ChristianBauerEng

Not sure if this issue is related, but I'm performing axisymmetric simulations of a convergent nozzle. Throughout most of the domain the axis values seem fine, but there is a...

I see. The python wrappers are far more powerful than I thought. But I'm still a bit lost with respect to the API of the driver(s). Where's a good place...

If you don't mind I'll leave this question open for now. I'm planning to provide a simplified example setup later on, once I've had time to properly dive into the...

I believe this is the section you're talking about: https://github.com/su2code/SU2/blob/1b085062547ec5b066a28ddeeacf4907588f4f5a/SU2_CFD/src/solvers/CNSSolver.cpp#L552-L559 Without understanding the structure of the Jacobian (and the overall solver) it's a bit hard for me to understand what's...

I did some more testing and can confirm, that also when relaxing the convergence criterion the solution does not change. Therefore I'm positive that I can get good results also...

It turns out using convective BCs drastically changes the global flow pattern. Here you can see a comparison of pressures between a simulation with isothermal wall (top, 300 K) and...

I'm still not clear on how the convection BC is supposed to work. If I understand correctly up to this point HEAT_FLUX and HEAT_TRANSFER are largely identical: https://github.com/su2code/SU2/blob/1b085062547ec5b066a28ddeeacf4907588f4f5a/SU2_CFD/src/solvers/CNSSolver.cpp#L476-L479 Essentially in...

On it, I've disabled the Jacobian contributions in my branch. Initial tests look primising, but I'll need a longer-running test to validate it on my case. Will update once the...

After some testing I can report the following: - With Jacobian contributions disabled convergence is greatly increased and on-par with the isothermal/heat flux configurations - After around 12800 timesteps the...

Quite simple I imagine. Since I have everything in place now to open, manipulate and export the mesh in SU2 format (c.f. #1877 ) I could easily filter points and...