Learning Forum

[Jim Beet]

I am conducting a simulation of a spinning golf ball traveling through the air. The simulation method involves using a moving reference frame. Two fluid domains are created: an outer rectangular domain and a rotating spherical domain containing the golf ball.When examining the contours and vectors, it appears that the golf ball is only experiencing rotational motion and not the tangential motion from the inlet flow. The wake region behind the golf ball is pointing vertically downward, which is incorrect. 

I am using a steady-state, standard k-epsilon model. The inlet speed is set to 45 m/s, and the rotational speed of the domain containing the golf ball is set to 4000 rpm. The walls of the golf ball are set to a moving wall condition with 0 rpm relative to the adjacent rotating domain. I am trying to make the flow similar to the first picture ( which is how it suppose to be). Could you please help me resolve this issue?

 

[Rob]

What is the surface setting between the two fluid zones? 

[Jim Beet]

Hi rob, thanks for the reply. I am not sure on the surface setting between the two fluid zone, I designed the domain using boolean subtract on both outer and rotating domain. How do I inspect the fluid zones surface settings? 

[Rob]

In the solver you'll have either interior:fluid:fluid (or similar, fluid being the zone name) or interface. Because of the scale it's difficult to see whether you have flow through that boundary, and what that boundary does to the flow. 

[Jim Beet]

based on the boundary conditions it has both interior and interface

[Rob]

OK, so nonconformal interface between the two zones. Check the mass flux on both interface surfaces and also plot pathlines down the centreline (release from a line) so we can see what's going on. 

[Jim Beet]

Im not sure how to do the mass flux and pathlines, is there any source you can recommend for me to learn?

[Jim Beet]

Hi rob manage to do the path line. Im still figuring out how to check the mass flux. From the pathlines, it seems that the flow from the inlet went around the rotating flow instead of merging

[Jim Beet]

is this the mass flux you ask for? the value is somewhat similar but one is -ve signed.

[Jim Beet]

hi rob, please reply. I really need your help here 

[Rob]

I don't work weekends! UK office hours only, other than this week where I stop at close of Wednesday for a week off. 

There's something odd with that result. Not least that I'd expect pathlines to be continuous. Please can you post the Pathline panel?

[Jim Beet]

hi rob, I manage to make a  new pathline using line. the flow from the inlet joined the rotating flow and did not go through the golf ball

[Rob]

Turn off Relative and replot. 

[Jim Beet]

turned off the relative. Is there a way to make the inlet flow flow straight so that the wake region of the golf ball is horizontal?

[Rob]

You can turn off the spin! The pathlines follow the flow, so it looks like there is a 3d effect too if you look at where they finish up. 

[Jim Beet]

I cant turn off the spin as I want to simulate a rotating golf ball passing through the air

 

[Rob]

But you can if you want to check the influence on the flow.