[federernadal]

So I dont have a multiple reference frame license.

I am trying to rotate a fan to see fan air patterns on a design.

So here is a model of a centrifugal fan –

So I drew up an enclosure around it, imprinted a face for the inlet and set up the fan body as a solid domain, while the enclosure body as a fluid domain. The default interface is set up.

Enclosure –

Fan (I made a new coordinate system as shown and set it to rotate about the Z axis) –

The solution converges, but it didn’t do what I wanted it to do. Air just blows right past the fan.

How would I get the correct fan profiles? Any approach I could use?

[Michael B]

Something to consider is when you rotate a domain in this fashion it is not moving the mesh, merely the fluid physics for instance. So rotating a solid domain I imagine would just be moving the solid physics around the object which wouldn't do anything to my knowledge. As for why the flow is traveling through the object, i'm not sure. I usually would only mesh the fluid domain and this would be fine but you would still have the same problem in that you couldn't rotate the object in this fashion. In order to not use MFR you would need a dynamically moving mesh which I don't have any experience with unfortunately but is a lot harder than MFR.

[sathishkrazy]

Can we simulate a rotating region (fan) in steady state without creating/seperating it from steady zone, like without creating MRF.

Example:

Say we dont have enough clearance between rotating region and stationary region to create interface. what would be the possible solution for this?