code_saturne User's Forum

I'm looking to rotate a mesh using a single instance case of CS.
To do this I've tried specifying omegaz in usini1 but it has no effect. However, if I specify this with icorio=1 the frozen rotor method works.
Furthermore setting up two instances with half the mesh each and giving them equal values of omegaz with give my desired effect but seems a rather strange way around the problem.
Is there something in place that will prevent specifying a rotational vector for the unsteady set up of a single instance case?
 
Thanks,
James

Actually, the rotation is accounted for in the code when the indicator imobil is set to 1. At the moment, and because we hadn't seen any particular needs to do differently, this variable is set to 1 only in the case of coupling two or more domains. So if you want to activate the rotation in the case of a single instance, you'll have to modify it by yourself in usini1.f90 (close to the icorio definition for example).
David

Thanks David,
I guess you are correct that there is no obvious needs although I have seen cases such as rotating barrels / jars where I can see a use. I intended to rotate a cylindrical mesh to simulate an object rotating in free-stream. Nonetheless I achieved my aim with the ALE module - I expect the methods of rotating domain by either method is quite similar.
James

Indeed, if you force the nodes displacement with the ALE module, it should nearly be the same as with the rotation method (for rotor / stator interaction). There may be small differences due to the way the mass flux update (due to the rotation of the mesh) is handled in both cases.
David