rotating domain
 

Hi everyone! Is it possible to set a rotating solid domain? I have a simulation in which two domain are involved: one fluid stationary and one solid rotating. How can I specify rotation? I thought to refer solid domain to a rotating frame of reference but I didn't found how define either RFR and a frame of reference for a solid domain!! Do someone have soma hints?

Thank's a lot!

Re: rotating domain
 

I do not know either.

The only thing I would say is to define a rotating domain, which includes at least all the space the solid would occupy along its rotating path. Then the solid body is just an empty space inside this rotating domain. Why do we need the solid domain? As a blade from the rotor is simulated, do you see anyone buid a domain for the blade?

Re: rotating domain
 

Hi BB, you are right. At the beginning I tried to setting up simulation using rotating boundaries, but I had a lot of problems with mesh deformation and with definition of a CEL to describe motion. So someone suggested me to try with a solid domain referred to a RFR because mesh deformation doesn't work with a 360° rotation. Have you any suggestion?

Thank you for your response.

Re: rotating domain
 

u can assume a rotating domain of fluid aroud a solid wall, which is stationary.Then u dont need solid domain,just have a wall.

This is same as a solid rotating in a stationary fluid.

But choosing the rotating domain size is problem dependent.

u need frozen rotor interface in between your rotating domain and stationary domain of fluid,go through help for further details on this.

in steady case this will do but for time dependent case u need transient rotor stator interface with specified pitch angle,etc,etc.

gud luck.

Re: rotating domain
 

It is the same as simulating a rotating wheel (Now the wheel only has one fake "blade" which is the solid you refered). Put the wheel domain in rotating reference frame and use frozen rotor at all interfaces (top, bottom and side) for steady state simulation. For your case, it definitely needs trasient rotor stator interface and a trasient run. There is no pitch angle concept here, it is just 360 degree full round.

Re: rotating domain
 

It does not need mesh deformation. It needs MFR: the full round rotating domain including the solid; the stationary domain outside.

Re: rotating domain
 

Hi, thanks everyone for your response. I'm trying to define a RFR but I don't find the option to define it! I selected "Create a Frame of Reference", but in it there is not Rotating option. I CFX guide I didn't find how define it. Someone can help me please?

Thank you!

Re: rotating domain
 

It requires MFR license.

Re: rotating domain
 

I checked our license and we have the MFR module. In fact I'm able to define different frame of references for different fluid domains, but not for solid domain!

Re: rotating domain
 

The mesh coordinates don't change in a rotating frame, so there is no need to rotate the solid. There is a beta capability for rotating solid domains, but it is not required in this case. I assume you are solving heat transfer and the diffusion of energy through a solid domain is not affected by rotation. Just set up the solid domain as you normally would.

Where rotating solids are useful is if you have, say, a rotating cylinder in cross flow and you don't want to solve for the transient rotation. Making this a solid rotating frame adds an advection term to the heat transfer within the solid (in the direction of rotation), to reflect the fact that the solid material is rotating (while the mesh does not) and the rotation is transporting heat in that direction.

In any case, this is not what you have. Don't worry about rotating frames in your solid.

-CycLone