BCAL Reconstruction Meeting 2015-06-25
Revision as of 11:39, 14 March 2017 by Marki (Text replacement - "http://argus.phys.uregina.ca/gluex" to "https://halldweb.jlab.org")
Video Conferencing Information
Meeting Time: 11:15 a.m. EDT
- To join via a Web Browser, go to the page  https://bluejeans.com/907185247.
- To join via Polycom room system go to the IP Address: 22.214.171.124 (bjn.vc) and enter the meeting ID: 907185247.
- 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
- 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.
- 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
- High priority items
- BCAL and FCAL z coordinates: some are referenced to the center of the target, not the global coordinate system.
- Low priority items
- z-coordinate determination from up/down amplitude ratio
- Sampling fraction tables
- Code Cleanup
- Action Items
- Commissioning Status Quo
- Any other business
Attendees: Elton, Mark (JLab), Will L., Will M. (CMU), Sean (NWU), Ahmed, Andrei, Noemi, Tegan, Zisis (Regina)
- Announcements: the solenoid review will take place on July 14 and a Science & Technology Review will take place in late July. Eugene asked for slides on detector calibrations.
- Action Items:
- Coordinate systems: Will is working on the clusterizer. Mark reported that Matt had a strong opinion about how things should be done: calculations should be done in the target system and then transfer to global coordinates just before writing out the object. Everyone agreed. Will M. stated that the FCAL and BCAL showers are consistent with each other as a pair, and likewise for the FCAL-BCAL hits pairing. As long as the users 'grab' from showers, everything is ok. Avoid clusters unless absolutely needed to access these. Also, FCAL energies are more in line now. BCAL has the raw energies too but FCAL does not.
- Commissioning Status Quo
- BCAL - Effective Velocities and Time Offsets (George): New calculations based on z-track calculations and the four layer radii. He plots the z-coordinate of the points in the cluster versus the z-coordinates of the matched track for every channel and perform a linear fit on the outcome. George used both TGraph and TGraphError and shows comparisons on slide 3 of his presentation, top and bottom respectively. TGraphError has smaller errors but the values are further away from expectations. For the remaining slides he used TGraph. Structures in the data are apparent in the layers which become much smaller when using comics data or cuts on the beam data. The effective velocity seems to converge to 17cm/ns. There seem to be fluctuations in the 350-700 channel zone indicating an asymmetry between the top and bottom of the calorimeter. Mark suggested correlation plots (cluster, vs track coordinates, v_eff vs time offsets...) to further study this observation. The table on p 21 shows losses (due to cuts) between 30-38% (cosmics to data). These are large but at least consistent between them. Elton: are these uniformly distributed across modules, are they near the edges? Mark: we can throw events away provided we understand them. Is it a tracking problem or a BCAL problem? George said that B-field runs are more complex to analyze. Sean: look at the results vs different areas of the BCAL. George stated that radius information comes from HDDS. Whatever is in the constructor does not get saved. Mark suggested contacting David Lawrence.
- Time-walk corrections (Noemi): Noemi is now looking at data, following her experience with Elton's toy pulse generator. Work in progress.
- Layer efficiencies (Ahmed): Ahmed observed a curiosity in that points are formed for events where energy is below threshold. He is investigating.
- Cosmics (Andrei): B=0 comics runs do not have enough statistics for the analysis in order to extract updated calibration coefficients. He and Irina are now working on the B-field data.
- Simulations: DC3 (farm stress test) will start next week (most likely). Spring simulations will also be undertaken. Mark proposed that we switch the attenuation length from 310 to 525 cm and do it by Monday if possible. Use Tegan's code for the sims. Associated changes are the gains. Tegan stated that the shower energies are not correct (non-linear correction does not really do much now, based on the values used). Will L.: Non-linear corrections to energy: top values in code were used for no summing sims, and bottom for summing; this was done for the segmentation optimization work and therefore we should keep the 2nd set of numbers. Corrections are small at 90 degrees but bigger at smaller angles (larger z), maybe up to 5% (Will will check).Future: calibrate MC like data: gain constants per channel, Digi-level vs Hits-level. Tegan: more points will be kept with 525 cm values. Sean will help Tegan run tests to make sure things are good.
- Single-energy code (Tegan)
- Tegan's HDGEANT and mcsmear questions: mostly answered in meetings and by email.
- pi0 mass and gamma reconstructed discrepancies; example:  Skip, not too important for the moment.
- Shower curvature (Tegan, Andrei): points at polar angles smaller than 13 deg and greater than 119 deg are now included. Curvature code appears to reconstruct less shows but more points in the shower. Lower thetas lose some points. IU code ignores 4th layer points. Plots show difference between IU and Regina-curvature code. See more below.
- DBCALShower_factory_CURVATURE comparison to IU Shower code: Tegan's plots were discussed. Will L: z-position error plot looks better in curvature code. Tegan will look up details and email them around.
- First Look at BCAL Clusterization: Will M.'s plots: 1st cluster is split into 2 clusters but this is a red herring as in this view it is not obvious that the clusters are 100 cm apart in z. Clusterization is based on (0,0,65cm) vertex assumption (centre of target). After all cuts (slide 4) 2574 events survive. Future improvements may require an iteration from cluster to track and back to cluster. Next: tackle single-ended hits.
- Any other business