CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Under - relaxation factors in CFX help? (http://www.cfd-online.com/Forums/cfx/23588-under-relaxation-factors-cfx-help.html)

Adrijana January 31, 2007 08:58

Under - relaxation factors in CFX help?
 
I cannot find where to change the under - relaxation factors for my simulations in CFX-pre and/or CFX-Solver Manager. Please help?

opaque January 31, 2007 09:09

Re: Under - relaxation factors in CFX help?
 
Dear Adrijana,

What kind of under-relaxation factors are you referring to? Do you want to under-relax the discrete transport equations? The ANSYS CFX does not use explicit relaxation, but false transient timestepping. You will find that for a steady state run, you must set some kind of physical timescale, or let the solver autocompute one for you..

For additional information, you can refer to the theory section in the help.

Opaque

HekLeR February 6, 2007 20:42

Re: Under - relaxation factors in CFX help?
 
just run a smaller physical timescale instead.

cfx does not generally expose under-relaxation because the above is usually good enough.

Georges February 15, 2007 14:56

Re: Under - relaxation factors in CFX help?
 
Hi, if you're refering to Newton Pressure Under Relaxation factor you can change it through the defenition editor in post.

Reza_computational November 11, 2010 06:52

Quote:

Originally Posted by Georges
;80362
Hi, if you're refering to Newton Pressure Under Relaxation factor you can change it through the defenition editor in post.

please help me how i can find the definition editor in cfx-post. I need to get to converge the iteration...
thank u in advance
Reza

ghorrocks November 11, 2010 20:56

Adjusting these sort of solver parameters is almost always a bad idea in CFX. It is not like other segregated solvers where you had to adjust the under relaxation factors (and lots of other solver factors) to get convergence. If your simulation is not converging then 99% of the time the problem is resolved using the control on the front panel - after checking your physics and boundaries are correct, use a smaller timestep or a better initial condition to start things off.


All times are GMT -4. The time now is 06:07.