Failed reading b'OSERT'

[Ionut Cosmin Oncescu]

Hi, I am trying to read an op2 file with results from MSC Nastran 2017, and I am getting this error:

DEBUG:   op2.py:542                   combine=True
DEBUG:   op2.py:543                   -------- reading op2 with read_mode=1 (array sizing) --------
INFO:    op2_scalar.py:1556           op2_filename = 'launch.op2'
DEBUG:   op2_reader.py:270            mode = 'msc'
DEBUG:   op2_scalar.py:1735             table_name=b'GEOM1'
DEBUG:   op2_scalar.py:1735             table_name=b'GEOM2'
DEBUG:   op2_scalar.py:1735             table_name=b'GEOM3'
DEBUG:   op2_scalar.py:1735             table_name=b'GEOM4'
DEBUG:   op2_scalar.py:1735             table_name=b'EPT'
DEBUG:   op2_scalar.py:1735             table_name=b'MPT'
DEBUG:   op2_scalar.py:1735             table_name=b'OQG1'
DEBUG:   op2_scalar.py:1735             table_name=b'OES1C'
DEBUG:   op2_scalar.py:1735             table_name=b'OEFIT'
DEBUG:   op2_scalar.py:1735             table_name=b'OESRT'
ERROR:   subcase.py:528               STRESS/STRAIN
ERROR:   subcase.py:529               {'_encoding': 'utf8', 'load_as_h5': False, 'h5_file': None, 'size': 4, 'is_msc': True, 'is_nasa95': False, 'table_name': b'OESRT', 'approach_code': 14, 'isubcase': 1, 'table_code': 56, 'tCode': 56, 'sort_code': 0, 'sort_method': 1, 'device_code': 4, 'analysis_code': 1, 'sort_bits': [0, 0, 0], 'element_type': 95, 'load_set': 5, 'format_code': 1, 'num_wide': 9, 's_code': 0, 'thermal': 0, 'lsdvmn': 1, 'data_names': ['lsdvmn'], 'nonlinear_factor': nan, '_times_dtype': 'float32', 'thermal_bits': [0, 0, 0, 0, 0], 'element_name': 'QUAD4LC-composite', 'subtitle': 'OSL_101.SC1', 'subtitle_original': 'OSL_101.SC1', 'pval_step': '', 'superelement_adaptivity_index': '', 'label': 'SUBCASE 1', 'title': '', 'stress_bits': [0, 0, 0, 0, 0], 'is_stress_flag': True, 'is_strain_flag': False}
failed reading b'OESRT' isubtable=-4

It seems that it won't read my Stress/Strain data. What could be the cause of this?

Thanks.

[Steven Doyle]

Do you have an example?

--
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/c14fb314-9264-4736-9a54-12a9e57b7d32n%40googlegroups.com.

[Ionut Cosmin Oncescu]

Yes, I have attached an .op2 example that I am running with a composite plate. It seems the problem occurs when I am requesting Strength Ratio for composite Ply's (with PARAM SRCOMPS YES). The program can't read the "OESRT" table.

I looked online and the problem seems to be the fact that the Nastran failure index datablock (OEFIT) has the same format as the SR datablock. I tried using and alter that substitutes Failure Indices for SR values (from here: https://www.eng-tips.com/viewthread.cfm?qid=451430 )

If I comment the strength ratio output, everything works fine.

[Steven Doyle]

If you just want to skip the results, you can use the dev branch.

Can you also attach the bdf and f06 that goes with this?  The OESRT table is a little weird.  I think it has mixed types (sometimes a field is an integer and sometimes a float) and it's undocumented in the dmap guide.  

To view this discussion on the web visit https://groups.google.com/d/msgid/pynastran-discuss/5539d031-36a6-450e-97e3-45e643c116fen%40googlegroups.com.

[Ionut Cosmin Oncescu]

I have attached the bdf and f06 files in a zip archive.

I don't want to skip the strength ratio results. Until now I was reading them from the f06 file with a python script that I made and was looking to see if I can read them from .op2 with pyNastran.

Thanks!

[Steven Doyle]

Give the latest version a shot.  It's not vectorized yet, but it should work for ctria3s, cquad4s, ctria6s, and cquad8s.  Run the following to see the new outputs:

>>> test_op2 testTempStress.op2

Off the top of my head, I believe you want model.op2_results.strength_ratio.cquad4_composite_stress.  Not 100%, but it should be similar.

The key outputs are self.data (the 3 float values: sr_bonding, fi_bonding, and sr_bonding), failure_theory_flag, which are strings (2 columns) and element_layer, which is the same as the composite stress/strain results.  The write_f06 method is supported as well.

To view this discussion on the web visit https://groups.google.com/d/msgid/pynastran-discuss/e8a89a39-8d5e-4417-8938-0ac5e6cb2f91n%40googlegroups.com.