First Version Supporting Three Phase Flows

Added three-phase case to repository.

On branch membrane

modified:   test_build.sh

About to remove CGNS support

Ever since Egor left, nobody is using CGNS any more. Besides, CGNS doesn't seem to be a unique format, but rather a set of conventions how to write files in hierarchical format, but every software had its own variant of CGNS. Put on top that these libraries are approximately 10 times bigger than the rest of code combined, I see little sense to drag them around any longe.

So, anyhow, this version still contains CGNS in case we ever want to go back to using them.

(This release is tied to branch "membrane")

Last Version Before Gaussian Gradients

The code was relying on least squares method for computation of gradients, ever since it yielded first solutions. However, for flows with strong body forces, as well for the meshes with irregularities on the wall, the Gaussian theorem might prove to be an useful addition to the arsenal of numerical tools it had. So ... just before introducing the Gaussian gradients, I mark this release in the "membrane" branch to have a safe point to return to if things go royally bad with Gaussian method. It is a safety measure. The tests I am performing at the moment of drafting this release seem to show that Gaussian method works just fine.

Jet works again with membrane

There was a point during the development of the branch "membrane", at which I couldn't resolve jet any more. That was a big crisis and I spent many days trying to figure out what is wrong. There were many things wrong in these files:

Compute_Momentum.f90
Compute_Pressure.f90
Correct_Velocity.f90
Field_Mod/Grad_Pressure_Correction.f90
Main_Pro.f90
Rhie_And_Chow.f90

but what I essentially did was to take functions from development_branch, and adopt them to membrane as long as the jet was still OK. That worked, eventually, although some changes might be considered as a step back in terms of software architecture. Nonentheless, jet works again and crisis is over :-)

I am not sure about VOF, didn't have the nerve to check it.

Multiple Domains and Binary Vtu Files

This release sports simulations with multiple domains and also saves .vtu results in binary format.

First version with Choi's and Gu's correction for momentum interpolation on faces

Merge pull request #221 from DelNov/vis_t_improved_for_keps

Vis t improved for keps

Initial upload

The version uploaded on May 28, which will represent the basis for the first stable release.