Difference between revisions of "August 10, 2016 Calibration"

Revision as of 12:03, 10 August 2016

GlueX Calibration Meeting
Wednesday, August 10, 2016
11:00 am, EST
JLab: CEBAF Center, F326

Communication Information

Remote Connection

Slides

Talks can be deposited in the directory /group/halld/www/halldweb/html/talks/2016 on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2016/ .

Calibration Tasks

1. Spring 2016 Recon Checklist
2. Mcsmear updates - mcsmear 20160323
3. Data Validation
4. GlueX Detector Performance Document

Calibrations for Second Production Run

• Updated error matrices (all)
• Fix handling of individually-read-out TAGM columns (Alex B.)
• Tagger hit merging (Nathan/Alex B.)
• Updates to FCAL non-linearity (Adesh)
• Recalibrate gains for early runs ~<10600 (Adesh)
• Add timing information to FCAL clustering/inner ring calibrations (Adesh/Matt S.)
• BCAL TDC calibrations (Andrei)
• BCAL energy non-linearity correction (Will M.)
• FDC wire timing & reconstruction (Lubomir/Simon/Alex A.)
• FDC per-channel calibrations (Alex A./Mike S./?)
• TOF ADC energy and timing calibrations [dropped] (Beni)
• TOF alignment (Beni/Simon)
• Tagger energy calibration (Richard J./?)
• General timing realignment (Sean)

Agenda

1. Announcements
   • ROOT 6 transition - test your script!
2. Automated Calibrations
   • Tests planned before October run
   • Timing Calibrations
   • ADC gains?
   • Drift time-to-distance?
   • Anything else?
3. Calibration Plans for Second Production Run
   • Deadline: August 31?
4. Next Monitoring Run
5. Calibration Updates
6. Simulations
   • sim1.1 for data/simulation matching studies (Sean/Mark I.)
7. AOB

Minutes

Attending: Sean (NU); Adesh, Paul, Beni, Simon, Alex A., Alex B., Mark I., Nathan, Lubomir (JLab); Curtis, Will M., Mike S., Naomi (CMU); Zisis (Regina)

1. Announcements
   • The offline software is in the process of moving to requiring ROOT 6. This primarily affects the scripts/macros that are used for monitoring and calibration, due to the improved interpreter that requires well-formed C++11 compliant code. Yesterday Sean checked into sim-recon some fixes to make most scripts compile under ROOT 6. The correctness of the scripts needs to be verified, so he asked everyone to check their own code.
   • Mark mentioned that on Friday he released a new version of sim-recon. This includes the latest tracking reconstruction improvements and versions using ROOT 5 & 6 were built.
2. Automated Calibrations
   • With the experience we've gained this year, we are planning on running some fully automated calibrations in the Fall run. These will have appropriate diagnostic output and can be checked by experts and/or shifters. The core of these calibrations will be the timing calibrations. Sean is coordinating the consolidation of what has been done already. The tagger/PS calibrations need attention since a lot of development was done with them over the summer.
   • Zisis reported that Andrei is finishing off his calibration of the BCAL TDCs. Andrei will communicate with Sean and Elton in order to integrate his improvements into sim-recon. Zisis will check with Andrei to see if we can accomplish this by the end of the month.
   • Naomi has some linear relations she is using to calibrate the CDC overall gain factor, and she agreed to come up with a script to automate this procedure. She also reported that an apparently faulty thermocouple on the CDC was replaced.
   • Mike is studying the CDC time-to-distance relationship with the new reconstruction code to see what calibration improvements can be made. He is studying the run variation of this calibration. Lubomir reported that the same FDC time-to-distance calibration was acceptable for the whole spring run.
3. Calibration Plans for Second Production Run
   • Improvements for sim-recon for the next REST production run should be finalized by the end of the month. We reviewed the list of current projects.
   • Alex B. and Nathan are finalizing the hit merging in the tagger reconstruction. The big issues are dealing with the interface between the TAGH and TAGM, and handling gaps in the TAGH counters.
   • Adesh has finished the minor improvements in the spring gains on his list, and is finalizing the precision timing calibrations. After performing a per-channel timing alignment, he still needs to align the FCAL with the rest of the detector, and will contact Mike and Sean about this. Studying the addition of timing information into the clustering algorithm will be done at a later date.
   • Will has non-linearity corrections for the BCAL, but there are a few issues that are not understood, and he is looking to simulation for help. First, he needs to tune the simulation to get reasonable results. He is aiming to finish this by the end of the month, but his current numbers are certainly an improvement.
   • Alex A. has started looking into a per-channel FDC calibration, as discussed in yesterday's Analysis meeting.
   • Beni has shelved the TOF ADC calibration because of its complicated rate dependence. He is looking into changing the timewalk correction to be done with the amplitude instead of the integral, and preparing for the next run, where the calibrations will need to be completely redone due to the different running parameters. Beni and Mark I. will generate a run plan for the beginning of the October run.
   • In yesterday's Analysis meeting, Paul reported some strange effects in the tagger energy reconstruction. Richard is looking into this.
   • Sean is looking into some potential charge-dependence of the detector timing.
4. Next Monitoring Run
   • Alex A. reported that there were some cache disk problems yesterday, so the analysis launch should finish today or tomorrow. He wants to launch the monitoring jobs soon after that. This launch will include ROOT 6 and David's new EVIO parser.
   • If you have anything you want included in this launch, contact Alex today.
5. Calibration Updates
   • Naomi reported on a quick look at a quick look at using the pulse amplitude instead of the integral for CDC dE/dx. It does seem to give better separation. Lubomir pointed out that in the current CDC readout mode, the amplitude is divided by 8, so for greater precision they could consider using one of the FDC readout modes.
6. Simulations
   • Mark reported that sim1.1 is done. He will update the webpages. It is ready for people to perform data/MC comparisons.
   • ROOT trees for efficiency studies were generated.
   • Sean will make some of the standard monitoring plots.
   • After the monitoring run, we will organize an analysis run on this data, which will help determine the size of the MC we should generate for analysis purposes. Only a percent or so of the data taken was simulated, though, so it is encouraged to run your own code on these files.