[andrewgould]

I am brand new to openFOAM and CFD in general. I am trying to model a propeller that has vortices at the tips of the propeller. I just want to model the propeller without anything else. This means that I want to keep my geometry constant and have a rotational flow about it. After, looking over many threads on this site plus doing the tutorials that openFOAM comes with, I am stuck. I would really appreciate a nudge in the right direction so that I can learn.
Thanks for your time,
Andrew

[jphandrigan]

I hope you get some advice... because I think I'm in the same boat as you.

[daniels]

Hi Andrew, hi Paul,

we used MRFsimpleFoam to compute a proleller in open water condition. I think it is the easiest and fastest way. You will find MRFsimpleFoam in the tutorial folder of the current release.

Daniel

[marcosch]

Hi Daniel,
how realistic is MRFsimpleFoam for propeller simulation?

Afaik you have to assume that all variables (pressure, velocities in cylindrical coordinates) are uniform on the interface between the static and the rotating reference frame.

This may be approximately true in open water conditions although I guess you have to make the rotating frame pretty large so that the interface "feels" little influence whether a blade is passing by or not.

Is the MRF approach of any use for propellers mounted on a hull? Is my understanding correct, that MRF will not be able to correctly account for the nonuniform flow field surrounding the ship?

Marco

[daniels]

Hi Marco,

my recommendation is only valid for Andrews problem:

"I just want to model the propeller without anything else."

Openwater in my terminology refers to undisturbed inflow, not under a free surface. For simulation of propeller behind a hull you can use sliding interface or ggi technique in OF.

[marcosch]

Hi again,
just for my understanding, is it correct that the radius of the rotating frame is critical for an accurate solution?

If the radius is too small, the properties at the interface are artificially unified although they should show substantial variation in reality.

If the radius is too large, you are on the safe side concerning variations on the interface, but you don't win much compared to a single reference frame.

Does this make sense?

Is there any empirical rule how to choose the diameter of the rotating frame in relation to the propeller? Or do you have to increase the size until the solution converges?

Marco