dynamicMesh, dynamicFvMesh and meshPhi
 

[LENGTHY PREAMBLE]

So for those of you watching at home, I've been trying to get attachDetach boundary conditions working on 21x. I was able to get incompressible flow to work, and now am struggling to get compressible flow to work. Have a look at the thread and cases I've been working on to see what I'm dealing with

attach Detach Topology Modifier: Proper Turbulent Boundary Conditions

I thought I had to make some modifications to the basic thermo models (updating the boundary values for quantities like compressibility and enthalpy). but I think the biggest problem I have is that the mesh flux field isn't being properly updated after the topology change.

[REAL DISCUSSION]

This is something I would like to know more about in general, especially since I believe it would help me understand the dynamic mesh types much better. The main question I have is: What is the difference between dynamicMesh and dynamicFvMesh? I have been having trouble understanding exactly where one ends and the other begins.

Further to this discussion, how is the mesh flux field meshPhi handled in each?

The reason I ask is that I've tracked the problem down to the fvc::meshPhi(rho, U) operator. Comparing the values it returns with other wall boundaries, it should be 0 (no flux through walls), but I get all kinds of values from 1e-15 to 500. By removing this term from places where I need to reconstruct the flux, I can run both close/open/close and open/close/open simulation in compressible flow. Its only on the attach/detach boundaries, so if I can figure out how to get the those boundaries are calculated and how to make sure that the new boundaries are properly set to zero flux we should be in business.

Many thanks to those luminaries who can enlighten me on the subject. I'll do my best to repay the favor.