Difference between revisions of "BCAL Reconstruction Meeting 2014-06-26"

From GlueXWiki
Jump to: navigation, search
(Created page with "=Video Conferencing Information= ''<font color=red>Meeting Time: 11:15 a.m. EDT</font>'' # To join via a Web Browser, go to the page [https://bluejeans.com/907185247] https://b...")
 
(Minutes)
Line 37: Line 37:
  
 
= Minutes =
 
= Minutes =
 +
 +
''Attendees: Mark D, David (JLab), Will (CMU), Andrei, Tegan, Zisis (Regina)''
 +
 +
# Announcements
 +
## JANA low-level objects (DL and MD):  Mark has looked at the data in different modes.  The plan is to have the BCAL be the trailblazer for this JANA work.  At the moment the pulse integral is there but not much more.  Mark has identified the pulse integral and waveform modes and will write a short HOWTO for the Wiki; the fast electronics group has a nice 5-page write up on this.  At present we are thinking to start in raw mode with the data and later go to parsed mode; the switch can be done with a simple configuration file change.
 +
## The BCAL is out of alignment in HDDS, but 1/2 module, i.e. 3.75 deg, between MC and data.  This will be fixed.
 +
# Action Items
 +
## These items are deferred to after Tegan’s return to Regina from MIT.  He will be there July 13-26 to work with Justin on BDT as applied to reconstruction.  Zisis proposed that the Regina students (helped by Mark and Elton) will gradually populate thd ccdb with conditions and calibration parameters.
 +
# Updates
 +
## Tegan followed Will’s instructions (June 4 email).  Tegan is using code revision 13953, which according to Will is up to date.  Tegan ran 100 events and then 10,000 using the single photon gun.  Spikes are observed at zero, most of which are gone when the reconstructed cluster flag match==1 is applied.  Andrei proposed to set “failures” to a number other than zero (e.g. -9999 or similar) to avoid confusion by new users — UofR will do this at a later date.  The TTree produced by bcal_tree plugin provides truth (cheat) and reconstructed variables.
 +
## ‘’Variable details’’. nCells = no of cells participating in a reconstructed photon cluster; it is zero only when no cluster is reconstructed.  nPoints and ePoints are made by DBCALPoint() factory which combines information from both ends of the calorimeter. Etruthcell is the energy deposited in the homogenous material of the calorimeter, without attenuation.  ’Truth” means MC cheating data, which is not accessible to real data.  ePoint and Etruth should be compared.  Specifically, the 1st Etruth cell should be the same as the 1st Epoint cell (reconstructed).  Epoint is corrected for attenuation and includes the up plus down information.  The information of the incoming photon is stored in the kinematic parameters in hddm.
 +
## Cosmics simulations: Irina set up the standalone code, with the full BCAL having 40 energy deposition cells per side (with the feature to sum them up later according to our readout) and the trigger scintillators above and below.  No iron yoke is included.  Andrei just completed 15 million events  on the JLab cluster, which he will proceed to analyze.  We decided collectively that he should start with the extraction of mean energy deposited in readout cells (and then widths, shapes…) from these events and compare to the cosmic data we are collecting, and after we work out a plan on what parameters we want extracted, Irina/Andrei will correct the 3.75 deg rotation, implement changes and rerun.  We will convey to Elton the need to run cosmics data nightly and on weekends, as we will need adequate statistics per cell.  Cosmics data to date is being collected using the scintillator signals as trigger.
 +
# New Action Items: (for implementation by Regina): see updated list above
 +
## Cosmics MC analysis: extract means first
 +
## Reconstruction code
 +
### Sampling fraction tables
 +
### Shower curvature tables: compare KLOE to IU codes: check gap in cluster handling between codes.
 +
### time-walk: check fits, decide on 1 or 2 iterations)
 +
### timing: extracted from F1TDC and compared to fADC
 +
### z-coordinate determination from up/down amplitude ratio
 +
### set variable “failures” to a number other than zero (e.g. -9999 or similar)
 +
### comb code for constants to be pulled out into a database
 +
# Other business: none

Revision as of 12:54, 27 June 2014

Video Conferencing Information

Meeting Time: 11:15 a.m. EDT

  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 F326 is 757-269-6460 (usual room)
  • JLab Phone in CC L207 is 757-269-7084
  • Phone in the Regina Video-conference Suite is 306-585-4204

References

  1. BCAL Reconstruction Issues
  2. BCAL Reconstruction Algorithms
  3. Will's reconstructed energy info

Action Items

  1. "Curvature effect": incorporate Andrei's and Irina's parametrization of this effect into the KLOE and IU codes and establish that this closes the gap in cluster handling between the two codes. (Andrei/Irina)
  2. Start developing BDT variables for cluster handling in the BCAL Reconstruction. (Tegan)
  3. Constants in code should be transferred to database; list of parameters for ccdb storage.

Agenda

  1. Announcements
  2. Action Items
  3. Updates
  4. Any other business

Minutes

Attendees: Mark D, David (JLab), Will (CMU), Andrei, Tegan, Zisis (Regina)

  1. Announcements
    1. JANA low-level objects (DL and MD): Mark has looked at the data in different modes. The plan is to have the BCAL be the trailblazer for this JANA work. At the moment the pulse integral is there but not much more. Mark has identified the pulse integral and waveform modes and will write a short HOWTO for the Wiki; the fast electronics group has a nice 5-page write up on this. At present we are thinking to start in raw mode with the data and later go to parsed mode; the switch can be done with a simple configuration file change.
    2. The BCAL is out of alignment in HDDS, but 1/2 module, i.e. 3.75 deg, between MC and data. This will be fixed.
  2. Action Items
    1. These items are deferred to after Tegan’s return to Regina from MIT. He will be there July 13-26 to work with Justin on BDT as applied to reconstruction. Zisis proposed that the Regina students (helped by Mark and Elton) will gradually populate thd ccdb with conditions and calibration parameters.
  3. Updates
    1. Tegan followed Will’s instructions (June 4 email). Tegan is using code revision 13953, which according to Will is up to date. Tegan ran 100 events and then 10,000 using the single photon gun. Spikes are observed at zero, most of which are gone when the reconstructed cluster flag match==1 is applied. Andrei proposed to set “failures” to a number other than zero (e.g. -9999 or similar) to avoid confusion by new users — UofR will do this at a later date. The TTree produced by bcal_tree plugin provides truth (cheat) and reconstructed variables.
    2. ‘’Variable details’’. nCells = no of cells participating in a reconstructed photon cluster; it is zero only when no cluster is reconstructed. nPoints and ePoints are made by DBCALPoint() factory which combines information from both ends of the calorimeter. Etruthcell is the energy deposited in the homogenous material of the calorimeter, without attenuation. ’Truth” means MC cheating data, which is not accessible to real data. ePoint and Etruth should be compared. Specifically, the 1st Etruth cell should be the same as the 1st Epoint cell (reconstructed). Epoint is corrected for attenuation and includes the up plus down information. The information of the incoming photon is stored in the kinematic parameters in hddm.
    3. Cosmics simulations: Irina set up the standalone code, with the full BCAL having 40 energy deposition cells per side (with the feature to sum them up later according to our readout) and the trigger scintillators above and below. No iron yoke is included. Andrei just completed 15 million events on the JLab cluster, which he will proceed to analyze. We decided collectively that he should start with the extraction of mean energy deposited in readout cells (and then widths, shapes…) from these events and compare to the cosmic data we are collecting, and after we work out a plan on what parameters we want extracted, Irina/Andrei will correct the 3.75 deg rotation, implement changes and rerun. We will convey to Elton the need to run cosmics data nightly and on weekends, as we will need adequate statistics per cell. Cosmics data to date is being collected using the scintillator signals as trigger.
  4. New Action Items: (for implementation by Regina): see updated list above
    1. Cosmics MC analysis: extract means first
    2. Reconstruction code
      1. Sampling fraction tables
      2. Shower curvature tables: compare KLOE to IU codes: check gap in cluster handling between codes.
      3. time-walk: check fits, decide on 1 or 2 iterations)
      4. timing: extracted from F1TDC and compared to fADC
      5. z-coordinate determination from up/down amplitude ratio
      6. set variable “failures” to a number other than zero (e.g. -9999 or similar)
      7. comb code for constants to be pulled out into a database
  5. Other business: none