EM_levitation

[raback]

Hi Kevin,

You could study the Mesh Update variable at least in the complete setup. Or compare the visually.

I will give a webinar on solving coupled problems next week wednesday. This would make a nice addition to that, so I will try to sort out things before. Self imposed deadlines are sometimes good, and now I even promised this...

-Peter

[Rich_B]

Hello Kevin,

If FlowSolver can give the average velocity downward of the body due to gravity, and v = -gt, plotting the velocity versus time should be a valid check?

Rich.

[kevinarden]

If I plot v = -gt versus the velocity output in Elmer I do not do very well

velocity.png (50.08 KiB) Viewed 1331 times

[raback]

Oh, that looks bad. Maybe you can share the test case for this. I could try to have a look. -Peter

[kevinarden]

This was the Roland levitation case modified to just have gravity

mesh.zip

(32.88 KiB) Downloaded 130 times

case.sif

(5.34 KiB) Downloaded 126 times

[Roland]

Hi, here I am again with my recovered account(thanks Peter!).
I think that a could thing could be to try to levitate the charge as a liquid (with Navier Stokes, as there is also stirring in the liquid bulk) but with a very high viscosity on a first state, so that the liquid behaves like "thick honey", meaning that it does not deform a lot, but just levitates.
As I told you, I am currently working on the" electromagnetic sloshing tank" which is in progress and I keep you informed(as I suggested it, I can send you the model)
We keep us informed
Roland

[raback]

Hi

This one could be close.

There we several problems. One was that MeshSolve had nonlinear iterations which made it add the displacement every iteration.

The gravity should not be scaled by density. Ma bad.

Also, I added some summing up of quantities from the liquied body. Now these can we used to get the average position and velocity of the liquid body.

Would like to use a little more to get parameters right but maybe a little later...

-Peter

[kevinarden]

I turned off the magnetic forces and just let the ball drop due to gravity. It falls as expected v=-gt.

drop.png (17.53 KiB) Viewed 1298 times

[kevinarden]

The gravity overwhelms the magnetic forces. If you turn of the gravity and use just the magnetic forces, the problem solves nicely without having those badly distorted elements.

[Roland]

very interesting.
But is this model with a solid charge or a liquid charge in levitation?
Another question: when clicking on "hydrostatic pressure" in a Navier Stokes Body Force, must the gravity force -(g*density ) be added or not (I am not sure what to do...)?
Roland