|
[Sponsors] |
July 30, 2009, 04:23 |
Unstedy behavior
|
#1 |
New Member
Kostas
Join Date: Mar 2009
Posts: 23
Rep Power: 17 |
I try to solve an unsteady state of a problem having a initial steady state. I have determined the residuals at 1e-4. When start unsteady analysis the value of residuals produce and after reduce until reach the value of residual 1e-4. Then, time is changed. This behavior occur for 20-30 time steps but after stopped and by each time step the solution converged immediately and it is not appeared change in flow field.... Why occur that?? Is the problem by nature steady or occur something else?
|
|
July 30, 2009, 14:38 |
|
#2 |
Senior Member
Micael
Join Date: Mar 2009
Location: Canada
Posts: 156
Rep Power: 18 |
Please, print here what appear in your Fluent console so we can better understand what happen.
Have a good day. Micaël |
|
July 30, 2009, 17:33 |
|
#3 |
New Member
Kostas
Join Date: Mar 2009
Posts: 23
Rep Power: 17 |
So... I load the picture of residuals. As it is appeared, initialy there is a state in which each of time step the residual is increased and after reduce. However, after 49000 iterations the line of residual is changed and all of time steps converged directly in 1 iteration. With other words, like as dont occur change of flow field by run of time. Is it realy steady state the physical problem or reach in steady problem as a result of diffusion of solute?
|
|
July 30, 2009, 18:12 |
|
#4 |
Senior Member
Micael
Join Date: Mar 2009
Location: Canada
Posts: 156
Rep Power: 18 |
That behavior might be normal. If you disturb the system right at the beginning of the transient simulation, then it is normal that first time steps take many iterations. Then, after a while, the system might tend to a steady state and hence requires less iterations.
In other words: If you start your transient simulation in non-equilibrium state and that physically your system should tend to return at equilibrium, then it is possible that your case is normal. Micaël |
|
July 31, 2009, 03:43 |
|
#5 |
New Member
Kostas
Join Date: Mar 2009
Posts: 23
Rep Power: 17 |
Therefore, the physical problem is steady due to this result of residual. If was unsteady will be followed similar bahavor by start behavor.
But as a steady problem why dont appear the final solution by steady state analysis and it need to run with unsteady state analysis?? Help for this below: Consequently, if we want to take a right solution by a physical steady problem, we must try a steady state analysis for initial contition and after an unsteady analysis to take the right solution. If we stay at steady state the solution isnt right. Right? |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
transient behavior dependent on time step | Fab | Main CFD Forum | 2 | May 29, 2008 11:38 |
CFX-Solver, issue with convergence behavior | Andy | CFX | 7 | September 5, 2006 03:24 |
Steady behavior prediction from unsteady data !! | Mamun | Main CFD Forum | 4 | November 23, 2005 02:09 |
flow behavior in lid-driven microcavity flow | Xiaodong Niu | Main CFD Forum | 0 | May 12, 2003 05:30 |
UNSTEDY FLOW OVER A CIRCULAR CYLINDER | sanjib das sharma | FLUENT | 5 | March 16, 2001 01:49 |