[shahrooz.omd]

Hi everyone

I'm writing a code for a an axisymmetric flow inside a vertical tube using SIMPLER algorithm. The staggered grid is in such a way that pressure cell faces are on the physical domain boundaries. I have added extra pressure nodes with zero volumes on boundaries as well.

Specifying the value of these zero volume control volumes are pretty straight forward. zero pressure gradiant on walls, P = P_outlet in outlet and using Bernouli eq and stagnation pressure for inlet.

My main problem is that I'm not sure how to deal with first and last CVs in each direction. obviously I should not cut their link with boundary sides by setting aE = 0 for I = Imax cells but if I don't what should do for pseudo velicities then? do I need to calculate them or use v, and u themselves. this approach seems ok on walls since u=v=0 is known but what about inlet and outlet which I only know P_outlet and P_stagnation. I also don't know how am I supposed to calculate d(i,j) term for these CVs.

please help me or introduce me to articles / books. (I have studies Malalasekera's book and Patankar's book but didn't get my answer unfortunately)

my current approach is taking d = 1 on boundary nodes as Malalasekera did on his 1d SIMPLE example (which I don't know why to be honest! and is wrong in my opinion) and using last calculated value of v and u on boundary faces as pseudo velocities.