CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   sloshing (

Rezaslosh July 15, 2009 08:24

Hello all,

How can I give a sinosoidal motion(not angular) to a partially filled fuel tank?

michael_owen July 15, 2009 08:50

Apply your sinusoidal transverse acceleration in the buoyancy gravity vector using CEL.

ckleanth July 15, 2009 08:57

you could as well apply the sinusoidal mesh movement to the whole domain (obviously you need to write the movement with cartesian x, y coordinates with sin and cos expressions)

Rui July 15, 2009 09:59

That is, theoretically, obviously the same (you can't differentiate whether you are being accelerated or under a force field *)
But I wonder if both cases are numerically handled by CFX in the same way :confused:

* I've heard of a theory that says that the Earth gravity force doesn't exist, but in fact the Earth radius is expanding with an acceleration of 9.8m/s2.

michael_owen July 15, 2009 10:24

If you use ckleanth's method, which I have used many times, be sure to create a subdomain for the entire domain and specify the rigid body mesh motion on it, so that you avoid distortion of the internal mesh due to "mesh inertia", and the associated numerical errors associated with it. In other words, if you only specify the motion on the external surfaces, that motion will only diffuse into the internal mesh, distorting it, rather than carrying it along whole "for the ride", so to speak.

Rezaslosh July 29, 2009 03:24

Thank you all
I was on holydays. I will check the results and compare them with some available theoritical and experimental data and get back here soon.

All times are GMT -4. The time now is 05:13.