Page 1 of 1

JEFF-3.3 Fatal error in function ProcessEDistributions

Posted: Sun Mar 14, 2021 10:42 pm
by ondrejch
I switched from ENDF-VII to JEFF-3.3 and got this error while reading in the cross sections. Serpent version 2.1.32. No problems with JEFF-311, ENFD-VII, ENDF-VIII.
Processing cross sections and ENDF reaction laws...
Nuclide 61147.02c -- promethium 147 at 900K (Pm-147)
Nuclide 61148.02c -- promethium 148 at 900K (Pm-148)
Nuclide 61348.02c -- promethium 148m at 900K (Pm-148m)
Nuclide 61149.02c -- promethium 149 at 900K (Pm-149)

***** Sun Mar 14 15:34:39 2021:

- MPI task = 0
- OpenMP thread = 0
- RNG parent seed = 1615750228925
- Host name = node67

Fatal error in function ProcessEDistributions:

Error in cosines (mt 91, law 61)

Simulation aborted.

Re: JEFF-3.3 Fatal error in function ProcessEDistributions

Posted: Mon Mar 15, 2021 12:52 pm
by Ana Jambrina
It might be similar error that the one described in the post viewtopic.php?f=25&t=2776&p=8142&hilit= ... rror#p8142) - known issue with how NJOY processes angular distributions in JEFF(-3.2) library, 'processmudistributions.c'. In this case the error arises in 'processedistributions.c', but it is also related with a similar consistency check been how the highest value of the PDF of the scattering cosine differs from abs(1.0), 1 or -1.

Nonetheless, I have run Serpent 2.1.32 with JEFF-3.3 library downloaded from the OECD website (at - processed files at 900K, to be fully in-line with the error). The files are read and processed by Serpent without errors.

Which data files are you using?

Re: JEFF-3.3 Fatal error in function ProcessEDistributions

Posted: Tue Mar 16, 2021 11:33 pm
by ondrejch
Dear Ana, thank you for a prompt reply. I re-downloaded the data from and it all works now splendidly. Apparently the problem was with the old data.