 0.5
Section property Cross-sectional area A =100 mm
2


Test a : shallow strut, H = 25 mm
Test b : deep strut, H = 2500 mm
cross-sectional area = 100 mm
2
A
B
L = 2500 mm
P
H

Figure 9.1.1 shows a simple two-node truss element that is loaded vertically to illustrate snap-through
behavior. Point A is pinned to a rigid surface and point B is free to slide vertically Geometric nonlinear
analyses are performed for two configurations; shallow and deep struts. The arc-length scheme is adopted to
trace unstable equilibrium paths.

Geometric nonlinearity
9-2

Figure 9.1.2 Load vs. vertical displacement (Shallow strut)


Figure 9.1.3 Load vs. vertical displacement (Deep strut)



-15
-10
-5
0
5
10
15
-60 -50 -40 -30 -20 -10 0
L
o
a
d

(
N
)
Displacement (mm)
midas NFX
Reference
-60
-40
-20
0
20
40
60
80
-6000 -5000 -4000 -3000 -2000 -1000 0
L
o
a
d

(
N
)
x

1
0
0
0
0
0
Displacement (mm)
midas NFX
Reference


3
midas NFX Benchmark Series 9-3
Table 9.1.1 Load control solution (Shallow truss)
Reference midas NFX
Load [N]
B
v [mm]
B
v [mm]
1.9 -1.010589 -1.012491
3.8 -2.176156 -2.178650
5.7 -3.584407 -3.589748
7.6 -5.457451 -5.506271
9.5 -9.261434 -9.287744
11.4 -54.44518 -54.42430

Table 9.1.2 Load control solution (Deep truss)
Reference midas NFX
Load [N]
B
v [mm]
B
v [mm]
1.00E+06 -148.2087 -142.0112
2.00E+06 -314.4049 -285.8079
3.00E+06 -510.7903 -480.9255
4.00E+06 -774.6389 -694.0182
5.00E+06 -5608.692 -5644.465
6.00E+06 -5715.415 -5765.079



Geometric nonlinearity
9-4
9.2 Bifurcation problem for a simple truss element

REFERENCE NAFEMS [9-1]
KEYWORDS rod elements
MODEL FILENAME GeometricNonlinearStatic02.nfx





Figure 9.2.1 Simple truss model for bifurcation process

Material data
Youngs modulus
Poissons ratio
E = 500 10
3
N/mm
2
v = 0.0
Section property Cross-sectional area A =100 mm
2


H = 25 mm (
o
> 0 ,imperfect buckling)
cross-sectional area = 100 mm
2
A
B
L = 2500 mm
P
K
S
= 1.5 N/mm

o
H
Figure 9.2.1 shows a simple two-node truss element subjected to an axial load at point A while a spring of
stiffness
S
K is attached to the other end at point B. The truss is assumed to have an initial inclination. A
horizontal slider condition is given at point A and a vertical slider condition is given at point B. The arc-length
scheme is adopted to ensure convergence beyond critical load.


5
midas NFX Benchmark Series 9-5

Figure 9.2.1 Load vs. horizontal displacement (Imperfect buckling)

Table 9.2.1 Load control solution (Imperfect buckling)
Reference midas NFX
Load [N]
B
u [mm]
B
u [mm]
0 0.0000 0.0000
760 0.1096 0.1140
1520 0.3046 0.3139
2280 0.8035 0.8738
3040 3.5598 3.5793
3800 5000.3 5001.0



-4000
-3000
-2000
-1000
0
1000
2000
3000
4000
5000
0 1000 2000 3000 4000 5000
L
o
a
d

(
N
)
Displacement (mm)
midas NFX
Reference

Geometric nonlinearity
9-6
9.3 Snap-back problem for a simple truss element

REFERENCE NAFEMS [9-1]
KEYWORDS rod elements
MODEL FILENAME GeometricNonlinearStatic03.nfx





Figure 9.3.1 Simple truss model for snap-back process

Material data
Youngs modulus
Poissons ratio
E = 500 10
3
N/mm
2
v = 0.0
Section property Cross-sectional area A =100 mm
2


A
B
2500 mm
P
K
S1
= 1.5 N/mm
K
S3
= 1.0 N/mm
K
S2
= 0.25 N/mm
H
cross-sectional area = 100 mm
2
C
Figure 9.3.1 shows a simple two-node truss element that is horizontally loaded at point C to illustrate snap-
back behavior. The point load P is imposed indirectly to the pin-joint with horizontal spring via a modifier
spring with stiffness of
2 S
K , which is accountable for the snap-back response. The arc-length scheme is
adopted to ensure convergence.


7
midas NFX Benchmark Series 9-7

(a) Response at point A


(b) Response at point C
Figure 9.3.2 Load vs. horizontal displacement (Snap-back)


-3000
-2000
-1000
0
1000
2000
3000
4000
5000
6000
0 1000 2000 3000 4000 5000
L
o
a
d

(
N
)
Horizontal displacement, u
A
(mm)
midas NFX
Reference
-3000
-2000
-1000
0
1000
2000
3000
4000
5000
6000
0 2000 4000 6000 8000 10000
L
o
a
d

(
N
)
Horizontal displacement, u
C
(mm)
midas NFX
Reference

Geometric nonlinearity
9-8
Table 9.3.1 Load control solution (Snap-back)
Reference midas NFX
Load [N]
A
u [mm]
C
u [mm]
A
u [mm]
C
u [mm]
0 0.0000 0.0000 0.0000 0.0000
700 0.0990 700.00 0.1013 700.10
1400 0.2633 1400.3 0.2645 1400.3
2100 0.6261 2100.6 0.6531 2100.7
2800 1.9706 2802.0 2.3042 2802.3
3500 46.432 3546.4 116.67* 3616.7*
4200 5000.2 9200.2 5000.2 9200.2
*
linear interpolation values around the unstable point




9
midas NFX Benchmark Series 9-9
9.4 Rigid-body rotation

REFERENCE NAFEMS [9-1]
KEYWORDS bar elements, plain stress elements
MODEL FILENAME GeometricNonlinearStatic04A.nfx
GeometricNonlinearStatic04B.nfx







Figure 9.4.1 Rigid-body rotation

Material data
Youngs modulus
Poissons ratio
E = 210 10
9
N/m
2
v = 0.0
Section property Cross-sectional area
d = 0.1 m
t = 0.1 m









A B

L = 3.2
C
D
X
X
t = 0.1
d = 0.1
Section X-X
Units: m
Figure 9.4.1 shows a simple pinned bar model to test the ability of the strain-displacement relationships to
reproduce a strain free situation in which the structure undergoes a rigid body rotation. A set of pin-jointed
conditions are given at line AD and for all points rigid body rotation of 45 is specified about mid-point of line
AD. Stress and strain are evaluated for bar and plane stress elements while the structure undergoes rigid body
rotation.

Geometric nonlinearity
9-10
Table 9.4.1 Near zero stress and strain response obtained during rigid body rotation
BAR QUAD8
Angle Stress [N/m
2
] Strain Stress [N/m
2
] Strain
0 0.00E+00 0.00E+00 0.00E+00 0.00E+00
3 1.78E-07 8.48E-19 1.13E-05 5.40E-17
6 4.84E-07 2.30E-18 -7.76E-08 -3.69E-19
9 2.00E-07 9.52E-19 -4.60E-08 -2.19E-19
12 3.31E-07 1.58E-18 1.89E-06 9.01E-18
15 7.41E-07 3.53E-18 5.76E-07 2.74E-18
18 9.25E-07 4.40E-18 5.36E-05 2.55E-16
21 2.70E-07 1.29E-18 6.72E-08 3.20E-19
24 3.36E-07 1.60E-18 -7.36E-08 -3.51E-19
27 4.67E-07 2.22E-18 5.42E-08 2.58E-19
30 4.88E-07 2.33E-18 5.97E-06 2.84E-17
33 3.58E-07 1.70E-18 -9.20E-08 -4.38E-19
36 4.88E-07 2.33E-18 -2.48E-07 -1.18E-18
39 2.30E-07 1.09E-18 4.89E-06 2.33E-17
42 5.01E-07 2.39E-18 1.90E-05 9.03E-17
45 5.45E-07 2.59E-18 1.26E-05 5.98E-17
*
near-zero values are expected.





11
midas NFX Benchmark Series 9-11
9.5 Straight cantilever

REFERENCE NAFEMS [9-1]
KEYWORDS bar elements, plain stress elements
MODEL FILENAME GeometricNonlinearStatic05A.nfx
GeometricNonlinearStatic05B.nfx






Figure 9.5.1 Straight cantilever beam model

Material data
Youngs modulus
Poissons ratio
E = 210 10
9
N/m
2
v = 0.0
Section property Cross-sectional area
d = 0.1 m
t = 0.1 m



A B
P
3.2 m
C D X
X
t = 0.1 m
d = 0.1 m
section X-X
Q
M
Case A : P = 0, Q = 0, M = 3.4361 x 10
6
Nm
Case B : P = 1.709 x 10
6
N, Q = 0, M = 0
Case C : P = 3.844 x 10
3
N, Q = 3.844 x 10
6
N, M = 0
For a beam: P, Q and M are applied at mid-point of BC
For a continuum: Nodal forces are applied as follows
M: two horizontal forces; M/d (point B) and M/d (point C)
P: three vertical forces; P/6 (point C), 2P/3 (mid-point of BC) and P/6 (point B)
Q: three horizontal forces Q/6 (point C), 2Q/3 (mid-point of BC) and Q/6 (point B)
Figure 9.5.1 shows a straight clamped bar model to test the large rotations and displacements capabilities. One
end of the bar is built-in and the other end of the cantilever is subjected to three load cases; case A, B and C
consisting of combination of longitudinal, transverse loads and end moment (Figure 9.5.1). The cantilever
structure is modeled by 4, 8, 16, 32 bar or plain stress elements.

Geometric nonlinearity
9-12



Figure 9.5.2 Load-displacement curve obtained using bar elements subjected to end moment (case A)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized x-displacement (Ux/L)
CaseA32
CaseA16
CaseA8
CaseA4
Reference
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized y-displacement (Uy/L)
CaseA32
CaseA16
CaseA8
CaseA4
Reference
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized rotation (/(2))
CaseA32
CaseA16
CaseA8
CaseA4
Reference


13
midas NFX Benchmark Series 9-13



Figure 9.5.3 Load-displacement curve obtained using bar elements subjected to transverse load (case B)
0
1
2
3
4
5
6
7
8
9
10
-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized x-displacement (Ux/L)
CaseB32
CaseB16
CaseB8
CaseB4
Reference
0
1
2
3
4
5
6
7
8
9
10
-0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d
l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized y-displacement (Uy/L)
CaseB32
CaseB16
CaseB8
CaseB4
Reference
0
1
2
3
4
5
6
7
8
9
10
-1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized rotation(/(/2))
CaseB32
CaseB16
CaseB8
CaseB4
Reference

Geometric nonlinearity
9-14



Figure 9.5.4 Load-displacement curve obtained using bar elements subjected to both transverse and axial loads (case C)
0
5
10
15
20
25
30
-1.6 -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized x-displacement (Ux/L)
CaseC32
CaseC16
CaseC8
CaseC4
Reference
0
5
10
15
20
25
30
-0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized y-displacement (Uy/L)
CaseC32
CaseC16
CaseC8
CaseC4
Reference
0
5
10
15
20
25
30
-1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized rotation (/)
CaseC32
CaseC16
CaseC8
CaseC4
Reference


15
midas NFX Benchmark Series 9-15



Figure 9.5.5 Load-displacement curve obtained using plane stress elements subjected to end moment (case A)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized x-displacement (Ux/L)
CaseA32
CaseA16
CaseA8
Reference
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized y-displacement (Uy/L)
CaseA32
CaseA16
CaseA8
Reference
0
0.2
0.4
0.6
0.8
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized rotation (/(2))
CaseA32
CaseA16
CaseA8
Reference

Geometric nonlinearity
9-16



Figure 9.5.6 Load-displacement curve obtained using plane stress elements subjected to transverse load (case B)
0
1
2
3
4
5
6
7
8
9
10
-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized x-displacement (Ux/L)
CaseB32
CaseB16
CaseB8
Reference
0
1
2
3
4
5
6
7
8
9
10
-0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d
l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized y-displacement (Uy/L)
CaseB32
CaseB16
CaseB8
Reference
0
1
2
3
4
5
6
7
8
9
10
-1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized rotation(/(/2))
CaseB32
CaseB16
CaseB8
Reference


17
midas NFX Benchmark Series 9-17



Figure 9.5.7 Load-displacement curve obtained using plane stress elements subjected to
both transverse and axial loads (case C)
0
5
10
15
20
25
-1.6 -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized x-displacement (Ux/L)
CaseC32
CaseC16
CaseC8
Reference
0
5
10
15
20
25
-0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized y-displacement (Uy/L)
CaseC32
CaseC16
CaseC8
Reference
0
5
10
15
20
25
-1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
N
o
r
m
a
l
i
z
e
d

l
o
a
d

(
P
L
2
/
(
E
I
)
)
Normalized rotation (/)
CaseC32
CaseC16
CaseC8
Reference

Geometric nonlinearity
9-18
9.6 Curved cantilever

REFERENCE NAFEMS [9-1]
KEYWORDS bar elements, plain stress elements
MODEL FILENAME GeometricNonlinearStatic06A.nfx
GeometricNonlinearStatic06B.nfx





Figure 9.6.1 Curved cantilever beam model

Material data
Youngs modulus
Poissons ratio
E = 210 10
9
N/m
2
v = 0.0
Section property Cross-sectional area
d = 0.1 m
t = 0.1 m


t = 0.1
d = 0.1
section X-X
L = 3.2
M = 3.4361 x 10
6
Nm
X
X
C D
A B
M
Units: m
Figure 9.6.1 shows a curved cantilevered bar to test the large rotations and displacements capabilities. The free
tip is subjected to an end moment to produce unrolling 2D motion. The built-in condition is prescribed at
segment AD. The cantilever structure is modeled by 4, 8, 16, 32 bar or plain stress elements.


19
midas NFX Benchmark Series 9-19



Figure 9.6.2 Load-displacement curve obtained using bar elements at free end
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-0.25 0.00 0.25 0.50 0.75 1.00
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized x-displacement (Ux/L)
Bar 32
Bar 16
Bar 8
Bar 4
Reference
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized y-displacement (Uy/L)
Bar 32
Bar 16
Bar 8
Bar 4
Reference
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized rotation (/(2))
Bar 32
Bar 16
Bar 8
Bar 4
Reference

Geometric nonlinearity
9-20



Figure 9.6.3 Load-displacement curve obtained using plane stress elements at free end
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-0.25 0.00 0.25 0.50 0.75 1.00
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized x-displacement (Ux/L)
Quad 32
Quad 16
Quad 8
Quad 4
Reference
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized y-displacement (Uy/L)
Quad 32
Quad 16
Quad 8
Quad 4
Reference
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
N
o
r
m
a
l
i
z
e
d

m
o
m
e
n
t

(
M
L
/
(
2

E
I
)
)
Normalized rotation (/(2))
Quad 32
Quad 16
Quad 8
Quad 4
Reference


21
midas NFX Benchmark Series 9-21
9.7 Z-shaped 3D cantilever

REFERENCE NAFEMS [9-1]
KEYWORDS bar elements, shell elements
MODEL FILENAME GeometricNonlinearStatic07A.nfx
GeometricNonlinearStatic07B.nfx





Figure 9.7.1 Z-shaped 3D cantilever model

Material data
Youngs modulus
Poissons ratio
E = 200 10
3
N/mm
2
v = 0.3
Section property Cross-sectional area
b = 20 mm
t = 1.7 mm
Y
Z
X
C
D
A B
L
L = 60
L L
W = 20
H = 30
t = 1.7 P = 4000 N
Units: mm
P = 4000
P = 1078
P = 208
P = 100
P = 0
Figure 9.7.1 shows a Z-shaped cantilever laid along the oblique line of 45. Total load P at all points on free
end D in the positive z-direction is conservative (non-follower) load. The cantilever structure is modeled using
bar or quadrilateral shell elements. Tension stiffening and change in sign of bending moment can be observed.

Geometric nonlinearity
9-22

Figure 9.7.2 Solution for Z-shaped 3D cantilever model


0
500
1000
1500
2000
2500
3000
3500
4000
-12000 -9000 -6000 -3000 0 3000 6000 9000 12000
L
o
a
d

(
N
)
Moment at point B (Nmm)
BAR
SHELL4
SHELL8
Reference
0
500
1000
1500
2000
2500
3000
3500
4000
0 20 40 60 80 100 120 140 160
L
o
a
d

(
N
)
Vertical tip displacement (mm)
BAR
SHELL4
SHELL8
Reference


23
midas NFX Benchmark Series 9-23
9.8 Torsional buckling of cantilever

REFERENCE NAFEMS [9-1]
KEYWORDS shell elements
MODEL FILENAME GeometricNonlinearStatic08.nfx





Figure 9.8.1 Cantilever model for torsional buckling

Material data
Youngs modulus
Poissons ratio
E = 10 10
3
N/mm
2
v = 0.0
Section property Cross-sectional area
h =5 mm
t = 0.2 mm


Test a : P = 0.017 N (conservative)
Test b : P = 0.032 N (non-conservative)
Initial imperfection in the out-of-plane y-direction
Z
Y
X
P
H = 5 mm
t = 0.2 mm
L = 100 mm
A
B
C
E
D
Figure 9.8.1 shows a cantilever structure subjected to point load P in the negative z-direction at neutral axis of
the beam, point E. Conservative load with no initial imperfection makes a linear buckling problem whereas
non-conservative load that follows deformation with initial imperfection produces instability. The cantilever
structure is discretized using shell elements. The geometric imperfection is given in Table 9.8.1.

Geometric nonlinearity
9-24

Figure 9.8.2 Solution for torsional buckling of cantilever

Table 9.8.1 Out-of-plane (y-coordinate) imperfections over the length of the cantilever
x-
coordinate
(along
length of
beam)
y-
coordinate
at
z = 0
y-
coordinate
at
z = 0.833
y-
coordinate
at
z = 1.667
y-
coordinate
at
z = 2.5
y-
coordinate
at
z = 3.33
y-
coordinate
at
z = 4.167
y-
coordinate
at
z = 5.0
0 0 0 0 0 0 0 0
10.0 0 0 0 0 0 0 0
20.0 0 0 0 0 0 0 0
30.0 0 0 0 0.001 0.001 0.001 0.001
40.0 0.001 0.001 0.001 0.001 0.001 0.001 0.001
50.0 0.002 0.002 0.002 0.002 0.002 0.002 0.002
60.0 0.003 0.003 0.003 0.003 0.003 0.003 0.004
70.0 0.004 0.005 0.005 0.005 0.005 0.005 0.005
80.0 0.006 0.006 0.006 0.006 0.006 0.006 0.007
90.0 0.008 0.008 0.008 0.008 0.008 0.008 0.008
100.0 0.009 0.010 0.010 0.010 0.010 0.010 0.010


-0.032
-0.028
-0.024
-0.020
-0.016
-0.012
-0.008
-0.004
0.000
-0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04 0.05 0.06
L
o
a
d

(
N
)
y-displacement at point E (mm)
Conservative
Non conservative
Reference


25
midas NFX Benchmark Series 9-25
9.9 Pear-shaped cylinder under end shortening

REFERENCE NAFEMS [9-2]
KEYWORDS shell elements
MODEL FILENAME GeometricNonlinearStatic09.nfx





Figure 9.9.1 Pear-shaped cylinder under end shortening

Material data
Youngs modulus
Poissons ratio
E = 1.0 10
7
N/mm
2
v = 0.3
Section property Thickness t = 0.01 mm



X
Y
Z
A
R = 1.0
R = 1.0
45
o
L = 0.8 thickness = 0.01 Units: mm
Figure 9.9.1 shows a pear-shaped cylinder subjected to a uniformly applied displacement under the elastic
large deformation, i.e. z-directional end shortening. Considering symmetry, only one-quarter of the cylinder is
simulated. The collapse load is determined using various types of shell elements.

Geometric nonlinearity
9-26

Figure 9.9.2 Axial reaction load vs. y-displacement at front bottom node

Table 9.9.1 Collapse load obtained using various types of shell elements
Collapse load [N]
Reference 2437
Element type
QUAD-4 2638
QUAD-8 2516
TRIA-3 2898
TRIA-6 2477


0
500
1000
1500
2000
2500
3000
-0.04 -0.035 -0.03 -0.025 -0.02 -0.015 -0.01 -0.005 0
A
x
i
a
l

l
o
a
d

(
N
)
y-displacement (mm)
midas NFX
Reference


27
midas NFX Benchmark Series 9-27
9.10 Curved elastic cantilever under transverse end load

REFERENCE NAFEMS [9-2]
KEYWORDS bar elements
MODEL FILENAME GeometricNonlinearStatic10.nfx





Figure 9.10.1 Curved elastic cantilever under transverse end load

Material data
Youngs modulus
Poissons ratio
E = 1.0 10
7
N/mm
2
v = 0.0
Section property Cross-sectional area
b = 1.0 mm
h = 1.0 mm

Table 9.10.1 Displacements at end point at various load levels
Reference midas NFX
Load [N]
x
u [mm]
y
u [mm]
z
u [mm]
x
u [mm]
y
u [mm]
z
u [mm]
300 7.100 12.14 40.43 6.908 12.09 40.31
450 10.82 18.70 48.68 10.56 18.67 48.60
600 13.62 23.78 53.59 13.46 23.73 53.56
3000 25.00 47.70 68.56 25.06 47.71 68.69


A
B
P = 3000 (conservative)
R = 100
45
o
Y
X
Z
1.0
cross section
1.0
Built-in
Units: mm
P = 3000
P = 600
P = 450
P = 300
P = 150
P = 0.00
x
y
z
Figure 9.10.1 shows a curved elastic cantilever under transverse end load. This problem is concerned with
three-dimensional large rotation-large displacement with combined twisting, bending and stretching actions.
Conservative load is applied at end point and the nonlinear response is observed.

Geometric nonlinearity
9-28

Figure 9.10.2 Load-tip displacement curves in the direction of global coordinates


-60
-45
-30
-15
0
15
30
45
60
75
90
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
T
i
p

d
i
s
p
l
a
c
e
m
e
n
t
s

(
m
m
)
Load (N)
u_x
u_y
u_z
Reference


29
midas NFX Benchmark Series 9-29
9.11 Buckling of a flat plate subjected to in-plane shear

REFERENCE NAFEMS [9-2]
KEYWORDS shell elements
MODEL FILENAME GeometricNonlinearStatic11.nfx






Figure 9.11.1 Flat plate model subjected to in-plane shear

Material data
Youngs modulus
Poissons ratio
E = 6.4 10
6
N/mm
2
v = 0.3
Section property Thickness t = 12.5 mm

L = 1000
L
thickness t = 12.5
All edges simply supported
Uniform shear load along all edges
Maximum load/edge = 2.12 x 10
8
t
t t
t
Units: mm
Edge load = 1.074 x 10
8
x 3 magnification
Figure 9.11.1 shows a flat plate subjected to in-plane shear. Nonlinear buckling response considering large
rotation and displacement is sought. An initial imperfection in the middle surface of the plate is defined by:
( , )
0.5
( 2, 2)
f x y
z t
f L L
= where ( , ) sin sin
mn
m n
m x n x
f x y a
L L
t t
=

and
11 22 13 31 33 42
1.0, 0.2897, 0.0706, 0.0691, 0.0384 and 0.0032 a a a a a a = = = = = =

Geometric nonlinearity
9-30

Figure 9.11.2 Total edge load vs. central deflection

Table 9.11.1 Central z-deflection
Reference midas NFX
Load [N]
z
u [mm]
z
u [mm]
1.1074 x 10
8
15.02 14.86
2.1117 x 10
8
38.67 38.21
*
interpolated values


0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
0 4 8 12 16 20 24 28 32 36 40
T
o
t
a
l

l
o
a
d

p
e
r

e
d
g
e

(
N
)
x

1
0
8
Central Z-deflection (mm)
midas NFX
Reference


31
midas NFX Benchmark Series 9-31
9.12 Hinged spherical shell under pressure loading

REFERENCE NAFEMS [9-2]
KEYWORDS shell elements
MODEL FILENAME GeometricNonlinearStatic12.nfx







Figure 9.12.1 Hinged spherical shell under pressure loading

Material data
Youngs modulus
Poissons ratio
E = 69 N/mm
2
v = 0.3
Section property Thickness t = 100 mm

Table 9.12.1 Target solution
Reference midas NFX
Pressure [N/mm
2
] Central deflection [mm] Central deflection [mm]
0.0624 75.11 68.62
0.0468 151.5 154.7
0.0299 237.5 234.9
0.1012 303.1 304.1
*
interpolated values

Y
X
Z
L = 1570
thickness = 100
Units: mm
Figure 9.12.1 shows a hinged large spherical shell subjected to a uniform surface follower pressure. This
problem can be characterized by large displacement nonlinear response with snap-through behavior. The shell
mid-surface is defined in terms of global Cartesian coordinates where
4
2.0285 10 [ (1570 ) (1570 )] Z X X Y Y

= + .
The arc-length scheme is adopted to capture the snap-through behavior.

Geometric nonlinearity
9-32

Figure 9.12.2 Applied pressure vs. central deflection


0
0.02
0.04
0.06
0.08
0.1
0.12
-330 -300 -270 -240 -210 -180 -150 -120 -90 -60 -30 0
A
p
p
l
i
e
d

p
r
e
s
s
u
r
e

(
N
/
m
m
2
)
Central defelction (mm)
midas NFX
Reference


33
midas NFX Benchmark Series 9-33
9.13 Pinched hemispherical shell

REFERENCE NAFEMS [9-2]
KEYWORDS shell elements, solid elements, layered solid elements
MODEL FILENAME GeometricNonlinearStatic13.nfx





Figure 9.13.1 Hemispherical shell model and its deformed shape

Material data
Youngs modulus
Poissons ratio
E = 6.825 10
7
psi

v = 0.3
Section property Thickness t = 0.04 in

Table 9.13.1 Displacement
x
u at point A and
y
u at point B obtained using shell elements

18
o
x
P = 100
Symmetry
Free
y
A
R = 10, t= 0.04
Symmetry
Free
z
B
P = 100
Units: in

A
x
u [in]
B
y
u [in]
Reference 5.952 3.427
Number of elements per side 4 8 12 4 8 12
Element
type
TRIA-3 7.10145 6.02279 5.80750 3.39043 3.39043 3.39043
QUAD-4 4.24946 5.54358 5.79628 2.61147 3.22147 3.36121
TRIA-6 0.77068 2.44332 3.87531 0.67140 1.86193 2.58335
QUAD-8 4.41358 5.68836 5.82769 2.64918 3.30901 3.38504
Figure 9.13.1 shows a pinched hemispherical shell with inward and outward forces at point A and point B,
respectively. The hemisphere has 18 hole at the top and the quadrant of the hemisphere is modeled utilizing
symmetric boundary conditions. Solid model has single layer in the thickness direction. Displacement at points
A and B are determined.

Geometric nonlinearity
9-34
Table 9.13.2 Displacement
x
u at point A and
y
u at point B obtained using solid elements

Table 9.13.3 Displacement
x
u at point A and
y
u at point B obtained using layered solid elements
*
obtained using higher order layered hexahedral elements with reduced integration


Figure 9.13.2 Displacement-load curve at loading points obtained using shell elements

0
1
2
3
4
5
6
7
0 10 20 30 40 50 60 70 80 90 100
D
i
s
p
l
a
c
e
m
e
n
t

(
m
)
Load (N)
Point A
Point B
Reference

A
x
u [in]
B
y
u [in]
Reference 5.952 3.427
Number of elements per side 4 8 12 4 8 12
Element
type
PENTA-6 0.00406 0.01514 0.03277 0.00421 0.01558 0.03320
HEXA-8 1.07985 3.80530 4.91809 0.86833 2.40960 2.95427
PENTA-15 0.48059 2.18989 3.73388 0.46399 1.68889 2.49961
HEXA-20 4.60393 5.69317 5.85576 2.86908 3.32026 3.39509

A
x
u [in]
B
y
u [in]
Reference 5.952 3.427
Number of elements per side 4 8 12 4 8 12
Element
type
PENTAL-6 0.37832 2.47292 4.15472 0.34637 1.74139 2.57284
HEXAL-8 4.43582 5.56950 5.80171 2.70573 3.24622 3.37106
PENTAL-15 0.48071 2.18880 3.73223 0.46405 1.68804 2.49907
HEXAL-20* 4.36964 5.67284 5.8339 2.75724 3.29608 3.38722


35
midas NFX Benchmark Series 9-35
9.14 Reticulated space trusses

REFERENCE Ryoji Miike et al. [9-3]
KEYWORDS rod elements
MODEL FILENAME GeometricNonlinearStatic14.nfx





Figure 9.14.1 Reticulated space trusses model

Material data
Youngs modulus
Poissons ratio
E = 3030 N/mm
2
v = 0.0
Section property Cross-sectional area A = 317 mm
2


P
Y
X
Z
cross section area = 317 mm
2
A
Figure 9.14.1 shows a reticulated space truss model loaded downward at point A. The truss structure is
modeled using 24 rod elements and pinned boundary condition is assigned at six end points. The arc-length
scheme is adopted to capture the snap-through behavior.

Geometric nonlinearity
9-36

Figure 9.14.2 Load-vertical displacement curve at point A


-400
-300
-200
-100
0
100
200
300
-50 -45 -40 -35 -30 -25 -20 -15 -10 -5 0
A
p
p
l
i
e
d

l
o
a
d

(
N
)
Vertical displacement (mm)
midas NFX
Reference


37
midas NFX Benchmark Series 9-37
9.15 Hinged panel subjected to a point load

REFERENCE Chang [9-4]
KEYWORDS shell elements
MODEL FILENAME GeometricNonlinearStatic15.nfx





Figure 9.15.1 Hinged panel model subjected to a point load

Material data
Youngs modulus
Poissons ratio
E = 3.10275 10
3
N/mm
2
v = 0.3
Section property Thickness t = 6.35 mm



X
Z
R = 2540
Y
t = 6.35
2L = 508
A
= 0.1 rad
B
P =1000
Units: mm
Hinged
Hinged
Fig. 7.15.1 shows a hinged panel subjected to a point load. This problem is defined in the context of finite
strain analysis and buckling behavior. Due to symmetry of the model, only one-quarter of the shell is analyzed.
The arc-length scheme is adopted to trace the equilibrium path.


Geometric nonlinearity
9-38
Table 9.15.1 Z-displacement at point A and B obtained using shell elements


Figure 9.15.2 Load-deflection curve of the hinged panel


-1000
-900
-800
-700
-600
-500
-400
-300
-200
-100
0
100
200
300
400
500
-33 -30 -27 -24 -21 -18 -15 -12 -9 -6 -3 0 3
L
o
a
d

(
N
)
Deflection (mm)
Point A
Point B
Reference

A
Z
u [mm]
B
Z
u [mm]
Reference -30 -26
Element
type
TRIA-3 -29.3823 -26.1443
QUAD-4 -30.9110 -25.7016
TRIA-6 -30.8191 -26.5135
QUAD-8 -31.2231 -26.5315


39
midas NFX Benchmark Series 9-39
9.16 Circular plate under uniform surface pressure

REFERENCE Timoshenko et al. [9-5]
KEYWORDS axisymmetric elements
MODEL FILENAME GeometricNonlinearStatic16.nfx





Figure 9.16.1 Circular plate model under uniform surface pressure

Material data
Youngs modulus
Poissons ratio
E = 2.0 10
11
N/m
2
v = 0.3
Section property Thickness t = 0.0025 m

Table 9.16.1 Central deflection


X
Y
Z
t = 0.0025
a
X
Y
a = 0.25
1 2 3 4 5 6 7 8 91011
1 2 3 4 5 6 7 8 9 10
p = 6585.175 N/m
2
C
Units: m

Y
u [m]
Reference -0.00125
Element
type
TRIAX-3 -0.00003
QUADX-4 -0.00125
TRIAX-6 -0.00122
QUADX-8 -0.00121
Figure 9.16.1 shows a circular plate built-in around its outer rim under uniform surface pressure. The circular
plate is modeled by various axisymmetric elements. Central deflection of the circular plate is determined and
compared with the reference solution.

Geometric nonlinearity
9-40

Figure 9.16.2 Pressure-central deflection curve



0
1000
2000
3000
4000
5000
6000
7000
-0.0014 -0.0012 -0.001 -0.0008 -0.0006 -0.0004 -0.0002 0
P
r
e
s
s
u
r
e

(
N
/
m
2
)
Central deflection (m)
midas NFX
Reference


41
midas NFX Benchmark Series 9-41
9.17 Lees frame buckling problem

REFERENCE NAFEMS [9-6]
KEYWORDS bar elements
MODEL FILENAME GeometricNonlinearStatic17.nfx







Figure 9.17.1 Frame buckling model

Material data
Youngs modulus
Poissons ratio
E = 71.74 10
9
N/m
2
v = 0.3
Section property Cross-sectional area
t = 0.03 m
d = 0.02 m

t = 0.03
Beam
Cross-Section
d = 0.02
Units: m
P
X
Y
d
B
L = 1.2
d A
C
0.8 L 0.2 L
x
y
z
Figure 9.17.1 shows a right angle frame model. Points B and C are constrained in all translations and X and Y-
rotations. All other nodes are constrained in the Z-translations, X and Y-rotations. A conservative shear force P
is applied at point A in increments up to a maximum value of
2
PL EI =31.887 . The displacement at point of
loading is determined. The arc-length scheme is utilized to capture the snap-back buckling response.

Geometric nonlinearity
9-42

Figure 9.17.2 Load-lateral deflection curve at point A


-20
-10
0
10
20
30
40
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
N
o
r
m
a
l
i
z
e
d

l
o
a
d
Normalized lateral deflection
midas NFX
Reference


43
midas NFX Benchmark Series 9-43
9.18 Pull-out of an open cylindrical shell

REFERENCE Sze et al. [9-7]
KEYWORDS shell elements
MODEL FILENAME GeometricNonlinearStatic18.nfx





Figure 9.18.1 Open cylindrical shell model

Material data
Youngs modulus
Poissons ratio
E = 10.5 10
6
N/mm
2
v = 0.3125
Section property Thickness t = 0.094 mm



P
R = 4.953
L = 10.35
Z
free edge
Y
X
C
B
A
P
free edge
Units: mm
t = 0.094
P = 40000 N
P
Figure 9.18.1 shows an open cylindrical shell pulled by a pair of diametrically opposite point forces. For
obvious symmetry consideration, only one-quarter of the cylinder is simulated. Responses at points A, B and C
are obtained and compared with those in the reference.

Geometric nonlinearity
9-44
Table 9.18.1 Deflection of the open cylinder at P = 20000 N



Figure 9.18.2 Load-deflection curves of the open ended cylinder model


0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
P
u
l
l
i
n
g

f
o
r
c
e

a
t

A

(
N
)
x

1
0
0
0
0
Displacements at points at A, B and C (mm)
midas NFX
Reference
Point A Point B Point C

A
z
u [mm]
Reference 2.46
Element type
TRIA-3 2.46
QUAD-4 2.46
TRIA-6 2.46
QUAD-8 2.48


45
midas NFX Benchmark Series 9-45
9.19 Slit annular plate under line force

REFERENCE Sze et al. [9-7]
KEYWORDS shell elements
MODEL FILENAME GeometricNonlinearStatic19.nfx






Figure 9.19.1 Slit annular plate model under line force

Material data
Youngs modulus
Poissons ratio
E = 21 10
6
N/mm
2
v = 0.0
Section property Thickness t = 0.03 mm

Table 9.19.1 Deflection of point A and B of the slit annular plate when line load arrived at 0.8

B
P = 0.8 per unit length
A
R
i
= 6
R
o
= 10
t = 0.03
Units: mm
P
B
A

A
z
u [mm]
B
z
u [mm]
Reference 13.618 17.257
Element
type
TRIA-3 12.882 16.542
QUAD-4 13.619 17.259
TRIA-6 13.244 16.827
QUAD-8 13.619 17.259
Figure 9.19.1 shows a circular annular plate with inner radius
i
R and outer radius
o
R
.
A slit cut along the
radial direction is given and a line force is applied at one end of the slit while the other end of the slit is fully
clamped. Vertical deflection at points A and B are obtained.

Geometric nonlinearity
9-46

Figure 9.19.2 Load- vertical deflection curves


0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 2 4 6 8 10 12 14 16 18
F
o
r
c
e
/
u
n
i
t

l
e
n
g
t
h

(
N
/
m
m
)
Vertical deflections at points A and B (mm)
Point A
Point B
Reference


47
midas NFX Benchmark Series 9-47
9.20 Pinched cylinder with rigid diaphragms

REFERENCE Sze et al. [9-7]
KEYWORDS shell elements
MODEL FILENAME GeometricNonlinearStatic20.nfx





Figure 9.20.1 Pinched cylinder model with rigid diaphragms

Material data
Youngs modulus
Poissons ratio
E = 30 10
3
N/mm
2
v = 0.0
Section property Thickness t = 1 mm


P
R = 100
L = 200
Z
Rigid diaphragm
Y
X
B
A
P = 12000 N
Rigid diaphragm
Units: mm
t = 1
Figure 9.20.1 shows a circular cylinder mounted over end rigid diaphragms. The cylinder is pinched by a pair
of diametrically opposite point forces. One-eighth of the cylinder is modeled and analyzed. The arc-length
method is utilized to obtain the equilibrium path.

Geometric nonlinearity
9-48

Figure 9.20.2 Load-deflection curves of the pinched cylinder with rigid diaphragms


0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
11000
12000
-20 -10 0 10 20 30 40 50 60 70 80 90
P
i
n
c
h
e
d

l
o
a
d

(
N
)
Displacements at points A and B (mm)
QUAD4
QUAD8
Reference
Point B
Point A


49
midas NFX Benchmark Series 9-49
9.21 Composite cylindrical shell under uniform load

REFERENCE Sze et al. [9-7]
KEYWORDS layered shell elements
MODEL FILENAME GeometricNonlinearStatic21.nfx







Figure 9.21.1 Composite cylindrical shell under uniform load

Material data
Youngs modulus

Shear modulus

Poissons ratio
E
L
= 25 10
6
N/mm
2
E
T
= 1 10
6
N/mm
2
G
TT
= 0.2 10
6
N/mm
2
G
LT
= 0.5 10
6
N/mm
2
LT
v =
TT
v = 0.3
Section property Thickness t = 2.54 mm

R
Z
X
Y

L/2
L/2
C
q = 2.4
R = 2540
L = 508
t = 2.54
= 0.2 radians
Units: mm
Figure 9.21.1 shows a composite cylindrical shell of lamination [90/0] loaded by a uniform conservative
downward force per unit mid-surface area. All edges are clamped. One quarter of the cylindrical shell is
modeled. Composite shell elements are utilized to obtain the geometrically nonlinear response of the
cylindrical shell. The subscript L and T in the material data denote the properties in the longitudinal and
transverse directions with respect to the fiber direction.

Geometric nonlinearity
9-50

Figure 9.21.2 Distributed force vs. center deflection of the composite cylindrical shell



0
0.5
1
1.5
2
2.5
3
0 1 2 3 4 5 6 7
F
o
r
c
e

p
e
r

u
n
i
t

a
r
e
a

(
N
/
m
m
2
)
Central deflection (mm)
midas NFX
Reference


51
midas NFX Benchmark Series 9-51
9.22 Semi-cylindrical laminated shell

REFERENCE Sze et al. [9-7]
KEYWORDS layered shell elements
MODEL FILENAME GeometricNonlinearStatic22.nfx








Figure 9.22.1 Semi-cylindrical laminated shell model

Material data
Youngs modulus

Shear modulus
Poissons ratio
E
L
= 2068.5 N/mm
2
E
T
= 517.125 N/mm
2
G
TT
= G
LT
= 795.6 N/mm
2
LT
v =
TT
v = 0.3
Section property Thickness t = 3 mm


X
Z
Y
P
R = 101.6
A
L = 304.8
t = 3.0

Y
Units: mm
Figure 9.22.1 shows a semi-cylindrical laminated shell with a semi-circular edge clamped and the other free
hanging. All the Z translations of the two straight edges are restrained, i.e. 0
y
W u = = . The stacking sequences
under consideration are [0/90/0] and [90/0/90] where 0 refers to the longitudinal direction. One-half of
the semi-cylindrical shell is modeled. The subscript L and T in the material data denote the properties in the
longitudinal and transverse directions with respect to the fiber direction.


Geometric nonlinearity
9-52

Figure 9.22.2 Load-deflection curves of semi-cylindrical laminated shells


Figure 9.22.3 Deformed shape of the semi-cylindrical laminated shells


0
100
200
300
400
500
600
700
0 20 40 60 80 100 120 140 160
P
o
i
n
t

f
o
r
c
e

(
N
)
Downward deflection at A (mm)
midas NFX
Reference
[90/0/90]
[0/90/0]
|w
A
|=150 |w
A
|=150
[0/90/0] [90/0/90]


53
midas NFX Benchmark Series 9-53
References

[9-1] NAFEMS, Background to Finite Element Analysis of Geometric Non-linearity Benchmarks, Ref . R0065,
NAFEMS, Glasgow, 1999
[9-2] NAFEMS, A Review of Benchmark Problems for Geometric Non-linear Behaviour of 3-D Beams and
Shells, Ref . R0024, NAFEMS, Glasgow, 1993
[9-3] Ryoji Miike, Ichoro Kobayashi and Yoshikazu Yamada, Virtual large displacement theorem for framed
structures, J. Eng. Mech., Vol. 116(2), pp. 411-428, 1990
[9-4] C. C. Chang, Periodically Restated Quasi-Newton Updates in Constant Arc-Length Method,
Computers and Structures, Vol. 41 No. 5, pp. 963-972, 1991
[9-5] S. Timoshenko, S. Woinowsky-Knieger, Theory of Plates and Shells, McGraw-Hill, New York, 1959, pg.
401, eq. 232
[9-6] NAFEMS, Selected Benchmarks for Non-Linear Behavior of 3D-Beams, NAFEMS, Glasgow, 1989
[9-7] K. Y. Sze, W.K. Chan and T. H. H. Pian, An eight-node hybrid-stress solid-shell element for geometric
non-linear analysis of elastic shells, International Journal for Numerical Methods in Engineering, Vol.
55, pp. 853-878, 2002



1
midas NFX Benchmark Series 10-1
10.1 2D plane strain plasticity

REFERENCE NAFEMS [10-1]
KEYWORDS plane strain elements
MODEL FILENAME MaterialNonlinearStatic01.nfx





Figure 10.1.1 2D plane strain model

Material data
Youngs modulus
Poissons ratio
Perfect plasticity
Isotropic hardening
E = 250.0 10
3
N/mm
2
v = 0.25
Y
o = 5.0 N/mm
2
E
T
= 50.0 10
3
N/mm
2



L = 1.0 mm
A B
C D
u
y
u
x L = 1.0 mm
Figure 10.1.1 shows a 2D plane strain model undergoing elastic-plastic deformations. Two plasticity models
are considered; a perfect plasticity model and an isotropic hardening model. Plane strain elements are
incorporated to evaluate the nonlinear response. The loading conditions are summarized in Table 10.1.1.

Material nonlinearity
10-2

(a) Perfect plasticity


(b) Isotropic hardening
Figure 10.1.2 Stress variation obtained using plane strain elements


-5.0
0.0
5.0
10.0
15.0
20.0
25.0
0 1 2 3 4 5 6 7 8
S
t
r
e
s
s

(
N
/
m
m
2
)
Step
_xx
_yy
_zz
_eff
Reference
-5.0
0.0
5.0
10.0
15.0
20.0
25.0
0 1 2 3 4 5 6 7 8
S
t
r
e
s
s

(
N
/
m
m
2
)
Step
_xx
_yy
_zz
_eff
Reference


3
midas NFX Benchmark Series 10-3

(a) Perfect plasticity


(b) Isotropic hardening
Figure 10.1.3 Stress path obtained using plane strain elements







-10.0
-8.0
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
-10.0 -5.0 0.0 5.0 10.0

y
y

-

z
z

xx
-
zz
Yield surface
midas NFX
Reference
-10.0
-8.0
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
-10.0 -5.0 0.0 5.0 10.0

y
y

-

z
z

xx
-
zz
Yield surface
midas NFX
Reference

Material nonlinearity
10-4
Table 10.1.1 Displacements prescribed in 8 increments (R = 2.5 10
-5
)

Table 10.1.2 Stress obtained at step 4, perfect plasticity

Table 10.1.3 Stress obtained at step 4, isotropic hardening


Step Disp. change
x
o [mm]
y
o [mm] Stress state
Step 1 ux = R R 0.0 first yield
Step 2 ux = R 2R 0.0 plastic flow
Step 3 uy = R 2R R elastic unloading
Step 4 uy = R 2R 2R plastic reloading
Step 5 ux = -R R 2R plastic flow
Step 6 ux = -R 0.0 2R plastic flow
Step 7 uy = -R 0.0 R elastic unloading
Step 8 uy = -R 0.0 0.0 plastic flow
Num. element
xx
o [N/mm
2
]
yy
o [N/mm
2
]
zz
o [N/mm
2
]
eff
o [N/mm
2
]
Reference 16.31210 19.71433 13.97358 5.000000
QUAD-4 1 16.32229 19.71016 13.96755 5.000000
TRIA-3 2 16.32229 19.71016 13.96755 5.000000
QUAD-8 1 16.32229 19.71016 13.96755 5.000000
TRIA-6 2 16.32229 19.71016 13.96755 5.000000
Num. element
xx
o [N/mm
2
]
yy
o [N/mm
2
]
zz
o [N/mm
2
]
eff
o [N/mm
2
]
Reference 16.85673 20.24083 12.90243 6.361633
QUAD-4 1 16.86400 20.23665 12.89935 6.361184
TRIA-3 2 16.86400 20.23665 12.89935 6.361184
QUAD-8 1 16.86400 20.23665 12.89935 6.361184
TRIA-6 2 16.86400 20.23665 12.89935 6.361184


5
midas NFX Benchmark Series 10-5
10.2 2D plane stress plasticity

REFERENCE NAFEMS [10-1]
KEYWORDS plane stress elements
MODEL FILENAME MaterialNonlinearStatic02.nfx





Figure 10.2.1 2D plane stress model

Material data
Youngs modulus
Poissons ratio
Perfect plasticity
Isotropic hardening
E = 250.0 10
3
N/mm
2
v = 0.25
Y
o = 5.0 N/mm
2
E
T
= 50.0 10
3
N/mm
2



L = 1.0 mm
A B
C D
u
y
u
x L = 1.0 mm
Figure 10.2.1 shows a 2D plane stress model undergoing elastic-plastic deformations. Two plasticity models
are considered; a perfect plastic model and an isotropic hardening model. Material nonlinear analyses are
carried out using plane stress elements in accordance with the loading condition summarized in Table 10.2.1.

Material nonlinearity
10-6

(a) Perfect plasticity


(b) Isotropic hardening
Figure 10.2.2 Stress variation obtained using plane stress elements


-10.0
-5.0
0.0
5.0
10.0
0 1 2 3 4 5 6 7 8
S
t
r
e
s
s

(
N
/
m
m
2
)
Step
_xx
_yy
_eff
Reference
-10.0
-5.0
0.0
5.0
10.0
0 1 2 3 4 5 6 7 8
S
t
r
e
s
s

(
N
/
m
m
2
)
Step
_xx
_yy
_eff
Reference


7
midas NFX Benchmark Series 10-7

(a) Perfect plasticity


(b) Isotropic hardening
Figure 10.2.3 Stress path obtained using plane stress elements


-10.0
-8.0
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
-10.0 -8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0 10.0

y
y
-

z
z

xx
-
zz
Yield surface
midas NFX
Reference
-10.0
-8.0
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
-10.0 -8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0 10.0

y
y
-

z
z

xx
-
zz
Yield surface
midas NFX
Reference

Material nonlinearity
10-8
Table 10.2.1 Displacements prescribed in 8 increments (R = 2.080126 10
-5
)

Table 10.2.2 Stress obtained at step 4, perfect plasticity

Table 10.2.3 Stress obtained at step 4, isotropic hardening


Step Disp. change
x
o [mm]
y
o [mm] Stress state
Step 1 ux = R R 0.0 first yield
Step 2 ux = R 2R 0.0 plastic flow
Step 3 uy = R 2R R elastic unloading
Step 4 uy = R 2R 2R plastic reloading
Step 5 ux = -R R 2R plastic flow
Step 6 ux = -R 0.0 2R plastic flow
Step 7 uy = -R 0.0 R elastic unloading
Step 8 uy = -R 0.0 0.0 plastic flow
Num. element
xx
o [N/mm
2
]
yy
o [N/mm
2
]
eff
o [N/mm
2
]
Reference 3.308084 5.751942 5.000035
QUAD-4 1 3.320648 5.750553 5.000000
TRIA-3 2 3.320648 5.750553 5.000000
QUAD-8 1 3.320648 5.750553 5.000000
TRIA-6 2 3.320648 5.750553 5.000000
Num. element
xx
o [N/mm
2
]
yy
o [N/mm
2
]
eff
o [N/mm
2
]
Reference 5.685183 8.888501 7.797050
QUAD-4 1 5.790009 8.838760 7.776335
TRIA-3 2 5.790009 8.838760 7.776335
QUAD-8 1 5.790009 8.838760 7.776335
TRIA-6 2 5.790009 8.838760 7.776335


9
midas NFX Benchmark Series 10-9
10.3 3D plasticity

REFERENCE NAFEMS [10-1]
KEYWORDS solid elements
MODEL FILENAME MaterialNonlinearStatic03.nfx






Figure 10.3.1 3D solid model

Material data
Youngs modulus
Poissons ratio
Perfect plasticity
Isotropic hardening
E = 250.0 10
3
N/mm
2
v = 0.25
Y
o = 5.0 N/mm
2
E
T
= 50.0 10
3
N/mm
2


L = 1.0 mm
A
B
C
D
u
z
u
x
E F
G H
Z
X
Y
u
y
L = 1.0 mm
L = 1.0 mm
Figure 10.3.1 shows a 3D cube model undergoing elastic-plastic deformations. Two plasticity models are
considered; a perfect plasticity and an isotropic hardening model. The boundary condition is depicted in Figure
10.3.1 and the loading condition is summarized in Table 10.3.1. 3D continuum elements are utilized to obtain
nonlinear responses.

Material nonlinearity
10-10

(a) Perfect plasticity


(b) Isotropic hardening
Figure 10.3.2 Stress variation obtained using solid elements


-5.0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
0 1 2 3 4 5 6 7 8 9 10 11 12
S
t
r
e
s
s

(
N
/
m
m
2
)

Step
_xx
_yy
_zz
_eff
Reference
-5.0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
0 1 2 3 4 5 6 7 8 9 10 11 12
S
t
r
e
s
s

Step
_xx
_yy
_zz
_eff
Reference


11
midas NFX Benchmark Series 10-11

(a) Perfect plasticity


(b) Isotropic hardening
Figure 10.3.3 Stress path obtained using solid elements




-10.0
-8.0
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
-10.0 -8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0 10.0

y
y

-

z
z

xx
-
zz
Yield surface
midas NFX
Reference
-10.0
-8.0
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
-10.0 -8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0 10.0

y
y

-

z
z

xx
-
zz

Yield surface
midas NFX
Reference

Material nonlinearity
10-12
Table 10.3.1 Displacements prescribed in 12 increments (R = 2.5 10
-5
)

Table 10.3.2 Stress obtained at of step 6, perfect plasticity

Table 10.3.3 Stress obtained at step 6, isotropic hardening


Step Disp. change
x
o [mm]
y
o [mm]
z
o [mm]
Step 1 ux = R R 0.0 0.0
Step 2 ux = R 2R 0.0 0.0
Step 3 uy = R 2R R 0.0
Step 4 uy = R 2R 2R 0.0
Step 5 uz = R 2R 2R R
Step 6 uz = R 2R 2R 2R
Step 7 ux = -R R 2R 2R
Step 8 ux = -R 0.0 2R 2R
Step 9 uy = -R 0.0 R 2R
Step 10 uy = -R 0.0 0.0 2R
Step 11 uz = -R 0.0 0.0 R
Step 12 uz = -R 0.0 0.0 0.0
Num. element
xx
o [N/mm
2
]
yy
o [N/mm
2
]
zz
o [N/mm
2
]
eff
o [N/mm
2
]
Reference 22.27493 24.70006 28.02500 5.000002
HEXA-8 1 22.26845 24.71130 28.02025 5.000000
PENTA-6 2 22.26845 24.71130 28.02025 5.000000
HEXA-20 1 22.26845 24.71130 28.02025 5.000000
PENTA-15 2 22.26845 24.71130 28.02025 5.000000
Num. element
xx
o [N/mm
2
]
yy
o [N/mm
2
]
zz
o [N/mm
2
]
eff
o [N/mm
2
]
Reference 21.84202 25.24927 27.90871 5.267201
HEXA-8 1 21.86400 25.23665 27.89935 5.238801
PENTA-6 2 21.86400 25.23665 27.89935 5.238801
HEXA-20 1 21.86400 25.23665 27.89935 5.238801
PENTA-15 2 21.86400 25.23665 27.89935 5.238801


13
midas NFX Benchmark Series 10-13
10.4 Kinematic hardening plasticity

REFERENCE NAFEMS [10-1]
KEYWORDS plane strain elements
MODEL FILENAME MaterialNonlinearStatic04.nfx





Figure 10.4.1 2D plane strain overlay model

Material data
Youngs modulus
Poissons ratio
Perfect plasticity

E = 250.0 10
3
N/mm
2
v = 0.25
Y
o = 1000.0 N/mm
2
; w
1
= 0.1724 (element 1)

Y
o = 5.0 N/mm
2
; w
2
= 0.8276 (element 2)




L
1
= 1 mm
A B
C D
E
W
1
X
Y
W
2
A B
C D
L
2

=

1

m
m
u
x
X
Y
Figure 10.4.1 shows a 2D plane strain overlay model which undergoes prescribed extension and compression
along x-axis as presented in table 8.4.1. One element has nearly elastic property and the other has perfectly
plastic property. The combination of plastic property produces kinematic hardening effect.

Material nonlinearity
10-14

Figure 10.4.2 Stress variation per load step (Kinematic hardening)


Figure 10.4.3 Stress path (Kinematic hardening)


-20.0
-15.0
-10.0
-5.0
0.0
5.0
10.0
15.0
20.0
0 1 2 3 4 5 6 7 8
S
t
r
e
s
s

(
N
/
m
m
2
)
Step
_xx
_yy
_zz
_eff
Reference
-10.0
-8.0
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
-10.0 -8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0 10.0

y
y
-

z
z

xx
-
zz
Yield surface
midas NFX
Reference


15
midas NFX Benchmark Series 10-15
Table 10.4.1 Displacements prescribed in 8 increments (R = 2.5 10
-5
)

Table 10.4.2 Stress obtained at step 2, kinematic hardening


Step Disp. change
x
o [mm]
Step 1 ux = R R
Step 2 ux = R 2R
Step 3 ux = -R R
Step 4 ux = -R 0.0
Step 5 ux = -R -R
Step 6 ux = -R -2R
Step 7 ux = R -R
Step 8 ux = R 0.0
Num. element
xx
o [N/mm
2
]
yy
o [N/mm
2
]
zz
o [N/mm
2
]
eff
o [N/mm
2
]
Reference 12.24140 6.379450 6.379450 5.862071
QUAD-4 1 12.24138 6.379310 6.379310 5.862069
TRIA-3 2 12.24138 6.379310 6.379310 5.862069
QUAD-8 1 12.24138 6.379310 6.379310 5.862069
TRIA-6 2 12.24138 6.379310 6.379310 5.862069

Material nonlinearity
10-16
10.5 Two-bar assembly plasticity

REFERENCE NAFEMS [10-1]
KEYWORDS rod elements
MODEL FILENAME MaterialNonlinearStatic05.nfx







Figure 10.5.1 Two-bar assembly model

Material data
Youngs modulus
Poissons ratio
Coefficient of thermal
expansion
Perfect plasticity
Kinematic hardening
E = 10.0 10
3
N/mm
2
v = 0.0
o = 1.0 10
-5
/C

Y
o = 10.0 N/mm
2
E
T
= 1.0 10
3
N/mm
2
Section property Cross-sectional area A = 1 mm
2




Bar 1 (T
1
)
P = 15.0
Bar 2 (T
2
)
Temperature
Half-cycle
A = 1 mm
2
A = 1 mm
2
Figure 10.5.1 shows a pair of identical bars whose ends are constrained to move together axially. A constant
tensile force is applied so that the bars reach near-yielding point. And cyclic temperature variation is applied on
the bar assembly which changes the plastic state. Three load cases are tested; ratcheting (perfect plasticity with
1
100 T A = and
2
0 T A = ), shakedown (kinematic hardening with
1
100 T A = and
2
0 T A = ), and alternating
plasticity (kinematic hardening with
1
300 T A = and
2
0 T A = ).


17
midas NFX Benchmark Series 10-17

(a) Perfect plasticity (Ratcheting)

(b) Kinematic hardening (Shakedown)

(c) Kinematic hardening (Alternating plasticity)
Figure 10.5.2 Load variation per cycle

-5
0
5
10
15
20
25
0 5 10 15 20
L
o
a
d

(
N
)
Step
Bar 1 (midas NFX)
Bar 2 (midas NFX)
Bar 1 (Reference)
Bar 2 (Reference)
-5
0
5
10
15
20
25
0 5 10 15 20
L
o
a
d

(
N
)
Step
Bar 1 (midas NFX)
Bar 2 (midas NFX)
Bar 1 (Reference)
Bar 2 (Reference)
-5
0
5
10
15
20
25
0 5 10 15 20
L
o
a
d

(
N
)
Step
Bar 1 (midas NFX) Bar 2 (midas NFX)
Bar 1 (Reference) Bar 2 (Reference)

Material nonlinearity
10-18

(a) Perfect plasticity (Ratcheting)

(b) Kinematic hardening (Shakedown)

(c) Kinematic hardening (Alternating plasticity)
Figure 10.5.3 Strain variation per cycle


0
0.002
0.004
0.006
0.008
0.01
0.012
0 5 10 15 20
S
t
r
a
i
n
Step
Bar 1 (midas NFX)
Bar 2 (midas NFX)
Bar 1 (Reference)
Bar 2 (Reference)
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
0.0035
0.004
0 5 10 15 20
S
t
r
a
i
n
Step
Bar 1 (midas NFX)
Bar 2 (midas NFX)
Bar 1 (Reference)
Bar 2 (Reference)
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
0 5 10 15 20
S
t
r
a
i
n
Step
Bar 1 (midas NFX)
Bar 2 (midas NFX)
Bar 1 (Reference)
Bar 2 (Reference)


19
midas NFX Benchmark Series 10-19

(a) Perfect plasticity (Ratcheting)

(b) Kinematic hardening (Shakedown)

(c) Kinematic hardening (Alternating plasticity)
Figure 10.5.4 Load-strain curve

-5
0
5
10
15
20
25
0 0.002 0.004 0.006 0.008 0.01 0.012
L
o
a
d

(
N
)
Strain
Bar 1 (midas NFX)
Bar 2 (midas NFX)
Bar 1 (Reference)
Bar 2 (Reference)
-5
0
5
10
15
20
25
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035
L
o
a
d

(
N
)
Strain
Bar 1 (midas NFX)
Bar 2 (midas NFX)
Bar 1 (Reference)
Bar 2 (Reference)
-5
0
5
10
15
20
25
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009
L
o
a
d

(
N
)
Strain
Bar 1 (midas NFX)
Bar 2 (midas NFX)
Bar 1 (Reference)
Bar 2 (Reference)

Material nonlinearity
10-20
10.6 Rigid punch plasticity

REFERENCE NAFEMS [10-1]
KEYWORDS plane strain elements, solid elements
MODEL FILENAME MaterialNonlinearStatic06A.nfx
MaterialNonlinearStatic06B.nfx






Figure 10.6.1 Rigid punch on a deep plate

Material data
Youngs modulus
Poissons ratio
Perfect plasticity
Isotropic hardening
E = 1.0 10
3
N/mm
2
v = 0.3
Y
o = 1.0 N/mm
2

E
T
= 0.1 10
3
N/mm
2


A B
C D
Z
160
200
E
F
20
80
rigid
punch
Units: mm
Figure 10.6.1 shows a rigid punch pressed into a deep elastic foundation of finite width supported on the
frictionless plane. 2D plane strain condition is adopted for the foundation. The elastic perfectly plastic and
isotropic hardening material models are tested. Punch force is represented by the downward vertical
displacement, o
.
Prescribed displacement is given in 6 steps: o = 0.04, 0.12, 0.14, 0.16, 0.18, 0.24 mm.


21
midas NFX Benchmark Series 10-21

Figure 10.6.2 Solution for the load-deflection curve

Table 10.6.1 Reaction force at deflection = 0.24 mm

0.0
25.0
50.0
75.0
100.0
125.0
150.0
0.00 0.05 0.10 0.15 0.20 0.25
L
o
a
d

(
N
)
Applied deflection of punch (mm)
Perfect plastic
Isotropic hardening
Reference
Perfect plastic Isotropic hardening

R
F (N)
R
F (N)
Reference 110.530 123.230
QUAD-4 112.876 124.335
TRIA-3 119.525 129.139
QUAD-8 110.978 123.370
TRIA-6 111.481 123.607
HEXA-8 112.728 124.259
PENTA-6 119.525 129.139
HEXA-20 110.982 123.358
PENTA-15 111.481 123.607

Material nonlinearity
10-22
10.7 Thermal ratcheting of uniform beam

REFERENCE NAFEMS [10-2]
KEYWORDS plane stress elements
MODEL FILENAME MaterialNonlinearStatic07.nfx







Figure 10.7.1 Thermal ratcheting of uniform beam

Material data
Youngs modulus
Poissons ratio
Yield stress
Hardening coefficient
Coefficient of expansion
E = 200.0 10
3
N/mm
2
v = 0.0
Y
o = 100.0 N/mm
2
| = 0.02
o = 1.0 10
-5
/C



Temperature
distribution
T = -400 (10 half cycles)
Applied mechanical load (constant) 900 per unit width of beam
O
10
P P
T
Temperature
-400
1 10
0
Figure 10.7.1 shows a uniform beam with constant end load and cyclically varying linear through-thickness
temperature gradient. Plane sections are constrained to remain straight and parallel without curvature. The
material behavior is subjected to von Mises elasto-plastic model with linear kinematic hardening. Stress and
strain variations on the lower and upper surfaces are determined per load step. Each load step corresponds to a
half-cycle temperature variation.


23
midas NFX Benchmark Series 10-23

(a) Lower surface


(b) Upper surface
Figure 10.7.2 Stress variation per load step

-100
-50
0
50
100
150
200
0 1 2 3 4 5 6 7 8 9 10
S
t
r
e
s
s

(
N
/
m
m
2
)
Step
midas NFX
Reference
-100
-50
0
50
100
150
200
0 1 2 3 4 5 6 7 8 9 10
S
t
r
e
s
s

(
N
/
m
m
2
)
Step
midas NFX
Reference

Material nonlinearity
10-24

(a) Lower surface


(b) Upper surface
Figure 10.7.3 Strain variation per load step


-0.002
-0.001
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
0.011
0.012
0 1 2 3 4 5 6 7 8 9 10
S
t
r
a
i
n
Step
Total strain
Elastic strain
Plastic strain
Reference
-0.002
-0.001
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
0.011
0.012
0 1 2 3 4 5 6 7 8 9 10
S
t
r
a
i
n
Step
Total strain
Elastic strain
Plastic strain
Reference


25
midas NFX Benchmark Series 10-25
10.8 An extended straight truss

REFERENCE Crisfield [10-3]
KEYWORDS rod elements
MODEL FILENAME MaterialNonlinearStatic08.nfx





Figure 10.8.1 A straight truss model

Material data
Youngs modulus
Poissons ratio
Isotropic hardening
equation
E = 2.0 10
6
N/m
2
v = 0.3
c = 22222 + 200 (0 0.0009)
c = 220 (0.0009 )
Section property Cross-sectional area A = 1 m
2



Figure 10.8.2 Stress variation along axial displacement
X
Z
1 m
prescribed
displacement 0.004 m
cross-sectional area = 1 m
2
0.0
50.0
100.0
150.0
200.0
250.0
0 0.5 1 1.5 2 2.5 3 3.5 4

a
x
i
a
l
(
N
/
m
2
)
Axial displacement (mm)
midas NFX
Reference
Figure 10.8.1 shows a straight truss model fixed at one end in all directions. The other end is constrained to
move in axial direction and is extended to prescribed value using displacement controlled nonlinear analysis.
The material behavior is subjected to von Mises elasto-plastic model with linear isotropic hardening combined
with perfectly plastic model.

Material nonlinearity
10-26
10.9 Square plate under uniformly distributed load

REFERENCE NAFEMS [10-2]
KEYWORDS shell elements
MODEL FILENAME MaterialNonlinearStatic09.nfx






Figure 10.9.1 Square plate model

Material data
Youngs modulus
Poissons ratio
Perfect plasticity
E = 30000.0 N/mm
2
v = 0.3
Y
o = 30.0 N/mm
2

Section property Thickness t = 0.4 mm

Table 10.9.1 Load per unit area for the plastic state

40
40
0.4
Units: mm
Simply supported Clamped edge

lim
P [N/mm
2
]
lim
P [N/mm
2
]
Reference 0.01877 0.03852
QUAD-4 0.01890 0.03710
TRIA-3 0.01890 0.03750
QUAD-8 0.01910 0.03960
TRIA-6 0.01870 0.03670
Figure 10.9.1 shows a square plate subjected to uniformly distributed load. The load is applied up to the
yielding point. The plate material follows the von Mises elastic perfectly plastic material model without
hardening. The shell section is integrated in the thickness direction using 13-point Simpson integration rule.
Two boundary edge conditions are tested: simply supported and clamped edge.


27
midas NFX Benchmark Series 10-27

(a) Simply supported boundary condition


(b) Clamped edge boundary condition
Figure 10.9.2 Pressure-center deflection curve using quadrilateral element

0.000
0.005
0.010
0.015
0.020
0.0 1.0 2.0 3.0 4.0 5.0 6.0
P
r
e
s
s
u
r
e

(
N
/
m
m
2
)
Deflection at Center of Plate
midas NFX
Layered formulation (reference)
Gross yield formulation
P_Limit
0.000
0.005
0.010
0.015
0.020
0.025
0.030
0.035
0.040
0.045
0.0 1.0 2.0 3.0 4.0 5.0 6.0
P
r
e
s
s
u
r
e

(
N
/
m
m
2
)
Deflection at Center of Plate
midas NFX
Layered formulation (reference)
Gross yield formulation
P_Limit

Material nonlinearity
10-28
10.10 Uniformly loaded circular plate

REFERENCE Owen et al. [10-4]
KEYWORDS axisymmetric elements
MODEL FILENAME MaterialNonlinearStatic10.nfx






Figure 10.10.1 Circular plate model

Material data
Youngs modulus
Poissons ratio
Yield stress
E = 1.0 10
7
psi

v = 0.24
Y
o = 16000.0 psi




P
Y
X
Z
P
Y
R = 10
h = 1
Units: in
C
mesh shape
Figure 10.10.1 shows a circular plate model simply supported around its outer rim under uniform pressure. The
circular plate undergoes material nonlinear deformations. The material follows elastic, perfectly plastic von
Mises model. The limit pressure is determined using various types of axisymmetric elements.


29
midas NFX Benchmark Series 10-29

Figure 10.10.2 Pressure-center deflection curve

Table 10.10.1 Limit pressure obtained using axisymmetric elements
*

2
2
6 . 5 2
,
4
y y
lim y
M h
P M
R
o
~


0.0
50.0
100.0
150.0
200.0
250.0
300.0
0.0 0.5 1.0 1.5 2.0
P
r
e
s
s
u
r
e

(
p
s
i
)
Deflection at center of plate (in)
QUAD8X
TRIA6X
P_Limit

lim
P [psi]
Reference 260.8
QUAD-8 261.2
TRIA-6 263.2

Material nonlinearity
10-30
10.11 Two coaxial tubes

REFERENCE Crandall et al. [10-5]
KEYWORDS solid elements
MODEL FILENAME MaterialNonlinearStatic11.nfx





Figure 10.11.1 Two coaxial tubes

Material data (Inner
tube: steel 1020-RC)
Youngs modulus
Poissons ratio
Yield stress
E
steel
= 1.84375 10
5
N/mm
2
steel
v = 0.3
/ Y steel
o = 590 N/mm
2

Material data (Outer
tube: aluminum alloy
2024-T4)
Youngs modulus
Poissons ratio
Yield stress
E
alloy
= 7.6 10
4
N/m
2
alloy
v = 0.3
/ Y alloy
o = 380 N/mm
2



R
1
= 4.069718
R
2
= 2.478169
t = 0.5
Units: mm
= 0.1
L = 10
R
1
R
2
Figure 10.11.1 shows two coaxial tubes made of steel (1020-CR) and aluminum alloy (2024-T4) compressed
between two rigid flat ends. Both tubes are assumed to follow the von Mises elastic, perfectly plastic material
model with associated flow rule.


31
midas NFX Benchmark Series 10-31

Figure 10.11.2 Reaction force-applied displacement curve using hexahedron element

Table 10.11.1 Reaction force at the bottom surface with the whole model in plasticity state


7048.4
8690 8690
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000
0.00 0.02 0.04 0.06 0.08 0.10
S
u
m

o
f

r
e
a
c
t
i
o
n

f
o
r
c
e

(
N
)
Applied displacement (mm)
midas NFX
Reference yield points

R
F [N]
Reference 8690.0
HEXA-8 8681.1
PENTA-6 8673.3
HEXA-20 8681.1
PENTA-15 8683.7

Material nonlinearity
10-32
10.12 Residual stress problem

REFERENCE Crandall et al. [10-5]
KEYWORDS rod elements
MODEL FILENAME MaterialNonlinearStatic12.nfx






Figure 10.12.1 3 member frame model

Material data
Youngs modulus
Poissons ratio
Perfect plasticity
E = 30.0 10
6
psi

v = 0.3
Y
o = 30000.0 psi
Section property Cross-sectional area A = 1.0 in
2


Table 10.12.1 Residual stress results of center member after loading-unloading

A = 1.0 in
2
A = 1.0 in
2
A = 1.0 in
2
P

elastic
o [in]
res
o [psi]
Reference 0.1000 -5650
midas NFX 0.0998 -5650
Figure 10.12.1 shows a chain hoist attached to the ceiling through three tie rods which is made of cold-rolled
steel with yield strength
y
o . The three-rod structure is loaded in the vertical direction and then unloaded.
While the structure is loaded, the center rod member yields firstly while side members remain elastic. When
the frame is fully loaded, all three rods become fully plastic.


33
midas NFX Benchmark Series 10-33

Figure 10.12.2 Stress-deflection curve of the center member


-5.650.E+03
-1.0E+04
0.0E+00
1.0E+04
2.0E+04
3.0E+04
4.0E+04
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
A
x
i
a
l

s
t
r
e
s
s

o
f

c
e
n
t
e
r

r
o
d

(
p
s
i
)
Deflection at point A (in)
midas NFX
Reference

Material nonlinearity
10-34
10.13 Nonlinear equation solution tests

REFERENCE NAFEMS [10-1]
KEYWORDS plane strain elements
MODEL FILENAME MaterialNonlinearStatic13.nfx








Figure 10.13.1 3 overlay elements mesh model

Material data
(element 1)
Youngs modulus
Poissons ratio
Perfect plasticity
E = 100.0 10
3
N/mm
2
v = 0.25
Y
o = 3.0 N/mm
2

Material data
(element 2)
Youngs modulus
Poissons ratio
Perfect plasticity
E = 60.0 10
3
N/mm
2
v = 0.25
Y
o = 6.0 N/mm
2

Material data
(element 3)
Youngs modulus
Poissons ratio
Perfect plasticity
E = 40.0 10
3
N/mm
2
v = 0.25
Y
o = 8.0 N/mm
2



P/2
P/2
P = 1.0
Figure 10.13.1 shows a mesh with three overlaid elements, each with yield stress. This problem is tailored to
test the performance of nonlinear equation solver. Three elements are welded together at the nodes with
identical element connectivity. Three different stiffness update schemes are tested. Namely, initial stiffness,
modified Newton-Raphson and full Newton-Raphson methods are tested for number of iterations to reach
convergence. Nonlinear analyses are carried out with 6 increments and 0.01% internal force norm is used as
convergence criteria.


35
midas NFX Benchmark Series 10-35

Figure 10.13.2 Target equilibrium curve obtained by initial stiffness matrix method
*
Values and bracketed values are numbers of iteration required to reach convergence using midas NFX and the reference
number, respectively.


Figure 10.13.3 Target equilibrium curve obtained by modified Newton-Raphson method
*
Values and bracketed values are numbers of iteration required to reach convergence using midas NFX and the reference
number, respectively.


1(1)
1(1)
13(10)
13(10)
40(27)
40(27)
0.0
3.0
6.0
9.0
12.0
15.0
18.0
0.00E+00 6.00E-05 1.20E-04 1.80E-04 2.40E-04

x
(
N
/
m
m
2
)

x
(mm)
midas NFX
Reference
P_limit
1(1)
1(1)
2(3)
2(3)
6(3)
2(3)
0.0
3.0
6.0
9.0
12.0
15.0
18.0
0.00E+00 6.00E-05 1.20E-04 1.80E-04 2.40E-04

x
(
N
/
m
m
2
)

x
(mm)
midas NFX
Reference
P_limit

Material nonlinearity
10-36

Figure 10.13.4 Target equilibrium curve obtained by full Newton-Raphson method
*
Values and bracketed values are numbers of iteration required to reach convergence using midas NFX and the reference
number, respectively.


1(1)
1(1)
2(3)
2(3)
3(2)
2(2)
0.0
3.0
6.0
9.0
12.0
15.0
18.0
0.00E+00 6.00E-05 1.20E-04 1.80E-04 2.40E-04

x
(
N
/
m
m
2
)

x
(mm)
midas NFX
Reference
P_limit


37
midas NFX Benchmark Series 10-37
10.14 Stiffened cylindrical panel

REFERENCE NAFEMS [10-6]
KEYWORDS shell elements
MODEL FILENAME MaterialNonlinearStatic14.nfx











Figure 10.14.1 Stiffened cylindrical panel under compressive end load


Y
Z
9
o
L
d
x
y
9
o

z
R
Load = 18649
Symmetry
Definition of nodal forces to simulate
evenly distributed load
Total load
= 12.875P
P
4P
P
4P
2P
0.07955P
0.31818P
0.15909P
0.31818P
D, solution point
R=400
L=400
d=10.0
Thickness=1.0
X
GCS
Figure 10.14.1 shows a stiffened cylindrical panel under compressive end load. The underlying material is
elastic, perfectly plastic von Mises material model. Automatic incrementation of the applied load using arc
length procedure is adopted until the global Z displacement at point D exceeded 0.2. There is an initial
imperfection in both the cylindrical panel and the stiffener.
Cylindrical panel (cylindrical system, RZu )
2
0.569sin where -9 9, 0 200
7200
Z
R Z
ut
u
| |
A = s s s s |
|
\ .

Stiffener (local cartesian system, xyz)
0.3sin sin where 0 200, 0 10
400 20
x y
z x y
t t | | | |
A = s s s s
| |
\ . \ .


Material nonlinearity
10-38
Material data
Youngs modulus
Poissons ratio
Perfect plasticity
E = 2.1 10
5
N/mm
2
v = 0.3
Y
o =350 N/mm
2

Section property Thickness t = 1.0 mm

Table 10.14.1 Limit point solution


Figure 10.14.2 Load-displacement curve


0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2.2
2.4
2.6
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2
A
p
p
l
i
e
d

l
o
a
d

(
N
)
x

1
0
0
0
0
Z-displacement (mm)
midas NFX
Reference
Reference midas NFX
Limit point 1
Load [N] 23742 24333
Z-disp [mm] 0.1679 0.1723
Limit point 2
Load [N] 15456 16352
Z-disp [mm] 0.1377 0.1436


39
midas NFX Benchmark Series 10-39
10.15 Necking of a circular bar

REFERENCE Simo et al. [10-7]
KEYWORDS axisymmetric elements, solid elements
MODEL FILENAME MaterialNonlinearStatic15A.nfx
MaterialNonlinearStatic15B.nfx







Figure 10.15.1 A circular bar model with geometric imperfection and its deformed shape

Material data
Shear modulus
Bulk modulus
Initial flow stress
Residual flow stress
Linear hardening
coefficient
Saturation exponent
u = 80.1938 10
3
N/mm
2

k = 164.206 10
3
N/mm
2

Y
o = 450 N/mm
2
Y

= 715 N/mm
2
K = 129.24 N/mm
2

o = 16.93

X
Y
Z
Geometric
Imperfection
R = 6.413
L/2 = 26.667
Units: mm
Figure 10.15.1 shows a circular bar with geometrical imperfection (0.982 % of radius) to be imposed at
symmetric plane to trigger strain localization. The model is considered in the context of large strain problem
and is subjected to a uniform extension of 7 mm. A general nonlinear hardening law is adopted of the following
form:
0
( ) : ( )[1 exp( )], 0
y
k K K K o o o oo o

= + + > .

Material nonlinearity
10-40

Figure 10.15.2 Stress-strain curve


Figure 10.15.3 Necking ratio versus elongation ratio obtained using low-order axisymmetric quadrilateral elements

Table 10.15.1 Radius ratios obtained using axisymmetric and solid elements
0
100
200
300
400
500
600
700
800
900
1000
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
P
l
a
s
t
i
c

s
t
r
e
s
s

(
N
/
m
m
2
)
Plastic strain
0
0.25
0.5
0.75
1
0 0.1 0.2
R
a
d
i
u
s

/

R
a
d
i
u
s
0

L/L
0
midas NFX
Reference
Experimental:
2499R
2515ST
2501R
2502R
Element type
(Axisymmetric)
R/R0
Element type
(Solid)
R/R0
QUADX-4 0.4048 HEXA8 0.3587
TRIAX-3 0.7049 PENTA6 0.6964
QUADX-8 0.4000 HEXA20 0.3593
TRIAX-6 0.3906 PENTA15 0.3646


41
midas NFX Benchmark Series 10-41
10.16 Pressurized rubber disc

REFERENCE Oden [10-8]
KEYWORDS solid elements
MODEL FILENAME MaterialNonlinearStatic16.nfx








Figure 10.16.1 Pressurized rubber disc

Material data
Deviatoric strain energy
coefficients
Poissons ratio
C
10
= 80 lb/in
2
C
01
= 20 lb/in
2
v = 0.4995

t = 0.5 in
Pressure = 50 psi
Symmetry
R = 7.5 in
Figure 10.16.1 shows a rubber disc pinned around its outside edge. The disc is subjected to pressure so that it
bulges into a spherical shape. For obvious axisymmetric consideration, a pie of disc is modeled with solid
elements. The Mooney-Rivlin material model is used to represent nearly incompressible rubber elasticity. The
pressure load is non-conservative. Additional constraining equations are used to so that the edge remains
straight.

Material nonlinearity
10-42

Figure 10.16.2 Thickness strain vs. central displacement


Figure 10.16.3 Pressure vs. deflection curve



0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.5 1 1.5 2 2.5 3
T
h
i
c
k
n
e
s
s

/

O
r
i
g
i
n
a
l

t
h
i
c
k
n
e
s
s
U
z
/ R
0
midas NFX
Reference
0
10
20
30
40
50
60
0 2 4 6 8 10 12 14 16 18 20
P
r
e
s
s
u
r
e

(
p
s
i
)
U
z
(in)
midas NFX
Reference


43
midas NFX Benchmark Series 10-43
10.17 Inflation of a spherical rubber balloon

REFERENCE Souza et al. [10-9]
KEYWORDS solid elements
MODEL FILENAME MaterialNonlinearStatic17.nfx







Figure 10.17.1 Inflation of a spherical rubber balloon

Material data
Deviatoric strain energy
coefficients




1
o = 1.3
1
u = 6.3 N/mm
2
2
o = 5.0
2
u = 0.012 N/mm
2

3
o = -2.0
3
u = -0.1 N/mm
2


Figure 10.17.1 shows a spherical rubber membrane-like structure inflated under internal pressure. The three-
term Ogden material model is used to represent nearly incompressible rubber elasticity. The solid mesh
discretizes one octant of the sphere with symmetry boundary conditions imposed along the edge. The pressure
load is non-conservative. Arc-length method is employed to allow equilibrium to be found beyond the
instability point.

Material nonlinearity
10-44

Figure 10.17.2 Normalized pressure vs. expansion ratio

*
Normalized internal pressure and expansion ratio are defined as
0 0
2 p pr t
-
= and
0
r r = , where r and
0
r are the
current and initial radii of the balloon respectively, t
0
is the initial thickness of the rubber membrane and p is the current
internal pressure.


0
1
2
3
0 1 2 3 4 5 6 7 8
N
o
r
m
a
l
i
z
e
d

p
r
e
s
s
u
r
e
,

p
*

Expansion ratio,
midas NFX
Reference (Ogden, 1972)


45
midas NFX Benchmark Series 10-45
10.18 Inflation of an axisymmetric ellipsoidal balloon

REFERENCE Souza et al. [10-9]
KEYWORDS solid elements
MODEL FILENAME MaterialNonlinearStatic18.nfx







Figure 10.18.1 Inflation of an axisymmetric ellipsoidal balloon

Material data
Deviatoric strain energy
coefficients




1
o = 1.3
1
u = 6.3 N/mm
2
2
o = 5.0
2
u = 0.012 N/mm
2

3
o = -2.0
3
u = -0.1 N/mm
2




r
2
= 145
r
1
= 95
t = 3
Units: mm
Figure 10.18.1 shows an axisymmetric ellipsoidal balloon to be inflated. The initial geometry of the balloon
corresponds to the shape of an official rugby ball. This ball is assumed to be made of the three-term Ogden
material model. Solid elements are used to discretize one octant of the sphere with symmetry boundary
conditions imposed along the edge. The pressure load is non-conservative. Arc-length method is employed to
allow equilibrium to be found beyond the instability point.

Material nonlinearity
10-46

Figure 10.18.2 Pressure-expansion diagram for rugby ball


0.000
0.050
0.100
0.150
0.200
0.000 40.000 80.000 120.000 160.000
p
r
e
s
s
u
r
e
,

p

(
N
/
m
m
2
)

radii r
1
, r
2
(mm)
midas NFX
Reference
r
1
r
2


47
midas NFX Benchmark Series 10-47
10.19 Contact between rigid body and a hyperelastic
body

REFERENCE Zhi-Qiang Feng et al. [10-10]
KEYWORDS solid elements
MODEL FILENAME MaterialNonlinearStatic19.nfx






Figure 10.19.1 A rigid cylinder into a hyperelastic body model

Material data Shear modulus G = 10.0 N/mm
2


R = 50
80
280
= 50
rigid cylinder
bottom side is fixed
Figure 10.19.1 shows the indentation of a rigid cylinder into a hyperelastic body. The Blatz-Ko hyperelastic
material model is used to represent sponge-type material behavior which requires only one material constant,
shear modulus. The results are obtained by carrying out nonlinear analysis with displacement control. Friction
coefficients of 0.4 u = and 0.0 u = are tested.

Material nonlinearity
10-48

Figure 10.19.2 Load-displacement curve



Figure 10.19.3 Deformed shape of the sponge type body ( = 0.4)

0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
0 10 20 30 40 50
F
o
r
c
e

(
N
)

Displacement of the cylinder (mm)
midas NFX
Reference
= 0.4
= 0.0


49
midas NFX Benchmark Series 10-49
References

[10-1] NAFEMS, Background to Material Non-Linear Benchmarks, Ref . R0049, NAFEMS, Glasgow, 1998
[10-2] NAFEMS, Selected Benchmarks for Material Non-Linearity, Ref . R0026, NAFEMS, Glasgow, 1993
[10-3] M. A. Crisfield, Non-linear Finite Element Analysis of Solids and Structures, England, John Wiley &
Sons Ltd., 1994
[10-4] D.R.J. Owen and E. Hinton, Finite Elements in Plasticity Theory and Practice, Pineridge Press
Limited, Swansea, U.K., 1980
[10-5] S. H. Crandall and N. C. Dahl, An Introduction to the Mechanics of Solids, McGraw-Hill Book Co., Inc.,
New York, NY, 1959
[10-6] NAFEMS, A Review of Benchmark Problems for Geometric Non-linear Behaviour of 3-D Beams and
Shells, Ref . R0024, NAFEMS, Glasgow, 1993
[10-7] J. C. Simo, T. J. R. Hughes, Computational Inelasticity, Springer-Verlag, New York, pp. 326-335 1998
[10-8] J. T. Oden, Finite Elements of Nonlinear Continua, McGraw-Hill, 1972
[10-9] E. A. de Souza Neto, D. Peric and D. R. J. Owen, Computational Methods for Plasticity, Wiley, New
York, 2008
[10-10] Zhi-Qiang Feng, Francois Peyraut, Nadia Labed, Solution of large deformation contact problems with
friction between Blatz-Ko hyperelastic bodies, International Journal of Engineering Science, Vol. 41, pp.
2213-2225, 2003



1
midas NFX Benchmark Series 11-1
11.1 Contact patch test

REFERENCE NAFEMS [11-1]
KEYWORDS solid elements
MODEL FILENAME ContactNonlinearStatic01.nfx






Figure 11.1.1 Two-body contact model

Material data
Youngs modulus
Poissons ratio
E
punch
= E
foundation
= 100 MPa

punch
v =
foundation
v = 0.3


q =0.04 MPa
body 1
body 2
contact
interface
( = 0.0)
1
0
0
0
1000
Units: mm
Figure 11.1.1 shows a frictionless two-body model for contact patch test. The bottom side of the foundation is
constrained in the vertical direction, with the center of this side constrained in all directions. Uniform pressure
of 0.04 N/mm
2
is applied on the top surface of body 1. Nonlinear analysis is carried out to obtain contact
pressure on the contact interfaces with various regular and irregular meshes. Effect of geometric nonlinearity is
neglected.

Contact analysis
11-2


Figure 11.1.2 Various meshes from top left clockwise; regular matching, regular non-matching, irregular matching and
irregular non-matching


Figure 11.1.3 Contact pressure distribution obtained using various meshes




30000
35000
40000
45000
50000
0 0.25 0.5 0.75 1
C
o
n
t
a
c
t

p
r
e
s
s
u
r
e

(
N
/
m
2
)
x-coordinate (m)
regular match
regular non-match
irregular match
irregular non-match


3
midas NFX Benchmark Series 11-3
11.2 Rigid punch contact

REFERENCE NAFEMS [11-1], Gladwell [11-2]
KEYWORDS solid elements
MODEL FILENAME ContactNonlinearStatic02.nfx






Figure 11.2.1 Rigid punch contact model

Material data
Youngs modulus
Poissons ratio
E
foundation
= 100 MPa

foundation
v = 0.3



punch
contact
interface
( = 0.0)
30
48
6
4
Units: mm
foundation
Figure 11.2.1 shows a rigid punch pressed into a frictionless elastic foundation. The bottom surface of the
foundation is constrained in the vertical direction with the center of the surface constrained in all directions.
The horizontal sides of the foundation are constrained in the horizontal direction. Prescribed vertical
displacement of 0.05 mm is applied on the punch. Contact pressure is determined on the contact interface and
compared with theoretical solution. The results are obtained without considering geometric nonlinear effects.

Contact analysis
11-4

Figure 11.2.2 Contact pressure distribution


0
1
2
3
4
5
6
7
8
9
10
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
C
o
n
t
a
c
t

p
r
e
s
s
u
r
e

(
N

m
)
Distance from punch center (mm)
midas NFX
Reference


5
midas NFX Benchmark Series 11-5
11.3 Hertzian contact

REFERENCE NAFEMS [11-1], Gladwell [11-2]
KEYWORDS solid elements
MODEL FILENAME ContactNonlinearStatic03.nfx






Figure 11.3.1 Pressed elastic cylinder model

Material data
Youngs modulus
Poissons ratio
E
cyl
= 500 MPa

v = 0.3


rigid foundation
8 mm
contact
interface
deformable
cylinder
F/2
Figure 11.3.1 shows an elastic cylinder pressed onto a rigid foundation. The rigid foundation is constrained in
all directions. The cylinder is constrained in the horizontal direction along the inner edge to simulate the
behavior of a half cylinder. The results are obtained for three load values; F = 25, 50, and 100 N. Geometric
nonlinear analyses are carried out to determine contact pressure along the contact interface and the results are
compared with the theoretical solution.

Contact analysis
11-6

Figure 11.3.2 Contact pressure distribution



0
5
10
15
20
25
30
35
40
45
50
0.00 0.25 0.50 0.75 1.00 1.25 1.50
C
o
n
t
a
c
t

p
r
e
s
s
u
r
e

(
M
P
a
)
Distance from center (mm)
midas NFX
Reference
25 N
50 N
100 N


7
midas NFX Benchmark Series 11-7
11.4 Sliding wedge with linear springs

REFERENCE NAFEMS [11-1]
KEYWORDS solid elements
MODEL FILENAME ContactNonlinearStatic04.nfx







Figure 11.4.1 Sliding wedge model

Material data
Youngs modulus
Poissons ratio
E = 206 GPa

v = 0.3

Table 11.4.1 Horizontal displacement obtained for four different friction coefficients and spring stiffnesses
Friction coefficient Force [N]
Spring stiffness
[N/m]
Horizontal displacement [m]
Reference midas NFX
0.0 1500 132.6 1.0 1.0
0.1 1500 98.0 1.0 1.0
0.2 1500 62.6 1.0 1.0
0.3 1500 26.5 1.0 1.0

1.0
6.0
1
.
2
K
F
0
.
7
1
.
3
G
4.0
contact
interface
tan = 0.1
F = 1500 N, G = 3058 N
Units: m
1.0
2
.
0
0
.
8
Figure 11.4.1 shows a sliding wedge on an elastic foundation under gravity load. The bottom side of the
foundation is constrained in all directions. The left side of the wedge is connected to 9 linear 2-node springs in
horizontal direction. Horizontal distances are determined for four cases of frictional constants and spring
stiffnesses as summarized in Table 11.4.1. Geometric nonlinear analyses are carried out to obtain the horizontal
displacements and compared with the reference values.

Contact analysis
11-8
11.5 Cantilever beam and a rigid curvilinear surface

REFERENCE NAFEMS [11-1], Ayari et al. [11-3]
KEYWORDS solid elements
MODEL FILENAME ContactNonlinearStatic05.nfx







Figure 11.5.1 Cantilever beam model

Material data
Youngs modulus
Poissons ratio
E = 2.1 10
6
N/mm
2
v = 0.3

Table 11.5.1 Vertical displacement at cantilever tip,
y
u and horizontal distance from root to point of separation,
S
x
q [N/mm
2
] 0.050 0.075 0.100
y
u [mm]
Reference 23.55 24.03 24.27
midas NFX 20.60 22.05 22.81
S
x [mm]
Reference 410 518 583
midas NFX 425 525 600



q
L = 1000
10
contact
interface
R = 20xL
Units: mm
X
Y
Figure 11.5.1 shows a cantilever beam loaded by a uniform pressure over a frictionless rigid curvilinear
surface. The rigid surface is constrained in all directions at its reference node. The left side of the cantilever is
constrained in all directions. Vertical deflection of the cantilever tip and horizontal distance from the root to the
point of separation are determined for three cases of pressure level; q = 0.05, 0.075 and 0.1 N/mm
2
by
carrying out geometric nonlinear analyses.


9
midas NFX Benchmark Series 11-9
11.6 Sliding and rolling of a ring on a rigid surface

REFERENCE NAFEMS [11-1]
KEYWORDS solid elements
MODEL FILENAME ContactNonlinearStatic06.nfx







Figure 11.6.1 Sliding and rolling of a ring

Material data
Density
Youngs modulus
Poissons ratio
p =7850 kg/m
3

E = 210 GPa

v = 0.3


contact
interface
G = 2443 N
= /2

A steel ring under gravity load is allowed to slide and roll on a rigid surface, as depicted in Figure 11.6.1. The
friction coefficient between the two bodies is 1.0. The rigid plate is constrained in all directions. Geometric
nonlinear analysis is carried out to determine the horizontal displacement of the center point as the steel ring is
rotated 90 degrees. The horizontal displacement and maximum vertical displacement of center of the ring are
compared to theoretical solution.

Contact analysis
11-10
Table 11.6.1 Horizontal and maximum vertical displacement of center of the ring
Theory midas NFX
H
u [mm]
785.398(circle)
778.823(polygon)
778.818
V
u [mm]

12.536 12.536

Horizontal displacement:
- circle: 2 / 4 785.398
H
u R t = =

- polygon: ( ) 7 sin 0.5 / 7 778.823
H
u R t = =

Maximum vertical displacement:
( ) 1 c o s 0 . 5 / 7 1 2 . 5 3 6
V
u R t = =




Figure 11.6.2 Chattering of vertical displacement



0
2
4
6
8
10
12
14
0 15 30 45 60 75 90
V
e
r
t
i
c
a
l

d
i
p
l
a
c
e
m
e
n
t

Y
A
(
m
m
)
Rotation angle (degree)
midas NFX
Reference(max. value)


11
midas NFX Benchmark Series 11-11
11.7 Two contacting rings

REFERENCE NAFEMS [11-1]
KEYWORDS solid elements
MODEL FILENAME ContactNonlinearStatic07.nfx






Figure 11.7.1 Two contacting rings

Material data
Youngs modulus
Poissons ratio
E = 210 GPa

v = 0.3

Table 11.7.1 Maximum movement of the inner ring
Reference midas NFX
Distance S [mm] 122.5 122.2
max
Y [mm] 16.40 16.45

P = 500 N
t = 11.5
= /2
S
X
Z
Y
Units: mm

Figure 11.7.1 shows two contacting rings with friction constant of 1.0. The outer ring is rigid, constrained in all
directions. The inner ring is subjected to its own weight. The response of the inner ring is obtained by
geometric nonlinear analysis when the inner ring is forced to rotate about its center. The distance along the arc
and the maximum vertical displacement of the center point of the inner ring is compared with the reference
solution.

Contact analysis
11-12

Figure 11.7.2 Vertical displacement as the inner ring rotates



0
2
4
6
8
10
12
14
16
18
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
I
n
n
e
r

r
i
n
g

c
l
i
m
m
i
n
g

d
i
s
t
a
n
c
e

(
m
m
)
Rotation of inner ring (rad)
midas NFX
Refrence


13
midas NFX Benchmark Series 11-13
11.8 Crushing of a pipe

REFERENCE Peech et al. [11-4]
KEYWORDS solid elements
MODEL FILENAME ContactNonlinearStatic08.nfx






Figure 11.8.1 Pipe section and its deformed shape

Material data
Youngs modulus
Poissons ratio
E = 27.0 10
6
lbf/in
2
v = 0.3

Table 11.8.1 Plastic curve
Strain Stress [psi]
0.00000 35000
0.00350 40000
0.00830 43750
0.01330 46250
0.01820 50000
0.02810 52500
0.03800 55000
1.00000 215000
Units: in
L = 1.0
t = 0.349
outer dia. = 4.5
Figure 11.8.1 shows a pipe section as it is crushed. The rigid plates are constrained in all directions. A quarter
of the pipe section is modeled utilizing symmetry and associated symmetric boundary conditions are applied.
The pipe material is von Mises plastic with isotropic hardening. The plastic curve is given in Table 11.8.1.
Reaction force per unit length is determined versus relative anvil displacement.

Contact analysis
11-14

Figure 11.8.2 Reaction force per unit length vs. relative anvil displacement


0
2000
4000
6000
8000
10000
12000
0 0.5 1 1.5 2 2.5 3
F
o
r
c
e

p
e
r

u
n
i
t

l
e
n
g
t
h

(
l
b
f
/
i
n
)
Relative anvil displacement (in)
midas NFX
Reference (Experiment)


15
midas NFX Benchmark Series 11-15
References

[11-1] NAFEMS, Benchmark Tests for Finite Element Modelling of Contact, Gapping and Sliding, Ref .
R0081, NAFEMS, Glasgow, 2001
[11-2] G.M.L. Gladwell, Contact Problems in the Classical Theory of Elasticity, Sijthoff and Noordhoff,
Alphen aan den Rijn, 1980
[11-3] M.L. Ayari and V.E. Saouma, Static and Dynamic Contact/Impact Problems Using Fictitious Forces,
International Journal for Numerical Methods in Engineering, 32, pp. 623-643, 1991
[11-4] J.M. Peech, R.E. Roener, S.D. Porofin, G.H. East, and N.A. Goldstein, "Local Crush Rigidity of Pipes
and Elbows," Proc. 4th SMIRT Conf. paper F-3/8, North Holland, 1977



1
midas NFX Benchmark Series 12-1
12.1 Impact load on a rod by a mass at a constant velocity

REFERENCE Beer et al. [12-1]
KEYWORDS solid elements
MODEL FILENAME ExplicitDynamic01.nfx







Figure 12.1.1 Impacting mass body on a rod

Material data (rod)
Youngs modulus
Poissons ratio
Density
E = 2.1 10
11
N/m
2

v = 0.0
p = 1.0 kg/m
3

Material data (mass)
Youngs modulus
Poissons ratio
Density
E = 2.1 10
11
N/m
2

v = 0.0
p = 10000.0 kg/m
3


V
o
= 20
10
5
1
1
1
1
Units: m
Y
Z
X
A dynamic system consisting of a rod and a mass is shown in Figure 12.1.1. Fixed boundary condition is
assigned to root of the rod. A nonlinear transient analysis with impact loading due to initial velocity of 20
m/sec is performed on the model. The maximum total strain energy, maximum displacement and maximum
normal stress of the rod are determined.

Explicit dynamic analysis
12-2

Figure 12.1.2 Time history of total strain energy


Figure 12.1.3 Time history of axial displacement

0.0E+00
5.0E+05
1.0E+06
1.5E+06
2.0E+06
2.5E+06
0.0E+00 2.5E-03 5.0E-03 7.5E-03 1.0E-02
S
t
r
a
i
n

e
n
e
r
g
y

(
N
-
m
)
Time (sec)
midas NFX
Reference (max. value)
0.0E+00
2.0E-03
4.0E-03
6.0E-03
8.0E-03
1.0E-02
1.2E-02
1.4E-02
1.6E-02
0.0E+00 2.5E-03 5.0E-03 7.5E-03 1.0E-02
D
i
s
p
l
a
c
e
m
e
n
t

(
m
)
Time (sec)
midas NFX
Reference (max. value)


3
midas NFX Benchmark Series 12-3

Figure 12.1.4 Time history of normal stress

Table 12.1.1 Maximum strain energy, displacement and normal stress
Theory midas NFX
Max. strain energy [Nm] 1.999 10
6*
1.854 10
6

Max. displacement [mm] 13.80
*
13.12
Max. normal stress [Pa] 289.7 10
6*
287.4 10
6


*
maximum strain energy:
S L K
E K E = where
K
E is kinematic energy and
L
K is impact energy loss factor
2
1 1
3 2
rod rod
L
mass mass
m m
K
m m
| | | |
= + +
| |
\ . \ .

*
maximum displacement:
max
2
S
E L EA o =
*
maximum normal stress:
2
max
2
S
mv E AL E E AL o = =


0.0E+00
5.0E+07
1.0E+08
1.5E+08
2.0E+08
2.5E+08
3.0E+08
3.5E+08
0.0E+00 2.5E-03 5.0E-03 7.5E-03 1.0E-02
N
o
r
m
a
l

s
t
r
e
s
s

(
N
/
m
2
)
Time (sec)
midas NFX
Reference (max. value)

Explicit dynamic analysis
12-4
12.2 Flexure of a deep beam

REFERENCE Flanagan et al. [12-2]
KEYWORDS solid elements
MODEL FILENAME ExplicitDynamic02.nfx







Figure 12.2.1 A deep beam model

Material data
Youngs modulus
Poissons ratio
Density
E = 1.0 10
9
N/m
2
v = 0.0
p = 1000 kg/m
3


Table 12.2.1 Peak response of the vertical displacement at A

A
Z
u [m]
Reference 0.06401
Original mesh 0.06483
Two part mesh with MPC constraint 0.06483
Two part mesh with RBE 0.06483
Two part mesh with tied contacts 0.06483

0.1
0.4
A
line of symmetry
thickness = 0.025
Units: N, m
q = 720000
Figure 12.2.1 shows a simply supported deep beam. A constant pressure load of magnitude 720000 N/m
2
is
applied to the top surface of the beam as a step function in time. Peak response of the vertical displacement of
the center line of the beam (point A) is determined and compared with the reference value. To demonstrate the
usage of various methods of constraints and joining separate meshes in midas NFX, the same response are
obtained with meshes containing multi-point constraints (MPC), rigid body elements (RBE), and tied contacts.
The results are summarized in Table 12.2.1.


5
midas NFX Benchmark Series 12-5

Figure 12.2.2 Time history of vertical displacement of the centerline


-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0.00
0 5 10 15 20
V
e
r
t
i
c
a
l

d
i
s
p
l
a
c
e
m
e
n
t

(
m
)
Time (sec)
midas NFX
Reference
x 10
-3

Explicit dynamic analysis
12-6
12.3 Simply supported thick square plate

REFERENCE NAFEMS [12-3]
KEYWORDS shell elements
MODEL FILENAME ExplicitDynamic03.nfx






Figure 12.3.1 A quarter model of the thick square plate


Material data
Youngs modulus
Poissons ratio
Density
Mass proportional damping
Stiffness proportional damping
E = 210 10
9
N/m
2
v = 0.3
p = 8000 kg/m
3
o = 5.772 sec
-1
| = 6.929 10
-5
sec
Section property Thickness t = 1.0 m



5.0 m
5.0 m
y
q = 1 MN/m
2
0
x y z
u u o = = =
for all nodes
0
z
u =
along all four edges
0
x
o = along edges
0 x = 10 x =
0
y
o =
0 y = 10 y =
and
along edges
and
x
Figure 12.3.1 shows a thick square plate. An explicit transient forced vibration analysis is carried out with 2%
of critical damping in the dominant first mode with analytical frequency value, f =45.897 Hz . The pressure
load of 1.0 MN/m
2
is suddenly applied as a step function in time. Vibration responses including peak
displacement, peak stress and static displacement are determined and compared with those found in the
reference. Quadrilateral shell element meshes with different levels of refinements are used to obtain the
dynamic responses.


7
midas NFX Benchmark Series 12-7
Table 12.3.1 Forced vibration responses
Peak displacement Peak stress Static displacement
p
o [mm]
p
t [sec]

p
o [N/mm
2
]

s
o [mm]
Reference 4.524 0.0108 62.11 2.333
4x4 4.575 0.0109 57.03 2.328
3x3 4.564 0.0112 54.97 2.324
2x2 4.498 0.0121 48.66 2.309


Figure 12.3.2 Peak displacement obtained using 4 4 mesh

Figure 12.3.3 Time history of and static center displacement obtained using 4 4 mesh
0.0109, 4.575
0.00
1.00
2.00
3.00
4.00
5.00
0.000 0.005 0.010 0.015 0.020
C
e
n
t
e
r

d
i
s
p
l
a
c
e
m
e
n
t

(
m
m
)
Total time (sec)
0.0
1.0
2.0
3.0
4.0
5.0
0.0 0.5 1.0 1.5 2.0
C
e
n
t
e
r

d
i
s
p
l
a
c
e
m
e
n
t

(
m
m
)
Total time (sec)

Explicit dynamic analysis
12-8
12.4 Laminated strip under three-point bending

REFERENCE NAFEMS [12-4]
KEYWORDS layered shell elements
MODEL FILENAME ExplicitDynamic04.nfx







Figure 12.4.1 Laminated strip model

Material data
Youngs modulus
Shear modulus
Poissons ratio
Density
Mass proportional damping
Stiffness proportional damping
E
1
=100 GPa

,
E
2
= E
3
=5 GPa
G
12
=3 GPa, G
13
= G
23
=2 GPa
v
12
= v
13
=0.4, v
23
=0.3
p = 0.1 kg/mm
3
o = 8.4 sec
-1

| = 1.0 10
-4
sec


X
Y
0
o
fiber direction 10
X
Z
10 10 15 15
1
10 N/mm
A B
E
C,E
D
0.4
0.1
0.1
0.1
0.1
0.1
0.1
E
D
C
90
o
0
o
90
o
0
o
0
o
90
o
0
o
Units : mm
Figure 12.4.1 shows laminated strip with [0/90/0/90/0/90/0] layup under three-point bending. Transient
dynamic analysis is carried out using layered shell elements to obtain the steady-state response of the laminate.
During the analysis, the inertial effects are damped out to obtain steady state responses. The resulting vertical
displacement and the inplane bending stress are obtained at point E. These values are compared with the results
of linear and geometrically nonlinear static analyses and the reference value. Corner stress is underestimated
with reduced integration elements since stress is only computed at center with these elements.


9
midas NFX Benchmark Series 12-9

Figure 12.4.2 Time history of the vertical displacement at point E

Table 12.4.1 Bending stress
11
o and Z deflection
Z
u

at point E obtained using layered shell elements
*
obtained using layered quadrilateral elements with reduced integration


-2
-1.5
-1
-0.5
0
0 0.4 0.8 1.2 1.6 2
V
e
r
t
i
c
a
l

d
i
s
p
l
a
c
e
m
e
n
t

(
m
m
)
Time (sec)

E
11
o

[MPa]

E
Z
u

[mm]
Reference 683.9 -1.06
Analysis type
Linear static 679.3 (625.9*) -1.05
Nonlinear static 674.0 (623.1*) -1.05
Explicit dynamic 622.8* -1.05

Explicit dynamic analysis
12-10
12.5 Clamped cylindrical panel under impulsive load

REFERENCE Morino et al. [12-5]
KEYWORDS shell elements
MODEL FILENAME ExplicitDynamic05.nfx







Figure 12.5.1 Explosively loaded clamped cylindrical panel model

Material data
Youngs modulus
Poissons ratio
Density
Perfect plasticity
E = 72.4 10
3
N/mm
2
v = 0.33
p = 2.672 10
-6
kg/mm
3
Y
o = 303 N/mm
2

Section property Thickness t = 3.18 mm

Table 12.5.1 Maximum deflection along the center line
X
Y
60
R = 74.6 mm
30
Inward V
0
= 143.51 m/sec
L = 319 mm
t = 3.18 mm
symmetry
clamped
clamped
clamped
X
Y
Z

Y
u

[mm]
Experiment 31.8
QUAD-4
8 x 16 26.9
16 x 32 31.0
32 x 64 31.8
A clamped cylindrical shell panel is exposed to sudden velocity condition to simulate explosive load by
detonation. As depicted in Figure 12.5.1, half of the panel is discretized by quadrilateral shell elements with
symmetric condition imposed on one end. A von Mises elastic, perfectly plastic material model is used. The
maximum deflection along the center line obtained using several meshes with different levels of refinement are
compared with the reference value.


11
midas NFX Benchmark Series 12-11
12.6 Impact of a rod on rigid wall

REFERENCE Karaoglan et al. [12-6]
KEYWORDS solid elements
MODEL FILENAME ExplicitDynamic06.nfx







Figure 12.6.1 A rod impact model

Material data
Youngs modulus
Poissons ratio
Density
Hardening Modulus
Yield Stress
E = 17 msi

v = 0.35
p = 0.3224 lbm/in
3
T
E =14.5 ksi
Y
o = 58 ksi


Table 12.6.1 The deformed length and radius at t=80 usec after impact



L = 1.27 in, r = 0.126 in
Length [in]

Radius [in]
Reference 0.84 0.28
midas-NFX 0.84 0.26
Figure 12.6.1 shows a cylindrical rod model to simulate a high velocity impact event in which the cylindrical
rod collides with a rigid wall. The collision is modeled by imposing zero axial displacement prescribed at one
end of the rod, while imposing initial axial velocity of 8937 in/sec to all other nodes. A von Mises elastic-
perfectly plastic material model with isotropic hardening is used. The length and radius obtained 80 micro-
seconds after impact is obtained and compared with the reference values. The initial mesh and deformed
shapes at 40 and 80 micro-seconds after impact are shown in Figure 12.6.2.

Explicit dynamic analysis
12-12


Figure 12.6.2 Deformed shape of the rod at t=0, 40 and 80 usec




13
midas NFX Benchmark Series 12-13
12.7 Frictional contact between pad and rotating drum

REFERENCE Wriggers [12-7]
KEYWORDS shell and solid elements
MODEL FILENAME ExplicitDynamic07.nfx











Figure 12.7.1 Frictional braking of a rotating rigid body

X
Y

Rigid drum :
R = 200 mm

0
= 60 rad/sec
p = 0.350 MPa
Brake pad :
100 100 50 mm
3
Friction :
= 0.10, 0.15, 0.20
Solid pad with hyperelastic material model is in contact with a rotating rigid drum with initial angular velocity
of
0
e

as depicted in Figure 12.7.1. Three values of frictional coefficients ( u ) of 0.1, 0.15 and 0.2 are
considered. With normal pressure, p applied on the solid pad, the rigid drum is brought to rest by frictional
contact. Incorporating simplified analytic dynamic model for the pad-drum system, time required for the rigid
drum to come to rest

t A are 0.15, 0.1 and 0.075 seconds, respectively, corresponding to three values of
frictional coefficients.
Time required for drum to come to rest is obtained by
0
/ t e o A = where o

is the angular acceleration
obtained by dividing torque by rotational inertia, / T I o =

and is the torque exerted by the pad represented by
the frictional force multiplied by the radius,
n
T f R u = . Amount of rotation the rigid drum undergoes during
the braking event can also be found to be
2
0
0.5 / u e o = .

Explicit dynamic analysis
12-14
Material data (drum)
Youngs modulus
Poissons ratio
Density
Rotary inertia
E = 2.0 10
7
N/m
2

v = 0.0
p = 2330 kg/m
3
I = 0.175 kg m
2

Material data (pad)
Deviatoric strain energy coefficients

Volumetric strain energy coefficients
Density
C
10
= 0.552 10
6
N/m
2
C
01
= 0.138 10
6
N/m
2
D
1
= 6.89655172 10
6
N/m
2
p = 5000 kg/m
3

Section property Thickness (drum) t = 0.01 m


Figure 12.7.2 Time history of the rigid drum rotation


Figure 12.7.3 Time histories of the rigid drum reaction force (=0.20)
0
1
2
3
4
5
0.00 0.05 0.10 0.15 0.20
R
o
t
a
t
i
o
n

(
r
a
d
)
Total time (sec)
midas NFX
Reference
= 0.10
= 0.15
= 0.20
-12000
-9000
-6000
-3000
0
3000
0.00 0.05 0.10 0.15 0.20
R
e
a
c
t
i
o
n

f
o
r
c
e

(
N
)
Total time (sec)
Tx
Ty
Reference
= 0.20


15
midas NFX Benchmark Series 12-15

Figure 12.7.4 Time history of the rigid drum angular velocity


-20
0
20
40
60
0.00 0.05 0.10 0.15 0.20
A
n
g
u
l
a
r

v
e
l
o
c
i
t
y

(
r
a
d
/
s
e
c
)
Total time (sec)
midas NFX
Reference
= 0.10
= 0.15
= 0.20

Explicit dynamic analysis
12-16
12.8 Pre-tensioned belt on a rigid drum

REFERENCE Bhavikatti et al. [12-8]
KEYWORDS membrane elements
MODEL FILENAME ExplicitDynamic08.nfx








Figure 12.8.1 Three-dimensional belt on a rigid drum


Material data (drum)
Youngs modulus
Poissons ratio
Density
E = 2.1 10
11
N/m
2

v = 0.0
p = 10000 kg/m
3

Material data (belt)
Youngs modulus
Poissons ratio
Density
Mass proportional damping
E = 3.5 10
11
N/m
2

v = 0.2
p = 7500 kg/m
3

o = 1256 sec
-1

Section property
Thickness (drum)
Thickness (belt)
t = 0.009 m
t = 0.009 m
X
Z

F
B
= 50000 N
Friction :
= 0.20
Rigid drum :
R = 500 mm
Belt : w=100, t=9 mm

const
= 20 rad/sec
F
A
An elastic belt is wrapped around a rigid drum as depicted in Figure 12.8.1. One end of the belt is fixed and
tension force of 50kN is applied on the other end. Steady-state resisting moment of the drum is determined
while the drum is turning. The friction coefficient u

between the belt and the drum is 0.2. The resisting
moment is compared with an analytic solution, which is obtained in the following manner:

/ / 50000
A B A
F F F e
u|
= = where |

is the wrap angle in radians. The resisting moment,
R
M can be obtained
by: ( ) 21862
R A B
M F F R N m = = .


17
midas NFX Benchmark Series 12-17

Figure 12.8.2 Time history of the belt reaction moment


-2.5
-1.5
-0.5
0.5
1.5
2.5
3.5
4.5
0.000 0.004 0.008 0.012 0.016 0.020 0.024 0.028 0.032
R
e
a
c
t
i
o
n

m
o
m
e
n
t

(
N

m
)
x

1
0
0
0
0
Total time (sec)
midas NFX
Reference

Explicit dynamic analysis
12-18
12.9 Clamped shallow spherical cap

REFERENCE Beytschko et al. [12-9]
KEYWORDS shell elements
MODEL FILENAME ExplicitDynamic9.nfx







Figure 12.9.1 Shallow spherical cap model and applied step pressure load

Material data
Youngs modulus
Poissons ratio
Density
Perfect plasticity
Isotropic hardening
E = 10.5 10
6
psi
v = 0.3
p = 2.45 10
-4
lbf sec
2
in
-4
Y
o = 2.4 10
5
psi
E
T
= 2.1 10
5
psi
Section property Thickness t = 0.41 in


= 26.67
R = 22.27 in
t = 0.41 in P
time
P
600 psi
Figure 12.9.1 shows the shallow spherical cap problem. The outer boundaries are clamped and a uniform load
is applied over the cap. A quarter of the model is discretized with shell elements utilizing symmetry. A von
Mises elastic, plastic model with linear isotropic hardening is used. Three types of meshes with different levels
of refinement as summarized in Table 12.10.1 are used to determine the time history of center displacement.
The responses are compared with the reference values.


19
midas NFX Benchmark Series 12-19
Table 12.9.1 Meshes used for the spherical cap problem
Level of refinement Number of quadrilateral shell elements
Coarse 75 (5x5x3)
Medium 147 (7x7x3)
Fine 243 (9x9x3)



Figure 12.9.2 Time history of the center displacement

-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6
V
e
r
t
i
c
a
l

d
i
s
p
l
a
c
e
m
e
n
t

(
i
n
)
Total time (msec)
Coarse
Medium
Fine
Reference

Explicit dynamic analysis
12-20
12.10 Hemispherical load

REFERENCE Belytschko et al. [12-10]
KEYWORDS shell elements
MODEL FILENAME ExplicitDynamic10.nfx






Figure 12.10.1 Hemispherical shell model subjected to point loads

Material data
Youngs modulus
Poissons ratio
Density
Perfect plasticity
E = 6.825 10
7
psi
v = 0.3
p = 0.001 lbm in
-3

Y
o = 6.0 10
5
psi
Section property Thickness t = 0.04 in


X
Y
Z
F
X
= 1 lb
F
Y
= -1 lb
0
at x = 0
x y z
u u u = = =
Units: in
R = 10
0
at y = 0
y x z
u u u = = =
Figure 12.10.1 shows a spherical shell subjected to outward point loads on the x-axis and inward point loads on
the z-axis. The shell mesh discretizes one octant of the sphere with symmetry boundary conditions along the
edge. The model is assumed to be composed of perfect plastic material model. Time history of the vertical
displacement at the top point is determined and compared with the results found in the reference.


21
midas NFX Benchmark Series 12-21

Figure 12.10.2 Time history of vertical displacement at the top point


0.00
0.05
0.10
0.15
0.20
0.25
0.00 0.01 0.02 0.03 0.04 0.05
V
e
r
t
i
c
a
l

d
i
s
p
l
a
c
e
m
e
n
t

(
i
n
)
Time (sec)
midas NFX
Reference

Explicit dynamic analysis
12-22
12.11 Twisted cantilever beam

REFERENCE Belytschko et al. [12-10]
KEYWORDS shell elements
MODEL FILENAME ExplicitDynamic11.nfx





Figure 12.11.1 Twisted cantilever beam model

Material data
Youngs modulus
Poissons ratio
Density
E = 2.9 10
7
psi
v = 0.22
p = 2.0 10
-4
lbm in
-3

Section property Thickness t = 0.32 in


Figure 12.11.2 Vertical deflection at loading point
Units: in
12
1.1
t = 0.32

F
Z
= 1 lb
0
0.002
0.004
0.006
0.008
0.01
0.012
0.000 0.002 0.004 0.006 0.008 0.010
T
i
p

d
i
s
p
l
a
c
e
m
e
n
t

(
i
n
)
Time (sec)
midas NFX
Reference
Figure 12.11.1 shows a twisted beam structure. Fixed boundary condition is assigned to root of the beam.
Concentrated load is applied to the other end in z direction in the form of unit step function in time. Time
history of the vertical displacement at the tip is obtained by carrying out explicit transient dynamic analysis.


23
midas NFX Benchmark Series 12-23
12.12 Billet upset

REFERENCE Avitzur et al. [12-11]
KEYWORDS solid elements
MODEL FILENAME ExplicitDynamic12.nfx










Figure 12.12.1 1/8 steel rod model forged between two dies and its deformed shape

Material data
Youngs modulus
Poissons ratio
Density
E = 1.0 10
7
psi

v = 0.33
p = 2.5 10
-4
lbm in
-3


Table 12.12.1 Plastic strain vs. stress
Strain 0.000 0.005 0.010 0.050 0.100 0.200 0.700 4.000
Stress
[psi]
4790 6510 7420 10600 12500 14800 20100 30800
R = 0.75
h = 1.13
v
z
(t)
Units: in
A steel rod is forged between two dies. The finite element mesh of one eighth of the model as depicted in
Figure 12.12.1 is used utilizing symmetric boundary conditions. The Coulomb friction constant is 0.1 between
the steel rod and the die. A von Mises elastic, plastic model (Table 12.12.1) with isotropic hardening is used for
the steel rod. The die compresses the steel rod with velocity given as a half sine curve in time as shown in
Figure 12.12.2. Explicit dynamic analyses are carried out with and without mass scaling to demonstrate the
utilization of mass scaling to accelerate dynamic analysis in which the inertial effects on the solution is
relatively small. By adopting mass scaling with target time step of 0.1 microseconds, identical results can be
obtained as compared to those obtained without mass scaling. Computational cost is effectively cut in half as
evidenced by the number of time increments. Table 12.12.2 compares the computational cost as well as
effective plastic strain and equivalent stress at the end of the analyses.

Explicit dynamic analysis
12-24

Figure 12.12.2 Time dependent velocity of the press


Figure 12.12.3 Time history of the forging height

-900
-800
-700
-600
-500
-400
-300
-200
-100
0
100
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2
V
e
l
o
c
i
t
y

(
i
n
/
s
e
c
)
Time (msec)
-0.9
-0.8
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2
D
i
s
p
l
a
c
e
m
e
n
t

(
i
n
)
Time (msec)
midas NFX
Reference


25
midas NFX Benchmark Series 12-25
Table 12.12.2 Comparison of results obtained from explicit analyses with and without mass scaling (time, 2.2 sec t u = ).
Without mass scaling
With mass scaling
(target 0.1 sec t u A = )
Number of increments 53286 22779
Maximum effective
plastic strain
1.460 1.457
Equivalent stress [psi] 22511 22441



Explicit dynamic analysis
12-26
12.13 Actuated beam connected by revolute joints

REFERENCE Bauchau et al. [12-12]
KEYWORDS bar elements, joint elements, rigid elements
MODEL FILENAME ExplicitDynamic13.nfx










Figure 12.13.1 Actuated beam model


Material data
Beam axial stiffness
Beam bending stiffness
Beam shearing stiffness
Beam mass per unit span

Link axial stiffness
Link bending stiffness
Link shearing stiffness
Link mass per unit span
EA = 91.2 10
6
N
EI = 14.2 10
3
N m
2

GJ = 91.2 10
6
N
A p

= 2.1 kg / m

EA = 44 10
6
N
EI = 0.3 10
3
N m
2

GJ = 14 10
6
N
A p

= 1.6 kg / m


M
T
=25 kg
X
Y
Z
t/s
1.2 1.2
A
Units: m
0.2
0.2 45
o
B
M
T
Figure 12.13.1 shows the actuated beam problem. A cantilevered beam with a tip mass is actuated at the mid-
point M by a crank and link mechanism. The link connects the beam at point M to the crank at point B. The
crank is connected to the ground at point A and rotates about point A with constant angular velocity of t rad/s.
All structural components are free to rotate at the connection points. Therefore, revolute joint elements are
employed to model the joint kinematics. Elastic properties of the beam and the link are summarized in the table
below. The crank is considered to be rigid and modeled using rigid element.
Explicit dynamic analysis is carried out to obtain the transient response of the system during the 2 second
period. Time histories of the vertical and horizontal displacements of the beam tip are compared with those
included in the reference.


27
midas NFX Benchmark Series 12-27

Figure 12.13.2 Time histories of horizontal and vertical displacements at point T.

-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
0 1 2
T
i
p

D
i
s
p
l
a
c
e
m
e
n
t

(
m
)
Time (sec)
midas NFX, Uy
Reference, Uy
midas NFX, Ux
Reference, Ux

Explicit dynamic analysis
12-28
References

[12-1] F.P. Beer and E.R. Johnston Jr., Mechanics of Materials, McGraw-Hill, Inc., New York, 1981
[12-2] D.P. Flanagan and T. Belystchko, "A Uniform Strain Hexahedron and Quadrilateral with Orthogonal
Hourglass Control," J. Comp. Meth. Appl. Mech. Eng., vol. 17, pp. 679-706, 1982
[12-3] NAFEMS, Selected Benchmarks for Forced Vibration, Ref . R0016, NAFEMS, Glasgow, 1993
[12-4] NAFEMS, Composite Benchmarks, Ref . R0031, Issue 2, NAFEMS, Glasgow, 2001
[12-5] L. Morino, J.W. Leech and E.A. Witmer, "An Improved Numerical Calculation Technique for Large
Elastic-Plastic Transient Deformations of Thin Shells: Part 2 - Evaluation and Applications," Journal of
Applied Mechanics, vol. 38, pp. 429-436, 1971
[12-6] L. Karaoglan and G.S. Springer, Axial Impact of Composites, Computers and Structures, Vol. 49, No.
5, pp. 806-823, 1993
[12-7] P. Wriggers, Computational Contact Mechanics, John Wiley & Sons, 2002
[12-8] S.S. Bhavikatti and A.V. Hegde, Problems and Solutions in Engineering Mechanics, New Age
International Ltd., 2005, Rev. 2
nd
ed.
[12-9] T. Belytschko, J.I. Lin, and C. Tsay,Explicit Algorithms for the Nonlinear Dynamics of Shells,
Computational Methods in Applied Mechanics and Engineering, vol. 42, pp. 225251, 1984
[12-10] T. Belytschko and I. Leviathan, Physical Stabilization of the 4-Node Shell Element with One Point
Quadrature, Computer Methods in Applied Mechanics and Engineering, vol. 113, pp. 321-350, 1994
[12-11] B. Avitzur, C.H. Lee and T. Altan, Handbook of Metal Forming Procedures, John Wiley & Sons, pp.
952-954, 1983
[12-12] O. A. Bauchau and C. L. Bottasso, On the design of energy preserving and decaying schemes for
flexible, nonlinear multibody-systems, Computer Methods in Applied Mechanics and Engineering, vol.
169, pp. 61-79, 1999