Hi, I am quite new to Houdini and am having a bit of trouble understanding what it is trying to say to me.

I have created simple geometry in Blender, exported as .obj, then imported it into Houdini using File node. Next TetConform and Null.

Then in DOP network I have basic network setup for FEM, and in femsolidobject I get this:


Warning
there are 32080 triangles with nonpositive mass density
Warning
there are 32080 triangles with nonpositive thickness
Message

Shell material property statistics
effective stretch stiffness: min 0, avg 0, max 0
effective shear stiffness: min 0, avg 0, max 0
effective strong bend stiffness: min 0, avg 0, max 0
effective damping ratio: min 0, avg 0, max 0
effective mass density: min 0, avg 0, max 0
effective thickness: min 0, avg 0, max 0

Message

Solid material property statistics
material model, corotated linear: count 0
material model, stable neo-hookean: count 41525
effective shape stiffness: min 40000, avg 40000, max 40000
effective volume stiffness: min 1e+06, avg 1e+06, max 1e+06
effective damping ratio: min 0.1, avg 0.1, max 0.1
effective mass density: min 1000, avg 1000, max 1000

I have seen this message a few times already and don't really get what it is about. Ofcourse as long as there is this warning the FEM solver wont simulate.

Can anyone tell me what this is about?

Sounds like your geo has inverted polygons

Try using Reverse SOP before you pass it to TetConform

Thank you for answering.

It does not seem that the normals are the problem.
It does not matter if I put "Normal" node before Tet Conform or if I put reverse. The message stays the same. It will work if I use Tet Embed, but it is not exactly the same.

Also normal display shows, that Normals are correct.

I'm attaching screenshot with normals, and also zipped .hiplc + .fbx file.

Do you have any other ideas what could be wrong?

Same "warning" for me on similar vanilla setup (mesh-->remesh-->TetConform into femsolidobject) that I didn't encounter before. Is there a workaround?

Try unpacking that mesh. That helped for me.

Hmm. I'm confused. There are no packed primitives in my case: Box-->Remesh-->TetConform isn't packed.
Oh, it looks like a bug in TetConform: if I enable "Preserve Input Geometry" in TetConform, it complains. If I disable that, then it is fine.