I have just finished a new test with the 2nd order 0.23million mesh (1.8
million in pyFR simulation).. I mapped the results from my previous pyfrs
at 40.95s to the new case starting at t = 0.00s. I used bdf2 in physical
time marching and Euler in pseudo time marching. Delta_t = 0.0001 and
Delta_tau = 0.00001. So the pseudo time step is 1 over 10 of the physical
time step. 10 iteration per physical time step.
I ran that from t = 0 to 9s, and pick up data at t = 8 and 9s. The data
time interval is 0.01. It seemed still the diffusion part is very stiff to
converge. (figure attached). The velocity convergence stabled around 0.02.
I post processed the result and compared with the paper one. It is
Hence, it is quite still very confusing whether diffusion term convergence
does matter in your ACM.
Very strange thing is that in the last physical time step, the convergence
suddenly reach a very low level (1e-6 or 1e-5) for both pressure and
velocity. So how exactly these convergence scheme works?
You received this message because you are subscribed to the Google Groups "PyFR
Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
To post to this group, send an email to email@example.com.
Visit this group at https://groups.google.com/group/pyfrmailinglist.
For more options, visit https://groups.google.com/d/optout.