CFX Pre. Definition of rotating/non-rotating boundaries for a radial compressor

AntonZ · October 12, 2015, 05:31

Hi,
I am a newbie in ANSYS/CFX/CFD, so sorry for very basic questions

I have a compressor wheel (Domain), with
Subdomains: Inlet, Outlet, Hub, Schroud etc.
I have defined everything through QuickSetup Mode.
The wheel (Domain) is rotating at 37000 rpm.

The thing I can't clearly understand is why some parts boundaries are rotating, some are static and some are counterrotating.

The logic says if the wheel (domain) is rotating, then I define "domain motion" at 37000 rpm, and everything is static in that coord. system, except of the shroud, which is "counter rotating", which makes it static in absolute coord. system.

Or, alternatively, if domain coord. system is static, then all parts should be rotating, and the shroud is static in both domain and absolute coord. system.
But the shouldn't be all 3 options at the same time - rotating, static, counterrotating.

Where am I wrong?

ghorrocks · October 12, 2015, 07:19

I do not quite understand your question - but the concept is reasonably straight forward.

Obviously the rotor spins and the shroud is still. But a better way of thinking about it is to imagine a rigid mesh (ie not deforming and only able to spin) describing the fluid volume around the rotor and shroud. The mesh around the rotor obviously spins with the rotor and the mesh in the shroud is stationary, but you can see that there is some flexibility in where you can define the transition from rotating to stationary.

Then on the rotating frame of reference you can either have a stationary wall (which rotates with the reference frame) or a counter rotating wall (which is stationary in the absolute frame).

And on the stationary frame you can either have a stationary wall (which is stationary in the absolute frame) or a rotating frame (which rotates.. obviously).

AntonZ · October 12, 2015, 11:26

I mean,
when I am doing CFX-Pre it automatically (New Case -> Simulation Type -> Turbomashinery) I get
all 3 types - static, rotating and counter rotating.

So what is the sense of making counter rotating part on a rotating frame, instead of a simply stationary part?

ghorrocks · October 12, 2015, 20:35

As I said, think about what the mesh has to do to describe the motion. That tells you what needs to rotate and what needs to be stationary. Then you assign static, rotating and counter rotating based on that.

October 12, 2015, 05:31	CFX Pre. Definition of rotating/non-rotating boundaries for a radial compressor	#1
AntonZ New Member Anton Join Date: Oct 2015 Posts: 13 Rep Power: 11	Hi, I am a newbie in ANSYS/CFX/CFD, so sorry for very basic questions I have a compressor wheel (Domain), with Subdomains: Inlet, Outlet, Hub, Schroud etc. I have defined everything through QuickSetup Mode. The wheel (Domain) is rotating at 37000 rpm. The thing I can't clearly understand is why some parts boundaries are rotating, some are static and some are counterrotating. The logic says if the wheel (domain) is rotating, then I define "domain motion" at 37000 rpm, and everything is static in that coord. system, except of the shroud, which is "counter rotating", which makes it static in absolute coord. system. Or, alternatively, if domain coord. system is static, then all parts should be rotating, and the shroud is static in both domain and absolute coord. system. But the shouldn't be all 3 options at the same time - rotating, static, counterrotating. Where am I wrong?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Modeling a Fan by the Multiple reference frame (MRF) method in CFX.	saisanthoshm88	CFX	11	February 17, 2021 12:30
Compressor Blade Analysis in Ansys CFX	LynchMobster47	Main CFD Forum	0	April 10, 2012 23:43
[ICEM] Proper way to name boundaries on 2D model for use in CFX?	RossFS	ANSYS Meshing & Geometry	4	November 10, 2011 03:38
question about governing equation in CFX using rotating/non rotating reference frame	rystokes	CFX	0	January 12, 2010 07:14
Import Fan 3D Model - Workbench or CFX?	Stewart Long	CFX	2	October 28, 2008 05:05

October 12, 2015, 07:19		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	I do not quite understand your question - but the concept is reasonably straight forward. Obviously the rotor spins and the shroud is still. But a better way of thinking about it is to imagine a rigid mesh (ie not deforming and only able to spin) describing the fluid volume around the rotor and shroud. The mesh around the rotor obviously spins with the rotor and the mesh in the shroud is stationary, but you can see that there is some flexibility in where you can define the transition from rotating to stationary. Then on the rotating frame of reference you can either have a stationary wall (which rotates with the reference frame) or a counter rotating wall (which is stationary in the absolute frame). And on the stationary frame you can either have a stationary wall (which is stationary in the absolute frame) or a rotating frame (which rotates.. obviously).

October 12, 2015, 11:26		#3
AntonZ New Member Anton Join Date: Oct 2015 Posts: 13 Rep Power: 11	I mean, when I am doing CFX-Pre it automatically (New Case -> Simulation Type -> Turbomashinery) I get all 3 types - static, rotating and counter rotating. So what is the sense of making counter rotating part on a rotating frame, instead of a simply stationary part?

October 12, 2015, 20:35		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	As I said, think about what the mesh has to do to describe the motion. That tells you what needs to rotate and what needs to be stationary. Then you assign static, rotating and counter rotating based on that.