I have two concentric cylinders; the inside cylinder sitting in the fluid is static, but the outer cylinder is rotating. Here the inner surface of the outer cylinder is contacting the fluid and outside surface is contacting the air. We want to see the wall shear stress on the surface of the inner cylinder.
Any recommendation for generating the rotating region for the rotating (i.e., outer) cylinder?
Wouldn't it be better to just apply angular motion to the outer cylinder and let the simulation solve? That way, you won't face the conflict of applying rotating region that 'touches' two fluid materials.
But if you insist on using rotating region to solve the problem, i would suggest the following:
- prepare your simulation model such that it will have the two cylinders (inner & outer cylinder)
- there are two cylindrical volumes that represents 'water'
- assign the volume that is in contact with the rotating solid as "rotating region" (any solid in immediate contact will rotate with the region)
- assign the inner cylinder as different material from the outer cylinder
*If you didn't, rotating region causes parts with the same material to rotate together.
*In other words, your simulation will have both inner and outer cylinder rotating together leaving only the water volume static.
- i would consider leaving the 'air' region out of the simulation if it has no direct effect on the result itself
i'm attaching picture below to show my way to setup the simulation, perhaps there are better way to do it.
Hope it helps!
Thanks for your reply. Here're my concerns for both of you suggestions.
The reason that I'm trying to use the "rotating region" is based on my previous experience that angular motion normally takes a lot longer time. Plus, the solution doesn't make sense.
Your suggestion of geometry setting for the rotating region is not correct, i.e., since the outer cylinder is rotating, the rotating region should completely enclose the rotating part (which is the outer cylinder), but your setup doesn't. With that setup I bet the solver will generate the error message saying "...solver exited unexpectedly..."
Nevertheless, I greatly appreciate your time for this.
Thanks for your reply. I've personally tested the setup in 2014. The results clearly displays the type of output you're expecting, though the setup is a bit unconventional.
if you read wikihelp carefully, it states that whenever a rotating region 'touches' solid material, the solid will rotate with it:
- A rotating region cannot directly contact a non-rotating solid region, even if the solid is not inside of the rotating region. An example is a solid annulus surrounding the outside of rotating region. The result is that the solid annulus (which is supposed to be static) will rotate.
in other words, you don't have to fully immerse the solid to let it rotate. I suggest using "cut plane" results with shaded mesh as it runs, and you'll see the solid part mesh rotates.
I'm attaching a .cfz file for you to run the simulation yourself and see if it works (note: it is for 2014 version)
If you're still sceptical, and you're not concerned about the material for the outer cylinder at all, an alternative is to assign the outer cylinder material as rotating region straight away.