BCAL Readout Segmentation Task group Meeting: Jul 22, 2011
Contents
Meeting Info
Time:
- 3:30 JLab
- 1:30 Regina
Location:
- CC F326-327
Vid-con:
- ESNet 8542553
Agenda
Review action items from previous meeting
- Verify KLOE used 5 rings total (Zisis)
- Confirmed. Email sent 7/15/2011
- Circulate info on polar angle sensitivity to sampling fluctuations (Matt)
- Confirmed. Email sent 7/14/2011
- Implement timewalk corrections and continue timing resolution study (David)
- See today's slides
- Present results on extra showers from course segmentation (Andrei)
- See today's slides
New Agenda items
- Extra Showers from Course Segmentation (Irina S. and Andrei S.)
- BCAL Signal Timing Distributions (David L.)
- Toy Monte Carlo Studies (Elton S.)
Minutes
Participants
Beni Z., Elton S., David L., Andrei S., Irina S., George L., Zisis P. (partial), Matt S. (partial)
- We briefly read through the list of action items from last weeks's meeting at the top of this page and then went into the new agenda items.
- Zisis suggested that he will review all of the presentations given since March in an exercise similar to what David did the preceding week. He will summarize the findings in a way that does not misrepresent anyone's work or draw non-consensus conclusions. This will become the new historical summary record of the work done from mid-March to mid-June, superceeding the one Dave produced for the June 14th meeting.
Extra Showers From Course Segmentation
Andrei presented some results of simulations ran by Irina and analyzed by him. These focused on the extra showers in the BCAL (reported by Andrei at the June 17th meeting)
Slide 2:
A tagged version of the code from June 2, 2011 was used. A disclaimer was made that because the reconstruction code is known to have some problems (at the very least, non-optimized parameters for cluster splitting) as well as lacking some features (detailed signal timing evolution) no conclusions regarding segmentation choice should be drawn from the results presented.
Slide 3:
A reminder of the initial problem via 4 plots. These indicated that for the case of course segmentation using the default reconstruction (Matt's algorithm?) many more events with multiple reconstructed showers were seen than for the fine segmentation. The bottom set of plots indicates this problem is tightly coupled with the dark pulses since once they are turned off, the discrepancy all but disappears.
Slide 4:
Descriptions of 3 possible sources of additional showers in events with a single incident particle. One of these is dismissed due to the dark pulse test.
Slides 6-8:
Plots indicating the relative spatial and energy relations between primary shower (one with most energy in event) and secondary showers for Matt's code.
Slides 10-12:
Plots indicating the relative spatial and energy relations between primary shower (one with most energy in event) and secondary showers for KLOE-based code.
Slide 13:
Conclusions drawn from the current study. Given the disclaimer at the beginning, these conclusions were not debated.
BCAL Signal Timing Distributions
David presented some slides on his ongoing study to extract signal timing information for the BCAL from simulating the electronic pulse shape based on input from hdgeant
Slides 2-5:
A review of how the threshold is calculated and the exact pulse that is convoluted with the energy distributions to get a full signal shape. Multiple methods have been used to calculate the threshold in mV, but a decision was made to 44.7mV as it seemed a reasonable compromise.
Slide 6:
Timewalk corrections
Slides 8-10:
These indicated timing resolutions for the mean time and the time difference for single cells for the two segmentation schemes. Both of these indicated a slightly better resolution for the fine segmentation scheme than for the course.
- In the discussion, it was speculated that this could be due to the smaller spatial extent of the shower in a smaller cell leading to correspondingly tighter timing
- Elton stated that the leading edge should drive the single cell timing resolution so one should expect equal resolutions for cells with significant energy deposition.
- David indicated that asymmetries in the timewalk corrected distributions shown on slide 8 need to be double checked to make sure they are not an indication of a problem that ultimately affects the resolution plots.
- It was also pointed out that when combining the single cell timing info into a shower time/position, the fine segmentation will improve more relative to the course by virtue of having more measurements.
Toy Monte Carlo Studies
Elton showed a few plots from some studies based on a toy Monte Carlo he's been working on.
- The toy M.C. simulated showers by depositing energy along and around a particle trajectory at a given angle.
- Two angles were studied 90 degrees and 20 degrees
- Two configurations of the inner layers were studied: 3+3 and 1+2+3
- The timing distribution at each end of the BCAL was derived from the time/position of the distributions of the calculated showers.
- For the 90 degree trajectory, no significant difference was seen between the two segmentation schemes as very little energy was deposited in the first cell.
- George noted that the inclusion of preshower material (such as the Al plate) could change that significantly
- For the 20 degree case, Elton showed we would effectively get one additional measurement for the finer segmentation.
- Elton noted that there would be 2 measurements that would be nearly identical between the two segmentation schemes because the relevant cell boundaries are in the same place and the timing is driven by the leading edge. The additional measurement would then only improve the overall timing resolution of the reconstructed shower.
MEETING TIME
It was suggested we change the time and day of this meeting to Fridays starting at 4:00pm JLab time. Dave agreed to check with Matt before implementing the change.
Action Items
- Continue timing studies based on electronic pulse shapes (David)
- Check with Matt on meeting time change (David)
