Difference between revisions of "Mar 23, 2017 Calorimeter"

From GlueXWiki
Jump to: navigation, search
(Tentative Agenda)
(Minutes)
 
(3 intermediate revisions by 2 users not shown)
Line 35: Line 35:
 
# Action Items
 
# Action Items
 
# Run preparations
 
# Run preparations
 +
## Activities during the down
 
## LED Monitoring
 
## LED Monitoring
 
# FCAL update
 
# FCAL update
Line 46: Line 47:
 
# Reconstruction and Simulation
 
# Reconstruction and Simulation
 
## [https://logbooks.jlab.org/entry/3467614 BCAL Hadronic Efficiencies] (Elton)
 
## [https://logbooks.jlab.org/entry/3467614 BCAL Hadronic Efficiencies] (Elton)
 +
## [https://logbooks.jlab.org/files/2017/03/3468296/charged_clust_updateMar23.pdf BCAL charged clustering update] (Will)
 
# NIM articles
 
# NIM articles
 
# Any other business
 
# Any other business
Line 51: Line 53:
 
= Minutes =
 
= Minutes =
  
''Attendees:
+
''Attendees: Elton, Jon, Mark, Adesh (JLab); Sean (NW); Will (CMU); George, Christina (Athens); Andrei, Ahmed (UofR); Matt (IU)
 +
 
 +
# Announcements
 +
# Action Items
 +
#* Adesh: The 2 ns jumps happen at low energy. Need to investigate this further
 +
# Run preparations
 +
## Activities during the down
 +
##* Comic data taking concluded yesterday
 +
##* Sasha is updating firmware on all crates
 +
##* Fcal would like to take random triggers to study electronic noise without beam. This will take a couple of hours.
 +
##* All data should be taken before major mechanical activity in the hall
 +
##* Expect to leave electronics on. Bcal chillers will be parked at 18 degrees.
 +
## LED Monitoring
 +
##* Looking at spring 2017 data.
 +
##* Finds band structure (quasi quantization) of rms values. Jon: Sees bands in Fcal base voltages reported at 16 bits that are the result of averages of 10 bit numbers. Could a similar mechanism be at work.
 +
##* Plotted rms over the run vs channel number. Investigate outliers
 +
##* High rms of some channels include zero pulses for some runs. Need to find why this happens. Short runs are excluded, so we expect all channels to have entries.
 +
# FCAL update
 +
## Calibration (Adesh)
 +
##* Skim files almost completed for all runs. Files copied to IU
 +
##* Checking stability of constants, e.g. pedestals, over time.
 +
## Bases (Jon)
 +
##* Base failures seem to have slowed. 2 failures in the past two weeks.
 +
# BCAL update
 +
# Calibrations
 +
## pi0 calibrations
 +
## Timing calibrations
 +
### [https://logbooks.jlab.org/entry/3468218 Initial timing calibrations] (Mark)
 +
###* Low level calibrations have been completed as a base for gain calibrations.
 +
###* Repeating work that George did for spring 2016. Use Run 30895. Some differences, including moving from TGraphs to THF2 histograms, which are more easily scalable into our software.
 +
###* Fitting z (track) vs Deltat (tup - tdown). Means develop artifacts. Now use FitSlices to determine peak values of distribution vs Deltat. Use these to fit to pol2
 +
###* p0 shows a bimodal distribution with Delta z ~ 30 ns ~ 4 ns. Suspect hardware. Discussions with Ben Raydo suggest that the hardware explanation is plausible. Special firmware was developed for Hall B to avoid this problem. Depending on origin of problem, shifts can happen either on a run-by run-basis or upon power.
 +
###* Need to iterate calibration once tracking fixes are completed.
 +
###* Sean: concern that constants may not be stable over the entire spring period. Mark: Will check stability using files remaining on the gluon machines.
 +
###* Shape of residuals is similar for all channels, which suggests the systematics comes from other sources, e.g. tracking.
 +
### [https://halldweb1.jlab.org/wiki/images/9/91/Semenov-timesimulation-170323.pdf Simulation of BCAL Timing] by A.Semenov & I. Semenova (UofR)
 +
###* Find that resolutions are reasonably well reproduced compared to data
 +
###* Use detailed MC to check timing response in MC and compare to data
 +
###* MC: Find vU - vD ~ 0.3 cm/ns.  V decreases with layer number
 +
###* Data: Find vU - vD ~ 1 cm/ns. V increases with layer number, unlike MC. Not understood why
 +
###* Checked sigma_z vs E, find ~ 1.2 cm/sqrt(E) + 3 cm, with little systematics.
 +
## [https://halldweb.jlab.org/wiki/images/9/9d/Resolution_histos_23_03_2017.pdf pi0 Dalitz decay: Comparing thrown and reconstructed quantities] - George
 +
##* Moved from genr8 (-> geant3), 9 GeV photons, to geant4, polarized photon spectrum
 +
##* Compare gen vs reconstructed both both modes
 +
##* Central part of resolution looks like it is ~ 100 MeV, but large tails
 +
##* Matt: This should be sufficient to investigate efficiencies.
 +
##* Elton: What kinematic fitting is used? Vertex + p4 to select event sample.
 +
##* Matt: Concern that this could bias sample. Could look into Vertex + pi0 mass + zero missing energy.
 +
##* George: Investigate method to determine single photon resolution vs energy.
 +
# Reconstruction and Simulation
 +
## [https://logbooks.jlab.org/entry/3467614 BCAL Hadronic Efficiencies] (Elton)
 +
##* Not shown, refer reference for latest updates.
 +
## [https://logbooks.jlab.org/files/2017/03/3468296/charged_clust_updateMar23.pdf BCAL charged clustering update] (Will)
 +
##* Charged particle clusterizer
 +
##* Use wire-based tracks to match to cluster seed. Use axis between track projection and cluster seed. Follow this line to add points to cluster. At this point only use phi dimension; still need to investigate z dependence.
 +
##* Mark: Would be interesting to investigate the dependence of the results on the layer of the seed.
 +
##* Find improvement in finding a single shower (when a single shower is generated) with this algorithm compared to neutral clusterizer
 +
##* Still need to study various systematics, but new algorithm is an improvement over simply using the neutral clusterizer.
 +
##* Feedback is welcome. Zisis and Andrei will send comments later today.
 +
# NIM articles
 +
#* Elton and Zisis discussed updates on the Bcal NIM after the meeting.
 +
# Any other business

Latest revision as of 12:46, 23 March 2017

Video Conferencing Information

Meeting Time: 10:00 a.m.

  1. To join via a Web Browser, go to the page [1] https://bluejeans.com/907185247.
  2. To join via Polycom room system go to the IP Address: 199.48.152.152 (bjn.vc) and enter the meeting ID: 907185247.
  3. To join via phone, use one of the following numbers and the Conference ID: 907185247.
    • US or Canada: +1 408 740 7256 or
    • US or Canada: +1 888 240 2560
  4. Upon connection all microphones are automatically muted. To unmute your mike on a Polycom or equivalent unit, enter *4. Unmuting on a computer is trivial as there is a microphone button than can be clicked.
  5. More information on connecting to bluejeans is available.

Participant Direct Lines

  • JLab Phone in CC L207 is 757-269-7084 (usual room)
  • JLab Phone: in CC F326 is 757-269-6460
  • Phone in the Regina Video-conference Suite is 306-585-4204
  • Athens Phone: in Christina's office is 011-30-210-727-6947

Action Items

  1. Determine number of saturation/underflows that exist in the two LED bias monitoring settings (Ahmed)
  2. Understand the nature of the peak in the (Fcal propagated to target -RF) satellite peak of -2 ns (Adesh)
  3. Disable red LED string in dark room to avoid leaving it on in the future. (Adesh)
  4. Quantify the LED pulser sector jumps within one run (Ahmed)

References

  1. FCAL HDFCAL log book
  2. BCAL HDBCAL log book

Tentative Agenda

  1. Announcements
  2. Action Items
  3. Run preparations
    1. Activities during the down
    2. LED Monitoring
  4. FCAL update
  5. BCAL update
  6. Calibrations
    1. pi0 calibrations
    2. Timing calibrations
      1. Initial timing calibrations (Mark)
      2. Simulation of BCAL Timing by A.Semenov & I. Semenova (UofR)
    3. pi0 Dalitz decay: Comparing thrown and reconstructed quantities - George
  7. Reconstruction and Simulation
    1. BCAL Hadronic Efficiencies (Elton)
    2. BCAL charged clustering update (Will)
  8. NIM articles
  9. Any other business

Minutes

Attendees: Elton, Jon, Mark, Adesh (JLab); Sean (NW); Will (CMU); George, Christina (Athens); Andrei, Ahmed (UofR); Matt (IU)

  1. Announcements
  2. Action Items
    • Adesh: The 2 ns jumps happen at low energy. Need to investigate this further
  3. Run preparations
    1. Activities during the down
      • Comic data taking concluded yesterday
      • Sasha is updating firmware on all crates
      • Fcal would like to take random triggers to study electronic noise without beam. This will take a couple of hours.
      • All data should be taken before major mechanical activity in the hall
      • Expect to leave electronics on. Bcal chillers will be parked at 18 degrees.
    2. LED Monitoring
      • Looking at spring 2017 data.
      • Finds band structure (quasi quantization) of rms values. Jon: Sees bands in Fcal base voltages reported at 16 bits that are the result of averages of 10 bit numbers. Could a similar mechanism be at work.
      • Plotted rms over the run vs channel number. Investigate outliers
      • High rms of some channels include zero pulses for some runs. Need to find why this happens. Short runs are excluded, so we expect all channels to have entries.
  4. FCAL update
    1. Calibration (Adesh)
      • Skim files almost completed for all runs. Files copied to IU
      • Checking stability of constants, e.g. pedestals, over time.
    2. Bases (Jon)
      • Base failures seem to have slowed. 2 failures in the past two weeks.
  5. BCAL update
  6. Calibrations
    1. pi0 calibrations
    2. Timing calibrations
      1. Initial timing calibrations (Mark)
        • Low level calibrations have been completed as a base for gain calibrations.
        • Repeating work that George did for spring 2016. Use Run 30895. Some differences, including moving from TGraphs to THF2 histograms, which are more easily scalable into our software.
        • Fitting z (track) vs Deltat (tup - tdown). Means develop artifacts. Now use FitSlices to determine peak values of distribution vs Deltat. Use these to fit to pol2
        • p0 shows a bimodal distribution with Delta z ~ 30 ns ~ 4 ns. Suspect hardware. Discussions with Ben Raydo suggest that the hardware explanation is plausible. Special firmware was developed for Hall B to avoid this problem. Depending on origin of problem, shifts can happen either on a run-by run-basis or upon power.
        • Need to iterate calibration once tracking fixes are completed.
        • Sean: concern that constants may not be stable over the entire spring period. Mark: Will check stability using files remaining on the gluon machines.
        • Shape of residuals is similar for all channels, which suggests the systematics comes from other sources, e.g. tracking.
      2. Simulation of BCAL Timing by A.Semenov & I. Semenova (UofR)
        • Find that resolutions are reasonably well reproduced compared to data
        • Use detailed MC to check timing response in MC and compare to data
        • MC: Find vU - vD ~ 0.3 cm/ns. V decreases with layer number
        • Data: Find vU - vD ~ 1 cm/ns. V increases with layer number, unlike MC. Not understood why
        • Checked sigma_z vs E, find ~ 1.2 cm/sqrt(E) + 3 cm, with little systematics.
    3. pi0 Dalitz decay: Comparing thrown and reconstructed quantities - George
      • Moved from genr8 (-> geant3), 9 GeV photons, to geant4, polarized photon spectrum
      • Compare gen vs reconstructed both both modes
      • Central part of resolution looks like it is ~ 100 MeV, but large tails
      • Matt: This should be sufficient to investigate efficiencies.
      • Elton: What kinematic fitting is used? Vertex + p4 to select event sample.
      • Matt: Concern that this could bias sample. Could look into Vertex + pi0 mass + zero missing energy.
      • George: Investigate method to determine single photon resolution vs energy.
  7. Reconstruction and Simulation
    1. BCAL Hadronic Efficiencies (Elton)
      • Not shown, refer reference for latest updates.
    2. BCAL charged clustering update (Will)
      • Charged particle clusterizer
      • Use wire-based tracks to match to cluster seed. Use axis between track projection and cluster seed. Follow this line to add points to cluster. At this point only use phi dimension; still need to investigate z dependence.
      • Mark: Would be interesting to investigate the dependence of the results on the layer of the seed.
      • Find improvement in finding a single shower (when a single shower is generated) with this algorithm compared to neutral clusterizer
      • Still need to study various systematics, but new algorithm is an improvement over simply using the neutral clusterizer.
      • Feedback is welcome. Zisis and Andrei will send comments later today.
  8. NIM articles
    • Elton and Zisis discussed updates on the Bcal NIM after the meeting.
  9. Any other business