CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Problem Setup: Bi-Axial Rotation Possible? (http://www.cfd-online.com/Forums/cfx/20970-problem-setup-bi-axial-rotation-possible.html)

Harmeet February 4, 2005 05:58

Problem Setup: Bi-Axial Rotation Possible?
 
Hi everyone!

Would like to check with you please, about my problem setup.

I have a vertical cylinder that has its fluid inlet at the bottom and the outlet at the top.

My vertical cylinder, however, is spinning about 2 axes: spinning about the vertical axis and tumbling about the horizontal axis. In other words, bi-axial rotation.

I believe that this is a steady state problem. The question is:

- It seems to me that CFX CAN'T do rotation about 2 axes. Only rotation about 1 axis. Is there a way around?

- If rotation about 2 axes can actually be done, how do I set this up? Thanks a million people!

Harmeet.


Glenn Horrocks February 6, 2005 16:36

Re: Problem Setup: Bi-Axial Rotation Possible?
 
Hi,

Can you sum the two rotation vectors to give you a single total rotation vector? Otherwise you will have to use the moving mesh and write the mesh movement yourself.

Glenn Horrocks

Harmeet February 7, 2005 00:07

Re: Problem Setup: Bi-Axial Rotation Possible?
 
Hi Glenn,

Thanks for the response. Summoing the two rotation vectors, for example:

5 rad/s about x axis (i component) 7 rad/s about y axis (j component)

- find magnitude of sqrt(i^2 + j^2) - then find angle using tangent - so now I have a new rotational magnitude and direction. Right?

Hmmmmm... I tried this, but it didn't work. Sigh. It should but I have no idea.

How would I set up a moving mesh thingy? Could you kindly point me in the right direction please? Thanls Glenn!

Harmeet.

Glenn Horrocks February 7, 2005 16:41

Re: Problem Setup: Bi-Axial Rotation Possible?
 
Hi,

This is pretty basic mechanics. Yes, if your rotation vectors do not move then you can sum them to give a single stationary total rotation vector. If this is new to you then you should read a dynamics textbook to understand how to analyse rotational motion.

For moving mesh, have a look at tutorial 20.

Regards, Glenn Horrocks

Robin February 8, 2005 09:04

Re: Problem Setup: Bi-Axial Rotation Possible?
 
Hi Harmeet,

This has been done. I'll do some digging and find it for you.

Regards, Robin

Sandeep February 9, 2005 05:03

Re: Problem Setup: Bi-Axial Rotation Possible?
 
Dear Robin,

I would also be interested in this example, as one of my problems includes rotation about 2 axes.

Thanks

Sandeep

Robin February 11, 2005 15:27

Re: Problem Setup: Bi-Axial Rotation Possible?
 
Hi Harmeet,

To do this is actually fairly simple. Firstly, it will have to be transient, but you would expect that for a problem like this.

Create a rotating domain that also contains your second rotating part. Make sure the second rotating part is within a subdomain and specify a transient rotor/stator interface between this subdomain and the rest of the rotating domain. Within the subdomain, specify an expression for mesh motion which is equal to the rotation of the part. Note that the coordinates will remain relative to the rotating domain.

For details on using the moving mesh capability, have a look at tutorial 20.

Regards, Robin


All times are GMT -4. The time now is 03:13.