A weird error in FEFF 8.1 (and II)

[Angel Garcia-Adeva <angarcia@lanl.gov>]

Hi everybody,

I have been running some FEFF calculations taking into account the 
suggestions by Matt, Jeff, and Alex. I just though it would be a good 
idea to post here how it went just in case anybody else runs into the 
same problems I did.

OK, the first thing I tried was changing the value of nphx1. That 
didn't solve the problem. For more than 10 unique potentials, the 
calculation still crashed. Then I tried what Alex suggested: using a 
very small value for RMAX of 0.1 Å and the calculation run without any 
problem. So it seems clear that the problem was in the pathfinder 
routine. I was wondering, however, how this small value of RMAX will 
affect the results. In our simulation, initially, we were using a value 
of FMS of 9.3 Å and a value of RMAX of 9.756 Å. That corresponds to a 
cluster of 297 atoms for the FMS calculation, and 16 more atoms if one 
takes into account the value of RMAX. I would expect this to have no 
effect on the calculated LDOS, as there is always a line in the 
rhol??.dat files that says 0/0 paths. Anybody could tell me if I am 
correct?

Thanks again for all your help.

Angel