[Speng]

Hi,
both the "centrifugal blade" and "centrifugal stage" test cases are messed up. Among other things the config files are using old boundary conditions for the inflow and outflow BC. Makes it hard to use them as a reference if you're a newbie

For example:

%
% Riemann boundary marker(s)
% Format inlet: ( marker, TOTAL_CONDITIONS_PT, Total Pressure, Total Temperature, Flow dir-x, Flow dir-y, Flow dir-z)
% Format outlet: ( marker, type, STATIC_PRESSURE, Static Pressure, 0.0, 0.0, 0.0, 0.0)
MARKER_RIEMANN= (inflow, TOTAL_CONDITIONS_PT, 366000.0, 515.00, 0.52101, -0.85355, 0.0)
%
% Non reflecting boundary marker
% Format inlet: ( marker, TOTAL_CONDITIONS_PT, Total Pressure, Total Temperature, Flow dir-x, Flow dir-y, Flow dir-z)
% Format outlet: ( marker, type, STATIC_PRESSURE, Static Pressure, 0.0, 0.0, 0.0, 0.0)
MARKER_NRBC= (outmix, MIXING_OUT, 0.0, 0.0, 0.0, 0.0, 0.0, inmix, MIXING_IN, 0.0, 0.0, 0.0, 0.0, 0.0, outflow, STATIC_PRESSURE, 95750.0, 0.0, 0.0, 0.0, 0.0)
%
% Kind of Average (ALGEBRAIC_AVERAGE, AREA_AVERAGE, MIXEDOUT_AVERAGE)
MIXING_PROCESS_TYPE= AREA_AVERAGE
%
% Mixing plane interface
MARKER_MIXINGPLANE= (outmix, inmix)
%

[TKatt]

Hi,


Unfortunately we have some setups in the Testcases folder that are not in the regression tests. All cases in the regression at least start work and give some results because all new options have to be updated. If a case is not in the regression tests, nothing gets updated over time and at some point they dont work.


Turbomachinery as a whole didn't see much love over the last years as there was no-one around in the community who cared to work on this and keep it up to date.


If it is just a matter of cfg naming changes and you make it work I would kindly ask you to open a PR to update those files. If that is a case I am more than happy to help you on that (in case you need).


This does not really help but it might at least be informative,
Tobi

[Speng]

I tried updating the cfg for one of the cases (the "centrifugal blade") and it ran but it diverged so I doubt I did it right. Can you let me know how to open a PR? I can also send you the cfg as I have it now so maybe you can identify the problem

Originally Posted by TKatt

Hi,


Unfortunately we have some setups in the Testcases folder that are not in the regression tests. All cases in the regression at least start work and give some results because all new options have to be updated. If a case is not in the regression tests, nothing gets updated over time and at some point they dont work.


Turbomachinery as a whole didn't see much love over the last years as there was no-one around in the community who cared to work on this and keep it up to date.


If it is just a matter of cfg naming changes and you make it work I would kindly ask you to open a PR to update those files. If that is a case I am more than happy to help you on that (in case you need).


This does not really help but it might at least be informative,
Tobi

[TKatt]

Hi Speng,


you can open a PR by:


- forking github.com/su2code/su2
- push changes to your personal fork on github
- open a PR from your fork towards the develop branch of the original repository



Unfortunately, unless it is a very simple thing I am not of much help when it comes to fxing turbomachinery stuff.


Opening a PR certainly makes more people aware of this and is also highly appreciated, but of course there is no guarantee that your issue will get fixed.


hope that helps, Tobi