[sabago]

Dear Foamers,

I installed OF 2.4 in order to try the multi-grading feature. I have the following as part of my blockMeshDict

blocks
(
    hex (0 1 2 3 4 5 6 7) (1 100 1) 
    simpleGrading 
    ( 
        1                  // x-direction expansion ratio 
        ( 
            (0.12 0.12 1)    // 12% y-dir, 12% cells, expansion = 1 
            (0.1 0.1 3)    // 10% y-dir, 10% cells, expansion = 3 
            (0.78 0.78 1) // 78% y-dir, 78% cells, expansion = 1 
        ) 
        1                  // z-direction expansion ratio 
    ) 

);

but when I run blockMesh, I get the error below (which disappears as soon as I eliminate the multi-grading.

--> FOAM FATAL IO ERROR: 
wrong token type - expected Scalar, found on line 38 the punctuation token '('

file: /home/sandra/OpenFOAM/sandra-2.4.0/GSOFC/constant/polyMesh/blockMeshDict.blocks at line 38.

    From function operator>>(Istream&, Scalar&)
    in file lnInclude/Scalar.C at line 93.

FOAM exiting

Any help will be much appreciated

[Antimony]

Hi,

Perhaps you could try with OF-2.4.x instead? It seems like a couple of bugs in the multi-grading from 2.4.0 were resolved in 2.4.x

http://www.openfoam.org/mantisbt/view.php?id=1809

Hope this helps.

Cheers,
Antimony

[sabago]

Thank you Antimony!

I installed OF 2.4.X as well and now the multi-grading is working with the same case that wouldn't work in OF 2.4.0.

Best,
Sandra

[sabago]

blockMesh works fine in 2.4.x but my case stops running after a few iterations. It is the case that I have copied from 2.4.0 and it runs fine there. I have copied and recopied both the solver and case and it culminates in an error in 2.4.x

Is this a bug?

Best,
Sandra

[Antimony]

Hi,

What error message does it show? Can you post it here?

Cheers,
Antimony

[sabago]

Hi Antimony,

I've pasted the error below. But a few things to note are
1. The point in the code where the error is coming from seems to be the weakest point in my code. Even errors with the fvSchemes will invoke it and whenever I modify my solvers, it's the first part to break usually
2. In this case, I don't think it's my solver or case for they work fine in 2.4.x but as soon as I add the multi grading and blockMesh, and then run my code, I get this error after about 10 iterations.

here07
here09
#0  Foam::error::printStack(Foam::Ostream&) at ??:?
#1  Foam::sigFpe::sigHandler(int) at ??:?
#2  ? in "/lib/x86_64-linux-gnu/libc.so.6"
#3  ? in "/lib/x86_64-linux-gnu/libm.so.6"
#4  exp in "/lib/x86_64-linux-gnu/libm.so.6"
#5  Foam::exp(Foam::Field<double>&, Foam::UList<double> const&) at ??:?
#6  ? at ??:?
#7  ? at ??:?
#8  __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#9  ? at ??:?
Floating point exception (core dumped)

Best,
Sandra

[Antimony]

Hi,

I see that you have a floating point exception, as a result of which I am tempted to say that some quantity reaches zero and is used as the divisor. You might want to check the minimum area of an element via checkMesh and ensure that you don't have any zero area element that has come about because of the mesh grading.

Hope this helps.

Cheers,
Antimony

[sabago]

Hi again,

1. Thank you for your input but I seem to see a bigger issue. When I revert to simple grading and blockMesh, I still get the error when I run the solver so it doesn't make sense that the issue is from the grading coz reverting should fix it, no? So far, I have had to recopy the case to 2.4.x to make sure it's not my case that's causing problems and it runs fine if I just copy it and blockMesh without adding the multi-grading.
2.Here's the equation below; the parameters Rg (gas constant) and T(Temperature) are constants in the tranportProperties so I'm certain that there's no division by zero, otherwise, the error would say "FPE, division etc", no?

jbvagCell = iagCell*((Foam::exp(alphaaa*F*EttaaCell/(Rg*T)))-(Foam::exp(-alphaca*F*EttaaCell/(Rg*T))));

Best,
Sandra

[Antimony]

Hi,

My apologies. Floating point exception could be either division by zero or an extremely large number. As you have an exp term, again, it is tempting to say that that is what is causing it to blow up.

I would suggest that you write out the results at the end of every step (since you mentioned that it runs up to 10 iterations and then crashes) and view the results to help troubleshoot the problem. You will then know (hopefully!) which quantity is causing the problem.

Apart from this though, at the moment, I can't think of any other solution.

Cheers,
Antimony

[sabago]

Thanks Antimony,

How about the issue where i revert to simple grading and still error after i blockMesh and run the solver?

Sandra

[Antimony]

Hi,

I would suggest that you run my previous comment first with the regular blockMesh and figure out the problem.

Once that is settled, moving to a graded blockMesh (and troubleshooting it if necessary) should be easier.

Cheers,
Antimony