I have posted a new technote gluex-doc-1012 to the portal on the subject of SiPM gain non-uniformities and how they contribute to the energy resolution. The treatment is all analytical, with a Monte Carlo check for one set of parameters. Doing it analytically is a bit tedious, but it has the advantage that you end up with formulas that show the sensitivity over the entire range of possible design parameters. The bottom line is that there is considerable room for variability in the tile-to-tile gain of the SiPM modules without much effect on the final energy resolution, provided that the light guides spread out the light from individual fibers broadly enough over the face of the detectors. Richard Jones
S/MIME Cryptographic Signature