|
[Sponsors] |
January 17, 2008, 11:43 |
Constant field inside the volume
|
#1 |
Guest
Posts: n/a
|
Dear colleagues,
I would like to have a steady state simulation with the constant field of some quantity in all domain. I can calculate the field in any point inside the volume, but the calculation is slow enough. If I do such calculation every iteration, it will slow down the simulation very much. Instead, I would like to calculate the field only one time, store it into, for example, user variable, and use it in the all iterations. But I can't imagine how to do that. Could anybody help me? Thank you! |
|
January 17, 2008, 14:16 |
Re: Constant field inside the volume
|
#2 |
Guest
Posts: n/a
|
We need more information to help you:
Q1: What else is involved in your steady state simulation? I assume there's a steady state flow calculation involved. Describe it a bit. Q2: Also, what is the 'field'? How is it to be calculated: (a) - from an equation that needs to be *solved*, e.g. such as a diffusion equation for a concentration? or (b) - can it just be *computed directly* once you have the velocity / pressure / temperature that CFX solves for already in your simulation? Does the field influence the flow pattern? Let us know, and then we can give better help. andy |
|
January 17, 2008, 15:09 |
Re: Constant field inside the volume
|
#3 |
Guest
Posts: n/a
|
You need no more information. Please reread my question.
I would like to have additional scalar variable in my simulation and initialize it only once with the exturnal calculated values (it has no matter, what are these value and meaning). I need just to initialize it once. But now solver recalculate the values of such variable every iteration step. |
|
January 17, 2008, 17:54 |
Re: Constant field inside the volume
|
#4 |
Guest
Posts: n/a
|
Ben,
Andy was right to ask the question. Depending on what type of scalar field it is will determine appropriate ways of doing it. Also, knowing the background of an issue always helps put it in context. If you define a CEL expression and send it to a user scalar variable which does not include properties which vary during the simulation it will be calculated once and kept constant during the simulation. I think this is what you are asking about. Glenn Horrocks |
|
January 18, 2008, 02:23 |
Re: Constant field inside the volume
|
#5 |
Guest
Posts: n/a
|
Thank you, Glenn!
Ok, I'll try to explain my needs. I have additional volumetric scalar variable in the domain. I add transport equation for this variable. The source in the subdomain(==domain) includes my user function. I tried to describe my user function through the function that interpolate table data, calculated outside CFX. The arguments for the function are only (x, y, z) coordinates. This is the scalar field I talked earlier. Interpolation function works quite slowly. If I ran the simulation it lasts many times longer comparing to the case without additional variables or the case without interpolation function. That means my interpolation function is calculated every time step, not once. Thank you. |
|
January 20, 2008, 05:08 |
No Ideas? *NM*
|
#6 |
Guest
Posts: n/a
|
|
|
January 20, 2008, 17:59 |
Re: Constant field inside the volume
|
#7 |
Guest
Posts: n/a
|
Hi,
My understanding was that if the inputs to a function were constant and could therefore just be evaluated once at the start of a simulation and kept for future use. I would talk to your support office about this. Glenn Horrocks |
|
January 21, 2008, 04:48 |
Re: Constant field inside the volume
|
#8 |
Guest
Posts: n/a
|
Thank you for the answer.
But why do you think that "If you define a CEL expression and send it to a user scalar variable which does not include properties which vary during the simulation it will be calculated once and kept constant during the simulation."? Is that stated somewhere in the CFX manuals/specs? |
|
January 22, 2008, 18:07 |
Re: Constant field inside the volume
|
#9 |
Guest
Posts: n/a
|
Hi,
I discussed this issue with a developer some time ago. It might also be in the documentation, I have not checked. Glenn Horrocks |
|
February 6, 2008, 06:19 |
Re: Constant field inside the volume
|
#10 |
Guest
Posts: n/a
|
I think it's still an issue.
|
|
February 6, 2008, 17:32 |
Re: Constant field inside the volume
|
#11 |
Guest
Posts: n/a
|
Hi,
If you don't take my word for it then contact CFX support and ask them. Glenn Horrocks |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Sample a volume field | gwierink | OpenFOAM Post-Processing | 15 | September 22, 2022 12:49 |
Initialize volume temperature field with a computed field | magnounibo | FLUENT | 2 | December 11, 2009 04:06 |
How to create a fluid volume inside a solid volume | sanin | FLUENT | 4 | November 13, 2008 16:17 |
for loop inside a cell_loop? | MHDWill | FLUENT | 0 | September 26, 2007 22:24 |
transient temperature field with constant velocity | Törnquist | CFX | 0 | September 16, 2003 05:22 |