UDF for calculating angular velocity in a rotating tank

vinerm · June 18, 2020, 13:26

The image you have shared is not very clear about which one is axis and which one is radial direction.

vavnoon · June 18, 2020, 14:21

Sorry, the following is more clear. The tank rotates around the y-axis and the plane including the polar coordinates coincides with the x-z plane.

vinerm · June 19, 2020, 06:40

The results are as they should be. There is no vortex, it is just the effect of the axis of rotation. If you want vortex, then do not rotate the cell zone. Only set the outermost wall as a moving wall.

vavnoon · June 20, 2020, 04:29

Dear Vinerm,

Could you please give me more explanation?
I think in this case the mesh rotates at the same speed as the tank and it doesn't have any deformation.
Using mesh motion, we don't have mesh deformation. The cells experience a rigid motion. Furthermore, using the material definition as fluid (having viscosity property) and wall existence, it is expected that the fluid motion be not rigid. As we know, the relative motion between fluid and mesh is considered in Navier-Stokes eq. in the sliding mesh.

I think if I just rotate walls, the mesh will be deformed.

What do you think in this regard?

Thank you so much for your time

vinerm · June 20, 2020, 15:42

By providing rotation to whole mesh zone, you are stating that whole fluid should rotate as a rigid body. Whether it should be mesh motion or moving wall, depends upon the objective of the simulation as well as the physics. In case there is any impeller that is making the fluid rotate, then you should use that impeller and give either frame motion or mesh motion to the region containing the impeller. However, if it is a kind of a drum or barrel containing fluid and is rotating, then using mesh motion is incorrect. You need to give motion to wall.

Moving wall is not deforming mesh approach. The mesh is always fixed.

Navier-Stokes equation has no concept of mesh of any type. Eulerian frame always has fluid motion relative to mesh. But whether the mesh should move or not depends on the physical phenomenon.

vavnoon · June 28, 2020, 02:35

Dear Vinerm,

Many thanks for your reply.

If the cross-section is non-circular, am I able to use just moving wall?

Could you please guide me?
Thanks

vavnoon · June 30, 2020, 10:57

Does it need to use DEFINE_GRID_MOTION macro?

Thank you

June 18, 2020, 13:26	Image	#21
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 36	The image you have shared is not very clear about which one is axis and which one is radial direction. vavnoon likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

June 19, 2020, 06:40	Results	#23
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 36	The results are as they should be. There is no vortex, it is just the effect of the axis of rotation. If you want vortex, then do not rotate the cell zone. Only set the outermost wall as a moving wall. vavnoon likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

June 20, 2020, 15:42	Mesh Motion and Fluid Motion	#25
vinerm Senior Member Vinerm Join Date: Jun 2009 Location: Nederland Posts: 2,946 Blog Entries: 1 Rep Power: 36	By providing rotation to whole mesh zone, you are stating that whole fluid should rotate as a rigid body. Whether it should be mesh motion or moving wall, depends upon the objective of the simulation as well as the physics. In case there is any impeller that is making the fluid rotate, then you should use that impeller and give either frame motion or mesh motion to the region containing the impeller. However, if it is a kind of a drum or barrel containing fluid and is rotating, then using mesh motion is incorrect. You need to give motion to wall. Moving wall is not deforming mesh approach. The mesh is always fixed. Navier-Stokes equation has no concept of mesh of any type. Eulerian frame always has fluid motion relative to mesh. But whether the mesh should move or not depends on the physical phenomenon. vavnoon likes this. __________________ Regards, Vinerm PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
UDF for inlet velocity alongwith UDF for properties	akshayy224	FLUENT	0	February 1, 2018 10:17
[Transient Simulation] Angular Velocity dependent on different parameters than time	ThalesMachado	CFX	1	July 28, 2017 07:36
UDF for time dependent angular velocity calculation	shashankmechguy	Fluent UDF and Scheme Programming	2	October 24, 2016 05:43
UDF for defining a body force in Singel ROtating Reference Frame	teymourj	Fluent UDF and Scheme Programming	9	August 18, 2016 16:33
UDF velocity and temperature	Raj	FLUENT	3	February 1, 2009 19:29

June 20, 2020, 04:29		#24
vavnoon Member vav noon Join Date: May 2020 Posts: 49 Rep Power: 6	Dear Vinerm, Could you please give me more explanation? I think in this case the mesh rotates at the same speed as the tank and it doesn't have any deformation. Using mesh motion, we don't have mesh deformation. The cells experience a rigid motion. Furthermore, using the material definition as fluid (having viscosity property) and wall existence, it is expected that the fluid motion be not rigid. As we know, the relative motion between fluid and mesh is considered in Navier-Stokes eq. in the sliding mesh. I think if I just rotate walls, the mesh will be deformed. What do you think in this regard? Thank you so much for your time

June 28, 2020, 02:35		#26
vavnoon Member vav noon Join Date: May 2020 Posts: 49 Rep Power: 6	Dear Vinerm, Many thanks for your reply. If the cross-section is non-circular, am I able to use just moving wall? Could you please guide me? Thanks

June 30, 2020, 10:57		#27
vavnoon Member vav noon Join Date: May 2020 Posts: 49 Rep Power: 6	Does it need to use DEFINE_GRID_MOTION macro? Thank you