suggest solution for reverse flow
 

Hi Users

i am working on CFD analysis of thermocompressor which comprises two inlet steams pressure at 10 bar and 1.5 bar and output pressure is 3.5 bar.For meshing i used ICEM software and postprocessing is done in FLUENT.
i used unstructured volume meshing,quality of which is more than 0.3 which is sufficient. continuity value i took as 1e-6,after run given solution converges ,used k Epsilon model,
Initially i kept energy equations OFF.
After various RUN , i am getting reverse flow .
please give me suitable suggestion to eliminate reverse flow.or else suggest me suitable alternating model.
please reply fast.
Thanking YOU.

What are your boundary conditions?

While I don't know what your geometry or boundary conditions are, reversed flow is usually reported when an outlet (i.e., the one at 3.5 bar) begins drawing mass into the system. This seems like a reasonable condition, given that your "inlet" has a lower pressure than the outlet. I don't know about thermocompressors, but just ensure that the boundary conditions you've imposed (along with system dynamics/gravity/etc) wouldn't actually cause reversed flow.

Another thing to check: often, in flows with heavy swirl or pulsation, a few of the cells on an outlet will have some degree of backflow. This is consistent with "real life" and isn't really anything to be worried about. Thus:
1) Compare the number of faces which have reported backflow to the number of faces on the outlet
2) Assess whether the flow has pulsation

If 1 is small and 2 is a "yes" or a "maybe," then I wouldn't be worried. Additionally, often times the error resolves itself as the system reaches a natural steady state, as opposed to the initial guess you provided.

ComputerGuy