[boffin5]

In my renewed efforts to simulate a rotating wheel, I have created a graphic (attached) that shows my notional approach. My challenge here is to account for the deformation of the tire, in a relatively simple way. Actually modelling a squished tire is out of scope (and maybe impossible for a CFD simulation?).

Further, I would like to use the MRF method, rather than just a rotating patch, as the idea is that doing it this way would provide for more flexibility in using the case as a template for other simulations.

Problems:

- All the MRF tutorials I have seen have 2 meshes, one for the rotating part, and one for the stationary. Having a ground slab with a divot introduces a third mesh. Can MRF work with more than 2 meshes? Eventually I would like to add a car body, which would add another stl file.

- The MRF tutorials show a mesh for the rotating area, but in this case I would like to use a solid wheel that would not be meshed. is this antithetical with the MRF methodology?

- I am showing a moving wall boundary condition at the ground level, but is there something fundamentally wrong with doing this, since the ground level surface includes a divot that remains fixed with respect to the tire?

Since there are countless groups of people who have done just what I am trying to accomplish, for formula racing cars, Formula Student, etc., I would think that tutorials would be available, but I haven't seen anything that addresses my concerns. If anyone can advise, I would love to hear it!

[arjun]

Originally Posted by boffin5

(and maybe impossible for a CFD simulation?).

!

Already done. The company i used to work for was a tire company (among many other things). They are doing it at the moment. Its possible to do it with wildkatze too but they are using some other software (I am not allowed to name).

Not a big deal.

[boffin5]

Thanks anyway, arjun.
After spending some time, I found that the idea behind MRF concerns a moving mass of air, as interfacing with a stationary mass of air.
Hence, my understanding as shown in the picture was all wrong. So I am now just using a moving rotating wall for the wheel problem, and will defer the MRF approach to the future.