CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   MFR or simply rotating wall??? (http://www.cfd-online.com/Forums/cfx/23275-mfr-simply-rotating-wall.html)

 Italianboy November 5, 2006 14:45

MFR or simply rotating wall???

I want to simulate a blower-volute system. i know that i can do that using MFR and GGI between volute domain (stationary) and blower domain (rotating). But i am wondering if it is possible to create only one domain and set on cfx_pre the blower walls as rotating walls. What are the differences between the two methods???

Many thanks.

 Glenn Horrocks November 5, 2006 16:20

Re: MFR or simply rotating wall???

Hi,

If the blower is rotating then the only way is using MFR. Just setting a wall velocity on the blades but keeping them stationary is not meaningful.

Glenn Horrocks

 Italianboy November 5, 2006 16:34

Re: MFR or simply rotating wall???

Thank you Glenn, but i don't understand why just setting a wall velocity on the blades keeping them stationary is not meaningful.

If you want to study the relative velocity, i agree with you; but if you just want to know for example the difference in pressure between the inlet and the outlet, setting only the wall velocity for blades should be enough.

I am trying to study MFR theory again but i am not able to find an algebrical explanation.

Many thanks again.

Bye

 Camilo Costa November 5, 2006 18:02

Re: MFR or simply rotating wall???

In CFX tutorial have one example whit MFR!!!

 Italianboy November 6, 2006 02:43

Re: MFR or simply rotating wall???

I know Camilo, the question was another.

Thanks

 Glenn Horrocks November 6, 2006 16:08

Re: MFR or simply rotating wall???

Hi,

Keeping the mesh stationary but imposing a wall velocity is not meaningful. Try this mental experiment: Flow along a flat plate where the flat plate has a velocity perpendicular to the flow. Fluid which goes near the front of the plate is deflected upwards due to the wall velocity. This means no fluid can follow the surface of the plate as it has all been deflected upwards and continues to the exit with no influence of the wall. Some strange form of separation forms and I bet it won't converge.

As I said, this setup is not meaningful.

Glenn Horrocks

 Bian November 6, 2006 16:28

Re: MFR or simply rotating wall???

MFR is totally different from moving wall!!!

In rotating frame, the governing N-S equation for the flow has centrifugal force in it. Everyone forgot?

If your volute is round (or nearly), you might simplify the problem by simulating all in the rotating frame, and setting reverse direction rotating wall for the volute wall.

Bian

 All times are GMT -4. The time now is 22:38.