Skip to content

Add selectable Alart-Curnier mortar friction#33410

Draft
lindsayad wants to merge 14 commits into
idaholab:nextfrom
lindsayad:pr/selectable-ac-mortar-friction-32856
Draft

Add selectable Alart-Curnier mortar friction#33410
lindsayad wants to merge 14 commits into
idaholab:nextfrom
lindsayad:pr/selectable-ac-mortar-friction-32856

Conversation

@lindsayad

Copy link
Copy Markdown
Member

Summary

  • Add an explicit Alart-Curnier friction projection option for quasistatic Coulomb mortar contact.
  • Retain Hueber-Stadler-Wohlmuth as the default formulation; Alart-Curnier is opt-in.
  • Exercise both formulations in unit, normal contact, and performance workloads without duplicating entries by MPI, thread, or mesh mode.
  • Specialize the stable performance candidate names and document the local solver-work and PerfGraph findings.

Dependencies

Refs #32856

lindsayad added 14 commits July 22, 2026 15:08
Add opt-in physical normal and tangential scaling strategies based on contact geometry and acoustic stiffness. Account for variable and equation scaling, install the normal multiplier KSP right scale, and cover the strategies with contact unit tests and design documentation.

Refs idaholab#32856
Use the corrected augmented normal pressure, exact open-state degeneracy handling, normalized HSW rows, equation compensation, and tangential KSP right scaling. Retain HSW as the only production formulation at this review boundary.

Refs idaholab#32856
Expose degree-one Alart-Curnier friction as an opt-in alternative to the legacy degree-two HSW formulation. Preserve HSW as the default and retain formulation-specific row scaling.

Refs idaholab#32856
Add a controlled physical-scale workload and three solver-sensitive analyst workloads, each with stable HSW- and AC-candidate identities. Fail retained analyst trajectories on solve retries and validate the controlled case with repository-local CSV checks.

Refs idaholab#32856
Document Alart-Curnier as an opt-in alternative because the analyst workloads give mixed default-selection evidence. Keep performance findings in MortarPerformance.md rather than the scaling design document.

Refs idaholab#32856
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