
[Sponsors] 
July 26, 2013, 01:56 
solution is converged but a problem

#1 
Senior Member
Saeed
Join Date: Jan 2013
Posts: 177
Rep Power: 4 
''Hello
I am modeling an open channel flow with two inlet for air and water (velocity inlet). Without using "Open channel flow" In middle of channel i have two pressure inlet for entraining air into the flow. In steady state, i do not have problem but in unsteady state, i ran the solver and everytime the flow arrive to middle of channel, fluent write: "solution is converged" but it is keeping the solve and does not stop. what is that problem? I should say the residuals are arriving to residual criteria and they are 0.001 for all please see attached image Thanks 

July 26, 2013, 02:20 

#2 
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,992
Rep Power: 30 
max it / time step = 20
dt = 0.01 s number time steps = 300 So from my side, I understand that you never reached convergence before this time step (t=1.2s) and solver switched to next time step each time after 20 iterations. But residuals continued to fall until all your convergence criteria were satisfied (which is now the case at t=1.2s). So you get convergence, but you don't reach the max number of time steps, and the solver will continue to iterate untill solver time will be 300 * 0.01 = 3s. But as you already get converged solution at t=1.2s, now solver will need less than 20 iterations per time step, because you will get faster convergence.
__________________
In memory of my friend Hervé: CFD engineer & freerider 

July 26, 2013, 02:40 

#3 
Senior Member
Saeed
Join Date: Jan 2013
Posts: 177
Rep Power: 4 
Thanks for your quick reply.
I have the message "solution is converged' before 1.2 s? i should let it to continue? Because the flow is not arrive to outlet. I am using vof implicit. Can you tell me how can i calculate the time step size and number of time step and max iterate in vof implicit? since the courant number is unactivate in vof implicit ? I do not know what are they? There is no guide in fluent manual for implicit. 

July 26, 2013, 03:28 

#4  
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,992
Rep Power: 30 
I don't think
Quote:
for dt you could take 0.5*min_cell_length/max_velocity_expected max it = 20 is ok. As I said, once you get converged solution for one time step, then it will converge within 5 or 10 iterations If number of time step is reached, but the flow hasn't arrived to outlet, then increase numer of time step...
__________________
In memory of my friend Hervé: CFD engineer & freerider 

July 26, 2013, 04:37 

#5 
Senior Member
Saeed
Join Date: Jan 2013
Posts: 177
Rep Power: 4 
What is difference between the time in fluent model and time in laboratory model?
Suppose i have a 2 meters long channel that velocity into the channel is 4 m/s, then the flow cross from channel in 0.5 second. should i enter for example 0.001 for time step size and number of time step: 0.5/0.001=500 ? Is that right? If yes, what about max iteration? If no please explain more? I do not have any experience in unsteady modeling. 

July 26, 2013, 05:07 

#6 
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,992
Rep Power: 30 
yes it is, but it is not so important, since you can continue iterating when you reached given number time steps. (start from latest time step)
Most important is dt (time step size)
__________________
In memory of my friend Hervé: CFD engineer & freerider 

July 27, 2013, 01:59 

#7 
Senior Member
Saeed
Join Date: Jan 2013
Posts: 177
Rep Power: 4 
Is there any problem if the flow encounter with up boundary in open channel? Because it is look like every things are fix and solution not converge after more iterations.


July 31, 2013, 00:40 

#8  
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,992
Rep Power: 30 
Quote:
Maybe you can post pictures of your model.
__________________
In memory of my friend Hervé: CFD engineer & freerider 

Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Convergence  Centurion2011  FLUENT  24  May 9, 2015 08:02 
Unable to get converged solution using SimpleFoam  jr33  OpenFOAM Running, Solving & CFD  5  February 16, 2013 05:43 
Solution converged after 1 iteration!!  mecarlg  FLUENT  1  April 12, 2010 11:02 
Calculate a converged flow solution  kumar  FLUENT  2  May 24, 2007 01:33 
Numerical solution to the rotating disk problem?  johny  Main CFD Forum  7  September 5, 2005 05:53 