FEM→MPM stress mapping from PLAXIS to Anura3D (MAPF) causes instability / particle “explosion”

[Aidamehr]

Hi all,
I’m trying to initialize an Anura3D MPM model using stresses exported from PLAXIS via the FEM–MPM linking workflow. I created a simple geometry and prepared the MAPF file following tutorial example 13.4, but when I run the MPM model the particles quickly become unstable / “explode”.

Could you please advise on these points?

Stress conventions: Are the stress directions/signs the same in PLAXIS and Anura3D (compression sign, shear sign/order, global axis definitions)?

Smoothing length: Should the smoothing length be ~90% of the FEM mesh size everywhere, or should it be related to particle spacing / local element size?

Stability tips: What are the most common causes of this behavior when importing stresses (e.g., unit mismatch, effective vs total stress, non-equilibrated stress field with body forces/BCs, too large time step)?

I can share the MAPF file and a minimal input case if needed.

Best regards,
Aida

[Aidamehr]

Could you please guide me?