[Nek5000-users] Problem with passive scalar fields - subroutine outfld

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

Dear Donato,

Thank you for bringing to my attention the issue with outpost/outpost2.

It is almost certain that you want to use outpost2 - that was designed by one
of our developers for exactly the scenario you describe where you have many
passive scalars.  (Though I don't think it was tested also with conjugate heat
transfer.)

Regarding the two flags in the IFTMSH line, the first one is actually
for the "mesh" which is field 0, i.e., in moving boundary problems the
moving part of the mesh could be either just fluid or fluid and solid.
Thus, the flags are technically addressed to fields

 1 2 3 4 ... for IFNAV/IFADVC

and 

 0 1 2 3 ...  for IFTMSH

So, would things be functioning correctly for you if you use only
outpost2() ?

There should be no issue in restarting from fields generated by this
process.

Paul

________________________________________
From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov]
Sent: Wednesday, April 02, 2014 12:33 PM
To: nek5000-users at lists.mcs.anl.gov
Subject: [Nek5000-users] Problem with passive scalar fields -   subroutine      outfld

Dear Paul,

Thanks for your answer.
I already had flags as following:
    T T T T T T T T T T T T IFNAV & IFADVC (convection in P.S. fields)
    F F T T T T T T T T T T IFTMSH (IF mesh for this field is T mesh)
because all passive scalar fields are temperature fields in my case. Is

it correct?
(By the way, why 2 flags for field 1, velocity, in the IFTMSH line? It
seems to be that the first one doesn't matter, i.e. I don't see
differences in results using "T F T T ..." instead of "F F T T ...").

Performing some tests (also setting p93=p94=p95=0. in .rea in some
cases), I noted that if subroutine outfld is called during the
simulation, then PS fields change as described in my previous mail.
Indeed, if in .rea file is p15>p11 (i.e. outfld called at last step
only) then PS fields are correct.
And PS fields are correct also if vx,vy,vz,pr,t are dumped with outpost2
during the simulation while outfld is not called.
This is the reason why I thought that the problem is subroutine outfld
instead of outpost2.

And I confirm that only some of the PS fields are changed: running two
different simulations that differ only for the number of PS fields, a
same field may be correct in one simulation and not in the other.

My only solution, for now, is to run the simulation with p15>p11 in
.rea, saving variables directly with outpost2.
Is it a problem to restart a case with a field file obtained with
outpost2(vx,vy,vz,pr,t,ldimt,'res') instead of outfld? Field files
obtained seem to have different size, but results for restart seem to be
the same.

Best reagards,
Donato




-----Original Message-----
Date: Mon, 31 Mar 2014 04:16:30 +0000
From: nek5000-users at lists.mcs.anl.gov
To: "nek5000-users at lists.mcs.anl.gov"


Dear Donato,

I've not looked closely at this yet.  However, I suspect it might be
related to the projection.
Can you run your example with param 94 = 0 in the .rea file ?  This
should turn off the velocity/passive scalar projection.

Another question ---

in your .rea file, where it says "IFADVC" there should be two lines
like:


  T F F F F F F F F F F IFNAV & IFADVC (convection in P.S. fields)
  F F T T T T T T T T T T IFTMSH (IF mesh for this field is T mesh)


For your cases, you probably want them:


  T T T T T T T T T T T T IFNAV & IFADVC (convection in P.S. fields)
  F F T F  F F F F F F F F F  IFTMSH (IF mesh for this field is T mesh)

This would do the following:

1) It turns on advection for all fields (fluid, temperature, and passive
scalars)

2) It sets the mesh field to be Fluid
    It sets the velocity field to be Fluid
    It sets the temperature to be Temperature (i.e., fluid + solid)
    It sets the remaining passive scalars to be Fluid

For conjugate heat transfer, normally only the temperature acts over
both the fluid and solid element set, which is why I set those flags as
above.

Let me know if this resolves the issue for you...

Regards,

Paul

_______________________________________________
Nek5000-users mailing list
Nek5000-users at lists.mcs.anl.gov
https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users