Skip to content

AI changes for MID4 on PSHELL - #347

Merged
victorkemp merged 1 commit into
MystranSolver:devfrom
victorkemp:mid4
Oct 25, 2026
Merged

victorkemp merged 1 commit into
MystranSolver:devfrom
victorkemp:mid4

Conversation

@victorkemp

Copy link
Copy Markdown
Collaborator

A. MITC4+: support PSHELL MID4

  • Parser: the bulk data reader rejected MID4 (error 1194). It now accepts it.
  • Sign: SHELL_B was +T^2 [G4]. It is now -T^2 [G4], the Nastran convention, as in MYSTRAN_Documentation ABD_to_PSHELL (validated there against MSC and ComLab). WRITE_PCOMP_EQUIV already used it. With the old sign, MITC4+ reproduced the unsymmetric PCOMP only when G4 was negated.
  • Other elements: a PSHELL MID4 on QUAD4 MIN4/MIN4T or TRIA3 now stops with new error 1966 ("only supported for CQUAD4 with QUAD4TYP MITC4+"). Those elements only add the coupling for an unsymmetric PCOMP, so before this MID4 would have been silently ignored.
  • Force recovery: N = Ae + Bk and M = Be + Dk. A free coupled strip loaded at the reference plane reported M = Dk (My = -0.2) instead of 0. The same recovery feeds the MITC4+ buckling membrane forces.
  • EBM (MATERIAL_PROPS_2D) is reset per element, and SHELL_BALP/TALP (PSHELL branch) are reset, so no values carry over from a previous element.

Tests:

  • The closed-form coupled strip.
  • An independent CLT PSHELL equivalent of an unsymmetric [0/45/-45/90] PCOMP at a 30 deg material angle. Every DOF matches the PCOMP.
  • Coupled strip-column buckling, within 0.1% of exact.

@victorkemp
victorkemp merged commit 3f47f7e into MystranSolver:dev Oct 25, 2026
1 check passed
@victorkemp
victorkemp deleted the mid4 branch October 25, 2026 02:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant