[Shomaz ul Haq]

Dear all. Hi. Hope all are well. I wanted to ask I was having a problem with convergence and then I looked at the FAQs. I was using 500 iterations and 10e-4 as convergence criteria. I followed tips one by one in order.

1. I output the parameters important to my simulation to monitor points and displayed them in the solver manager. The parameters were changing to an accuracy tolerance suitable for my simulation.
2. Then I used a physical time step (PTS) using the residence time in simulation domain in CFX-Post by placing a streamline and looking at the "Time" variable on it. I used the maximum value of time. The solution did not converge. I increased PTS cautiously and smartly. Even then nothing changed.
3. Finally when I used a local timescale factor of 5 and ran solver, solution converged smoothly within 250 iterations for 10e-4 criteria.

But the thing is as mentioned in the FAQ local timescale factor should not be used during the last few iterations but instead it is recommended to use PTS.

I have two problems here.

1. First that what PTS should I use now? The last one or the one with maximum residence time?
2. Secondly when I changed option from local timescale factor to PTS and then looked for putting a value of PTS in "edit run in progress", I couldn't find a new generated line for PTS value. Also local timescale factor value was still there. When I tried validating or saving it gave the command file error message "Error: The essential parameter 'Physical Timescale' is missing from /FLOW:Flow Analysis 1/SOLVER CONTROL/CONVERGENCE CONTROL".

Please help. Would be grateful. Thanks.