CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Questions about rotate immersed solid and mesh deformation (

wjy-c June 15, 2013 10:57

Questions about rotate immersed solid and mesh deformation
Hi, Iím trying to simulate water splash in box baused by gear rotating. Iím new to CFX and I have tried several times, but the results turn out that the water volume fraction is no right., so I will be appreciated for your help.
(1) I read the CFX help document, I noticed that the immersed solid would not interact properly with mutiphase fluid. Since I concerned about the water and airdistribution caused by rotating gear, does the help document means I cannot use immersed solid for my simulation?
(2) I try a gear in a box model and the simulation result seems correct. But I did not define mesh deformation, Iím wondering wether or not the meshdeformation must be defined in situation like mine, I suppose the mesh deformation is required, but how to explain my successful simulatin in the first time.
English is not my native language, so I may understand and express mistakely , so please corret me if I made some mistake. Thank you very much!

ghorrocks June 16, 2013 06:34

The documentation states that immersed solids do not interact properly with multiphase flow. I do not thing immersed solids is a good way to do this simulation. If the gear rotates then you might be able to do it with a rotating frame of reference. This is quite easy and fully supports multiphase flows. Note you do not need mesh deformation for this approach - which is a big help as it complicates the simulation a lot.

wjy-c June 17, 2013 07:56

thank you for your reply. my friend recently told me that I could creat a rotating multiphase domain which contain the gear and regard the rotating gear as wall boundary. I figure conduct my simulation successfully as my friend's advice. consider the rotating frame of refernce as you mentioned, I do not quite understand what does the "reference" mean, do you mean the same solution as my friend said?

ghorrocks June 17, 2013 08:11

You have to be very careful to use the correct words. Some of the features in CFX have similar names but are quite different. So be very careful to use the correct name for the CFX features you are talking about.

The rotating domains of reference model is mainly used for turbomachinery modelling, but is useful for just about anything else which rotates as well. It is valid with just about all the physics models so you can use complex multiphase models with it.

You might be able to do this with moving mesh but it will be much more complex and slower.

You might be able to do this with a rigid body solver but it will be much more complex.

I do not think you can do this with an immersed solid approach as it does not properly support multiphase.

wjy-c June 17, 2013 11:33

It's very kind of you to offer me those advices. I understand what you mean. Really thank you.

All times are GMT -4. The time now is 22:01.