[Nek5000-users] nek on new setup (fwd)

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

Hi Mike,


In the current repo version of conduct.f, the default for the array t(1,1,1,1,1) in axisymmetric case (ifaxis=.true.) is a scalar (isd=1). 

Only in MHD axisymmetric cases it is treated as a vector component -- meridional velocity in axisymmetric flow w/ a swirl :

conduct.f:55

         isd = 1
         if (ifaxis.and.ifmhd) isd = 2 !This is a problem if T is to be T!

So for your problem, you still need to modify this file and set isd=2 if ifaxis=.true. irrespective of ifmhd but watch out if you decide to add a real temperature/passive scalar later :)

We hope we will have a better solution for it it some point.

Best,
Aleks

 



----- Original Message -----
From: nek5000-users at lists.mcs.anl.gov
To: users at nek5000.mcs.anl.gov
Sent: Friday, June 11, 2010 3:32:25 PM GMT -06:00 US/Canada Central
Subject: Re: [Nek5000-users] nek on new setup (fwd)



Hi Mike,

I'm fwd'g to the nek5000 users list (are you on this?).

I think Aleks can answer the question about axisymmetry +
meridional velocity.

Paul


---------- Forwarded message ----------

On an unrelated note, is there anything special that needs to be done for a 2D 
axisymmetric problem where temperature is used to represent the meridional 
velocity?  In the old version, I made a manual change to one of the .f 
routines.  I'm guess that is all "automated" in the present version, but I 
wanted to make sure.

Thanks.
--Mike