Absorption calculation stability

[pawel21]

Hi,

I deal with absorption from multi shell in BLR.
I found strange behaviour: when increasted r - distance of the blob from the Black Hole, absorption sometimes it increases sometimes it increases or sometimes go to very very low value.
Here plot absorption vs r for E = 1e5 MeV.

I am attaching notebook with my code.
absorption_blr.ipynb.zip

[jsitarek]

just to add some more information on this. I suspect that it is connected with #86 - the integrals that are used in agnpy just sum up over a given grid of points, if one of the grid points of the BLR shell ends up close to the one of the photon path grid points you end up with a large addition to the optical depth and strong absorption. This can sometimes happen for a single line, and when many lines are summed up at different distance it becomes even more possible to get it in one of the lines.