Unable to load results due to glue table

[kgraham9]

When I attempt to load results with an edge to face glue, the library seems unable to bypass the OBG1 table.

[kgraham9]

note that the edge is on a plate element and the face is on a solid element. is there a chance that the difference in element shape is causing the skip routine to fail?

[SteveDoyle2]

Hard to say without an example.

I took a look at the code and it looks like OBC1 is supported in the dev version. I'd try that.

[kgraham9]

Hey Steve, can you point me to where the code is handling the OQG tables? as far as I can tell, when solid elements are glued to a plate element, the format of the OBG1 table changes making it different than the OQG table. the end result is a (table?) code gets read in as a marker.

[SteveDoyle2]

OQG1 is handled in:
from pyNastran.op2.tables.oqg_constraintForces.oqg import OQG

OBC1 is handled in:
from pyNastran.op2.tables.oqg_constraintForces.obc import OBC

The definition of what functions are called is found in:
from pyNastran.op2.op2_interface.op2_scalar import OP2_Scalar
if you search for a table, you'll find two functions. Both are called and the relevant data for the header or data table is passed in.

[SteveDoyle2]

Oh, you have OBG1, not OBC1. OBG1 is not supported. It should still skip correctly.

[kgraham9]

any suggestions on debugging? for other skipped tables such as OQGGF1 when the marker of -5 is pulled it skips correctly. for the OBG1 table, the read_record_ndata4 subroutine is called incorrectly and generates an error. I dont think that the passer is being set properly for the table but im struggling to fix it

[kgraham9]

when i comment out the OBG1 table entirely it results in:
imarker=0; markers=[-5, 1, 0]; marker_actual=(0, 0, 2) table_name=b'OBG1'

[kgraham9]

I cannot send an op2 with organizational constraints but any edge of a plate element glued to the face of a solid element should result in a similar issue

[SteveDoyle2]

Are you on the latest or the version off pypi? The table skipping method was updated recently to fix a bug, so it might just work. to debug, I print the data using op2.show_data and op2.show_n data depending to see what’s going on. It’s a lot easier to do on a small problem.

…

On Tue, Jun 3, 2025 at 3:39 PM kgraham9 ***@***.***> wrote: *kgraham9* left a comment (SteveDoyle2/pyNastran#832) <#832 (comment)> I cannot send an op2 with organizational constraints but any edge of a plate element glued to the face of a solid element should result in a similar issue — Reply to this email directly, view it on GitHub <#832 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAICUWOT6LSAAKCPI6FNNZL3BYPYNAVCNFSM6AAAAAB6GQUUJSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMZXGQ4TANZWGA> . You are receiving this because you commented.Message ID: ***@***.***>

[kgraham9]

I already made modifications to allow the reading of elastic strain. does the latest version also handle that? might need to redo my modifications if the latest version doesnt handle that

[SteveDoyle2]

The elastic strain from which table and which element? There’s a lot of ways to read strain.

…

On Wed, Jun 4, 2025 at 8:53 AM kgraham9 ***@***.***> wrote: *kgraham9* left a comment (SteveDoyle2/pyNastran#832) <#832 (comment)> I already made modifications to allow the reading of elastic strain. does the latest version also handle that? might need to redo my modifications if the latest version doesnt handle that — Reply to this email directly, view it on GitHub <#832 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAICUWKBHVAPNRRUKS6X3D33B4JATAVCNFSM6AAAAAB6GQUUJSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSNBQGU2TMOBUGQ> . You are receiving this because you commented.Message ID: ***@***.***>

[kgraham9]

cquad4_elastic_strain and ctria3_elastic_strain primarily

[kgraham9]

Also here is something to highlight how the code is breaking. Not sure if it’s helpful. the first shows a table skipping properly, the second gets to the point where it identifies that it needs to skip and then reads in a data block in the read_record_ndata4 routine that causes it to error out

[SteveDoyle2]

I can’t tell from that output. I’d need an example. I need to see the error with context of what is happening to fix it.

…

On Wed, Jun 4, 2025 at 10:52 AM kgraham9 ***@***.***> wrote: *kgraham9* left a comment (SteveDoyle2/pyNastran#832) <#832 (comment)> IMG_5122.jpeg (view on web) <https://github.com/user-attachments/assets/7ee04411-4242-4824-9edd-c21a599191e2> IMG_5123.jpeg (view on web) <https://github.com/user-attachments/assets/682f241e-49ad-4500-b1ef-415208076181> Also here is something to highlight how the code is breaking. Not sure if it’s helpful — Reply to this email directly, view it on GitHub <#832 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAICUWJJM3HNPMTMBZA3G7D3B4W4JAVCNFSM6AAAAAB6GQUUJSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSNBQHA4DQNZXHA> . You are receiving this because you commented.Message ID: ***@***.***>

[kgraham9]

are those elastic strains supported in the new version?