PENDF and GENDF tapes for MF34 processing

[AARyzhkov]

Dear NJOY team,

I encountered an issue concerned with how some angular distribution (P1) uncertainties (MF34) are processed (NJOY2016.77 was used). MF34 is the other File, allowed for processing with a PENDF tape, besides MF33. Comparing the results of processing with a PENDF tape and the corresponding GENDF tape shows some differences at lower energies (using PENDF yields zeros) for O16 from ENDF/B-VIII.0 and ENDF/B-VIII.1. After the comparison with the corresponding ENDF tape, it seems that using a PENDF tape for MF34 may yield significantly different uncertainties from the ENDF data too (the ENDF data were taken from IAEA's NDS). I am providing two images to demonstrate the discrepancies for O16 from ENDF/B-VIII.0 and ENDF/B-VIII.1 in the 28-group ABBN approximation. The same issue can be spotted at lower energies in the case of ENDF/VIII.0
I also have doubts about the resonance region, where the PENDF-based uncertainties are generally lower than the ENDF values while it is reasonable to expect something looking like an average, similar to the GENDF tape. To make it clearer, I am pinning the ENDF/B-VIII.1 results in the 299-group ABBN approximation. Besides, at an energy of 1 MeV, the GENDF-based result shows a lower value too. In addition, the images demonstrate another anomaly for both tapes at an energy of 0.1 MeV, where the uncertainties plunge to zero though the ENDF tape is fine

Kind regards,
Alexander Ryzhkov