MF35 does not process if MF34 is processed

[sombrereau]

In reference to discussion about calculating angular distributions in Pu239 with ENDFVIII data

NJOY's errorr module WILL produce a MF35 covariance tape if the errorr module called before the errorr call to produce a MF34 (case the works), but WILL NOT if the call to produce a MF34 covariance tape if called before the call to produce a MF35 covariance tape (case that does not work).

I've attached the input decks to this message and I'm using ENDFVIII data downloaded from the IAEA website

weird.

[sombrereau]

This behavior happens with Pu242 -- which is strange because you also need to add the additional reactions to resolve a -- no information for mt251 issue, similar to the Pu239 nuclear data. Maybe important ... maybe a red herring?