Cavitation 2D geometry

[javier101191]

Hi everyone,

I have a problem with a 2D geometry where get static pressures of -95000 [Pa], and with the cavitation module enabled, when i look at cavitation there is 0%, although i have changed in phase options of the water the vapor pressure to 50000 [Pa] (i know thats not real, but it was to check if cavitation module was working).

Could anyone help me?

[Jon.Wilde]

Hi,

I'm sure we can help but please can you share a CFZ file with us?

We may need a bit more description about what the model is and what you are trying to achieve too, if it is not clear from the cfz file 🙂

Thanks,

Jon

[javier101191]

Thanks!! here is the file

[Jon.Wilde]

I can see a couple of things here:

1. You need to ensure at least 4-5 elements through a gap (not including the mesh enhancement), your mesh is rather too coarse -  the advantage of a 2D model is that we can apply the mesh far more densely than a 3D
2. It looks like you Rotating Region disappears to near zero, is there any reason why it cannot have a larger diameter? We need it to capture a couple of fluid elements around the solid too
3. I would have though that the problems you are seeing are purely down to these two things

[javier101191]

lot of thanks for your answer, but i still dont know how the rotating region must be draw.

My question is, i have 3 materials, the rotor, the stator and the water between them, so, rotating region should be for example half part of the water material???, because, rotating region material is water too if it is touching water material??

lot of thanks again

[Jon.Wilde]

I think the guide here will answer everything: Rotating Region Best Practices

Yes though, the RR will be fluid and it will also rotate the solid part you have inside. The size depends on the mesh, we need a good couple of elements between the solid and the RR wall (page 9 of the pdf in the link above).

Thanks,

Jon

[javier101191]

Lot of thanks again, it its wonderful being able to have a so good technical support!

[javier101191]

Hi, i have a problem with a 2D geometry, first of all, i dont know if i am drawing by the right way the rotation region, because, i have 3 materials in real, a rotor(in), the fluid (water) and then the stator (out). When i look at the results, in the viscosity, the rotation region part has a different viscosity than the water part, and i though that ratation region took all the fluid properties, isn´t it??. In the problem i want to see the pressure distribution, and look if i can force cavitation. With 1500 rpm i think it should cavitate. Has the problem a good setup?? ADV5, 3 degrees per time step, cavitation enabled, etc.. Lot of thanks

[srhusain]

The rotating region has to be circular, inside and outside but your outer boundary is the toothed solid and this is invalid. you need to break up your existing rotating region part with the outer one being a fluid that touches the saw-tooth boundary and the inner one retaining the rotating region material.

Hope this helps.

[srhusain]

I apologize for posting wrong information in my previous post (Jon Wilde pointed this out). I need to dig in a bit further and will get back to you on this.

[javier101191]

doesnt matter srhusain, but i am getting really sick with this problem, i think it shoudnt be so difficult...
So Rotation Region was right??

[Jon.Wilde]

I would also suggest two changes:

Refine the mesh (just go crazy) 🙂 I have just assigned a 0.5 manual mesh to everything

Ramp up the RR over 50 steps

My results still aren't great though and we are not 100% sure why yet.

It could be that we need a smaller timestep - really we should use the element sizing to ensure that we are not skipping more than one element at a time. Or we can turn Intelligent Solution Control on, which I think will modify the timestep to suit.

Another thought is that we have no pressure assigned and I am not clear if we can run cavitation in this way, I will try to find out.

[srhusain]

Yes, the rotating region is indeed correct.

[javier101191]

Really thanks for your help, i have tried to use ADV5, ADV2, intelligent contro enabled, differents turbulence models, really smalls time-steps....etc, and i dont get good results, because it is a model in real, and in real world it cavitates with 1500 rpm and smaller rpms......
I dont know what more to do...

[javier101191]

pressure donest converge in any way

[srhusain]

Can you try assigning a pressure (zero gauge will do) on a small fluid surface? This might help stabilize the pressure by anchoring it- ideally, the surface would be very small so as not to cause appreciable mass leakage.

[Jon.Wilde]

I'm not sure if we can apply them internally to a RR (Syed?) but I do know that if we apply a P=0 to a solid surface, there can be no leakage. It would mean meshing the rotor though.

Just to add also, this is a more complex problem than it seems, it is often the way when something looks so simple 🙂

Keep going, we'll help.

[srhusain]

Good point- I was thinking of applying the pressure BC to a stator side wall edge.

[javier101191]

The problem is that i can only assign a P=0 to a edge, not to a surface.

[Jon.Wilde]

Applying to an edge is the right thing to do 🙂