CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   STAR-CCM+ (http://www.cfd-online.com/Forums/star-ccm/)
-   -   Problem with simulation (http://www.cfd-online.com/Forums/star-ccm/110688-problem-simulation.html)

bean December 18, 2012 07:55

Problem with simulation
 
1 Attachment(s)
I am trying to simulate the effect of having a microperforation, similar to that which you would use for laminar flow control.

I have refined the mesh for the hole, and the area around the boundary layer.

I have used the following models: Ideal Gas, K-epsilon Turbulence, Realizable K-epsilon Two-layer, RANS, Segregated flow, segregated fluid temperature, steady, 3D, turbulent and 2 layer all y+ wall treatment.

It is running incredibly slowly, so I tried changing it from segregated to coupled to see if that changed anything. Then this crashed, giving me a "floating point exception" error. It seems to me that currently this is a fairly simple simulation. I havent even put conditions on to make suction occur yet.

I am getting huge recirculations within the flow, as you can see from my attached picture.

Any glaringly obvious mistakes?

Thank you for your help

LuckyTran December 24, 2012 00:05

Switching to the coupled solver isn't going to make anything better (it's more unstable than the segregated solver and takes longer to solve).

Whenever you use ideal gas, temperature is usually the culprit. The solver is solving for the fluid temperature which affects the density and messes a lot of things up. It is better if you can remove this dependency. I always recommend using constant density instead of ideal gas. Do you even need to use the ideal gas model? Is the incompressible (constant density) gas more appropriate?

Regardless of whether you are using ideal gas / constant density the energy equation takes a long time to solve! I always try to freeze the energy solver whenever possible because of how long it takes.

siara817 December 25, 2012 12:33

Do you know how many mesh cells do you have?
It may be because of the high number of mesh cells. In addition, it depends on your computer. How many processors do you use?

bean December 30, 2012 12:06

Hi,

Thanks for your replies. Yeah I think it is just to do with the amount of mesh cells. I have approximately a million.

LuckyTran, you mentioned you try and freeze the energy solver. What do you mean by this please?


All times are GMT -4. The time now is 21:37.