 |
 |
 |
|
|
|
[Sponsors] | ||||
March 19, 2012, 15:49
|
divergence transient study
|
#1 |
|
Member
andres
Join Date: Sep 2010
Posts: 44
Rep Power: 15  |
Hi folks
I am performing a transient study and after 3000 iterations it takes place a divergence of residuals and so my results... Is is: Density-based Energy On Laminar Flow Courant Number 0.1 (which works perfect) Implicit Second Order Transient condicionts: Time Step Size = 0.1 seg Number Steps = 5000 Max Iter/ Time Step = 50 Everything runs well and behaves as supposed but when it reaches 3000 iterations....diverges! Any advice?? is there anything else I can do?? Any suggestion?? Thanks in advance Regards |
|
|
|
|
|
March 22, 2012, 13:05
|
|
#2 | |
|
Member
banty
Join Date: Mar 2012
Posts: 52
Rep Power: 14 |
Quote:
Plz Increase the maximum coarse level under multigrid panel (Advanced solution control) and lets see its works or not. |
||
|
|
|
||
|
March 22, 2012, 13:07
|
|
#3 |
|
Member
andres
Join Date: Sep 2010
Posts: 44
Rep Power: 15 |
I will do it, thanks
|
|
|
|
|
|
|
March 22, 2012, 16:21
|
|
#4 |
|
Member
andres
Join Date: Sep 2010
Posts: 44
Rep Power: 15 |
I set even 10000 in max coarse and it didn't change at all the results...
|
|
|
|
|
|
|
March 23, 2012, 12:36
|
|
#5 |
|
Member
banty
Join Date: Mar 2012
Posts: 52
Rep Power: 14 |
no, even 10 is sufficient. 10000 is not correct(unreal). u can play with multigrid option. (pre sweeps, post sweeps) . hopefully it will help.
http://https://www.sharcnet.ca/Software/Fluent12/html/ug/node794.htm r u using explicit method to time advancement? ( u set courant no 0.1.) try using implicit method and please post whether it is working or not. |
|
|
|
|
|
|
March 23, 2012, 12:47
|
|
#6 |
|
Member
andres
Join Date: Sep 2010
Posts: 44
Rep Power: 15 |
yes I am using implicit method (itīs default) , and it gives me the possibility to set courant number. I have tried from 0.1 to 5 (0.1 more precise but slower and 5 very coarse)
max coarse number didnīt affect my calculations nor residuals. I am going to surf this site and Iīll let you know! Thank you! |
|
|
|
|
|
|
March 23, 2012, 13:12
|
|
#7 |
|
Member
banty
Join Date: Mar 2012
Posts: 52
Rep Power: 14 |
Hello cfdhydraulic,
Plase don't make different thread for same problem as i think u have two or three threads for this problem. its confusing, right!!. |
|
|
|
|
|
|
March 23, 2012, 13:16
|
|
#8 |
|
Member
andres
Join Date: Sep 2010
Posts: 44
Rep Power: 15 |
yeah sorry some times I get confused with this forum... sorry about that! no happens again
|
|
|
|
|
|
|
March 23, 2012, 13:38
|
|
#9 |
|
New Member
Nik
Join Date: Mar 2010
Location: Michigan
Posts: 25
Rep Power: 16 |
Try reducing the time step size and simultaneous reduce the URF
Nik |
|
|
|
|
|
 |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Transient conduction possible in fluent? | jlefevre76 | FLUENT | 2 | February 5, 2013 09:53 |
| Transient state study | akj | Main CFD Forum | 0 | February 1, 2012 07:53 |
| Quarter Burner mesh with periosic condition | SamCanuck | FLUENT | 2 | August 31, 2011 11:34 |
| Best practice for transient simulations? | siw | CFX | 5 | October 30, 2010 05:45 |
| transient run divergence | john | Siemens | 1 | November 26, 2005 01:15 |
Linear Mode
