CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   please help me --> FOAM Warning : From function Time::operator++() (http://www.cfd-online.com/Forums/openfoam-solving/109306-please-help-me-foam-warning-function-time-operator.html)

adambarfi November 14, 2012 13:02

please help me --> FOAM Warning : From function Time::operator++()
 
hi every body

any body knows what is the meaning of this warning:

Code:

Courant Number mean: 6.18417 max: 57.2403
deltaT = 4.41444e-94
--> FOAM Warning :
    From function Time::operator++()
    in file db/Time/Time.C at line 982
    Increased the timePrecision from 44 to 45 to distinguish between timeNames at time 0.000971272
Time = 0.000971271576785370013146225787181720079388469458

when this warning appeared, after some iteration the following errors appear:

Code:

#0  Foam::error::printStack(Foam::Ostream&) in "/home/mostafa/OpenFOAM/OpenFOAM-2.1.0/platforms/linuxGccDPOpt/lib/libOpenFOAM.so"
#1  Foam::sigFpe::sigHandler(int) in "/home/mostafa/OpenFOAM/OpenFOAM-2.1.0/platforms/linuxGccDPOpt/lib/libOpenFOAM.so"
.
.
.

Thank you

mturcios777 November 14, 2012 13:57

If you read the warning, its telling you that the time precision had to be increased as your timestep has just become really really REALLY small (10^-94!). Your simulation crashes after for reasons likely related to that.

Your courant number is really huge at that point (56!) so you should really check what's going on in your simulation, as these are only symptoms of a deeper underlying problem.

adambarfi November 14, 2012 14:09

Quote:

Originally Posted by mturcios777 (Post 392110)
If you read the warning, its telling you that the time precision had to be increased as your timestep has just become really really REALLY small (10^-94!). Your simulation crashes after for reasons likely related to that.

Your courant number is really huge at that point (56!) so you should really check what's going on in your simulation, as these are only symptoms of a deeper underlying problem.

Thank you Marco,

I know that the time precision has been really small and my Courant number is huge but I don't understand where is the source of my mistake.

Does it originate from my solver?!!

mturcios777 November 14, 2012 14:18

That is the source of your problem. Unless you have some amazing numerical schemes and/or implicit methods at work, there is no way that a Courant number that big (or time-step that small) will remain stable for long:

http://en.wikipedia.org/wiki/Implicit_method
http://en.wikipedia.org/wiki/Courant...Lewy_condition

I would recommend looking at the output from your simulation at that point. I can almost guarantee something unphysical is happening.

petr.f. November 14, 2012 14:18

At first post some more detailed info (which solver do you use, what are the basic solver settings, characteristics of the solved case.. otherwise we might be as well guessing from crystal ball). It may be caused by a lot of things, e.g.
- does it happen at the beging of your simulation? If yes your initial timestep might be too big...
- what Courant number do you set in controlDict?
- boundary conditions...

mturcios777 November 14, 2012 14:20

Quote:

Originally Posted by petr.f. (Post 392117)
At first post some more detailed info (which solver do you use, what are the basic solver settings, characteristics of the solved case.. otherwise we might be as well guessing from crystal ball). It may be caused by a lot of things, e.g.
- does it happen at the beging of your simulation? If yes your initial timestep might be too big...
- what Courant number do you set in controlDict?
- boundary conditions...

This is also super important. Here are some guidelines to have a better change of having your question answered:

http://www.cfd-online.com/Forums/ope...-get-help.html

adambarfi November 14, 2012 14:40

Quote:

Originally Posted by petr.f. (Post 392117)
At first post some more detailed info (which solver do you use, what are the basic solver settings, characteristics of the solved case.. otherwise we might be as well guessing from crystal ball). It may be caused by a lot of things, e.g.
- does it happen at the beging of your simulation? If yes your initial timestep might be too big...
- what Courant number do you set in controlDict?
- boundary conditions...

I'm using a modified viscoelasticFluidFoam that I made it. a solver that can solve natural convection in a viscoelastic fluid.

Quote:

- does it happen at the beging of your simulation? If yes your initial timestep might be too big...
No, it happen at 7.38s.

Quote:

- what Courant number do you set in controlDict?
my maxCo is 0.3

I think I made a mistake somewhere in my solver, but I don't know where!!!!!

petr.f. November 14, 2012 15:18

Well, if you made a mistake in your solver, much more information would be needed (perhaps the whole code?).

But, if I were you, I would observe, where the time step begins to decrease significantly (in order of magnitude), set this moment as an end of the simulation and researched behaviour of all the variables (p, U, T, K, epsilon or omega respectively)...

adambarfi November 16, 2012 05:35

1 Attachment(s)
Dear petr,
I attach the log file

petr.f. November 19, 2012 06:20

Hi Mostafa, I've checked your log and it seems that problems begin to appear after approx. 10 iterations Just to be sure, I would try much smaller time step for the beginning (~ 1e-5, but I don't know any details about your simulation...). But because the divergence begins so quickly, I would guess there might be some error in the code... Anyway, I've seen a post about visco-elastic simulations in openfoam not so long ago - http://www.cfd-online.com/Forums/ope...fluidfoam.html . Try to find some hints there...

adambarfi November 19, 2012 07:12

Quote:

Originally Posted by petr.f. (Post 392918)
Hi Mostafa, I've checked your log and it seems that problems begin to appear after approx. 10 iterations Just to be sure, I would try much smaller time step for the beginning (~ 1e-5, but I don't know any details about your simulation...). But because the divergence begins so quickly, I would guess there might be some error in the code... Anyway, I've seen a post about visco-elastic simulations in openfoam not so long ago - http://www.cfd-online.com/Forums/ope...fluidfoam.html . Try to find some hints there...

thank you petr,
I will do it.

another question:
any body knows how can I change the transport and viscosity model?


All times are GMT -4. The time now is 22:34.