[Nek5000-users] dns mode vs perturbative one

[nek5000-users at lists.mcs.anl.gov]

Hi neks,
I'm trying to perform two simulations with Nek5000.

The first one is a DNS of turbulent plane Poiseuille flow with the classical dns code (npert = 0), and the second one is the same but in nonlinear perturbative form (npert = -1), using the classical parabolic profile as a base flow. I'm using the same spatial discretisation. The only difference is the time order integration: 3 in the first case and 2 in the second. I'm using the same parameter in the .rea file in the turbChannel example.

My questions are:
1) Why in the first formulation (npert=0) i'm able to use a cfl greater than 1, and for the second (npert=-1) I get an error in the gmres solver (it does not converge) for cfl greater than 0.4 after few time steps.
2) Even when I use a cfl lower than 0.4 for the perturbative form, the gmres solver fails again when the flow reaches a fully turbulent state and I don't understand why.

I hope to have been clear,
Thank.

-- Mr. Mirko FaranoPh.D. student at Politecnico di BariTel.: +39-0883529289 / +39-0805963462Mob.: +39-3202342719Department of Mechanics Mathematics and Management - Fluid Machinery and Energy SystemsVia Re David, 200 - 70125 Bari (Italy)mirko.farano at poliba.itIn joint supervision with theEcole Nationale Supérieure d'Arts et Métiers de Paris - ParistechTel.: +33-144246436Mob.: +33-789822134Dynfluid Laboratory151 Boulevard de l'Hôpital - 75013 Paris (France)mirko.farano at ensam.eu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20160315/6145999c/attachment.html>