Skip to content

CQUAD4/CQUAD8: a blank T1-T4 on the connection entry is the PSHELL T, not 0 - #328

Open
Valladarex wants to merge 1 commit into
MystranSolver:devfrom
Valladarex:pr/blank-thickness
Open

Valladarex wants to merge 1 commit into
MystranSolver:devfrom
Valladarex:pr/blank-thickness

Conversation

@Valladarex

Copy link
Copy Markdown

Fixes #327.

One commit, in SHELL_COMPOSITE_CARDS.f03, ELEMENT_DATA_GATHERING.f03 and MODEL_STUF.f03.

  • Readers: a blank Ti is stored as PLATETHICK_BLANK.
  • Element data: ELMDAT1 puts the PSHELL T in its place.
  • All four blank: a CQUAD4 continuation with all four Ti blank uses no grid thicknesses at all.
  • CTRIA3: not changed. It does not read its Ti; that code is disabled.

Test: MYSTRAN_Validation MystranSolver/MYSTRAN_Validation#40, cquad4_blank_ti_continuation.bdf: a unit-square CQUAD4 in uniform tension whose continuation has T1-T4 blank.

  • It must stretch P L / (E t w) = 1.4285714E-4 with the PSHELL T, with lateral contraction -nu times that.
  • On dev it stops with ERROR 1949. With this PR both checks pass.

Also checked on this commit alone (a CQUAD4 cantilever):

  • All four Ti blank: the same answer as the deck without a continuation.
  • .2, blank, .2, blank: the same answer as .2, .1, .2, .1. On dev the blanks were 0.
  • All four given: unchanged.

Validation suite: with both PRs, 1 failure of 3,257 checks, the same one current dev has without them ("mitc4 thermal strain free rectangle cylinder"). The 2 new checks pass.

… not 0

BD_CQUAD and BD_CQUAD8 read each Ti of the continuation with R8FLD, which gives 0 for a blank field, so a blank Ti
was a thickness of 0. A continuation with all four blank stopped with ERROR 1949 (membrane thickness 0); one with some
blank solved silently with the blanks at 0 (MIN4 uses the average: a thinner plate). A blank Ti is now stored as
PLATETHICK_BLANK (MODEL_STUF) and ELMDAT1 puts the PSHELL T there; a CQUAD4 continuation with all four Ti blank uses
no grid thicknesses at all. (CTRIA3 does not read its Ti: that code is disabled.)

Checks (a CQUAD4 cantilever, PSHELL T 0.1): T1-T4 all blank stopped with ERROR 1949, now gives the deck without a
continuation; T1-T4 = .2, blank, .2, blank gave the answer of the blanks at 0 (average 0.1), now the same as
.2, .1, .2, .1; all four given: unchanged.

Test (MYSTRAN_Validation): cquad4_blank_ti_continuation.bdf, a unit square CQUAD4 in uniform tension whose
continuation has T1-T4 blank: u = P L / (E t w) = 1.4285714E-4 with the PSHELL T, lateral -nu times that. Current dev
stops with ERROR 1949.
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