Hi everyone, I encounter a problem about boundary condition. my model is a long duct.there are many ribs in it. in order to decrease the grid number. i have to choose only a small segment of the duct to be my compute domain. As the people said before, i set the inlet and outlet boundary condition as periodic, and creat a momentum source. but it's always error so the solver can't compute it. i don't know why. could you tell me how to solve it? Thanks in advance. victor.

It should work. Try to initialise the calculation with a previous run where you have used an inlet/outlet condition. Remember to specify the momentum source (pressure gradient) with the right value and sign.

If you are using CFX 11, you don't need to specify inlet and outlet conditions and a momentum source.

You can use a periodic GGI connection between the ends of your duct section, and specify a pressure drop or a mass flow rate to get the periodic solution.

Johnson

Hi,Dr Flow Squad, Thank you for your answer. my set steps is followed as: 1, creat a domain periodic interface, set the inlet as the periodic side1, and set the outlet as the periodic side2. the interface model is "translational periodicity".and the "mass and momentum" option is "pressure change"; 2,creat a subdomain,but the subdomain is the same as the previous domain,and set the momentum source.

after this ,if the compute beginning,there will be a error: " ERROR #001100279 has occurred in subroutine ErrAction. Message: SYMASS_ZIFCS_EL : The solver ran out of temporary space while building a linked list for a domain interface. Try setting the expert parameter "topology estimate factor zif" to a value greater than 1.0. Values higher than 1.2 should not be necessary. "

do you know why? thank you very mucn! victor

hello Johnson, i have a try as your suggestion.first, i creat a domain periodic interface,the interface models is "general connection",the mesh connection method is "GGI",and choose the mass and momentum option is "pressure change".

if the solver begin, it tell me "2 isolated fluid regions were found in domain compute". then i turn off the "check Isolated Fluid Region",it can't work.

hi Dr Flow Squad, i also have a problem.if i set the inlet and outlet as periodic boundary conditon, i can't set the inlet pressure further. so i think the boundary condition is not enough to get a accurate result. what do you think about that? victor

Well your GGI set up sounds correct, but you have 2 disconnected 3D fluid regions in your model.

You need to check that internal regins are defined as part of GGIs, and that the two sides of your periodic boundary are not defined on disconnected regions.

Johnson

thank you Johnson. I will check up my setting as your suggestion. Best regards.