[carl_r]

I am trying to make a coupled thermal stress model related to a casting/die interface. What is described next is a simple model to demonstrate an issue I’m having specifying gravity on an object sitting on top of a rigid surface.

I have a cube (cast) sat on top of a rigid surface (die). The rigid die surface is made of 2D elements. I want to apply gravity (vertical direction, -Y direction, *Dload, Grav ) to the casting so it sits on top of the die surface.
The casting can expand/contract based on the temperature imported from a previous thermal only model. The model is made of 2 parts in an assembly.

Issue: when I run the model the casting either penetrates the rigid surface or moves vertically upwards. The displacement is a large amount (~>100km!, model only 10cm in height) in a few time steps (2.0E-4 s) before the model crashes.

I have investigated (not all combinations of setting) the below parameters but can’t make sense of the problem:
- the gravity magnitude
- the gravity direction
- the rigid surface normal direction (SPOS, SNEG)
- the initial clearance between the die and cast nodes

I have pinned the casting to the die to stop vertical displacement, however this leaves artifacts in the stress strain fields that I would like to avoid.

I have attached a .inp as a .txt file. Any assistance to get the cube to sit on the rigid surface under the force of gravity would be appreciated.

Cube_ThermStress_V3_SIMPLE_inp.txt