Solid stress in frequency response SOL111

49 views
Skip to first unread message

Nikolay A

unread,
Jun 10, 2024, 8:53:06 PM6/10/24
to pyNastran Discuss
Hi Steve,

many-many thanks for the package!

I encountered the issue when parsing STRESS data for any solid elements in frequency response (SOL111). The example files are attached.


Solver: Simcenter Nastran 2022.1

PyNastran version: 1.5.0+dev.no.checksum.error


Details:

  • a small model contains all element types I need for further analysis (CTRIA, CQUAD4, CHEXA, CPYRAM etc).
  • I'm running SOL111 in NX Nastran requesting STRESS(PLOT)=ALL with PARAM,POST,-1
  • PyNastran (1.5.0+dev from today) crashes when parsing "cpyram_stress"
  • When removing CPYRAM, I have op2_results.modal_contribution for 2D elements, but nothing for 3D

Questions:

  • How can I get 3D stress output?
  • Could you please point me to the example for looping over the stress output in frequency response, e.g. read all SXX at GRID=0 (centroid for SOLIDs) into dictionary {eid: SXX}?

Thank you!


Cheers,
Nikolay
SOL111-solid-stress.zip

steve

unread,
Jun 11, 2024, 1:34:54 AM6/11/24
to pyNastran Discuss
Looks like that form of solid stress/strain is not supported. I need the f06 results to get it right, so STRESS(PLOT,PRINT)=ALL.

Nikolay Asmolovskiy

unread,
Jun 11, 2024, 11:40:57 AM6/11/24
to pynastra...@googlegroups.com
Sure. Please find attached the updated zip.

--
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/09441b27-791f-4f5c-b368-aece180b33e8n%40googlegroups.com.
SOL111-solid-stress-with-print.zip

Ivan Gutierrez

unread,
Feb 4, 2026, 8:18:31 PMFeb 4
to pyNastran Discuss
Nikolay, where you able to solve the missing solid stress from SOL 111 using NX?

I have the same situation on my side. 

steve

unread,
Feb 6, 2026, 1:49:39 AMFeb 6
to pyNastran Discuss
It does work now. It requires the v1.5 dev version. 

The big thing holding up the new release is PyPi now requires two factor authentication and it's not clear how to do it. I did it once, but didn't write it down clearly.

INFO: op2_scalar.py:1932 op2_filename = 'nx_sine.op2'
params:
POST = -1
PRGPST = 'NO'
op2_results.eqexin: EQEXIN(nid, ndof, doftype); nnodes=55
op2_results.bgpdt: BGPDT(cd, xyz); nnodes=55
op2_results.force.cquad4_force[100]
op2_results.modal_contribution.ctria3_stress[100]
op2_results.modal_contribution.cquad4_stress[100]
op2_results.modal_contribution.ctetra_stress[100]
op2_results.modal_contribution.cpenta_stress[100]
op2_results.modal_contribution.chexa_stress[100]
op2_results.modal_contribution.cpyram_stress[100]
op2_results.modal_contribution.cbush_stress[100]
op2_results.modal_contribution.cquad4_composite_stress[100]
eigenvalues['']
matrices:
Matrix['BHH']; shape=(14, 14); type=scipy.sparse._coo.coo_matrix; dtype=float64; desc=symmetric
Matrix['KHH']; shape=(14, 14); type=scipy.sparse._coo.coo_matrix; dtype=float64; desc=symmetric

type=ComplexPlateForceArray ntimes=11 nelements=1; table_name='OEF1X'
eType
data: [ntimes, nnodes, 8] where 8=[mx, my, mxy, bmx, bmy, bmxy, tx, ty]
data.shape = (11, 1, 8)
element name: CQUAD4
sort1
freqs = [ 50. 60. 70. 80. 90. 100. 110. 120. 130. 140. 150.]; dtype=float32


Reply all
Reply to author
Forward
0 new messages