Hi Everyone,
Previously I was using OpenMC 0.7.1. and I have upgraded OpenMC to version 0.8.0. But when I run OpenMC for a basic pincell it gives error:
ERROR: Delayed neutron with energy-dependent group probability not implemented
application called MPI_Abort(MPI_COMM_WORLD, -1) - process 0
[unset]: write_line error; fd=-1 buf=:cmd=abort exitcode=-1
:
system msg for write_line failure : Bad file descriptor
The snapshot of this is attached. Please tell me what kind of error this is and how it can be resolved.
Best Regards,
Mirza
Hi Everyone,
Can anybody help how to resolve this issue?
Best Regards,
Mirza
Hi Mirza,
The error appears to be due to the cross section library you are using since this error is occurring when the ace files are being read in. What cross section library are you using?
Sam
Mirza,
There is a known issue with JEFF 3.2 data, which I suspect you are using. The delayed fission neutron group abundances in JEFF 3.2 are dependent on the incident neutron energy causing fission. Version 0.8.0 is not yet equipped to handle this situation. You have two options:
- I’ve attached a diff which you can use to modify your version of OpenMC so that it will ignore the energy-dependent of the delayed group abundances (it basically amounts to commenting out lines 729 and 737-740 in src/ace.F90). This will result in very small errors in delayed neutron production, with some groups receiving an incorrect yield at higher incident neutron energies.
- Use the latest developmental branch in which we’ve switched from using ACE data to a new HDF5 format. The conversion process is able to recognize the energy-dependent delayed group probabilities and handle them appropriately. This option will require that you convert your ACE data; however, I’ll point out that there is a utility script get_jeff_data.py that will automatically download the JEFF 3.2 data and do this for you. If you choose this route, let us know and we can help you along further.
Best regards,
Paul
jeff32-fix.diff (1.18 KB)
Hi Sam,
The cross section library I am using is JEFF3.2. If there is an issue with this library then which library should be used?
I am using this because it have data on many different temperatures.
Best,
Mirza
Hi Paul,
Thank you for your help. I would like to use latest developmental branch. I will be highly thankful to you if you explain this method in detail.
Best,
Mirza
Hi Paul,
I have commented line 729 and 737-740 in src/ace.F90 as you mentioned and compiled OpenMC again. And it worked !
Thank you so much for your sincere guidance.
OpenMC is now working fine and it generates tallies.out file and statepoint.50.h5. But when python tries to import that file to generate output then it gives incompatibility issue with h5py version. I will post this in group also. The snapshot of it is attached. Hope you can help.
Best,
Mirza
Hi Paul,
The issue has been resolved. Thank you so much.
Best Regards,
Mirza Younis Baig
