Hello,

I perform modal analysis on a "big model"

Memory allocated: 512Mb
+
----- DYNAMIC ANALYSIS INPUT DATA
M <----- Analysis Type (HARMONIC/SPECTRUM/MODES/TIMEHISTORY)
9 <----- Static Load Case for Nonlinear Restraint Status
0.3 <----- Stiffness Factor for Friction (0.0-Not Used)
0 <----- Max. No. of Eigenvalues Calculated (0 - Not Used)
70 <----- Frequency Cutoff (Hz)
LUMPED <----- Mass Model (LUMPED/CONSISTENT)
Y <----- Sturm Sequence Check on Computed Eigenvalues (Y/N)
6 <----- Estimated No. of Significant Figures in Eigenvalues
1E-12 <----- Jacobi Sweep Tolerance
1E10 <----- Decomposition Singularity Tolerance
15 <----- Subspace Size (0 - Not Used)
8 <----- No. to Converge Before Shift Allowed (0 - Not Used)
0 <----- No. of Iterations per Shift (0 - Pgm Computed)
0 <----- % of Iterations per Shift Before Orthogonalization
N <----- Force Orthogonalization After Convergence (Y/N)
N <----- Use Out-of-Core Eigensolver (Y/N)
2500000 <----- Frequency Array Spaces

My question is:
When the calculation is finished, I can see the mode shapes but I am not able to load "load unity Normalized" results: it is very long and the softwarte get “white” and stop working properly.

Could I open the listing from specific files with bloc-note?
What do you think of this problem? What would you do?

Best regards
MH

How many frequencies were extracted below your 70hz cut-off? You should not set the "Frequency Array" size to 250,000. Put that back to something reasonable - maybe 200 or 250.

If you have 250,000 frequencies below 70hz, your system is way too flexible to survive.

Hello,
If I put less (200 or 250) the calculation doesn't launch.

The thing is that the calculation stoped normally and i get the modes and the associated modes properly all the results I can get are consistants.

I am just not able, for each point and each mode, to load the "text" files for "load unity Normalized"...
Somewhere in the result files there is those values-> i get the movies.

I will reduce the cut-of frequency to get only several modes: it works but I will have to launch several calculation. And i am not sure that i can go untill 55HZ:

But can I reduce the study around several frequency values?
for exemple 11-15HZ and 45-55HZ -> (about 30 modes that concernes my study)

Thank you for your answer.
Best regards.

MH

Hello,

The calculation works with Frequency Array Spaces=250, you are right I misunderstood an error message...

Anyway, the thing is that the calculation stopped normally and i get the modes and the associated modes shapes (via movies) properly. All the results I can get are consistent.

I am just not able, for each mode, to load the "text" files for "load unity Normalized"...
Somewhere in the result files there are those values-> i get the movies.


I reduced the cut off frequency: until i get only 10 frequencies it works.
After 10 frequencies i am not able to read the "load unity Normalized" but everything else is working.

Can I reduce the study around several frequency values?
For example 11-12HZ (10 frequencies) 12-14 HZ (10 frequencies).... (About 30 modes that concerns my study)

Another idea: the frequencies are very close and the mode shape are quite similar for several frequencies. I am not really interested in the exact values but an idea of the "load unity Normalized" for each "mode shape type" is sufficient for me.

Could I obtain those values via a harmonic study for example?


Thank you for your answer.
Best regards.

MH

Open a Support Request in eCustomer and send in the job.

(The animation file you're looking for is binary - you can't view its content.)