I am trying to use dual time stepping for a compressible flow simulation (M = 5) but I always get the same error no matter what mesh I use. I have tried using the test example but I get the same error over and over again.
This is what the pseudostast.csv and the residual.csv look like.
To run that case you will need some kind of shock capturing scheme. Of those currently in PyFR entropy filtering is by far the best. Thus, I strongly suggest you switch to explicit time stepping?