How to set up the inlet boundary condition for a low pressure case?
 

Hi, everyone. I'm new to Fluent. I'm currently trying to simulate the flow field of a vacuum chamber operated at 1 Torr. Since the inlet gas is controlled by a mass flow controller, I thought it might be better to set the inlet as a mass flow inlet. The following are the settings for my case:

Solver: pressure-based
Viscous: Laminar
Operating pressure: 133 pa (=1 Torr)
Inlet: mass flow inlet with mass flow rate of 2.728E-05 kg/s, corresponding to 1000 sccm of Ar
Outlet: pressure outlet with gauge pressure of 0 pa

I got a convergent result but it seems weird to me. First of all, the velocity is too low. According to the outlet area (38.465 cm2) and the volume flow rate calculated based on ideal gas law (=1000 sccm * 760), the velocity at outlet should be 3.29 m/s. However, the area-weighted average velocity I got from Fluent is just 0.009 m/s. I checked the absolute pressure. It's correct, which is still 133 pa. But I found that the density is 1.62 kg/m3, which is exactly the value at 1 atm. Since I’m running a case at 1 Torr, shouldn’t the density be 0.002 kg/m3?

So, I decided to change the inlet from a mass flow inlet to a velocity inlet. I calculated the inlet velocity based on the inlet area (0.636 cm2) and the volume flow rate based on ideal gas flow (=1000 sccm * 760). The inlet velocity is 199 m/s (Is this reasonable?). I didn't get a convergent result this time. But I found that the density still remains 1.62 kg/m3 and the highest pressure in the simulation domain has jumped to 1E+05 pa.

Can anybody tell me what I did wrong? Thanks for your kind help in advance.

I think I have found out what's wrong with my case. I didn't change the setting for density. The default is 1.62 kg/m3.

Here comes new questions. Should I change the density to 0.002 kg/m3 (=1.62/760 kg/m3) directly? Or should I select the imcompressible ideal gas law? Since there is no heating source in my case, is it more convenient to set the density as 0.002 kg/m3?

It is better to define the density as a global constant. Using ideal gas law will increase computational cost to solve for density everywhere at each iteration, whereas it is a simple look-up if it is defined explicitly.

Thanks for your reply. I've run the case with a constant density (1.62/760 kg/m3). And yes, it's less computational expensive since it does not have to solve the energy equation. However, I found that the outlet velocity is overestimated because the pressure of a certain part of my simulation domain is not 133 pa. It jumps to as high as 1000 pa. I guess I have to try the ideal gas law and hope that I will get results. Thanks for your help again.