[soontobeengineerihope]

hey,

So im trying to analyze an exhaust manifold flow.
I think my used parameters are good or atleast close enough to what they would be in reality. I created a mesh and changed the prism layer thickness according to some help i got from other users.
If i try to separate surfaces by non contiguous no new volumes would be created, so this is good as far as i know.
My volume or geometry consists of 4 inlet planes , 1 output plane and one "pipe" which is formed according to the path the multi-component gas would be able to take.

I already ran a simulation, it went for about 10 iterations but then i got this messages:

1. A non-finite residual (Continuity) was added. Typical causes are overflow, underflow, or a division by zero.
2. Non-finite residual (CH) detected after iteration update of solvers. Typical causes are overflow, underflow, or a division by zero.


I just dont know where to look anymore, i already tried many different approaches to solve this problem, but as im quite new to using this software my ideas are running out. I saw that many of you can help others very efficiently, so i hope i can find some help too

Each solution or push in the right direction is highly appreciated. Thanks a lot in advance

[LuckyTran]

Basically, your case diverged on the 10th iteration. It sounds like you're using Star. Right click on your region and use the remove invalid cells option. Make sure there aren't negative volume cells because those will blow up instantly always. If you have negative volumes, this can be a bandaid fix.



Check your boundary conditions and initial conditions.
If you're using the segregated solvers then go to solvers and use a ramp for the urf's. I recommend as high as ramping over 1000 iterations.
If using the coupled solver, use a low courant number and also ramp this up.

[Ankit.Dave]

Originally Posted by LuckyTran

Basically, your case diverged on the 10th iteration. It sounds like you're using Star. Right click on your region and use the remove invalid cells option. Make sure there aren't negative volume cells because those will blow up instantly always. If you have negative volumes, this can be a bandaid fix.



Check your boundary conditions and initial conditions.
If you're using the segregated solvers then go to solvers and use a ramp for the urf's. I recommend as high as ramping over 1000 iterations.
If using the coupled solver, use a low courant number and also ramp this up.

To add on to what Lucky has mentioned. Please do check for non contiguous cells as well while reviewing Invalid cells. Many time contiguous cells may produce similar errors.