Pressure loss over a resistance changes after refining the mesh

[Florian.Wolfenstetter]

Hi!

I simulated a resistance by assigning a through-flow K-value to a surface. Everything seemed to be fine in the first try.

So I refined the mesh and still everything was ok. The pressure loss was 150 Pa.

But after the second refinement to get the most exact result which was possible, the pressure loss was suddenly 230 Pa. Everything else seems to be as before. Only the pressure loss is odd.

Could it be, that I took too many elements in the third simulation, so that the computer was not able to solve the simulation?

Or what else could be the reason?

[apolo_vanderberg]

Take a step back and lets look at that pressure loss.

A resistance is easy to do sanity check with a hand-calculation.

dP = 0.5 * K(loss coefficient) * density * Velocity * Velocity

use the velocity coming in to the resistance and the K value specified to determine what the dP should be and how does that compare to the 150 or 230Pa?

Also, is this a volume resistance or surface resistance?  What advection scheme are you using? As you went through meshing, how many elements were across the thickness (in the flow direction) of the resistance at each level?