Professional Documents
Culture Documents
AM2036
AM2036
PURPOSE:
This document shall form the basis for the preparation and exchange of major component FE models
within Airbus, and will be applicable to all future analysis structures and projects. The intention is to create
a common FE model of a complete aircraft structure by assembling the component parts as supplied,
without the need to alter FE interface details, and which will analyse first time without the need to re-work
the FE structure.
The model obtained will provide the starting point for all future analysis work and for all engineering
disciplines, thus avoiding duplication of effort.
This specification has been prepared by representatives of each site of Airbus.
SCOPE:
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Issue: E Page 1 of 68
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Table of contents
1 Objective.................................................................................... 3
8 Dynamic Model........................................................................ 19
APPENDIX C........................................................................................................... 64
Glossary ................................................................................................................. 67
Approval ................................................................................................................. 67
Issue: E Page 2 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
1 Objective
This specification shall form the basis for the preparation and exchange of major
component FE models within AIRBUS, and will be applicable to all future analysis
structures and projects.
The intention is to create a common FE model of a complete aircraft structure by
assembling the component parts as supplied, without the need to alter FE interface
details, and which will analyse first time without the need to re-work the FE structure.
This model will provide the starting point for all future analysis work and for all
engineering disciplines, thus avoiding duplication of effort.
The specification also identifies common validation procedures to assure that all the
analysis work undertaken by Airbus continues to be of the highest quality consistent with
world class design organisations.
The correct application of AM2036 will not guarantee that the model results are
acceptable to the project, only that the results will reliably reflect the response of the
structure and will not be affected by analysis anomalies.
This specification will be subject to review on a regular basis.
Issue: E Page 3 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Issue: E Page 4 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Issue: E Page 5 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Issue: E Page 6 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Issue: E Page 7 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Issue: E Page 8 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Issue: E Page 9 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Issue: E Page 10 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Issue: E Page 11 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
B. Mass model QA: The mass model should be verified by producing XY ~ XZ plots
using scaled mass points. Additionally the normal modes check defined in the
attached table should also be carried out using a mass case that includes payload,
fuel, and both structural and systems masses (M1 or M2).
Notes: → M1 = artificial mass case to test every mass and RBE3 connections.
→ M2 = realistic mass case.
Continued…
Issue: E Page 12 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
C. The following verification checks will be required to be undertaken with the model
using CONMx cards:
C.1. Normal modes analysis with mass model (using CONMx cards): The normal
modes should be calculated for M1 or M2 mass conditions: see table 5-1. Items 4
and 5.
C.2. Normal modes analysis with mass model (using CONMx cards) in A-set: The
normal modes should be calculated for a M2 mass status, using the A-set for the
conditions (∗) specified: see table 5-1 Item 6.
Normal modes assessment tests are defined as follows:
Boundary
CHECK Test Masses Method Frequencies Target
Conditions
1 Densities in Lanczos [ -1, 1] free/free 6 and only 6
MATx cards with actuators rigid body modes
8
(Kact ~ 10 N/m) (RBM)
2 Densities in Lanczos [ 0, 1] free/free 6 rigid body modes
MATx cards without actuators plus 0 Hz rotations
3 Densities in Lanczos [ 1, ∞] free/free local or unrealistic
MATx cards 15 first with actuators modes detection
modes
And for the Mass-model
Boundary
CHECK Test Masses Method Frequencies Target
Conditions
4 RBE3 and Lanczos [ -1, 1] free/free 6 and only 6
CONMx with actuators rigid body modes
(M1) (Mass model
verification)
And for the Dynamic model
Boundary
CHECK Test Masses Method Frequencies Target
Conditions
5 RBE3 and Lanczos [ 0, ∞] free/free Normal modes for
CONMx 15 first structure
(M2) modes (used as reference)
6 RBE3 and MGIVENS [ 0, ∞] free/free Complementary
CONMx or MHOU 15 first (for Analysis set
(M2) and A-set modes verification)
Table 5-1 - List of Normal Modes Checks
Continued…
(∗)
Any alternative to A-set definition should be agreed between all parties.
Issue: E Page 13 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Notes: → Where MSC.Nastran DMAP Alters have been used in any of the QA checks,
this Alter should be declared and agreed between all parties.
→ The highest frequencies for Rigid Body Modes, Checks 1 and 2, should be
as close to 0.0 as possible, but should not be greater than 1/100 of the
magnitude of the lowest predicted flight mechanics frequency for first flexible
mode. Values below 0.01 Hz are desirable; any values above this level must
be discussed between all parties.
Issue: E Page 14 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
E. Complete documentation shall be supplied for each component model. The following
items of documentation shall be included:
E.1. A statement identifying the engineering standard/design status used in the
preparation of the model.
E.2. Overall plots of relevant parts, especially for the interface areas.
E.3. Descriptions of the model Bulk Data deck to enable users to easily identify the data
and definition of the rules to decode the numbering sequence used for the GRID
and element numbering.
E.4. The interface grid points complete with co-ordinate frame, and the boundary
element characteristics for each interface relevant to the component. Where
relevant this interface document should also include an actuator installation sketch.
E.5. Verification that the model geometry checks identified in A.1. have been performed
correctly.
E.6. A brief description of the static test load cases used to verify the model.
E.7. A diagram or other description of the support conditions and a summary of the
generated reaction forces from the static test cases.
E.8. Results from the static test load cases in the form of the OLOAD, Maximum
Displacements and SPCFORCEs resultant tables.
E.9. The maximum Von Mises stresses generated by the "ZERO STRESS" thermo-
elastic test.
E.10. The rigid body modes, frequencies and mode shapes (plots), plus control surface
modes generated from the normal modes tests (items 1 and 2 in table 5-1).
E.11. Tabulated frequencies for flexible normal modes, plus plots of first three flexible
modes (item 3 in table 5-1). These modes should also be identified.
E.12. The energy matrix from the rigid body test (items 3, 4, 5 and 6 in the table 5-1).
E.13. Tabulated frequencies and flexible modes displacement plots for all DOFs involved
in RBE3 mass connections (Item 5 in table 5-1).
E.14. A list of all supplied filenames.
E.15. EI and GJ curves and other specific information (for pylons, Centre Section, etc.)
may be supplied as optional extras.
E.16. MSC.Nastran version and computer description, including Operating System level.
The distributed masses used as part of the tests should also be supplied for use in
the verification of the assembled model only. This data should not be used for any
other purpose.
Continued…
Issue: E Page 15 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
F. The following forms have to be completed and attached to the full documentation of
the FEM:
• FORM 1: "REFERENCES": In this form a FEM overall description is included:
e.g. configuration, geometry/property/material references, drawings (if exists),
document of dimensioning or whichever good reference that identify correctly
the model.
• FORM 2: "FILES RECORD": Describe each one of the datasets that define the
model and the different runs used in FORM 4. Because the main use of this
form is to verify the right transmission of the full component FEM, this form
has to be sent at the same time as the model.
• FORM 3: "MODEL CHECK LIST": Describe all the geometry checks
performed on the model. As well as indicating if the checks have been
performed, give details of any problems that may have been found, along with
maximum/minimum values where appropriate. Where necessary append plots
for clarity (e.g. co-ordinate systems).
• FORM 4: "ANALYSIS CHECK LIST": Describe mainly the results of static and
Rigid Body checks.
• FORM 5: "ANALYSIS CHECK LIST": NORMAL MODES": Describe all the
checks performed with the model using density in the MATx cards.
• FORM 6: "ANALYSIS CHECK LIST: DYNAMIC NORMAL MODES": Includes
the results of all the dynamic checks performed with the model using CONM2
cards.
• FORM 7: "ANALYSIS CHECK LIST: THERMO-ELASTIC STRESS": Includes
the results of the "ZERO STRESS" test.
Issue: E Page 16 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
(∗)
Any alternative to A-set definition should be agreed between all parties.
Issue: E Page 17 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
7 Mass Data
All parties shall provide the locations and the connection system for mass points for their
component(s), for masses subdivided into separate distributions, e.g. Structures,
Systems, etc.
Based on this information, the responsible team shall supply mass data for appropriate
configurations, and shall distribute this data to all other parties.
Mass data will be supplied for the complete aircraft, in kg, m and s using CONM2 cards.
Prior to delivery mass quality assurance checks shall be undertaken including density
plots for the XY~XZ plane.
If the model is to be used for optimisation, it may be necessary for the component
supplier to provide mass distributions for their components equivalent to the mass of the
FEM generated by the density parameter on the material cards. This distribution will be
required to be subtracted from the Fixed Structure mass distribution.
Issue: E Page 18 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
8 Dynamic Model
The team responsible for the assembly of the complete aircraft model shall transform the
model from N and mm to N and m, without any other modification, maintaining the Bulk
Data deck in its original sequence and also maintaining the same data formats: see
table 8-1.
The aeroelastic calculation shall use both the half-aircraft and the full-aircraft models to
properly represent the effects of the non-symmetrical considerations.
The responsible team will then prepare a symmetric half-aircraft model, left side, for the
specific needs of the Aeroelastics and Loads departments.
These aircraft models shall be distributed via a suitable electronic data transfer link. In
addition to the model bulk data, files containing the correct boundary constraint data for
the half model, for both symmetric and anti-symmetric boundary conditions, shall also be
supplied.
These models shall form the basis of any dynamic investigations that may be required.
For the half-aircraft model, mass connection elements shall be taken from those supplied
for the complete aircraft model, with suitable modifications to allow for the splitting of the
model.
Mass linkage to structure: Mass should be assigned to single "non structural points"
(GRIDs), linked to the structure using RBE3 elements connected to a set of "local hard
structural points". Where A-set (*) is to be used there are two possibilities:
a) The A-set should include part of the "local hard points". Therefore, the "non
structural mass points" should be the DEPENDENT nodes on the RBE3.
b) The A-set should include part of the "non structural mass points". Therefore, these
points should be the INDEPENDENT nodes on the RBE3 and unique for all
loading conditions.
(*)
Any alternative to A-set definition should be agreed between all parties
Issue: E Page 19 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
9 Update Procedure
For each complete model, a record of each component model shall be maintained by the
party responsible for the complete model assembly. This record to include the issue
number and the preparation date of the component.
Where relevant changes have been made to any component, a new complete model will
be assembled, verified, documented and distributed via a suitable electronic data
transfer link to all parties as defined elsewhere in this specification.
The following numbering system (in capital letters) will be used for identification of the
complete assembled models and for formal referencing.
a/c model/engine version/design standard/model issue/model type/model size
Where:
• Model type:
• S = Static Model
• D = Dynamic or Aeroelastic Model
• Model size:
• C = Complete Model
• H = Half Model
For example: A3XX100/01/08/01/D/C
Represents the first issue of a complete aircraft model of the A3XX-100 at Status 8, with
engine option 1. The model is for Dynamics usage and will therefore be in N and m.
Where configuration or other studies are undertaken using parts or all of one of these
models, these will be considered as independent activities, and will not be recorded as
part of this numbering.
Issue: E Page 20 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX A
Forms
REFERENCES
PART COMPONENT: A/C COMPONENT: PAGE:
Geometry
Meshing
Properties
Materials
Approved by:
Continued…
Issue: E Page 21 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX A
Forms, continued
QUALITY ASSURANCE OF FEM PROJECT:
FILES RECORD
PART COMPONENT A/C COMPONENT: PAGE:
COMPUTER
DATE OF
DATASET/FILE COMMENTS/DESCRIPTION
CREATION
NAME
Approved by:
Continued…
Issue: E Page 22 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX A
Forms, continued
QUALITY ASSURANCE OF FEM PROJECT:
MODEL SIZE
GRIDs 1-D 2-D 3-D SPRINGs RIGID BODYs
Approved by:
Continued…
Issue: E Page 23 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX A
Forms, continued
QUALITY ASSURANCE OF FEM PROJECT:
Approved by:
Continued…
Issue: E Page 24 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX A
Forms, continued
QUALITY ASSURANCE OF FEM PROJECT:
Approved by:
Dynamic Acceptance:
Continued…
Issue: E Page 25 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX A
Forms, continued
QUALITY ASSURANCE OF FEM PROJECT:
Approved by:
Dynamic Acceptance:
Continued…
Issue: E Page 26 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX A
Forms, continued
QUALITY ASSURANCE OF FEM PROJECT:
Defined Temperature:
Approved by:
Issue: E Page 27 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines
1 Introduction
The objective of this appendix is to clarify the correct application of the recommendations
written in AM2036, and identify areas where it will be necessary to expect definitions of
local project variations.
Although driven by the needs of Loads and Aeroelastics analyses, AM2036 outlines
basic principles of good practice that should be applied to all forms of analysis including
Linear Elastic.
Users of FE Analysis should also consider objectively whether their FEM might possibly
be used as part of an analysis investigation involving dynamic analysis, and if this
possibility exists, should then apply AM2036 rigorously to their model from the
beginning.
AM2036 has been directed towards models for linear static and dynamic applications.
Where additional analysis requirements exist it will be necessary for analysts to consider
additional validation checks based on those extra requirements.
The same physical checks are required to be performed for the assembly of component
models as have been applied to the individual components models. Similarly the data
supply and documentation is also identical. This is in line with Section 6 of the main
Document.
For each of the geometric and analysis checks, this appendix will be provide a best
practice method for performing them. These are only recommendations, the analysts
may use any method that is considered suitable to achieve the required results and
satisfactorily completes the checks.
The results from the assembled model will reflect the behaviour of the whole aircraft
(instead of only the separate components) and should be used as the basis for any
decisions regarding model acceptability or real aircraft characteristics.
Issue: E Page 28 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Continued…
Issue: E Page 29 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Issue: E Page 30 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Issue: E Page 31 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
To be made using the command "Equivalence". This command eliminates, for one list,
group, or the whole model, the nodes located in the same position, with a user defined
tolerance, erasing the undesired free edges. The coincident nodes linked by ELAS,
GAP, MPC, and rigid elements are not eliminated by MSC.Patran. The erased nodes are
highlighted and the command may be reversed with "UNDO".
To display the coincident nodes linked by ELAS and/or GAP elements, you may use the
menu "Utilities/Display/Mark 1D Elements". By putting the elements with length smaller
than a user defined tolerance value in a list, the user may create a list of nodes related
to these elements using the menu "List" command.
Issue: E Page 32 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
The QUAD4 element co-ordinate system used by MSC.Patran to obtain the element
normal is shown in figure App B.5.1.1.4.1.
The command "Finite Element/Verify" is used to check for element distortions. This
command measures the deviation between the actual element and the ideal one. The
worst element distortions, according to the MSC.Patran criteria, must be located outside
the zones with large stress concentration or high stress gradient.
Two parameters are checked separately:
Issue: E Page 33 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
2) Taper (QUAD4)
The taper is calculated according to figure App B.5.1.1.5.3 (ideal element is the
rectangle, taper = 1).
Taper recommended values are between 0.8 and 1.
Continued…
Issue: E Page 34 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Issue: E Page 35 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Issue: E Page 36 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
It must be confirmed that all ELAS elements have zero length and have coincident
nodes. This operation may be combined with the operations described in the section
Coincident Nodes (see 5.1.1.1). The menu "Utilities/Display/Mark 1D Elements" is used
to put elements with null length, i.e. smaller than a user defined tolerance value, into a
group. Using the menu "Utilities/Display/Show/Element info", you may check the
stiffness (K), the co-ordinates of the nodes, the co-ordinate system for displacements of
those nodes, and the degrees of freedom affected for all elements within the group.
MPCs, and Rigid Elements, are equations that define relationships between degrees of
freedom of nodes. It must be confirmed that the relationship between the degrees of
freedom used are correct, and do not produce changes to local load paths due to
erroneous or undesired stiffness. To verify the relationship between degrees of freedom
you may to use the menu "Finite Element/Show/MPC" and "Utilities/Display/Show
MPCs".
The orientation and location of all general and local co-ordinate systems must be
confirmed. Plots must be made showing all model co-ordinate systems used.
Plots showing element connectivities must be obtained to verify they are correct. This
verification is very important to ensure the correct interpretation of MSC.Nastran element
forces and the correct application of pressure loads. Correct connectivity is also required
to guarantee the correct operation of any software used for automated strength analysis.
The command to check connectivities for 2D (TRIA3 & QUAD4) elements is
"Utilities/FEM Elements/Modify 2D Connectivities (Display)". This instruction displays a
vector between the first and second node for each element. The command to display
connectivities for 3D (Solid elements) is "Verify/Element/Connectivity". This instruction
will mark those elements with negative volume.
Continued…
Issue: E Page 37 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Figures showing the element offsets in BAR/BEAM and QUAD4/TRIA3 must be obtained
to verify the correct application in the model. To display the BAR/BEAM with offset, the
function "Display/LBC/Element Properties/Beam Display 1D Line + Offset" option must
be used. To display the offset in nodes as a vector, the function "Properties/Show/Offset
@ Node 1-2" is used. To display the offset in plate elements, the function "Utilities/
Display/Display Plate Offset" is used. The offsets values of concentrated masses
(CONM2 cards) are obtained with the function "Properties/Mass Offset". The CONM2
cards in MSC.Patran are shown as a triangle located in the assigned node. There is,
however, a reported error in MSC.Patran where the position of mass is defined in
absolute co-ordinates (option CID= -1 in CONM2 card). In this situation MSC.Patran
takes these co-ordinates as the offset values.
For the correct interpretation of element forces, stresses, and strains it is necessary to
know the direction of the element co-ordinate system axes. With the function
"Utilities/Display/Finite Element Coordinate Frames" you may be able to display the
element co-ordinate systems for selected elements (select MSC.Nastran option).
Plots showing the element co-ordinate system for areas where consistent element
orientations are necessary should be obtained, checking the axes in 1D (BAR/BEAM)
and 2D (QUAD4/TRIA3) elements as necessary.
When a model uses composite materials, the material co-ordinate system, and the
relative angle between the material and the element co-ordinate systems must be
verified and plotted.
It is recommended that the material axes are defined using an auxiliary global co-
ordinate system in order to avoid problems due to irregular elements or future mesh
modifications. Do not define Xelement by specifying the angle between G1G2 and the
material axis in CQUAD4/CTRIA3 cards.
The projection of the material co-ordinate system over the elements can be obtained
with the command "Utilities/Display/Plot Material Orientation".
Continued…
Issue: E Page 38 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Issue: E Page 39 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
In order to perform the analysis check with MSC.Nastran V70.5 it is necessary to select
the following PARAM values:
a) PARAM,CHECKOUT,YES → To check the model geometry and the matrix
consistency, and to produce the load resultant for the applied loads, in the origin of
basic co-ordinate system (default). If you need the load resultant relative to a different
point use the following parameter.
b) PARAM,GRPNT,G0 → This defines the node about which to calculate the load
resultant, the reactions, the mass and inertia, etc. If you do not use this parameter the
default value is the origin of basic co-ordinate system.
c) PARAM,PRTMAXIM,YES → To show the maximum displacements, maximum
spcforces, maximum applied load and maximum mpcforces for each load case in
basic co-ordinate system. This printing is automatic in V68 and V69 when you ask for
displacement, spcforces in the "Case Control" but not in V70.5.
The MSC.Nastran execution shows also if there are RBEs not correctly defined or are
quasi singular.
Continued…
Issue: E Page 40 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Issue: E Page 41 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
This restriction also applies for dynamic validations, as it is the intention to use the same
model for both static and dynamic calculations. This will therefore be the normal
configuration.
A general sketch/plot of each set of boundary conditions used in the model must be
created for the purposes of validation and documentation.
The reaction resultant, i.e. spcforce resultant, at the reference point must be equal and
opposite in sign to the applied load resultant, i.e. oload resultant.
5.4 Loads
It must be confirmed that all loads applied to the model are applied in the correct
directions, have the correct magnitudes, and are at the correct nodal locations.
The following specific items must be taken into account:
• The total applied load in all 6 DOFs with respect to the reference point. This
point is by default the origin of the basic co-ordinate system.
• The loading distributions along the component main dimension for shear load,
bending moment, and torque moment derived using MSC.Nastran internal
loads.
• The comparison between applied and theoretical load distributions from
MSC.Nastran, at each load monitoring station, e.g. ribs or frame stations.
• The reaction resultant is equal and opposite of the applied load resultant. This
output is printed automatically in solutions such as SOL 101.
These checks will be included in a technical report and referenced on Form 4.
When it is necessary to introduce new nodes for loading introduction, it is recommended
that they are included by using an RBE3. This RBE3 is defined as follows:
The new node must be the "REFERENCE GRID" for the RBE3 with either a) 6
dependant degrees-of-freedom, or b) 6 independent degrees-of-freedom as shown on
page 19.
In the case of b), the independent DOFs must be taken from structural GRIDs and must
be selected to define completely the rigid body motion of reference point.
Continued…
Issue: E Page 42 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Issue: E Page 43 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Tables App B.5.5.1 to App B.5.5.3 show examples of the specific cards used for each of
the above analyses for Executive, Case Control, and Bulk Data decks.
ASSIGN OUTPUT2='~/neptuno/juanluis/a3xx/din/d1.op2',UNIT=12,delete
ID =A3XXHTP,STATUS10C
SOL 103
CEND
$--------------------------------------------------
$.. CASE CONTROL DECK
TITLE = A3XX STATUS10C: HTP FEM
LABEL = DYNAMIC VALIDATION SOL 103 LANCZOS: NORMAL MODES
SUBTITLE= RUN 1: EIGENVALUES CALCULATIONS RIGID BODY MODES [-1,1] Hz.
LINE=84
MAXLINES=4000000
$ Only SPCs in order to avoid singularities in the model (rotations only).
ECHO=NONE
SPC=510
MPC=510
$ Necessary in order to check the boundary conditions.
SPCFORC=ALL
MPCFORC=ALL
$ Necessary in order to create the normal modes.
DISP=ALL
$ Method for eigenvalues calculations.
METHOD = 10
$--------------------------------------------------
$.. BULK DATA CARDS
BEGIN BULK
$.. ANALYSIS MECA
PARAM POST -1
$ Method for eigenvalues calculations: Search modes from -1 to 1 Hz.
$ (Rigid Body Modes)
EIGRL 10 -1.0 1.0
$ UNITS USED IN THE MODEL:
$ Length unit: mm !
$ stiffness unit (E in MATs card): N/mm2 ----> Force unit = N. !
$ The Time unit = s
$ In order to use the coherent units is necessary to put mass in Ton.
$ but normally the CONM2 card are in kg (and the RHO of MATs in kg/mm**3)
$ ====> Therefore must be used PARAM,WTMASS,0.001. If the model mass is not the actual one it $ must
to adjust the value in order to obtain the correct order of magnitude of the $ eigenvalues (this value
do not produces changes in the output of model mass).
$ 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 .. 8 .. 9 .. 10 .
PARAM WTMASS 1.7-3
$ In order to put off PARAM AUTOSPC
PARAM,AUTOSPC,NO
$ Put on the "GRID POINT WEIGHT GENERATOR"
PARAM GRDPNT 0
$-------------------------------------------------------------------------------
$ All the SPC for the boundary conditions of the model must be removed, only the necessary in $ order
to avoid the rotations singularities must be presents in the model.
$ The GRID card must be not contain any permanent constrain in field 8.
$-----------------------------------------------------------------------------
$ Materials with RHO in Kg/mm**3
$ Note: It is possible to use a multiplier in RHO in order to obtain the actual mass of the $
structure
$
$ Aluminium Alloy
$ 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 .. 8 .. 9 .. 10 .
$ - RHO-
$MAT1 1 72000. 0.33 2.7-6
$..
Issue: E Page 44 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
ASSIGN OUTPUT2='~/neptuno/juanluis/a3xx/din/d2.op2',UNIT=12,delete
ID =A3XXHTP,STATUS10C
SOL 103
CEND
$--------------------------------------------------
$.. CASE CONTROL DECK
$..
TITLE = A3XX STATUS10C: HTP FEM
LABEL = DYNAMIC VALIDATION SOL 103 LANCZOS: NORMAL MODES.
SUBTITLE= RUN 2: EIGENVALUES CALCULATIONS RIGID BODY MODES + MOBILE SURFACES ROTATIONS [0,1] Hz
$ Actuators without connection
LINE=84
MAXLINES=4000000
ECHO=NONE
$ Only SPCs in order to avoid singularities in the model (rotations only).
SPC=510
MPC=510
$ Necessary in order to check the boundary conditions.
SPCFORC=ALL
MPCFORC=ALL
$ Necessary in order to create the normal modes.
DISP=ALL
$ Method for eigenvalues calculations.
METHOD = 11
$--------------------------------------------------
$.. BULK DATA CARDS
$..
BEGIN BULK
$.. ANALYSIS MECA
PARAM POST -1
$ Method for eigenvalues calculations: Search modes from 0 to 1 Hz.
$ (Rigid Body Modes + mobile surface rotations)
$ 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 .. 8 .. 9 .. 10 .
EIGRL 11 0.0 1.0
$ UNITS USED IN THE MODEL:
$ Length unit: mm !
$ stiffness unit (E in MATs card): N/mm2 ----> Force unit = N. !
$ The Time unit = s
$ In order to use the coherent units is necessary to put mass in Ton.
$ but normally the CONM2 card are in kg (and the RHO of MATs in kg/mm**3)
$ ====> Therefore must be used PARAM,WTMASS,0.001. If the model mass is not the actual one it
$ must to adjust the value in order to obtain the correct order of magnitude of the
$ eigenvalues (this value do not produces changes in the output of model mass).
$ 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 .. 8 .. 9 .. 10 .
PARAM WTMASS 1.7-3
$In order to put off PARAM AUTOSPC
PARAM,AUTOSPC,NO
$ Put on the "GRID POINT WEIGHT GENERATOR"
PARAM GRDPNT 0
$-------------------------------------------------------------------------------
$ All the SPC for the boundary conditions of the model must be removed, only the necessary in $ $
order to avoid the rotations singularities must be presents in the model.
$ The GRID card must be not contain any permanent constrain in field 8.
$-----------------------------------------------------------------------------
$ Materials with RHO in Kg/mm**3
$ Note: It is possible to use a multiplier in RHO in order to obtain the actual mass of the
$ structure
$
$ Aluminium alloy
$
$ 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 .. 8 .. 9 .. 10 .
$ - RHO-
$MAT1 1 72000. 0.33 2.7-6
$..
Issue: E Page 45 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
ASSIGN OUTPUT2='~/neptuno/juanluis/a3xx/din/d3.op2',UNIT=12,delete
ID =A3XXHTP,STATUS10C
SOL 103
CEND
$--------------------------------------------------
$.. CASE CONTROL DECK
$..
TITLE = A3XX STATUS10C: HTP FEM
LABEL = DYNAMIC VALIDATION SOL 103 LANCZOS: NORMAL MODES.
SUBTITLE= RUN 3: EIGENVALUES CALCULATIONS FLEXIBLE MODES [0, ∞] Hz (FIRST 15TH MODES)
LINE=84
MAXLINES=4000000
$ Only SPCs in order to avoid singularities in the model (rotations only).
ECHO=NONE
SPC=510
MPC=510
$ Necessary in order to check the boundary conditions.
SPCFORC=ALL
MPCFORC=ALL
$ Necessary in order to create the normal modes.
DISP=ALL
$ Method for eigenvalues calculations.
METHOD = 12
$--------------------------------------------------
$.. BULK DATA CARDS
$..
BEGIN BULK
$.. ANALYSIS MECA
PARAM POST -1
$ Method for eigenvalues calculations: Search FIRST 15TH modes from 0 to ∞ Hz.
$ (Flexible modes)
$ 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 .. 8 .. 9 .. 10 .
EIGRL 12 15
$ UNITS USED IN THE MODEL:
$ Length unit: mm !
$ stiffness unit (E in MATs card): N/mm2 ----> Force unit = N. !
$ The Time unit = s
$ In order to use the coherent units is necessary to put mass in Ton.
$ but normally the CONM2 card are in kg (and the RHO of MATs in kg/mm**3)
$ ====> Therefore must be used PARAM,WTMASS,0.001. If the model mass is not the actual one it
$ must to adjust the value in order to obtain the correct order of magnitude of the
$ eigenvalues (this value do not produces changes in the output of model mass).
$ 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 .. 8 .. 9 .. 10 .
PARAM WTMASS 1.7-3
$In order to put off PARAM AUTOSPC
PARAM,AUTOSPC,NO
$ Put on the "GRID POINT WEIGHT GENERATOR"
PARAM GRDPNT 0
$-------------------------------------------------------------------------------
$ All the SPC for the boundary conditions of the model must be removed, only the necessary in
$ order to avoid the rotations singularities must be presents in the model.
$ The GRID card must be not contain any permanent constrain in field 8.
$-----------------------------------------------------------------------------
$ Materials with RHO in Kg/mm**3
$ Note: It is possible to use a multiplier in RHO in order to obtain the actual mass of the
$ structure
$
$ Aluminium alloy
$
$ 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 .. 8 .. 9 .. 10 .
$ - RHO-
$MAT1 1 72000. 0.33 2.7-6
$..
$..
Issue: E Page 46 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Verifications:
1. In Test #1, the following items must be taken into account:
1.1. Check using output from the GRID point weight generator that the total mass and
C.G. are consistent. The following table shows an example:
DIRECTION
MASS AXIS MASS X-C.G. Y-C.G. Z.C.G.
SYSTEM (S)
When the mass or the C.G. is inconsistent, the following tasks must be performed:
• Verify the mass units.
• Check the density values in MATi cards.
• Adjust the mass of model to the actual mass with the parameter PARAM,
WTMASS, FACT in order to obtain the correct order of magnitude in the
eigenvalue calculation. See example in tables App B.5.1 - 3.
1.2. Verify the lists of eigenvalues, checking there are exactly 6 rigid body modes. The
frequency must be lower than 0.01 Hz. If not, perform the following tasks:
• When there are less than 6 modes, or the frequencies are higher than 0.01
Hz, it indicates that the structure is over-constrained. The "SPCFORCES"
listing must be checked in detail in a static run, SOL 101, to verify which
degrees of freedom are the reason of the problem and eliminate the incorrect
constraints. The problem may also be due to ELAS elements with no
coincident nodes.
• Modes from 0.01 to 0.1 Hz may be acceptable in very complicated structures
and when only rotational modes are affected. This must be agreed by all
affected parties.
1.3. Verify the "Grid Point Singularity Table" and confirm that the translation constraints
TX; TY, TZ are at the correct nodes and DOFs.
Continued…
Issue: E Page 47 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
2. In Test #2, the following items must be taken into account:
2.1. The modes obtained must be 6 rigid body modes similar to those obtained in the
previous test #1, plus one more per each modelled movable surface. These modes
should be plotted with MSC.Patran to show that they are real modes and not
spurious mode. The frequency range is from 0.0 to 0.01 Hz.
2.2. This test is used to detect and to correct the undesired connections between
movable surfaces and the main structure. It will also identify situations where the
hinge line points are not perfectly aligned.
2.3. This test checks that the moving surfaces have been modelled correctly, and will
detect problems with DOFs not detected in test #1.
Issue: E Page 48 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
Issue: E Page 49 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
7 Mass Data
The masses of the F.E. model will be checked using PARAM, GRDPNT, GID where GID
is the reference node identification. This card activates the GRID POINT WEIGHT
GENERATOR module, which will calculate the masses and the moment of inertia with
respect to the defined reference point. To obtain these results it is obviously necessary
to include the density parameter in the MATi material cards or alternatively the actual
mass distribution with concentrate mass cards, CONM2. When the actual mass
distribution is supplied using CONM2 cards, all density parameters on material cards
must be zero or blank without exception.
NB: If GID = 0, the reference point is the origin of the basic co-ordinate system.
8 Dynamic Model
Only one official dynamic model must exist, the only difference between the dynamic
model and the stress model must be the units, m for dynamic and mm for stress. There
must be no other differences.
9 Update Procedure
Any revision to component models which must then be included in revised assemblies,
or revisions to existing assemblies, will require the new assembly to be reissued under
the update procedure defined in the main document, Section 9.0.
Continued…
Issue: E Page 50 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
10 Comments to Appendix A
The following pages contain explanatory guidelines to completing the Forms required for
conformance to this AM procedure.
Two sets of guidelines are given; the first contains explanatory text defining the contents
of sections of the forms to assist the user in completing them.
The second set contains a completed example based on an Airbus HTP example.
For the convenience of the user, the two sets have been interleaved so that the
explanation guidelines for each form are immediately followed by the example for that
form.
NB: As Forms 5 & 6 and Forms 4 & 7 are identical in layout and manner of completion,
only Form 4 and 5 has been included in the example.
Continued…
Issue: E Page 51 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
REFERENCES
PART COMPONENT A/C COMPONENT: PAGE:
< Analyst should provide a simple description of the FE model, if necessary outlining the function to which it is
being put >
< A definition of the geometric axis system used for the model data, and the relationship of this system to that
given within the relevant DBD. Include a reference for the source of data used. >
Geometry
< Reference to a report or other form of data source, e.g. a CADs database or list of drawing references,
used to define the geometry data used within the model. If necessary append a separate list to the end of this
report and refer to the list here. >
Meshing
< Reference to a report or other form of data source used to define the standard of meshing density used
within the model. >
Properties
< Reference to a report or other form of data source used to define the standard of idealisation used within
the model. If necessary append a separate list to the end of this report and refer to the list here. >
Materials
< Reference to a report or other form of data source used to define the materials used within the model, and
the data values assigned to them. If necessary append a separate list to the end of this report and refer to the
list here. >
Approved by:
Continued…
Issue: E Page 52 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
REFERENCES
PART COMPONENT: A/C COMPONENT: PAGE:
HTP HTP 1
ANALYSIS IDENTIFICATION: VERSION:
A320EHTPG5 02
Finite Element Model general description:
General model of half A320 HTP, including torque box, elevator and leading edge.
The used elements are ROD, TRIA3, QUAD4, ELAS2, RBE1 and RBE3 (for EI and GJ calculations).
System 0, global aircraft axes. All the grids, except for the elevator, are referred to this system.
Other auxiliar systems are used to provide the correct directions for constrains, loads or material properties.
Meshing
MSC/PATRAN V9.0
Properties
Only torsion box properties have been modified, according to ATL sizing.
Materials
GR V Z19.775 HERCULES 8552/AS4 TH0 0.18 mm Tape has been used for the torsion box, since ATL
manufacturing process is going to be used.
Approved by:
Continued…
Issue: E Page 53 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
FILES RECORD
PART COMPONENT A/C COMPONENT: PAGE:
COMPUTER
DATE OF
DATASET/FILE COMMENTS/DESCRIPTION
CREATION
NAME
Enter list of ALL associated computer files for this model and the validation activity.
Include BULK DATA, submission macros, test loading and results. In particular those files that
form the deliverable to Airbus should be clearly marked and perhaps grouped together for
convenience.
Where files are stored in a common directory, give the directory path name at the top of the
table.
Approved by:
Continued…
Issue: E Page 54 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
FILES RECORD
PART COMPONENT: A/C COMPONENT: PAGE:
HTP HTP 2
fem0_6_e_2.dat 01-02-01 Half HTP Static Model with load test case.
Approved by:
Continued…
Issue: E Page 55 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
MODEL SIZE
GRIDs 1-D 2-D 3-D SPRINGs RIGID BODYs
< Against each entry in the table above, enter Name and Department/Group reference or Siglum of the
person responsible for completing the checklist item. Also, where relevant, give the most appropriate value,
e.g. for Element Twist quote the worst twist measured. If required, hard copy output can be appended and a
reference made to the relevant page numbers in this report. The person named should then sign the end
column. >
< For the rest of the form, complete the entries and sign below as Compiler. >
Approved by:
Continued…
Issue: E Page 56 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
Approved by:
Continued…
Issue: E Page 57 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
< Describe the method of constraining the model. Where necessary add appropriate diagrams to illustrate
< Describe the load case being used for model testing. >
< From the analysis of the test load case described above, complete the above table. >
< Complete one of these pages for EACH static test load case used. >
Approved by:
Continued…
Issue: E Page 58 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
OLOAD RES and SPCFOR RES values for Fx and Fy are roughly zero, so percentage difference between
them is not significant.
In rotations there are some great values, because NASTRAN plate elements do not have stiffness in the
normal direction of the plate, but these singularities do not have any influence in the model performances.
Approved by:
Continued…
Issue: E Page 59 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
BOUNDARY CONDITIONS DESCRIPTION: < Describe the model constraints. If necessary refer to
(text, figure or references) an appended diagram/plot. >
Check 1 Check 2 Check 3
Mass Reference
Mass Standard Value
Mass Output from GPWG
Mode Plots & Frequency List
Frequency [Hz]
Check 1 Check 2 Check 3
(Densities in MATx) (Densities in MATx) (Densities in MATx)
[-1,1] [0,1] [1,∞]
Mode No: free/free with actuators free/free without actuators free/free with actuators
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Comments:
< From the normal modes analysis, complete the table. Only entries in the UNSHADED areas will be used
for the assessment. >
< NB. This sheet requires a DYNAMIC acceptance signature in addition to the normal approval. >
Approved by:
Dynamic Acceptance:
Continued…
Issue: E Page 60 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
Approved by:
Dynamic Aceptance:
Continued…
Issue: E Page 61 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
BOUNDARY CONDITIONS DESCRIPTION: < Describe the model constraints. If necessary refer to
(text, figure or references) an appended diagram/plot. >
Check 4 Check 5 Check 6
Mass Reference
Mass Standard Value
Mass Output from GPWG
Mode Plots & Frequency List
Frequency [Hz]
Check 4 (M1) Check 5 (M2) Check 6 (M2)
(Masses in CONMx) (Masses in CONMx) (Masses in CONMx)
[-1,1] [0,∞] [0,∞]
Mode No: free/free with actuators free/free with actuators free/free with actuators
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Comments:
< From the normal modes analysis, complete the table. Only entries in the UNSHADED areas will be used
for the assessment. >
< NB. This sheet requires a DYNAMIC acceptance signature in addition to the normal approval. >
Approved by:
Dynamic Acceptance:
Continued…
Issue: E Page 62 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX B
General Explanations and Guidelines, continued
QUALITY ASSURANCE OF FEM PROJECT:
Defined Temperature:
Approved by:
Issue: E Page 63 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX C
1 Introduction
A Finite Element Model, used to analyse the mechanical forces and constraints coming
from thermal expansion, must satisfy the usual validity criteria for mechanical FEM (See
the previous pages of this manual).
The most important aspects are: zero frequency rigid vibration modes, loads-constraints
balance , limited displacements verifying the linear hypothesis, etc.
In addition to these criteria, a FEM has to satisfy a supplementary zero stress check,
which then validates the use of thermo-elastic loads.
The zero stress check requires zero mechanical stresses over the entire structure with
uniform expansion of the structure, uniform CTE (coefficient of thermal expansion) and
arbitrary temperature loading
constant value for all of the αi,i expansion coefficients and ensure that the αi,j (i#j)
coupling coefficients are set to zero.
To satisfy this "Zero Stress" check, a mechanically validated model must be modified,
according to the following rules:
Elements:
• ROD: No Restriction
• BAR - BEAM:
• The offsets of shear axis must not modify the distance between the
2 nodes of the beam (vectors W1 and W2)
• The offsets of neutral axis (BEAM) have no influence.
• TRIA-QUAD:
• If the offsets ZOFFS or Z0 are used, the parameter SNORM must be set
to 0.5 instead of 20. (Due to Nastran limitations)
• A warped QUAD results in a non-zero stress (Nastran limitation) but is
generally acceptable (under the accuracy level)
• TETRA-HEXA: No Restriction.
Continued…
Issue: E Page 64 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX C, continued
• GENEL-DMIG:
• Generally non valid Elements for a thermo-elastic analysis (except under
circumstances where their inclusion does not modify the displacements
in another part of the model)
• CELAS:
• CELAS between two coincident nodes: No Restriction
• CELAS of non zero length not valid: see GENEL-DMIG above.
D.O.F RELATIONS:
• RBE2:
• RBE2 between two coincident nodes for 6dof: No Restriction
• Otherwise: see GENEL
• RBE3:
• If the slave node has no associated rigidity (Mass or Load introduction
node): No Restriction
• Otherwise: see GENEL-DMIG
• MPC:
• If the slave node is a barycenter of two master nodes with coefficients in
the same ratio as the distances (Menu Airbus in Patran): No Restriction.
• Otherwise: see GENEL-DMIG
Within the Nastran input deck, the reference temperature is set to a constant value for all
materials. The parameter TEMP (INIT) = 20 and TEMP (LOAD) = 2 are inserted in the
Case Control Section and similarly TEMPD, 20, 20 and TEMPD, 2, 120 into the Bulk
Data Section.
(The indicated values are set to perform a uniform ∆T = 100°C).
Upon completion of the modifications, the whole model is run using Nastran.
Applying the TEMP(INIT) parameter is preferred over the use of TREF in the MATi and
the PCOMP, because it reduces the modifications to the Nastran input deck.
(Within Nastran, TEMP (INIT) prevails over TREF)
Setting in the Case Control parameter ESE=ALL allows the user to view the element
stresses and identify those which fail to satisfy the Zero Stress criteria (values exceeding
1 Mpa).
Continued…
Issue: E Page 65 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
APPENDIX C, continued
3 Temperature Loading
The temperature map is generated from a preliminary calculation on a Finite Element
Thermal Model. The temperature map results are transferred from the thermal FEM to
the mechanical FEM by the thermal team (ESAN). The team then applies temperatures
directly at the nodes of the mechanical FEM by TEMP cards.
It may be necessary to refine the mechanical meshing in zones of strong thermal
gradients, to obtain a better appreciation of the associated mechanical constraints.
If undertaken, the work must be done in agreement with the thermal team (ESAN) to
ensure coherent thermal and mechanical models in these zones.
For two parts of structure in contact but with different temperatures, it is necessary to
have 2 coincident nodes connected by an RBE2 element. This allows the user to apply
the 2 corresponding temperatures.
In instances of meshing refinement of the local model, e.g. when a partial model is
extracted from a global model for local studies, the temperature map must be
interpolated to all nodes of the refined model.
Issue: E Page 66 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Reference documents
AP2000 Procedural Documentation Handling and Writing
AP2080 Airbus Reference Language Main
MSC.Nastran Documentation (Definitions for BAR, ROD etc.)
Glossary
See AP2080 - Airbus Reference Language Main
Assembly A number of parts, subassemblies or any combination
thereof joined together to perform a specific function
and which can be disassembled without destruction of
designed use.
Component Any self-contained part, combination of parts,
subassemblies or units, which perform a distinctive
function necessary to the operation of a system.
Computer See ISO 2382/1-1984
Data See ISO 2382/1-1984
File See ISO 2382/1-1984
Input(Data) See ISO 2382/1-1984
Quality Assurance All those planned or systematic actions necessary to
provide adequate confidence that a product or service
will satisfy given needs.
Validation The act of confirming that a specific requirement has
been meet.
Approval
This document has been approved by:
(signatures or proof of agreement are archived together with the master document)
Function/Organization Approval/Name
Issue: E Page 67 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.
AM2036
Finite Element Analysis Manual
Record of revisions
or the following:
e-mail: airbus.documentation-office@airbus.com
This document and all information contained herein is the sole property of AIRBUS S.A.S. No
intellectual property rights are granted by the delivery of this document or the disclosure of its
content. This document shall not be reproduced or disclosed to a third party without the express
written consent of AIRBUS S.A.S. This document and its content shall not be used for any
purpose other than that for which it is supplied. The statements made herein do not constitute
an offer. They are based on the mentioned assumptions and are expressed in good faith.
Where the supporting grounds for these statements are not shown, AIRBUS S.A.S. will be
pleased to explain the basis thereof.
Issue: E Page 68 of 68
Airbus SAS 2004. All rights reserved. Confidential and proprietary document
Printed Copies are not controlled. Confirm this is the latest issue available through the Portal.