CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Two rotating axes ? (http://www.cfd-online.com/Forums/main/1230-two-rotating-axes.html)

 Michael Palm September 1, 1999 03:28

Two rotating axes ?

Hello,

does anyone know a cfd-code which can handle the following problem. In large space there's a rotating disk. On that disk at an arbitrary radial position there's a cylinder with rectangular cross section perpendicular to the disk area. This cylinder is rotating around is own axis.

Is there a code to set up this problem avoiding to much user coding ?

Thanks,

Michael

 Jim Park September 1, 1999 07:59

Re: Two rotating axes ?

To start, I assume that you want to calculate the flow inside the small cylinder.

Classical Mechanics was a long time ago, but ... I think you write the vector transformation from the inertial coordinate system (one in which f = ma applies) to the local coordinate system associated with the cylinder. Substitute that vector transformation into the inertial equations and the boundary conditions.

The equations emerge as the NS equations expressed in the local coordinates and velocities, PLUS a transformation-dependent "body" force that can be expressed and evaluated with user-supplied routines. The bc's will also appear in local coordinates with transformation quantities and can be evaluated with user routines. The information needed to evaluate the extra terms will be angular and linear rotations and accelerations plus distance between rotation axis, stuff that must be known to do the problem.

This is just the way that the Coreollis and centrifugal 'force' terms appear in rotating flow.

You asked about codes. I know that Flow-3D can handle this kind of thing, and I would imagine that others could as well. But you need to do the math first to define your acceleration (body force) terms.

Good Luck!

 All times are GMT -4. The time now is 06:05.