[mrsl]

Hey guys,

as this is my first post here, I would like to introduce myself: I am a research engineer in the area of combustion engines. Right now I am dealing with injecting - lets call it - gas A into the port of an engine.

For this I use a tube (15 mm inner diameter) with smal holes (4 mm inner diameter) in it. Gas is injected into the "main tube" and then exits through the small holes. The problem is that the flow accelerates to supersonic speeds as it passes the small holes. I think this causes the simulation to diverge. When I use another gas (gas B with ~ 8 times the density of gas A) and use the same mass flow BC (--> lower velocities) I have no problems with convergence. Gas B and 8 times the gas A mass flow causes also divergence. So the high velocities seem to cause the problem.
My mesh cell size is 0.25 mm in the area of the small holes and my timestep is approx. 0.05 ms.

I am sorry for the more or less little information but as there are industry partners involved I am not able to post much information. I would like to but I can't.

Basically my question is: Why the flow becomes supersonic if this is no laval nozzle and can this cause the divergence problem? Right now I am using 0.005 ms as a timestep to see if this helps.

Best regards and thanks in advance
Marcel