CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   STAR-CCM+ (http://www.cfd-online.com/Forums/star-ccm/)
-   -   rotating regions? (http://www.cfd-online.com/Forums/star-ccm/109138-rotating-regions.html)

sieginc. November 10, 2012 03:07

rotating regions?
 
I'm trying to define a rotating region in STAR, but am having trouble. I create a rotating reference frame, and create a new rotation motion under the motions tab in tools. However, when I go to assign the motion specification in regions the drop down only includes the default stationary. Is there something I'm missing?

cfdguy November 11, 2012 11:39

Your Motion should be set on 'Physics Values -> Motion Specification -> Reference Frame'.

sieginc. November 11, 2012 12:55

Quote:

Originally Posted by cfdguy (Post 391499)
Your Motion should be set on 'Physics Values -> Motion Specification -> Reference Frame'.

I've tried this, but to no avail. Under motions in tools you can create new "motions" such as rotation, rotation and translation, DFBI, etc. These then should appear in the drop down menu next to motion in the motion specification tab and then you select which reference frame.

I've seen how they do it in the tutorial in STAR for rotating systems, they leave the motion as stationary and use the reference frame. For some reason this doesn't work for me. I still want to know why when I create a new rotation motion in tools, that I do not have the option to select it in regions.

krishnag2002 November 11, 2012 20:41

Physics Models screenshot
 
Hello, Can you please give screen shot of physics models plz?

sieginc. November 13, 2012 22:44

Well I figured out it was because I had not selected the implicit unsteady solver. Now I am able to select "Rotation" from the drop down menu under motion specification.

The new issue I am having is that when I do that, the reference frame defaults to the lab reference. My solid regions do not stay in place because of this. How do I fix this? If I try to select a different reference frame, the motion defaults back to stationary.

The user guide states: "Currently it is not possible to use non-trivial combinations of a motion and a reference frame. If a motion is selected, the reference frame is automatically set to Lab reference frame. On the other hand, if a reference frame other than Lab is selected, the motion is automatically set to Stationary in that reference frame."

This seems like a very restrictive issue, what if you have a rotating object that's axis of rotation is not centered on the lab reference? This is my problem.

abdul099 November 14, 2012 17:49

Why should you need to activate a motion and a reference frame other than the lab reference frame in a region at the same time?
Either you're solving physics with a rotating region (moving vertices, so rotate the mesh). OR you're just solving physics like the region would rotate without actually moving vertices (the mesh stays always the same).

And to understand why a motion can only be chosen when you're running unsteady: A steady state simulation tries to get the solution after an infinite long time - so when you've specified the rotation rate, which orientation will your region have after an infinite long time? Since there is absolutely no solution for this, you can choose a motion only in unsteady simulations.

sieginc. November 16, 2012 01:06

Quote:

Originally Posted by abdul099 (Post 392148)
Why should you need to activate a motion and a reference frame other than the lab reference frame in a region at the same time?
Either you're solving physics with a rotating region (moving vertices, so rotate the mesh). OR you're just solving physics like the region would rotate without actually moving vertices (the mesh stays always the same).

What if there are rotating components whose axis of rotation does not lie at the same origin as the default lab reference when you import the geometry? Wouldn't this mean you can't select rotation motion, because the component would not rotate in place rather revolve around wherever the lab reference axis lies?

In the case I'm running I have two gears; I defined a local Cartesian coordinate at the center of each gear, and created rotation reference frames and selected those coordinate systems. I then went to regions and selected the rotating reference frame for each gear.

This is how it's done in the tutorial for rotating systems, but I must be missing something because my regions do not appear to be rotating. Could you take a peak at this sim file and see if it is possibly not set up correctly? Each gear is a solid region, as well as the internal volume of the pump and there are interfaces between the gear/fluid and fluid/housing.

http://www.sendspace.com/file/gm2aut

Some of the models are incorrect because I'm getting a floating point error overflow, but I've since corrected that.

abdul099 November 22, 2012 19:36

EITHER you're using a rotating motion, where you can specify the rotation axis and actually move vertices. Since you specify the axis, it has nothing to do with the lab reference frame.
OR you're running in a rotating reference frame, where you also specify the axis, but just solve your equations in a moving coordinate system.

I'll be some days off, so better don't rely on my answer since I can probably have no look on the sim-file in the next week.


All times are GMT -4. The time now is 07:40.