Learning Forum

[javat33489]

Hi all.

I'm simulating the rotation of a wheel with water pressure using 6DOF. Previously, I made a test model with one wheel, at the inlet I had a flow rate of 1 l/s at the outlet atm pressure and at the second inlet atm pressure. The wheel turned and sucked water from the entrance and from the second entrance. Everything was ok.

Previous post links:

https://forum.ansys.com/forums/topic/connecting-the-wheel-into-one-piece-fluent/

https://forum.ansys.com/forums/topic/the-pump-wheel-does-not-rotate-in-the-fluent-dynamic-grid/

Then I complicated the task, I increased the wheel, added more blades on top and inside. All settings are similar, because the task remains the same as in the first option. The boundary conditions have changed, now at inlet 1 the flow rate is 2 l/s, at the outlet the pressure is 12 MPa, and at the second inlet (inlet 2)  the pressure is 4 MPa. I need the wheel to spin and take water from port 2. I need to see if this will work.

 

I made a good grid, it is 7 million cells, while the skewness is 0.88.

SST model. Simple method.

But when I run the task, I immediately see repeated 6DOF has not converged warnings. 

I believe that the problem is in the second input 4 MPa. Then I decided to check and put 12 MPa there as well as at the outlet. I constantly see repeated 6DOF has not converged warnings right away.

I waited until all pressure and flow graphs would equalize, as well as the moment on the rotating wheel would be equal to 0. But this did not happen, the oscillations continue. Then I stopped the calculation and again made the pressure at the second inlet 4 MPa. The fluctuations continued only at another level of the chart and the 6DOF has not converged warnings continue.

Scr:

1600 iterations at 12 MPa inlet (2) pressure:

___

Next, I set the pressure at inlet 2 with a size of 4 MPa, as I needed initially.

Scr:

How to fix it? How to deal with fluctuations? Their amplitude is very large (see pictures).

[javat33489]

Help please

[Essence]

Hello,

Please share the pressure contours and what RPM are you expecting from the rotor? Try setting the outlet to atm pressure to see if the issue is certainly with setting the pressure to 12MPA.

[javat33489]

To find the problem, I decided to set 1 MPa sequentially. And I found the problem. After 8 MPa, a negative fluid current occurs at the outlet and the rotation wheel begins to oscillate strongly, judging by the moment. Earlier, I immediately set 12 MPa at the output and did not understand where the error was. I conclude that setting the outlet pressure above 8 MPa is critical for the pump and impeller.

 

[Rob]

How did you initialise the model? If the rotor speed doesn't link to the flow very well then you'll see flow separation and all sorts of weird effects. I still don't understand if you're using sliding mesh with the 6DOF solver to calculate the speed of rotation or if you're remeshing the domain. 

To add, 9MPa is a fairly high pressure change, what is the working fluid? 

[javat33489]

>>I still don't understand if you're using sliding mesh with the 6DOF solver to calculate the speed of rotation or if you're remeshing the domain. 

ok sir. Tell me what screenshots should I take and show you?

>>To add, 9MPa is a fairly high pressure change, what is the working fluid? 

The working fluid is water with a density of 1050. The task is exactly this: the output will be 12 MPa and the input will be 2-4 MPa. Perhaps this problem cannot be solved, I decided to check it out.

[Rob]

It can, you depending on the flow field you may need to be careful. How are you updating the mesh position of the rotor? Are you using the radial pressure option on the outlet? 

[javat33489]

I use a rotating domain, it is passive and has rotating blades, and the suitable and suitable parts are deformable.

I watched the animation of the blades spinning.

[Rob]

If it's sliding mesh you don't need remeshing. If it's remeshing you may find you're having issues in the narrow gaps. Not least as the blades move and the cell size on the blade tip doesn't match with the outer casing mesh.