Hi all

I am running a backward step turbulence model for 0,1 and 2 eqn models and am wondering what transient time step settings to use such as what would be a good number of steps to use and how many seconds for each step? If anyone has experience of this problem and could help it would be greatly apprichated!

Dear Crusader, It depends on the size of the domain. Generally speaking, I would try to get 3 - 4 flow thru times, and chose my timestep based on that. Also, your grid resolution is important and time to run the simulations and processing power and order of the numerical method. Really, it depends on what you are looking for...is it a comparison of the separation/recirculation zone length and model comparison, or is it more than that?

Best Wishes, Allen

-- ACS Consulting Dr. Allen E. Badeau Jr. 2023 Ices Ferry Dr. Morgantown, WV 26508 304-685-2310 cfdconsu@cfd-consulting.org www.cfd-consulting.org

ACS Consulting is the US representative for CHAM's CFD Software package PHOENICS...performing CFD better, faster, and cheaper!

Dear Allen

Thank you for your response. Yes it is essentually a comparison of the seperation and reattachment length at RE=1000, 6000 and 10000 while using a 0 equation model, then a 1 and then a 2. At the moment i am running an 800 iteration simulation with a trainsient run of 20 seconds and 2000 timesteps, tho I think that may be a bit much! Any more advice you can give will be much obliged!

If the turbulence is statistically stationary then PHOENICS can be run in steady mode, and there is no need to advance the solution in time, unless you are unable to secure a converged solution using the steady-state solution algorithm.