Problem 1137

Summary:	G4FissLib issue
Product:	Geant4	Reporter:	Cristian Bungau <cbungau>
Component:	processes/hadronic/models/lll_fission	Assignee:	dennis.herbert.wright
Status:	RESOLVED FIXED
Severity:	major	CC:	cbungau
Priority:	P5
Version:	9.3
Hardware:	All
OS:	All

Description Cristian Bungau 2010-08-19 19:53:12 CEST

Hello,

I previously found an issue in G4NeutronHPFissionData.cc which resulted in the code (having the 232Th isotope inside the materials) crashing if used with $G4NEUTRONHP_SKIP_MISSING_ISOTOPES set to 1, (only for geant4.9.3 onwards) because it stopped using the fission cross-section of 232U for 232Th (which in reality does not undergo fission). Please see bug report:

http://bugzilla-geant4.kek.jp/show_bug.cgi?id=1118

The bug was kindly fixed by Tatsumi, by adding a safety check line to G4NeutronHPFissionData.cc, which is:

if ( ( ( *theCrossSections )( index ) )->GetVectorLength() == 0 ) return result;

Now that I have to use the G4FissLib model for fission, I ran into the same problem, and I don't know how to modify G4FissLib.cc in order to correct this.

I would be very grateful if you could please suggest a fix for G4FissLib.

Thank you very much,

Best regards,

Cristian

Comment 1 dennis.herbert.wright 2010-09-02 20:16:39 CEST

Thanks for the report.
I'll take a look.

Comment 2 Cristian Bungau 2010-09-06 12:02:57 CEST

Hi,

Since the Th232 element is causing all this problems, when used with $G4NEUTRONHP_SKIP_MISSING_ISOTOPES set to 1, I thought replacing 

   if((*(G4Element::GetElementTable()))[i]->GetZ()>89)

with

 if((*(G4Element::GetElementTable()))[i]->GetZ()>90)

in G4FissLib.cc will fix the problem by avoiding looking for the Th232 data, which does not undergo fission anyway, but it didn't. 

It looks like more C++ knowledge (than I have) is needed to fix this.

I'm wondering if putting fission data in G4NDL3.13 for Th232 with zero (or 1e-10 b) fission cross-sections at all neutron energies will fix this.

Best regards,

Cristian

Comment 3 dennis.herbert.wright 2010-11-22 18:19:02 CET

The author of the code, Jerome Verbeke (verbeke2@llnl.gov) reproduced the problem in 9.3.  Updating to 9.3.p02, the problem goes away.   Since there were many fixes
in that patch, this problem was likely solved.   However, Jerome would like for you to send your code to him so he can check further.   In the mean time, I will mark this problem as solved.  If anything turns up in Jerome's tests of your code,
we'll re-open the report.

Comment 4 Cristian Bungau 2010-11-30 10:59:05 CET

Dear Dennis,

Thank you for your reply. I installed the second patch (I was using geant4.9.3.p01) and indeed the problem is solved in this patch.

Thank you for looking into this,

Best regards,

Cristian

(In reply to comment #3)
> The author of the code, Jerome Verbeke (verbeke2@llnl.gov) reproduced the
> problem in 9.3.  Updating to 9.3.p02, the problem goes away.   Since there were
> many fixes
> in that patch, this problem was likely solved.   However, Jerome would like for
> you to send your code to him so he can check further.   In the mean time, I
> will mark this problem as solved.  If anything turns up in Jerome's tests of
> your code,
> we'll re-open the report.