CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Farfield BC assignment (https://www.cfd-online.com/Forums/main/14734-farfield-bc-assignment.html)

BM February 4, 2008 14:56

Farfield BC assignment
 
I am simulating a self-breathing polymer electrolyte membrane fuel cell i.e. the cell is open to air and there is no forced convection. The solver (in-house code) is incompressible utilizing the SIMPLE algorithm for momentum flow field, solves the species transport and decoupled heat equation for temperature (2D). The density is coupled via ideal gas law.

The fuel cell, is internal to the domain, on one side (with the other side closure obtained by joining the farfield boundary) and hence the boundaries are treated as internal and fluxes calculated accordingly. My problem arises when I consider the farfield boundary. Note, there is no flow initially. There is a species and temperature gradient created near the fuel cell boundary due to consumption of oxygen in the air and release of heat of reaction, which causes a density gradient and hence, air flow. I use Bousinessq approximation to model this. Considering this the BC's at farfield have been set to,

P=P_inf

V=0 or dV/dn=0

Y=Y_inf

T=T_inf

These would be akin to a pressure outlet BC. However, I am having numerical problems with these BC's. The code does not crash but does not converge either. (My domain is atleast 20L in both directions and I have refined grid) I have never used farfield conditions to model coupled heat and mass transport for incompressible flows and am unable to find literature pertaining to it. I would really appreciate any hint if anyone has any suggested papers or has dealt with similar problem before. Thanks



All times are GMT -4. The time now is 09:00.