To understand what the issues are you need to understand what
the crossing time has to do with our stub-finding.
The third page in this set
shows what we calculate for the time at which a particle crosses
the BMU system. The first page in the set shows the difference
between the crossing time as found by the BMU and the time of a
BSU scintillator hit. When I demand that the two times agree,
I find the plots on the fourth page. It is easy to see that the
rear scintillator is more afflicted with halo particles which
arrive early, and in the fifth page you see that most of these
come from the West side.
My algorithm selects the best hit collection based on whether or
not the chi-squared is better than that of an overlapping hit
collection. It is easy to see that unless you correct for the
crossing time the results may be wrong. You can see the original
algorithm's chi-squared here,
where the difference between the upper plots is that the left one
is for stubs that crossed stacks and the right for stubs within a
single stack. For Bob's algorithm the
chi-squared plots look very different: All in the upper left,
3-hit stubs in upper right, 4-hit in the same stack in the lower
left and 4-hit crossing stacks in the lower right.
It is perfectly plain that there are some alignment issues that
needed to be dealt with here.
- After making the adjustments to hit positions for the crossing
times (these adjustments are made internally, and do not change the
hit collections), I found the chi-squared numbers still did not make
sense. I added some kludge terms to the timing corrections, which
were different for stubs going through a single stack and stubs crossing
stacks, and adjusted these until the chi-squared was a minimum. I regard
this as a temporary hack that should last only until I can find out the
origin of the misalignments. I don't expect the results to change when
I find the underlying problem, of course.
Both stub-finders wind up with a deficit of CdfMuons
in the rear South West side. Both find essentially the same number of
Modified 30-Oct-2002 at 14:17
Previous meeting notes
Next meeting notes
Main slide directory
if you have trouble accessing the information on this page.