CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   centrifugal compressor impeller with splitter blad (

deepesh nanda October 14, 2004 06:54

centrifugal compressor impeller with splitter blad
Hi, This is Deepesh(M.Sc student)from India (Bangalore). Boss, i m doing CFD analysis of the centrifugal compressor of micro-gas turbine. I tried using fluent but now shifted to CFX. The problem i m facing is that as the impeller is having splitter blade and its speed is 100000 RPM. By the time i reach to this RPM the flow at impeller exit become supersonic which actually is subsonic. I would like to know from u regarding BC'S u apply and the flow domain creation. currently i m using suction and pressure side of main blade and keeping splitter blade as it is. kindly reply fast with some suggestion. if u want i can send the cloud point datas. kindly oblige.

with regards, deepesh India (bangalore)

Abhijit Tilak October 16, 2004 11:08

Re: centrifugal compressor impeller with splitter
Hi Deepesh,

1e5 is one hell of a rpm. I don't have much experience with commerical CFD codes, but I heard CFX is good for turbomachinery flows. What do you mean by "by the time I reach this RPM " ? Are you simulating transient flow or using flase transients to stabilize the solution procedure ?? I use characteristics based boundary conditions in my code. I believe commercial codes are intelligent enough to switch between different BC's based on exit Mach number.

If you know for sure that the exit is subsonic, you can try setting the exit BC to be subsonic. I guess CFX allows you to do that. It would be helpful if you can elaborate your problem a little more.

Regards, Abhijit.

deepesh October 18, 2004 11:13

Re: centrifugal compressor impeller with splitter
thanks buddy for your reply, I mean by saying that by the time i reach to that RPM is that u can't give 100K RPM in one shot, u begin with 1K RPM and then increase it by 10K until u reach the 100K rpm. By the thanks for your interest shown, I have got the solution now, becoz. mine is 24hours experimentation and one of them has clicked now. To share with u, the extended channel at upstream and downstream has be now made the part of impeller flow domain itself which was previously 3diff channels. Anyhow i got the solution both in FLUENT and CFX. Once again thanks for your intereset.

wish u all the best, deepesh

All times are GMT -4. The time now is 09:33.