ExoMars Rover

Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 1 of 25

Company Registration No. 2449259
Registered Office: Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2AS, UK

getFile.jsp




Mechanical FEM Specification
ExoMars Rover Module Sub-Systems


CI CODE: E131000000


DTI EXPORT CONTROL RATING: 9E001, 9E002, 9E004
Rated by: M. Roe

This document is produced under ESA contract 19234/05/NL/CB, ESA export exemptions may therefore apply.
These Technologies may require an export licence if exported from the EU



Prepared by: Date:
J. Michelin

Checked by: Date:
A. Kiley

Approved by: Date:
J. Clemmet

Authorised by: Date:
M. Roe


Astrium Limited 2010

Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any
purpose other than that for which it is supplied and shall not in whole or in part be reproduced, copied, or communicated
to any person without written permission from the owner.

Astrium Limited
Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2AS, England


ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 2 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp



INTENTIONALLY BLANK



ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 3 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp

CONTENTS
1. INTRODUCTION...........................................................................................................................................6
2. ABBREVIATIONS AND ACRONYMS...........................................................................................................6
3. MODEL DATA AND IDENTIFICATION.........................................................................................................7
3.1 GENERAL INFORMATION....................................................................................................................7
3.1.1 Deployed/Stowed configuration ...................................................................................................7
3.1.2 Model identification.......................................................................................................................7
3.1.3 Linearity of the model ...................................................................................................................7
3.1.4 Version of NASTRAN software ....................................................................................................7
3.1.5 Model units...................................................................................................................................7
3.2 BOUNDARY CONDITIONS ...................................................................................................................7
3.2.1 Interface node label......................................................................................................................7
3.2.2 Internal interfaces/joints ...............................................................................................................8
3.3 NUMBERING RANGE............................................................................................................................8
3.3.1 Bulk data card numbering range..................................................................................................8
3.3.2 Bulk data duplicate card...............................................................................................................8
3.3.3 Continuation card .........................................................................................................................8
3.3.4 FEM size ......................................................................................................................................8
3.4 COORDINATE SYSTEM........................................................................................................................9
3.4.1 Reference system........................................................................................................................9
3.4.2 Local reference system................................................................................................................9
3.4.3 Deployable item systems .............................................................................................................9
3.4.4 Cards referring to reference system.............................................................................................9
3.5 NASTRAN PARAMETERS ..................................................................................................................10
3.5.1 Prohibited NASTRAN parameters..............................................................................................10
3.5.2 Required NASTRAN parameters ...............................................................................................10
3.5.3 AUTOSPC parameter.................................................................................................................10
3.5.4 Parameters affecting a set of data .............................................................................................10
3.6 PERMITTED AND PROHIBITED CARDS ...........................................................................................11
3.6.1 Permitted NASTRAN card..........................................................................................................11
3.6.2 Prohibited NASTRAN card.........................................................................................................11
3.7 NODES AND ELEMENTS....................................................................................................................11
3.7.1 Free nodes .................................................................................................................................11
3.7.2 Nodes connecting CELAS..........................................................................................................11
3.7.3 Spring elements .........................................................................................................................12
3.8 PROPERTIES ......................................................................................................................................12
3.8.1 Material orientation.....................................................................................................................12
3.8.2 E, G and relation .....................................................................................................................12
3.8.3 Bar element orientation ..............................................................................................................12
3.8.4 Model mass definition ................................................................................................................12
3.8.5 Structural mass distribution........................................................................................................12
3.8.6 Non-structural mass distribution.................................................................................................12
3.9 THERMAL ............................................................................................................................................13
3.9.1 Model thermo elastic properties.................................................................................................13
3.10 DAMPING...........................................................................................................................................13
3.10.1 Prohibited damping definition...................................................................................................13
3.10.2 Recommended damping definition...........................................................................................13
3.10.3 Symmetry of the damping matrix .............................................................................................13
4. CHECKS......................................................................................................................................................14
4.1 GEOMETRY CHECKS.........................................................................................................................14
4.1.1 Free boundaries .........................................................................................................................14


ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 4 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
4.1.2 Coincident nodes........................................................................................................................14
4.1.3 Element geometry ......................................................................................................................14
4.1.4 Skew...........................................................................................................................................14
4.1.5 Taper ..........................................................................................................................................14
4.1.6 Aspect ratio ................................................................................................................................14
4.1.7 Warp...........................................................................................................................................14
4.1.8 Element co-ordinate systems.....................................................................................................14
4.1.9 Coincident elements...................................................................................................................15
4.2 BOUNDARY CONDITIONS .................................................................................................................15
4.2.1 Magnitude of seismic mass........................................................................................................15
4.2.2 Magnitude of interface load CELAS...........................................................................................16
4.3 MODEL CONDITIONING CHECKS.....................................................................................................16
4.3.1 Over constraint check.................................................................................................................16
4.3.2 Diagonal terms of stiffness matrix..............................................................................................17
4.3.3 DFLR check................................................................................................................................17
4.3.4 Warning messages.....................................................................................................................17
4.3.5 Maxratio......................................................................................................................................17
4.4 FREE-FREE MODES...........................................................................................................................18
4.4.1 Number of free/free modes ........................................................................................................18
4.4.2 Frequencies of rigid body modes...............................................................................................18
4.4.3 SUPORT strain energy ..............................................................................................................18
4.5 MASS PROPERTIES...........................................................................................................................18
4.5.1 Mass according to axes .............................................................................................................18
4.5.2 Mass matrix at G, N, F and A set size........................................................................................18
4.5.3 Centre of gravity.........................................................................................................................18
4.6 STATIC LOAD CHECKS......................................................................................................................19
4.6.1 Force balance.............................................................................................................................19
4.6.2 Enforced unit displacement and rotation....................................................................................19
4.6.3 Epsilon error ratio.......................................................................................................................19
4.6.4 Elements with zero load.............................................................................................................19
4.6.5 Magnitude of SPCFORCE at grids constrained by AUTOSPC. ................................................19
4.7 DYNAMIC CHECKS.............................................................................................................................20
4.7.1 Low frequency acceleration .......................................................................................................20
4.8 SIMPLIFIED MODEL............................................................................................................................20
4.8.1 Simplified model accuracy .........................................................................................................20
4.9 THERMAL CHECKS ............................................................................................................................20
4.9.1 Reference temperature check....................................................................................................20
4.9.2 Thermo elastic analysis suitability checks .................................................................................20
5. ACCURACY OF FEM VERSUS HARDWARE............................................................................................21
5.1.1 Mass, inertia and centre of gravity accuracy..............................................................................21
5.1.2 Dynamic accuracy......................................................................................................................21
6. DELIVERABLES..........................................................................................................................................22
6.1 FILES DELIVERED..............................................................................................................................22
6.1.1 Type of delivered files ................................................................................................................22
6.1.2 Format of delivered files.............................................................................................................22
6.1.3 Files identification.......................................................................................................................22
6.2 NASTRAN INPUT FILES .....................................................................................................................22
6.2.1 NASTRAN input files..................................................................................................................22
6.2.2 Use of ALTER ............................................................................................................................22
6.2.3 NASTRAN version......................................................................................................................22
6.3 NASTRAN BULK DATA .......................................................................................................................22
6.3.1 Use of PARAM...........................................................................................................................22
6.3.2 NASTRAN files for thermo elastic model ...................................................................................22
6.4 ANALYSIS RESULTS ..........................................................................................................................22
6.4.1 Model check results....................................................................................................................22


ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 5 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
6.4.2 Normal mode summary..............................................................................................................22
6.5 MODEL DESCRIPTION REPORT.......................................................................................................23
6.5.1 FEM description .........................................................................................................................23
6.5.2 Report outline.............................................................................................................................23


TABLES

Table 1 Rover Module FEM Numbering...........................................................................................................8
Table 2 Permitted model inputs......................................................................................................................11
Table 3 Prohibited model inputs.....................................................................................................................11


FIGURES

Figure 1 Rover Module Co-ordinate System Alignment ...................................................................................9
Figure 2 Shell element normal checking coordinate systems........................................................................15
Figure 3 Suggested Model Description Report Table of Contents.................................................................23




ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 6 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
1. INTRODUCTION
This document describes a set of requirements for Finite Element Models (FEM) delivered by suppliers to
form part of the Rover Module FEM. Defined are a set of generic FEM requirements for static, dynamic and
thermo elastic analyses and specific requirements for Rover Module sub-systems. The requirements are
based primarily on Astrium internal specifications, with additional ExoMars ESA FEM specifications and
Thales-Alenia FEM requirements taken into account where applicable.

It is expected that sub-system FEMs be supplied as simplified non reduced FEMs (i.e. not Craig Bampton
or Guyan reduction models). This document does not therefore include requirements for reduced models. If
a reduced model is agreed then a further reduced model requirements document will be issued.


2. ABBREVIATIONS AND ACRONYMS
ASU Astrium Stevenage (Rover Vehicle sub-contractor)
CID Co-ordinate System ID
CTE Coefficient of Thermal Expansion
DM Descent Module
DOF Degrees of Freedom
EIDP End Item Data Pack
FE(A/M) Finite Element (Analysis/Model)
RBE Rigid Body Element
RM Rover Module
TAS-I Thales Alenia (prime contractor)
TED Thermo Elastic Distortion




ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 7 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
3. MODEL DATA AND IDENTIFICATION
3.1 GENERAL INFORMATION
3.1.1 Deployed/Stowed configuration
Deployable systems shall be supplied in their stowed and nominal deployed configuration.
3.1.2 Model identification
A finite element model shall be assigned a unique identification name or number and refer to a configuration
of issue where possible. For example this can be a mechanical interface control document of the equipment
including mass, inertia and centre of gravity or an issue of the system mass budget at key phases of the
design, PDR, CDR...
3.1.3 Linearity of the model
A full structural model shall be delivered using only linear elastic elements and properties.
3.1.4 Version of NASTRAN software
All data in any delivered finite element definition shall be compatible with MSC MD NASTRAN Version
2006r1 unless an alternative is agreed with ASU. When an alternative analysis code has been used the
results must be correlated with the delivered NASTRAN model.
3.1.5 Model units
NASTRAN solver and associated pre/post processors have no preference for units, it is the responsibility of
the analyst to ensure consistent units are used in the model.
All models shall be defined using S.I. units:
Newton (N) for force
Kilograms (kg) for mass
Metres (m) for length
Seconds (s) for time
Degrees Celsius (C) for temperature
with the derived units:
E-modulus [Newton/meter]
density [kg/meter
3
]

3.2 BOUNDARY CONDITIONS
3.2.1 Interface node label
All model interfaces shall be representative of the physical attachment points of the items and shall be
modelled as independent points with appropriate degrees of freedom, typically six degrees of freedom.
The label of each interface node shall be clearly identified.

The following comments shall be included within the model description document:
Interface grid points, identification and location,
Boundary condition, clamped degrees of freedom,
Grids included in the ASET or CSUPER when appropriate.



ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 8 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
3.2.2 Internal interfaces/joints
Any internal sub-system interfaces/joints (for example deployment hinges) shall be modelled with CELAS2
elements to allow force recovery.

3.3 NUMBERING RANGE
3.3.1 Bulk data card numbering range
Identification numbers of all NASTRAN cards (e.g. nodes, elements, coordinate systems, material
properties...) shall be in accordance with the numbering range set out in Table 1. Note that some sub-
systems may be the subject of further numbering range sub-division in subsequent issues of this document
(for example service module into frame and PCB boards). A record of identities, quantities of elements and
grids associated with all parts of the FEM shall be kept. This will assist in error tracing during the completion
of constraint and conditioning checks.
ITEM start ID end ID
ROVER MODULE 500000 699999
MAIN TUB STRUCTURE 510000 549999
SOLAR ARRAYS (fixed and deployed) 550000 559999
DRILL BOX ASSEMBLY 560000 569999
CAMERA MAST ASSEMBLY 570000 579999
INSTRUMENT ARM ASSEMBLY 580000 589999
WISDOM RADAR ASSEMBLY (including support bracket) 590000 599999
ANALYTICAL LABORATORY DRAWER 600000 609999
SERVICE MODULE 610000 629999
RADIATORS AND THERMAL 630000 639999
STEERABLE ANTENNA ASSEMBLY 640000 649999
RESERVED (not currently used) 650000 669999
MISC ITEMS 670000 679999
LOCOMATION ASSEMBLY (including wheels) 680000 699999

Table 1 Rover Module FEM Numbering
3.3.2 Bulk data duplicate card
No duplicate node or element identification numbers shall exist in the model.
3.3.3 Continuation card
The continuation card shall be written with + or * symbol. If a number is used in the continuation card, it
must be the card ID number. For instance +E120000 continuation card is only authorised for element n
120000.
3.3.4 FEM size
Sub-system FEMs supplied for integration into the RM shall use only the minimum number of DOFs required
to describe dynamic behaviour of the component at system level analysis; for example all modes likely to
impart significant force on the structure or affect RM behaviour and allow output of acceleration responses
at key equipment locations. Where a detailed model is reduced in complexity for delivery to ASU a
comparative check shall be made as described in section 4.8 .




ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 9 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
3.4 COORDINATE SYSTEM
3.4.1 Reference system
One primary parent rectangular right handed reference system shall be defined by the supplier for the sub
model. This parent system shall be aligned to the RM system (X
RB
, Y
RB
, Z
RB
) illustrated in Figure 1 with
positive X-axis in the direction of forward motion and positive Z-axis normal to (and upward from) the RM
ground plane. Where practical it is preferred that the origin of this system and all child systems be within the
physical boundary of the sub-system.

Figure 1 Rover Module Co-ordinate System Alignment
3.4.2 Local reference system
All local reference systems shall be related to this primary reference system and in general aligned as
closely as possible to the primary RM system.
3.4.3 Deployable item systems
A separate co-ordinate system hierarchy shall be used for deployable items to allow deployment by re-
orientation of a single system. Where deployment requires orientation at more than one location in stages,
each stage of deployment shall be possible by re-orientation of a single system.
3.4.4 Cards referring to reference system
All NASTRAN cards using reference systems (e.g. GRID, CONM2, CBUSH etc.) shall be defined with
respect to the primary reference system or a local reference system i.e. the use of the default identification
number "0" is not allowed for any card. Where possible it is recommended on the GRID card to use the
same co-ordinate system for reference and displacement.

X
Y
Z


ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 10 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
3.5 NASTRAN PARAMETERS
NASTRAN has available a number of parameters to provide some options on how the model is processed.
3.5.1 Prohibited NASTRAN parameters
The following Parameters shall not to be used:
BAILOUT, -1
MAXRATIO
EPZERO
WTMASS
MECHFIX,YES
The use of other default values in version 2006 shall be agreed with ASU.
3.5.2 Required NASTRAN parameters
The following parameters shall be used with the specified value:
BAILOUT, 0
K6ROT, 1.0
SNORM, 20.0
MECHFIX, NO
AUTOSPC, YES

To verify the use of K6ROT the following normal modes sensitivity analyses shall be performed:

K6ROT
1 Sensitivity using K6ROT *10
-1
0.1
2 Baseline value 1.0
3 Sensitivity using K6ROT *10 10.0

Acceptance criterion: frequency shift and modal mass of paired modes 0.1% compared to baseline value.
This means that the shift in frequency when modifying the K6ROT value should be lower than 0.1% taking
into account possible mode re-ordering.

The use of other default values in version 2006 shall be agreed with ASU.
3.5.3 AUTOSPC parameter
When a permanent single point constraint set has been generated from AUTOSPC,YES this set must be
kept separate from the model boundary constraint set (i.e. different SPC card, same SPC set allowed) and
is part of the definition of the model so must be provided to ASU.
3.5.4 Parameters affecting a set of data
Any input specification, which can affect the data of other models when merged, shall not be used. For
example assigning a value of greater than 0 for PARAM, COUPMASS invokes the creation of the coupled
mass matrix which can lead to different frequencies compared to the use of the lumped mass matrix.



ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 11 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
3.6 PERMITTED AND PROHIBITED CARDS
3.6.1 Permitted NASTRAN card
The following NASTRAN cards are permitted:
COORDINATE SYSTEMS CORD2R, CORD2C, CORD2S
NODES GRID, SPOINT
1D ELEMENTS CROD, CBAR, CBEAM
2D ELEMENTS CTRIA3, CQUAD4
3D ELEMENTS CPENTA, CTETRA, CHEXA
CONCENTRATED MASSES CONM2
INTERFACES SPRINGS CELAS1, CELAS2, CBUSH (CELAS2 preferred)
RIGID ELEMENTS MPC, RBAR & RBE2, RBE3 (RBE2 preferred over RBAR)
OTHERS PLOTEL, DMIG, GENEL
BOUNDARY CONDITIONS SPC, SPC1
PROPERTIES Any linear properties
Table 2 Permitted model inputs
3.6.2 Prohibited NASTRAN card
The following bulk data input is prohibited:

MSGMESH commands MSGMESH commands used to create model bulk data input.
COORDINATE SYSTEMS CORD1x
NODES EPOINT
1D ELEMENTS BAROR
2D ELEMENTS CTRIA6, CQUAD8
RIGID ELEMENTS*
Non zero length RBAR, RBE2 in models for thermal distortion
analysis unless using a CTE value
PROPERTIES Non linear properties, structural damping
OTHERS GRDSET
Table 3 Prohibited model inputs

3.7 NODES AND ELEMENTS
3.7.1 Free nodes
Free nodes (nodes not associated to a particular element) are prohibited.
3.7.2 Nodes connecting CELAS
To avoid possible over constraint, ill conditioning and erroneous load paths, grids connected by CELAS
elements shall:
be defined in the same co-ordinate system
reference the same output co-ordinate system


ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 12 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
3.7.3 Spring elements
CELAS elements shall be labelled in accordance with the associated dof of grids e.g. CELAS connecting dof
1 of two grids shall be labelled as xxxxx1, CELAS connecting dof 2 of two grids shall be labelled as
xxxxx2

If the CELAS is connecting two SPOINT, it shall be labelled xxxxx0.


3.8 PROPERTIES
3.8.1 Material orientation
Material orientation and its consistency shall be taken into account if the following conditions apply to FE
models:
composite materials are modelled with anisotropic material properties,
loads are to be applied to the FEM with respect to the material axes,
element forces or stresses are to be extracted with respect to material axes.
3.8.2 E, G and relation
For MAT1, only two of the three values of E, G and shall be filled except if this is coherent with the relation
:
( ) +
=
1 . 2
E
G
This should eliminate warning message 2251 associated with inconsistent values for E, G and .
3.8.3 Bar element orientation
The orientation of linear elements, bars and beams, require the definition of an orientation vector. The
orientation shall be defined by the specification of components of the orientation vector. The use of a grid
point to define the element orientation is not permitted.
3.8.4 Model mass definition
The FEM mass and mass distribution shall be consistent with the predicted mass (M
p
) of the sub-system or
equipment where:

M
p
=M
bee
+M
mm

M
bee
=mass, best engineering estimate
M
mm
=maturity margin

Any deviations from the predicted mass Mp should be documented.
3.8.5 Structural mass distribution
Where practical the density field of the appropriate MAT card shall be used to define structural mass only.
This shall be representative of the material considered.
3.8.6 Non-structural mass distribution
Non-structural masses shall be spread over the structure via NSM field of property cards or as specific
lumped mass using CONM2 elements. Variation of material mass density to simulate distributed non-
structural masses shall be avoided. If used this shall be clearly documented.




ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 13 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
3.9 THERMAL
3.9.1 Model thermo elastic properties
All properties necessary for thermo-elastic analysis applications shall be specified, i.e. Coefficient of
Thermal Expansion and reference temperature. A reference temperature of 20
0
C shall be used. The use of
non-zero length rigid elements (excluding RBE3s) is not permitted for thermal analysis unless RBE2s are
specified with the correct CTE value (i.e. matching the component to which the RBE2 is attached). A single
RBE2 shall not be used to connect two items of differing CTE. Where material CTE differs along each
material axis (for example carbon fibre lay-ups) an average CTE value shall be used.


3.10 DAMPING
3.10.1 Prohibited damping definition
Any viscous damping elements shall be avoided unless they are inherent in the design. Specification of
damping on CELAS, CBUSH and MAT cards is generally prohibited. In exceptional circumstances (with the
agreement of ASU), where material/element based structural damping needs to be specified, the supplier
shall clearly identify and justify the specification of damping on any element or material input data.
3.10.2 Recommended damping definition
When the FEM has been used for dynamic response analysis the supplier shall provide the critical modal
damping versus frequency (NASTRAN TABDMP1 input data), in the technical description document.
3.10.3 Symmetry of the damping matrix
If a damping matrix is provided it shall be a symmetric.



ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 14 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
4. CHECKS
4.1 GEOMETRY CHECKS
4.1.1 Free boundaries
Element free boundaries or edges of the model shall be checked to ensure there are no unintentional splits
in the mesh or missing elements.

4.1.2 Coincident nodes
A check shall be made for non-intentional node coincidence. Node coincidence in an FEM is acceptable
and often intentional, a common example is when spring elements are used to extract loads, and such
springs must be of zero length.

4.1.3 Element geometry
Element geometry checks shall be performed to determine taper, skew angle, warping and aspect ratios.
Elements that violate the generally accepted conditions may not necessarily be incorrect. Generally
accepted conditions for quality checks are as follows,
4.1.4 Skew
Skew is the angle between lines that join opposite mid sides of an element it is recommended that this angle
be greater than 30 degrees. An angle of 90 degrees represents no skew.
4.1.5 Taper
Taper is the ratio of the areas on the two sides of a diagonal. This ratio shall be less than 0.5.
4.1.6 Aspect ratio
Aspect ratio is the ratio of the length of any two sides on a CQUAD4 element. It is recommended that this
value is less than 4.
4.1.7 Warp
Warping of shell elements occurs when the connected grids are not in the same plane. For contoured
surfaces warping shall be generally limited to 5
0
and not exceed 15
0
. Flat surfaces shall be checked for
zero warpage.
4.1.8 Element co-ordinate systems
Element co-ordinate systems shall be consistent where possible including the element normal. Consistent
element co-ordinate systems for elements on the same elementary structure ensure:
Consistent application of the direction of element pressure.
Consistent interpretation of element forces.
Consistent interpretation of element stresses.

Most pre processors have facilities for showing element co-ordinate systems and normals.


ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 15 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp

Figure 2 Shell element normal checking coordinate systems
4.1.9 Coincident elements
Element coincidence shall be checked in all pre processors. Such errors will not be highlighted by
constraint and conditioning checks and typically occur when creating multiple meshes. A clear inventory of
element usage, numbering and quantity, as well as the use of groups can minimise the occurrence of these
errors.

4.2 BOUNDARY CONDITIONS
4.2.1 Magnitude of seismic mass
For sine and random analysis the large (seismic) mass method is often used. A mass much greater than
the mass of the structure being excited is used to achieve fixed base modes of vibration and to provide a
scaling factor for base force input.
The magnitude of the seismic mass required to represent the fixed base condition shall be verified by
comparing the frequencies and effective modal mass obtained from processing the model with fixed base
boundary conditions constrained defined by SPCs and with the seismic mass with SUPORT specification.
f
1


M
m
k


m
k

f
2


The differences between the eigenfrequencies (f
1
, f
2
) of the two models shall be less than 0.1%.
As a guide, the seismic mass should be of the order of 10
5
to 10
7
times the model mass and inertias about
the excitation reference point.
Failing
Element


ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 16 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
4.2.2 Magnitude of interface load CELAS
Supplementary stiff CELAS elements are required at interface points in order to recover interface loads.

The magnitude of the stiffness coefficients of CELAS elements shall be determined by comparing the
frequencies and effective modal mass obtained from processing the model with or without this
supplementary stiffness:
f
1


m
k
K


m
k

f
2


The difference between the eigenfrequencies (f
1
, f
2
) of the two models shall be less than 0.1%.
It can be generally achieved with a stiffness K of 10
10
N/m for translational dof and 10
8
N.m/rad for rotational
dof. Model conditioning checks shall consider these elements.


4.3 MODEL CONDITIONING CHECKS
4.3.1 Over constraint check
The purpose of the constraint check is to verify the model includes no automatically created or unintentional
constraints that will have an effect on the internal loads or rigid body behaviour.
This verification can be made with the NASTRAN GROUNDCHECK parameter (available since version
2001 of MSC/NASTRAN). The model is constrained by specification of sufficient degrees of freedom to
eliminate rigid body motion via a SUPORT bulk data input. Where possible the constraint conditions should
be representative of those to be used in subsequent analyses.
The results of this check reported in the f06 file include four matrices, called KRBi. These are calculated
from the stiffness matrix and a rigid body vector based on the geometry of the model:
[KRBi]= []
T
[K] [] (Joules)
Where [K] is the stiffness matrix and [] a rigid body mode vector calculated from the model geometry. This
product is equivalent to twice the strain energy.
Matrix KRBi are performed at various model sizes:
G set (KRBG matrix): all structural degrees of freedom,
N set (KRBN matrix): all structural degrees of freedom after implementation of multi-point constraint
relationships,
F set (KRBF matrix): all unconstrained (free) degrees of freedom, after applications of all constraints,
A set (KRBA matrix): analysis set.
The resultant matrices KRBi are equal to twice the strain energy in the structure following its motion as a
rigid body. With SI units the six terms on the diagonal of the respective energy matrix, KRBi, shall be:
G set: less than 1.0E
-2
J
N set: less than 1.0E
-2
J
F set: less than 1.0E
-2
J for translational dof and 5.0E
-2
J for rotational dof
A set: less than 1.0E
-2
J for translational dof and 5.0E
-2
J for rotational dof


ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 17 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
In the event the supplier is unable to meet these limits, ASU will require a justification for the reasons why
these limits are not met with a sound rationale for the acceptability of the FE model. The scope of
justification depends primarily on the magnitude of deviation between the achieved 2*Strain Energy levels
versus the above limits.

4.3.2 Diagonal terms of stiffness matrix
Warning messages associated with inconsistent material properties shall be investigated. The material
properties are used to compute the stiffness properties associated with the degrees of freedom of the
element. NASTRAN for example uses only Youngs Modulus for axial and bending stiffness and shear
modulus for the torsional stiffness of linear elements; Poissons Ratio is not used. For shell elements
Youngs Modulus and Poissons Ratio is used for in plane stiffness properties and shear modulus only for
transverse shear.

Printing the diagonal items of the stiffness matrix may be useful to assess the potential ill conditioning of
inconsistent material properties.

4.3.3 DFLR check
The NASTRAN procedure for this conditioning check is based on this method of determining the existence
of singularities. The F-set size stiffness matrix is partitioned by use of a SUPORT card as follows:

=
RR RFR
FRR FRFR
FF
K K
K K
K
It is the sub-matrix K
FRFR
which is decomposed and so any singularities in this matrix are associated with
soft areas within the structure. The degrees of freedom chosen for the R-set by entering them on the
SUPORT card should be those which most closely represent the true support conditions of the structure.
However only six degrees of freedom should be used in order to give a statically determinate support.
There are two matrices to be studied as a result of this check designated DLFR and MECHFR (equivalent to
MAXRATIO). Firstly the matrix DLFR outputs the diagonal terms of the decomposed stiffness matrix if they
are less than unity, as a small value indicates a point of low stiffness. Secondly the matrix MECHFR outputs
the ratio between the terms on the leading diagonal of the matrix KFRFR to those on the leading diagonal of
the decomposed stiffness matrix, if the value exceeds 1000. The DOFs identified by MECHFR indicate the
presence of potential mechanisms in the structure.

The accuracy to be achieved from this check shall be as follows:

DLFR Values : > 1.0 E-3 (positive)

4.3.4 Warning messages
Any warning messages occurring in the NASTRAN f06 file shall be reviewed to confirm they do not influence
the results obtained.

4.3.5 Maxratio
The maximum ratio represents the ratio between the highest and the lowest term of the stiffness matrix. If
the value of MAXRATIO is greater than the NASTRAN default value (10
7
in version 2001) an error message
is printed. If the value is less than the default value, this is not reported within the NASTRAN solution
sequence.

The value of MAXRATIO shall be less than 10
7
. In the event the supplier is unable to meet this criterion, a
justification shall be provided.




ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 18 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
4.4 FREE-FREE MODES
The purpose of the "free-free" check is to verify the rigid body modes of the model.
This verification is made with a NASTRAN SOL 103 dynamic analysis with the model in free-free conditions,
i.e. no constraint specified by SPC or SUPORT input.
4.4.1 Number of free/free modes
It shall be verified that the number of rigid body modes is 6. Supplementary rigid body modes (such as
mechanisms) shall be justified on a case-by-case basis.
4.4.2 Frequencies of rigid body modes
The ratio of the highest rigid body frequency to the frequency of the first elastic mode shall be less than 10
-4
.
Furthermore, all rigid body frequencies shall be <0.005Hz.
4.4.3 SUPORT strain energy
This check uses the strain energy to confirm acceptable mathematical conditioning of internal loads of the
model.

This check is made with a classical NASTRAN dynamic analysis (SOL 103) with the model constrained
using SUPORT input data. The strain energy is calculated and given in the "*.f06" file at every support point.
The value of the strain energy shall be less than or equal to 5.0E
-2
J.


4.5 MASS PROPERTIES
4.5.1 Mass according to axes
The mass distribution check performed with respect to the NASTRAN Grid Point Weight Generator
(GPWG) provides:
mass matrix of the structure (M0) at a reference point defined by PARAM GRDPNT,
mass of the structure and position of the centre of gravity with respect to the reference point in the
primary local coordinate system,
inertia matrix at the centre of gravity and in the primary local coordinate system of the reference
point [I(S)] or in the principal axis of inertia [I(Q)],
the reference frame transformation matrix (Q), between the local coordinate system of the reference
point and the principal axis of inertia, the matrix of direction cosines.
Mass figures shall be the same for axes X, Y and Z. The centre of gravity location shall correspond with
design mass estimates or the appropriate ICD to within 5% of the overall linear dimension in each axis. If
this is not the case differences shall be justified.
4.5.2 Mass matrix at G, N, F and A set size
The integrity of the mass matrix shall be checked at the NASTRAN G, N, F and A set size.
The rigid body mass matrix shall be determined:
[MRBi]= []
T
[M] []
Where [M] is the mass matrix and [] a rigid body vector calculated from the model geometry.

4.5.3 Centre of gravity
Diagonal terms of the centre of gravity matrix shall be less than 10
-6
m.
The two other terms (successively for each direction) must be equal.




ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 19 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
4.6 STATIC LOAD CHECKS
4.6.1 Force balance
The purpose of the static load check is to confirm that total forces at the interface of the model balance with
the applied load. The reaction forces for all loads analysed, gravity vector, point load or thermal load, shall
be checked to confirm they are at expected locations and that the forces and moments balance the applied
load.

The application of unit acceleration vector (1 m.s
-2
) independently in each axis shall be used to confirm that
the sum of the constraint forces divided by the applied acceleration equals the model mass.

Interface reaction forces should be computed using the NASTRAN command SPCFORCES=ALL.
NASTRAN provides in the f06 output file the RESULTANT LOAD, the sum of the applied loads at the
GPWG reference point; the default is the origin of the basic coordinate system. The RESULTANT
SPCFORCES, the sum of the SPCFORCES is also provided. For unit acceleration vectors (1 m.s
-2
) the
force must be equal to the mass.

4.6.2 Enforced unit displacement and rotation
A unit enforced displacement in all six degrees of freedom shall be applied at the base of the structure (i.e.
at approximately the centroid of the structure interface points) to verify no over constraint is present in the
model. With the model only constrained at the position of applied enforced displacement the model should
move as a rigid body when it is translated one unit of displacement or one radian of rotation. For a condition
of rigid body motion the element forces, stresses and grid point forces should all be zero.

Displacement results from the three translational subcases shall be 1.0 in the input direction and zero in the
other five directions. For the rotational cases the rotation in the input direction shall be 1.0 and 0.0 in the two
rotational directions. The translations from the rotational cases will not be zero.

4.6.3 Epsilon error ratio
The ratio of work done by residual loads vs. work done by external applied loads shall be less than 1.0E-8
(reported as Epsilon by Nastran).
4.6.4 Elements with zero load
The model constraint checks, conditioning checks and free-free modes checks do not confirm correct
connectivity of all load paths. To verify that valid load paths have not been omitted an applied load
representative of that to be used in analysis (e.g. gravity vector, point load or thermal load) shall be applied
to the model. Appropriate Element output data, loads or stresses, shall be obtained and processed to
identify elements (CELAS, RBE, CTRIA, CQUAD, BAR, BEAM) with zero load. Elements with zero load
or stress shall be reviewed to assess potential loss of connectivity.

4.6.5 Magnitude of SPCFORCE at grids constrained by AUTOSPC.
Forces associated with degrees of freedom constrained by AUTOSPC should tend to zero. In practice some
non-zero constraint forces will exist. The magnitude however shall be reduced to a quantity that produces
negligible errors both to the overall solution and to the model locally.




ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 20 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
4.7 DYNAMIC CHECKS
4.7.1 Low frequency acceleration
For dynamic analyses using modal superposition, it shall be verified that the behaviour at low frequencies is
correct. The application of unit acceleration vector (1 m/s
-2
) at low frequencies (near 0 Hz) independently in
each axis shall be used to confirm that the sum of the constraint forces divided by the model mass is at least
equal to 95 % of the applied acceleration.

Values less than 95 % indicate insufficient modes in the analysis. It can be corrected with the PARAM,
RESVEC or with more modes.


4.8 SIMPLIFIED MODEL
4.8.1 Simplified model accuracy
In cases where the FEM supplied to ASU for RM integration is a simplified (reduced number of elements)
version of a more detailed sub-system FEM the following additional check shall be performed to ensure it is
sufficiently representative. The results from a fixed base normal modes analysis of the simplified model
shall be compared against the detail model and the following criteria met:

Major mode frequencies (effective mass >5% rigid item mass) shall be within 3%
Effective masses of major modes shall be within 5%
All other mode frequencies up to and including the tenth flexible mode shall be within 5%
Effective masses of all other modes up to and including the tenth flexible mode shall be within 10%

Mass and other model checking criteria shall conform to the requirements outlined throughout section 4.


4.9 THERMAL CHECKS
To check that the model is suitable for thermal distortion analysis the following assessments shall be
performed.
4.9.1 Reference temperature check
To check for correct reference temperature definitions a uniform temperature of 20
o
C shall be applied to the
model and results checked to ensure all element stresses/loads and nodal displacements are numerically
zero.
4.9.2 Thermo elastic analysis suitability checks
The linear expansion coefficient for all the material properties in the model shall be assigned a value of 10
-5

m/m/
o
C and a reference temperature of 20
o
C. A uniform temperature change, T=100
o
C, shall be applied
to the model with the model iso-statically constrained.

The maximum rotation in the model shall be less than or equal to 10
-7
rad. The maximum Von Mises
stresses in the model shall be less then or equal to 100 Pa.

Furthermore with the model material thermal properties set to their correct (model delivery) values a uniform
temperature change, T=100
o
C, shall be applied to the model as described above. An inspection of the
FEM thermo elastic distortion and stresses shall be made to ensure no unexpected behaviour occurs
resulting from incorrect CTE assignments.





ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 21 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
5. ACCURACY OF FEM VERSUS HARDWARE
5.1.1 Mass, inertia and centre of gravity accuracy
The accuracy of the model versus the equipment hardware (*) shall be:
1% for the mass,
5% for the moments of inertia,
1% of the maximum distance for the centre of gravity versus substructure I/F.
(*) The reference for accuracy should be the appropriate ICD relevant to the project or a delivered
system budget. The latter should be taken as the reference in the absence of weighed data e.g.
early study work, PDR stage etc. On the production of hardware the reference for accuracy should
be the weighed item.
5.1.2 Dynamic accuracy
Dynamic accuracy of the FEM versus the hardware shall be computed when experimental modal testing has
been performed. Error shall be lower than:
20 % for static or thermo-elastic displacements,
5 % for main frequencies
(1)
of the item in test configuration,
20 % for significant responses performed during tests.
(1)
Main frequencies means that the model effective mass has more than 5 % of the rigid mass of the item
involved.



ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 22 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
6. DELIVERABLES
6.1 FILES DELIVERED
6.1.1 Type of delivered files
The model files to be delivered shall be:
NASTRAN input files,
NASTRAN bulk data,
modal analysis with representative boundary condition (.f06 summary of eigen-frequency listing etc),
6.1.2 Format of delivered files
All data files shall be in ASCII format.
6.1.3 Files identification
The model name, issue and date shall be clearly specified.
A brief summary of all delivered files shall be provided.

6.2 NASTRAN INPUT FILES
6.2.1 NASTRAN input files
NASTRAN input files (containing the executive and case control section) shall be supplied for all analysis
types. These shall be supplied separate to the accompanying bulk data files.
6.2.2 Use of ALTER
If NASTRAN written DMAP ALTERS are used the name and all used parameters shall be supplied and
described.
6.2.3 NASTRAN version
The version and vendor of NASTRAN used shall be declared in the header of the file.

6.3 NASTRAN BULK DATA
6.3.1 Use of PARAM
All the PARAM cards required to run the model shall be included.
6.3.2 NASTRAN files for thermo elastic model
For the thermo elastic model the complete set of bulk data cards for the zero stress test (section 4.9.2 )
shall be delivered.

6.4 ANALYSIS RESULTS
6.4.1 Model check results
Extracts from the NASTRAN f06 output files appropriate to the model checks shall be reported in the model
description report.
6.4.2 Normal mode summary
A table showing at least the first ten elastic modes (from a fixed base normal modes analysis) shall be
provided including frequencies, effective masses and brief mode shape descriptions.



ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 23 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp
6.5 MODEL DESCRIPTION REPORT
6.5.1 FEM description
The finite element model description report shall be written in English.
6.5.2 Report outline
The FEM report shall define as a minimum the model designation, properties, mass distribution, modelling of
all parts that the integrator needs to reference, attach to, adapt on integration to system level and results of
the model checks performed.

Figure 3 provides a suggested list of the subjects in the model description report.

A sketch of the co-ordinate system(s) associated with the primary appendages should be provided. In
general, all input data used for the model shall be clearly documented. Detailed plots of the model clearly
showing all nodes, elements, and connectivity of primary interfaces (and relevant numbering and types)
shall be supplied.

Compliance with all requirements specified in this document shall be declared.


LIST OF ACRONYMS
LIST OF SYMBOLS
1. SCOPE
2. DOCUMENTS
2.1 APPLICABLE DOCUMENTS
2.2 REFERENCE DOCUMENTS
3. DESCRIPTION
3.1 OVERALL DESCRIPTION
3.2 DESIGN LEVEL
3.3 MASS AND COM BUDGET
4. FE MODEL
4.1 FEM CODE AND PRE/POST-PROCESSORS
4.2 FEM UNITS
4.3 FEM CO-ORDINATE SYSTEMS
4.4 FE MODEL DESCRIPTION
4.4.1 Assumption and idealisations
4.4.2 Geometry (node and element numbering)
4.4.3 Interfaces
4.4.4 Boundary conditions
4.4.5 Properties and materials
4.4.6 Mass distribution summary
4.4.7 Other special modelling features
5. FE/ MODEL CHECKS
5.1 Gravity load check
5.2 Strain energy check
5.3 Rigid-body frequency check
5.4 Enforced displacement check
5.5 Thermo-elastic check
5.6 Modal Behaviour Summary

6. COMPLIANCE WITH FEM REQUIREMENTS

Figure 3 Suggested Model Description Report Table of Contents




ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 24 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp






INTENTIONALLY BLANK



ExoMars Rover
Vehicle
EXM.RM.RQM.ASU.0026
Issue A
Page 25 of 25



Astrium Limited owns the copyright of this document which is supplied in confidence and which shall not be used for any purpose other than that for which it is supplied and shall not
in whole or in part be reproduced, copied, or communicated to any person without written permission from the owner.

getFile.jsp

DOCUMENT CHANGE DETAILS

ISSUE CHANGE AUTHORITY CLASS RELEVANT INFORMATION/INSTRUCTIONS
A - - Initial Issue



DISTRIBUTION LIST


INTERNAL


EXTERNAL

A. Kiley G. Giordano (TAS-I)
R. Slade
M. Roe
J. Clemmet
J. Michelin
Configuration Management