Problem because of periodicity
 

Hi,

I have a problem modeling a rotating disc with heat sources using periodicity I tried to illustrate (see file attached).
The system is made out of a rotating disc that is located very close to a stationary wall. the gap between disc an wall is filled with fluid. The surface of the disc has 15 heat sources each 24 degree in tangential direction (periodic) (see view 1 and 2). The disc, heat sources and wall are periodic in rotational direction. Therefore I used the option "rotational periodicity" to improve simulation time.
The result is that only at the point of time the periodic elements overlap the wall is beeing heated up (view 3), which means that the problem is calculated as if there were only one heat source. To get correct results the simulation has to inlcude the effect of 15 heat sources (see view 3 t=t1). How could the problem be solved?
Thank you,
Joe

Hi,

Are you using V12 or a previous version?

Glenn Horrocks

I'm using V11

I still couldn't solve the problem.
Does anyone know anything about it?

hi,

couldn´t you turn it arround so that the wall is moving with respect to the disc?

neewbie

If you are using Transient Rotor stator then it is better to use the full model instead of periodic model

Hi,

Newbie has a good point - if your entry and exits are such that you can model it with the wall moving relative to the disk that will simplify things a lot. Of course there is a second assumption hidden in this, and that is doing this will remove the centripetal and coriolis terms. Whether this is significant or not will depend on how fast the disk is rotating.

Rangbal's comment about needing to use a full model with TRS is not correct. TRS works fine with periodicity.

Having a closer look at your drawing - do you need to have a rotating frame of reference at all? If the wall is stationary and disc rotates then you can put a tangential velocity on the disc to generate the effect of the rotating disc. For the rotating heat sources this can be applied as a CEL expression where the heat source is a function of space and time such that it defines the heating patches rotating on the disc. You can still use periodicity in this approach. Is this approach valid?

Glenn Horrocks