code_saturne User's Forum

If i set both in saturne and syrthes the automatic timestep, how will the simulation go on? they both pick the minimum timestep (from saturne or syrthes) or they will go on in a indipendent (and so not correct) way?

thx
Andrea

Hello,

we already discussed this issue. Please have a look at this thread:
http://code-saturne.org/forum/viewtopic.php?f=2&t=43

An automatic time step handling is not implemented by default, yet. It is possible to use automatic time steps, but the setup is not trivial and requires the editing of some files.

Regards,

Martin

Hello,

No promises yet, but I think I will have a solution for this in upcoming version 2.3 (early July). The solution will not be general for all variables of all potential couplings, but will handle the temperature variable, including in the case of multiple SYRTHES couplings with thermal time step mutipliers, so this should be enough for 99% of cases.

Best regards,

Yvan

Hi,
now i am using the 3.0.1 version, is it possible to use the automatic in the coupled saturne saturne??

Hello,

Sorry, the solution I was thinking about was a bit more complex than expected, or rather risky to implement just before starting validation of 3.0...

I don't have time right away, but I could provide you a patch (to apply to version 3.0 or the upcoming 3.1 as user subroutines) to test in 2 or 3 weeks if you are interested. Remind me then...

Regards,

Yvan

yes !! i will be very interested, i would really like to do the beta tester for that, since i'm simulating a oven with a recyrculing fan -> i have to set the mrf in the saturne way : coupling 2 cases with coriolis' source term in the fan domain.

Since the convergence is faster using an initial unsteady (100 or 200 iteration) with very small timestep than restarting in unsteady (for another 500/1000 iterations), i really would like to use the automatic time step to be sure that the first iterations solve the starting part of the simulation avoiding divergences.