[hamidciv]

hi dear all
i doing run and finished, unfortunately i dont control convergence criteria (residuals and imbalane) in during solver manager with monitor plot, now how i can control this criteria in cfx-post? i want control that finished solver be induce of reach to max iterations or convergence criteria Is satisfied?
thanks a lot for any comment.
best regards

[ghorrocks]

Convergence criteria affect the simulation. It does not do anything in post processing.

[Antanas]

Originally Posted by hamidciv

hi dear all
i doing run and finished, unfortunately i dont control convergence criteria (residuals and imbalane) in during solver manager with monitor plot, now how i can control this criteria in cfx-post? i want control that finished solver be induce of reach to max iterations or convergence criteria Is satisfied?
thanks a lot for any comment.
best regards

You can open CFX solver manager in stand alone mode, then press 'View run information for a completed Solver run' and select your .res file to see corresponding Solver output, Residuals plots etc.

Or you can simply look inside .out file to see the reason solver finished the run.

[hamidciv]

der antanas
unfortunately i dont underastand your mean, how i can open solver manager in stand alone mode?
about second method as shown figure , i have about 43.out file in my reult file induced stops in during run, how open out file and what reason exist for solver finished run with look inside out file?
if your possible, please explain more.

[ghorrocks]

Open the .out files in any text editor.

[Antanas]

Originally Posted by hamidciv

der antanas
how i can open solver manager in stand alone mode?

Via CFX Launcher

Originally Posted by hamidciv

der antanas
about second method as shown figure , i have about 43.out file in my reult file induced stops in during run, how open out file and what reason exist for solver finished run with look inside out file?
if your possible, please explain more.

Each .out file is text file which corresponds to separate solver run. If you want to know why did 43rd solver run stop, you should open 43rd .out file. Inside that file look for 'Termination and Interrupt Condition Summary' section near its end.

[hamidciv]

hi dear anatans
first of all , thanks for clear explanation
unfortunately still, i have two question, i went cfx laucher, cfx solver manager, view run information and loaded my .res file and then view RESIDUAL as shown follow.
Do you mean that was?(select your .res file to see corresponding Solver output, Residuals plots etc.) max res residual located top 1e-04 that define for residual target, but RMS resdidual target located under it, i define residual type RMS in cx pre, Is it is no problem?

as you mentined i open out file in note pad , but i dont see any text about terminate, i attached ,out file, if your possible please check it.
thanks for attention

[hamidciv]

Originally Posted by ghorrocks

Open the .out files in any text editor.

thanks in advance dear glenn

[ghorrocks]

i define residual type RMS in cx pre, Is it is no problem?

Now your question is a FAQ: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

[Antanas]

Originally Posted by hamidciv

max res residual located top 1e-04 that define for residual target, but RMS resdidual target located under it, i define residual type RMS in cx pre, Is it is no problem?

Actually it is up to you whether it is satisfactory. Because it is you who decides what accuracy is needed and who sets target values for residuals. As I remember it is recommended to set RMS <=1e-5 or MAX <=1e-4.

Originally Posted by hamidciv

as you mentined i open out file in note pad , but i dont see any text about terminate, i attached ,out file, if your possible please check it.
thanks for attention

Your file seems truncated. When you open .res file in CFX Solver manager it should also load corresponding solver output file (look at 'Out File' tab), you can investigate it there.

[hamidciv]

hi dear antanas
thanks a lot again for quick and clear answer
dear antanas, i find condition summary in outfile tab, i attached in here.
whether it means that before solver raeched maximum number of iteration or residual target, solver finished induced of recah to total time that i defined?
dear antanas , in transient simulation usually what number of iteration recommended?
I considering total time 52s with time steps 0.08, my run was very long, about 4 day, also my results is good.
Please guide me
Best wishes

[Antanas]

Originally Posted by hamidciv

hi dear antanas
thanks a lot again for quick and clear answer
dear antanas, i find condition summary in outfile tab, i attached in here.
whether it means that before solver raeched maximum number of iteration or residual target, solver finished induced of recah to total time that i defined?
dear antanas , in transient simulation usually what number of iteration recommended?
I considering total time 52s with time steps 0.08, my run was very long, about 4 day, also my results is good.
Please guide me
Best wishes

Solver finished because total simulation time was reached. In transient simulations there's no such solver limiter as max. number of iterations. There's max. number of coefficient loops which is max number of iterations within timestep. It is recommended to keep max. number of coefficient loops within 3-5. If residuals don't fall below target values within that number of coef. loops, it is recommended to decrease time step size rather than increase max. number of loops. However some problems may require larger number of coef. loops.

[hamidciv]

dear antanas , sorry , my mean was same cofficient loops that I say it wrong number of iteration.
Is there such a situation ( However solver is finished induced of reached to simulation time but my result is good in comapred to experimental results) ,simulation cant be correct not at all?
only have one problem was falling water surface in inlet, i expressed this problem in this forum, according to glenn , it was inducing of location define inlet in my model.
i appreciate you for all
hamid

[Antanas]

Originally Posted by hamidciv

dear antanas , sorry , my mean was same cofficient loops that I say it wrong number of iteration.
Is there such a situation ( However solver is finished induced of reached to simulation time but my result is good in comapred to experimental results) ,simulation cant be correct not at all?
only have one problem was falling water surface in inlet, i expressed this problem in this forum, according to glenn , it was inducing of location define inlet in my model.
i appreciate you for all
hamid

Sorry, I don't understand your question.

[hamidciv]

dear antanas , sorry for My bad writing
Despite this situation, in my simulation, whether it is wrong? (my mean this is :there is no chance of being correct results in my model And it is certainly wrong?)
If the answer is yes, why my results is good?
i defined coefficient loops in model 1 to 10 or same default in cfx, with regard my simulation time is high (52s), whether lack of converging in my model can be
induced of high coefficient loop?

[Antanas]

Originally Posted by hamidciv

dear antanas , sorry for My bad writing
Despite this situation, in my simulation, whether it is wrong? (my mean this is :there is no chance of being correct results in my model And it is certainly wrong?)
If the answer is yes, why my results is good?
i defined coefficient loops in model 1 to 10 or same default in cfx, with regard my simulation time is high (52s), whether lack of converging in my model can be
induced of high coefficient loop?

Who said that your model is wrong? I know nothing about what you model and how you do that. If you have experimental data and your results are in good agreement with 'em then what the problem?

Why do you think that you have lack of convergence? Because of RMS target = 1e-4? Again it's up to you to decide what accuracy is needed. Maybe in your case the values that you use are sufficient, considering good agreement with experiment.

[hamidciv]

What makes I think me, my simulation is wrong,The reason is that:

as you know , in my model, solver finished induced of reached to simulation time.

dear anatans , my main question is:

whether finishing of solver induced of reached to simulation time in transient simulation is not problem and it not affect on results?

as i know solver should finished Due to the following reasons:

1-reached to convergence criteria

2- raeched to max of iteration that in transient simulation as you epressed it is coefficient loops

but in my case , None of the above reasons is not satisfied.

i worry about it, Otherwise my result have good agreement with experimental model.

many thanks for your follow-up and spending time for me.

[Antanas]

Originally Posted by hamidciv

What makes I think me, my simulation is wrong,The reason is that:

as you know , in my model, solver finished induced of reached to simulation time.

dear anatans , my main question is:

whether finishing of solver induced of reached to simulation time in transient simulation is not problem and it not affect on results?

as i know solver should finished Due to the following reasons:

1-reached to convergence criteria

2- raeched to max of iteration that in transient simulation as you epressed it is coefficient loops

but in my case , None of the above reasons is not satisfied.

i worry about it, Otherwise my result have good agreement with experimental model.

many thanks for your follow-up and spending time for me.

You are wrong. In transient simulation overal solution process normally finishes when total time or max number of timesteps (depending on what you specify in 'Analysis type') are reached. Next, in transient simulation solver gets solution for a timestep through iterations and max. number of coefficient loops as well as residual targets are limiters for a timestep but not whole solution process. If you don't know such basic things, you'd better do some reading in CFX help files and FAQ.

[hamidciv]

dear antanas
thanks a lot for all.
best regards