[boger]

Under 1.5-dev, in interDyMFoam/pEqn.H, the flux is converted to relative before calling adjustPhi, but a change was made in 1.5-dev wherein phi is converted to relative (and back) within adjustPhi. Isn't this then a bug, with the mesh motion being subtracted twice?

While I was staring at this, I was wondering what the case is that has

1. incompressible fluids (constraint of interDyMFoam)
2. relative flux at the moving boundary (requiring this change to adjustPhi in 1.5-dev), and
3. no change of mass within the volume itself (such that sum flux=0 is a correct form for conservation of mass).

I guess a rigid mesh undergoing a solid body motion that included rotation would require this change. Is that the correct/only example?

Thanks,
David

[hjasak]

My fault - apologies. I have fixed it now and checked it in. This did not get detected for sucha long time since the solver you are looking at is almost never used nowdays - there is a much better one in the naval hydro pack.

Re other questions, only one comment: Yes, there will be a problem with adjustPhi if the total volume of the domain changes during a time-step. You are fine with the rest...

Thanks for that,

Hrv

[A.Devesa]

Hi all,

Originally Posted by hjasak

since the solver you are looking at is almost never used nowdays - there is a much better one in the naval hydro pack.

- I'm pretty new to OpenFoam and am using at the moment interDyMFoam from the 1.5-dev version, combined with GGI. I'd have liked to have some more inputs on why you say that this solver would be quite unused at present? Did i miss something about interDyMFoam or about this "naval pack" you're mentioning?

- I was running into a combined adjustPhi / decomposition problem yesterday, and would like to use a manual decomposition this time. However i found no immediate possibility to change my cellDist type file into what i want, since i cannot easily access to cells coordinates. Any hint here?

Thanks in advance!

[hjasak]

This has been fixed now, by Henrik Rusche. It is interesting to see how long this has been in the code - presumably because we still do not do enough moving mesh

Naval pack is a set of codes developed by Wikki for naval hydro, currently used by people we directly collaborate with.

Hrv