nastran condensation and mode shape recovery

[FSB1]

Hi,
I run a condensation of a model.
The .asm and the .pch files are results of this condensation.

I run a normal modes analysis, SOL103, of the condensed model.
I include the .asm and .pch in the end of the this script, as it is supposed to be.

Everything's perfect, with no errors. But when it comes to see the mode shapes in patran, I can't see anything but the residual structure (superelement 0)

Do you know how to recover the mode shapes of the condensed model (e.g. superelement 1)

 

[hezhj2000]

you need to add 'param,post,-1' in your residual run to get op2 file with results in superelements.

 

[FSB1]

Thanks for answering.

I think I'm already doing that
look this is my file - after condensation of the model I run a normal modes analysis:

SOL 103
CEND
TITLE = DEFAULT
ECHO = NONE
$
PARAM,GRDPNT,0
PARAM POST -1
$
method = 1
spc = 2
SUBCASE 1
SUPER = 1
TITLE=default
$
K2GG = KAAX
M2GG = MAAX
param,post,-1
$
SUBCASE 2
SUPER = 0
$
BEGIN BULK
PARAM PRTMAXIM YES
$
eigrl,1,,,10
$
include './residual_structure.bdf'
$
$ interface
include './interface.bdf'
$
spc1 2 123456 555
$ $ $ $ $ $ $ $ $ $ $
include './se.asm'
include './se.pch'
$
ENDDATA

 

[hezhj2000]

Did you put 'DISPLACEMENT=ALL' in both your superelement and residual models?
Also, the op2 file will have result data for both the superelement and residual, but it doesn't have the node location and element connectivity for the superelement because these information are not available to the residual model. So you need a model with all the grids and elements to generate plots for the whole model.

 

[FSB1]

I tried now with displacement=all and disp(plot)=all in both residual and superelement and the result is the same

if I can only see the mode shapes with a FEM model, when all the grids and elements are present,
there is no point in using reduced cb-models and superelements in a normal modes analysis ?

 

[hezhj2000]

There is an example in the Nastran installation folder for data recovery in normal modes analysis. It is something like Doc/seug/ch12/modal/dmigpch.

There are still points for superelements. 1. time saving although it is somehow tricky because it depends on the size of your residual. 2. data protection. For example aerospace suppliers can just send the DMIGs to the OEM to run an integration model. And then with the boundary motions from the OEM, the suppliers can recover the mode shapes for their parts. 3. modularize your parts. For example, if you won't change the design for a particular part, you can DMIG it, and just make changes on your other parts.