Problem at the interface between IGV and Rotor

Julian121 · May 4, 2015, 13:59

Dear All,
I am simulating the rotating stall phenomenon in an axial compressor as part of my MSc project using CFX. My goal is to show that the occurrence of rotating stall can be delayed by using a vane recessed casing treatment. I have a rotor, a stator, IGV, inlet and plus outlet as can be seen below. I have used steady state and transient solvers. However, I have faced a serious problem in which the result does not match with flow behavior in reality. It seems that the wakes do not get modeled correctly as the interfaces. According to velocity contour which has been attached here the velocity does not change in the IGV which should be correct but then it changes suddenly at the interface between IGV and rotor. I do not think this can be correct at all.
Please see the screenshot of the interface settings.
I heard that this problem can be modeled by two approaches. First, is to rotate the domain and keep the blades static and the second one is to rotate blades and keep the domain fixed. I have used the first approach. Is that correct?
I would appreciate it if anyone could advise me.

Opaque · May 4, 2015, 16:13

You seem to be using the steady state Stage model.

Do you understand the approximations made to the rotor-stator interface to come up with a steady model. ?

Julian121 · May 5, 2015, 03:24

Quote:

Originally Posted by Opaque

You seem to be using the steady state Stage model.

Do you understand the approximations made to the rotor-stator interface to come up with a steady model. ?

No, not exactly. I used the default settings.
I have run many cases in transient mode but this one is in steady state but still the velocity and pressure contours do not change at the interfaces. It seems that CFX sees the IGV, rotor and stator separately. I think this is because my interface settings are wrong.
As I said earlier, I assumed that the blades are fixed while the domain is rotating. Can this be the reason why I am getting wrong velocity contours?
Any idea what I should do?
Have you seen any related tutorial?

Thomas MADELEINE · May 5, 2015, 04:38

there are a lot of tutorials on this subject in CFX: TBR method for multistage, frozen rotor, etc...

You want to model the wave generated by the rotor which is a transient phenomenon. you can't have it with a steady state simulation.
If you have any book related to turbo-machinery simulations, it will help you a lot to know which model are useful to get the results you want.

Jiricbeng · June 25, 2015, 05:13

And are the contours : Velocity contour or Velocity in Stn Frame?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
urgent problem in the multi-stage multi-row computation	BalanceChen	Fidelity CFD	2	September 23, 2010 21:26

May 4, 2015, 16:13		#2
Opaque Senior Member Join Date: Jun 2009 Posts: 1,804 Rep Power: 32	You seem to be using the steady state Stage model. Do you understand the approximations made to the rotor-stator interface to come up with a steady model. ?

May 5, 2015, 04:38		#4
Thomas MADELEINE Senior Member Thomas MADELEINE Join Date: Oct 2014 Posts: 126 Rep Power: 11	there are a lot of tutorials on this subject in CFX: TBR method for multistage, frozen rotor, etc... You want to model the wave generated by the rotor which is a transient phenomenon. you can't have it with a steady state simulation. If you have any book related to turbo-machinery simulations, it will help you a lot to know which model are useful to get the results you want.

June 25, 2015, 05:13		#5
Jiricbeng Senior Member Jiri Join Date: Mar 2014 Posts: 218 Rep Power: 13	And are the contours : Velocity contour or Velocity in Stn Frame?