CFD Online Discussion Forums - Beginner questions - Compressibility or Rarefaction

Hello !

I try to model an isothermal laminar flow of a gaz (H₂) inside a long (5m) straight microchannel (square cross section 1mm²).

Boundaries :

Inlet : Flow rate
Outlet : Atmospheric pressure
It is for fuel cell application, so the gaz is consumed along the channel through the top of the domaine. (Is it called a source term ?)

My first problem is to understand if i need a compressible solver as the density of the gaz will change along the channel but the mach number is far from 1 (8E-3). I'm not sure even if i read that : http://www.cfd-online.com/Forums/mai...ible-flow.html

If yes, in the SIMPLE algorithm is it possible to update local densities (i.e on each control volume) with the ideal gas law ( $\rho=\frac{MP}{RT}$ ) after the pressure and velocities correction ?

I don't think it's possible because it doesn't take into account the control volume itself like this expression $\rho=\frac{m}{V}$

What did I miss ?

Thank you for your comments.