CFD at vacuum
 

Hi! Is it possible to use Fluent 6.3.26 for simulation of a gas flow at very low pressure (1-1000 Pa)? Are standard Navier-Stokes equations valid here? Does it exist any option for nearly vacuum flow in Fluent?

Re: CFD at vacuum
 

Hi Jiri,

the validity of the underlying physics used in CFD ( Navier Stokes equations)depends on the assumption that you can describe the gas as a continuum. This assumption typically breaks down if the length scale of gradients of the moments becomes comparable to the mean free path of the molecules ( described by Bird's breakdown parameter). For the case of Argon at room temperature, the mean free path is a few mm at 1Pa. Thus if you are interested in flow in a system with dimensions in the cm range, you will get solutions that are off. If you have a system with dimensions in the meter range, and small gradients, you might be ok. The type of errors you can get depends on the case, but a factor of two or more is not uncommon. For flows in the so-called transition regime and molecular regime, people have been using Monte Carlo techniques that work very well. Over here at In Silicio, we might be able to help you out. Send me a message at jacques.kools@insilicio.com if you want to discuss more.

I hope this is useful,

Jacques Kools