Fluid-Solid Interface Settings for a Rotating Water Container
 

Hello everyone.

I'm simulating a steel rotary drum, full of hot water passing through it to investigate the effect of rotating on passing water temperature drop (Please take a look at the attachment).Attachment 26010

The device is modeled as two adjacent domains, a solid domain contains drum, and a fluid domain contains hot water.
In Basic Settings of solid domain, domain motion is set to rotating and its rotating speed is imported.
In Basic Settings of fluid domain, domain motion is left to be stationary.
Boundary conditions of the system are: normal velocity and temperature of hot water at inlet, static pressure at outlet, heat loss to to surrounding by convection at drum outer wall. The problem is the fluid-solid interface settings, shared by these two domains. It needs to be such that the rotation of drum transfers to the water.

In interface settings, Interface Model is set to General Connection and Frame Change/Mixing Model is changed to Frozen Rotor. But after a converged solution, hot water flow is not affected by rotating drum at all, no matter how fast the fast the drum rotates.

I've read CFX documentations several times and don't know what is wrong about my interface settings.

Thank you for your kind responses, any help is greatly appreciated.

What version of CFX are you using? Rotation was only modellee for solid domains recently.

Also - why do you need the solid outer section? Why can't you just impose a tangential velocity on the outside of the fluid domain?

Hi Glenn, thanks for your reply.

I'm using Ansys CFX 14.5.

I've considered the solid domain because I don't know the proper heat boundary condition at the fluid-solid interface (i.e. the outer layer of fluid domain). Also, the solid domain has a heat regenerative manner which receives heat from hotter water at inlet and gives it back to the cooler water at outlet.

OK, if the solid participates in the heat transfer then you have to model it.

I suspect you will need to make the solid and fluid domains rotate together. Not sure, but worth a try.

Dear Glenn,

I used your idea and set the fluid side of fluid-solid interface to rotate as fast as the solid domain (actually to rotate with it), while the fluid zone is still stationary, and it worked perfectly!

Thanks a lot.