Problem reading OESPSD1 data

[Brian Russell]

I am running into an issue reading some op2 data from a random vibe analysis.  The issue seems to be coming from missing format code information from cbeam elements in the OESPSD1 table:

I'm hoping someone might be able to shed some light on why this is happening and if there is a way to fix the issue.  Thanks very much in advance.

Cheers,

Brian Russell

[Steven Doyle]

Brian.

What version of pynastran and Nastran are you using?  Also, do you have an small example (bdf/f06/op2) that has the error?  Fake numbers are totally fine.

--
You received this message because you are subscribed to the Google Groups "pyNastran Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pynastran-disc...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pynastran-discuss/4d323360-8ed3-4aae-8253-25c70cb3af96n%40googlegroups.com.

[Brian Russell]

Hi Steve, thanks so much for getting back to me so quickly and apologies for just getting back to you now.  I thought i had this set up to notify my email when you replied, but I must not have.  

Also wanted to thank you generally for developing pyNastran.  It's an extremely useful tool and we love being able to work with Nastran results in python.  Full disclosure I am really a more of a programmer who is helping out our FEM team by writing processing tools for them.

I'm not able to share any of our real model data, and so I put in a request for them to try to create a simple model with fake data that replicates the issue, but it might take them a little bit to get to that.  Will keep you posted when I hear from them.

Our Nastran version is  "Simcenter NASTRAN 2020.1 VERSION - 1915.0004"

and we are using pyNastran version 1.3.4

At the time I ran into that issue, I was trying to access rms data from random vibe results in the model and was able to exclude the results that were causing me the problem using:

op2 = read_op2(op2fileName, build_dataframe=False, exclude_results=["psd.cbeam_stress", "psd.cbush_stress", "crm.cbush_stress", "crm.cbeam_stress", "no.cbush_stress", "no.cbeam_stress", "no.ctetra_stress"], debug=True)

After that I was able to access the rest of the data in the op2.  Once I did, though I ran into another issue that I haven't been able to figure out.  The results of read_op2 do not seem to not include rms data for cquad4 elements, but do include psd and crm data for cquad4 elements.  I've tried this with a couple different models and can't see to get any rms data for cquad4 elements.  I've seen some other forum posts that make me wonder if this is a limitation of pynastran with NX model results

I do see an error/warning when read_op2 is ingesting the data that makes me hopeful there is something we are doing setting up our models that is causing this and we could fix:

Appreciate your help very much and hope you had a good holiday season.

Cheers,

Brian Russell

[steve]

Oh, sorry I finally saw this!  You got caught by the new spam filter.

RMS is surprisingly tricky to parse.  As such it tends to just crash when I turn it on.  Hopefully in the next few months it will be better.