[tufty]

Hi. After successfully playing with OpenFOAM for a couple of 2d decompositions of my problem domain, I now want to go for a fully-3d approach.

The model in question is relatively complex, featuring:

An enclosed "pot" (basically, a cylinder with one closed end and a tapered cylinder open to the atmosphere "stuck on" the other end. this much is easy enough to model using the "classic" square core and 4 segments approach.

"Free floating" inside the pot, with both ends open (i.e not touching either end), a baffle, with its exterior a tapered cylinder and the interior "streamlined". this is a 3d structure, not a simple 2d baffle, if that matters.

I figured I could extend the "square core with 4 segments" to "square core with 4 segments and an additional 4 segments", but it seems to be causing problems with the "interior" arc segments, blockMesh seems to think the arc points are colinear.

Is this a known problem, and if so, does anyone have any ideas as to how I could solve it?

Simon

[tufty]

Here's a snapshot of the pot as modelled in gmsh (not, I might add, as I was trying to do things, the msh file itself doesn't want to import)

[tufty]

Meh. Forget it. I managed to get gmsh to export an mesh gmshToFoam liked by - ahem - remembering to name my volumes as well as my surfaces.

Onwards and upwards

Simon