Discussion Forum

Hello,

I am trying to set up 3-D navier-stokes equation for a fluid confined in a box, using weak form PDE , instead of the comsol predefined fluid flow module.

I eventually want to make one of the walls moving, so that the result is not trivial.

I was successful in making 2d model work, but when I did everything same for 3d, it gave me this error message:

Failed to find consistent initial values.
Segregated Step 1
System matrix is zero.
Last time step is not converged.
- Feature: Time-Dependent Solver 1 (sol1/t1)


I've been looking around to seek solution to this, but there weren't any consistent answer to this problem. Could it be the solver issue, where I need to set up the solver in some different way?

Several notes on how I set up the simulation (both in 2d and 3d):
- 2 weak form PDEs: momentum (velocity), and continuity(pressure)
- Dirichlet boundary condition on all the walls for momentum
- Pointwise constraint on the continuity (pressure) equation
- Initial conditions all zero
- Allowed discretization, function type: Lagrange, quadratic for velocity, and linear for pressure.


Attaching both my 2d and 3d models.


Thanks!