Muon Slides 24-October-2001

I fought through some odd crashes, found some bugs in other code, and now believe I have the IMU geometry fixed, except for the CMX miniskirt overlap issue. Commitment will follow the 5000-event test with a virgin installation.
On another front, I used Bob's time distributions to clean up the time distribution of hits in a cell (with a stub) and its jumpered pair. The result is much tighter and more useful for cuts.
I use these cuts to try to determine where the front BSU sits in relation to these time differences. The sample of events with a front BSU hit is contaminated, of course, so I divide the plot of the time differences with a front BSU hit by that with a rear BSU hit, and the result shows that a cut somewhere between 10 and 15 nsec for the time differences is appropriate. If I do the more canonical dividing the front by the total I get efficiencies of about 80% and a cutoff at the same place.