Fatal error near end of run on large modal analysis...

[jellis-2]

OK I have a specific error that no one here at work has seen before.  I have a large model running a modal analysis on, single pass adaptive, and after the second run sometime during or after the displacement and stress results are happening I am getting the error "The design study terminated abnormally" which generally doesn't bother me in the beginning of an analysis because that means something is wrong and I can fix it but that error at the end of the run throws me off a bit.  Additionally I get a very specific internal error dialogue shown below...

*** A fatal error has occurred. ***
An internal error has occurred in file S:\sim\engine\src\common\post\resultsOut.cxx, line 96.

An internal engine error has occurred.  Please be sure
to run error checking before you run this study.  If
you have run error checking, then contact Customer Support.

Any clue what the error means?  Also my diagnostics look good other than a warning about a beam element that doesn't have recovery points defined, would that cause this error?  In that particular beam I am not worried about stresses.

Thanks in advance for any assistance on this one.

This thread is inactive and closed by the PTC Community Management Team. If you would like to provide a reply and re-open this thread, please notify the moderator and reference the thread. You may also use "Start a topic" button to ask a new question. Please be sure to include what version of the PTC product you are using so another community member knowledgeable about your version may be able to assist.

[346gnu]

Jacob,

The warning in the diagnostics tells you that although the beam has stiffness and mass, it doesn't have any points defined by the user at which to calculate beam results (deflections and stresses) to display. The user is not obliged to define recovery points for a sketched beam and not having them keeps the results files a tiny bit smaller.

As to whether this is the cause the fatal error I don't know but would experiment a bit (I have a nagging at the back of my mind that I have seen similar when using a sketched beam section).

- Can I repeat the failure with a model containing a single beam that has no recovery points? (the modal plots won't be exciting))

- Add recovery points and see if the issue goes away.

- If the problem goes away with the single beam, put recovery points into the beams in your big model and see if the problem goes away.

- Exchange your sketched beam in your big model for a approx. equivalent standard section and see if the problem goes away

- Pass all of the above on to technical support and keep us all posted

Regards

[346gnu]

I just ran a modal study of a single beam model with no recovery points.

It ran ok, I got mode shapes etc.

[jellis-2]

I talked to tech support and we didn't see anything that immediately jumped out at us so I sent them a model and are going to check it out but I did  rerun the model over the weekend (made no alterations) and I came back to a completed analysis.  I am happy to get this far and the results look as I would expect them too but I wonder if tech support will find anything as it is not fun to wait +9 hours to find a failure! I will post anything I get back from tech support.

[KenFarley]

I don't know if this is possible, but I have had similar failures in the past when I've run out of memory. I don't think Creo Simulate is completely returning unused memory, and tends to eat a lot of it. Once it runs out, it crashes. I have had analyses that fail, but when I'd reboot, shut down all unnecessary programs, etc. the case would finish okay. I don't know if this is your cause of trouble, but it's worth a look.