CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM Running, Solving & CFD (
-   -   how to handle this kind of crash (

keishawillstone June 30, 2009 07:58

how to handle this kind of crash
Hi there,
My application crashes due to initial field set to 0 for some variables(the division operator can not handle it).
So I gave value to the fields initially set to 0 and I ran it,it was ok with the division but I had another trouble with the operator exp()
Then I went on the internet and found out an alternative about how to handle this,I tried and came out with the same error message!!
Here is what I found on the internet:
1-found out from the printed stack trace which files contain the functions that crash.Copy these into your local directory(At this stage I found out that the function that crashes is GeometricScalarField.C so I copied it and put into the directory of my application named keisha)
2-Add the files to your Make/files
What I did in my Make files is like this:

EXE = $(.....)/keisha

3-in Make/options add
to EXE_INC and recompile
And I did it but it seems like it doesn't work out and gives this message
"No rule to make target'GeometricScalarField.C ' needed by 'GeometricScalarField.dep '" when I compile it
I don't know whether I did well but I am quite stuck!!!I dont know what to do!!!
Can somebody out there help me out please?
Best Regards

sachin June 30, 2009 12:04

hi there,
Could you elaborate what exactly are you trying ....and why

keishawillstone July 1, 2009 06:21

Actually,I'm elaborating a code for heat transfer model in an incinaretor,this code contains severals lines.for the compiling it's ok but when I run it,it crashes at this line:

volScalarField rchim = ag/((exp(bo2)-1)/(bo2*rcdiff));

bo2 is volscalarfiled of which the initial field is set to 0
the problem is: I tried to give value to the field but it seems like it doesnt either
then I found out a way to work it out(to take into account the division by zero) in the internet(as I explain in my first post) but I dont know but it doesnt work.So please can you help me?

sachin July 2, 2009 07:52

I m sorry... i still dont understand...
anyways whatever i understood ... is it really necessary to initialize with 0
what solver is it transient or steady ....what if you just keep it slightly more than 0 and run ...might take bit longer to converge or atleast get some idea if it runs some iterations
are you sure about the way it is being done if it is zero ....i doubt..can u point to that page from where u studied it...sounds something interesting

sachin July 2, 2009 08:22

i found that page ....but it is just debug ...whereas in ur case u know it is because of initial field....why debug more...guess changing initial field is gud idea

keishawillstone July 2, 2009 08:36

thank you,I will try to set it slightly more than zero,you right!!!
best regards,

All times are GMT -4. The time now is 06:04.