From: Chris Tunnell (tunnell@mail.utexas.edu)
Date: Tue Apr 04 2006 - 20:43:51 CDT
Long story short: Our simulation is correct since the PMT bug I talked about doesn't exist. What I saw makes sense physically or was a temporary program state.
Long story long: When I described what I thought was a PMT bug on the phone-call, I talked about two aspects of it: total internal reflected angles generated photo-electrons and also a conservation of energy problem. After some good conversation with Glenn, the most important thing that was learned was some optics:
When describing (classically) the interactions of light and a boundary, there are three possible things that can happen to the light: transmission, reflection and absorption. This can be seem algebraically by T+R+A = 1 where T, R, and A equal the transmission, reflection and absorption probabilities respectively.
At incident angles higher than roughly 45 degrees, I argued that there should be total internal reflection, thus the light never gets to the photocathode. Glenn astutely pointed out that total internal reflection is a misnomer since even though T=0, A does not have to be 0, thus there is total internal reflection OR absorption. This means that photoelectrons are generated at these high angles in reality. This boils down to not understanding absorption of light at interfaces on my part.
The second problem of energy conservation was not a problem because it was fixed later on in code I did not see. I thought there was energy conservation issues since I inserted debugging code and notice photo-electrons and associated photons co-existing at the same time. This, however, was a temporary program state. If a photo-electron was generated it is guaranteed that the photon is removed from the tracker before geant resumes control.
Sorry about the false alarm. Glenn did share a lot of literature and explained how a lot of his work was derived from other's work, which makes me much more confident in the PMT code. The delay in informing the workgroup of this was caused by me reviewing Glenn's reference and making sure I knew what was going on. If there are any questions, feel free to ask. I can even show you Glenn and my communication, or the literature I looked at.
Cheers,
Chris
This archive was generated by hypermail 2.1.6 : Wed Apr 05 2006 - 12:01:02 CDT