Premature ending to time-marching for Navier Stokes simulation

[plaad]

Hi All,

I am trying to validate a supersonic wind tunnel at my university which uses the following conditions: [Density, U, V, W, P]= [0.0297, 860.818, 0.0, 0.0, 436.585]. This corresponds to a Reynolds number of 7901092.85 to be specified. I have noticed my transient, Navier Stokes simulation solution will work with explicit time marching (using RK2, CFL=0.85) (see issue #76 that I also wrote about another issue), but it seems to stop after a given number of timesteps and ends the run the well before the specified max wall time allocated (OUT_TIME) for the run, number of iterations (NTMAX), or total wall clock time limit (WALLC) is reached. See the images attached for my run so see ending time and other details.

My current values are [OUT_TIME, NTMAX, WALLC] = [10000, 50000, 10000]. There is no Errors.dat file so the solution/field variables have not diverged nor is there a grid issue. Further, when run for the same time marching scheme, but inviscid (Euler), the solution proceeds and outputs the files as expected by these parameters I have set.

However, are there other reasons as to why the run may be prematurely ending for the viscous case/not proceeding as expected?

[plaad]

I have attached my input deck used as a txt file here with a new name for reference: UCNS3D_visc.txt