Neumann Boundary condition (outflow bc)
Hi all,
iam solving flow in a 2D duct using SIMPLE algorithm.at the inlet iam specifying velocity inlet(u_in) and at the outlet iam specifying outflow bc i.e du/dx=0.iam just extrapolating 'u' values to the boundary from immediate neighbour cells inside the domain.similarly for 'v' velocity .but for pressure iam not specifying any boundary condition.i just initialised it to zero everywhere.should i specify pressure?.if i specify wouldn't it become pressure outlet i.e Dirichlet bc?.iam getting correct velocity values but pressure values increase with number of iterations.I think it is something to do with pressure bc. can pressure be extrapolated like velocity? please clarify me? thanks for ur time. reg Thiyaga. |
tha anwser is yes
|
Hi Mr.Ztdep,
thanks for ur response.from ur answer i suppose that i should extrapolate pressure at the outlet.in that case what is the pressure bc at inlet. please clarify me in detail. again thanks for ur time. reg Thiyaga. |
Hi Thyagarajan,
You should probably think of giving a fixed value for total pressure at the inlet (alternatively to the outlet). It is important that pressure has to have a ref value somewhere in the domain or you will see pressure values increasing. I assume your flow problem does not have a shock occuring in it. In that case you should consider given restraints on the total temperature and thereby having a value for pressure. Hope this helped. Prapanj |
Quote:
|
Hi,
I too had solved the problem by using SIMPLE algorithm. There I used neumann boundary condition for pressure as dp/dn=0 at all planes except at outlet, there I took p=0.(Assuming atmospheric pressure). That worked well and got right solutions! You may try with the same if your equations are non dimensional. Regards Sandip |
All times are GMT -4. The time now is 05:33. |