Minutes of the CALCOM meeting, April 10, 1998. ============================================== Agenda: Drift chambers: --------------- N. Pivnyuk - Analysis of straight tracks in CLAS drift chamber J. McNabb - CMU analysis of straight tracks M. Mestayer - Future plans for DC calibration Cherenkov counters: ------------------ A. Vlassov - Analysis of Cherenkov counter efficiencies based on single photoelectron calibration Short reports: -------------- B. Asavapibhop - First analysis results from the photon run B. Niczyporuk - Observations in DC calibration (no report) ======================================================================== V. Burkert: ---------- There were some discussions on the relative priorities of the drift chamber calibration work. Having the correct DC geometry in RECSIS is obviously extremely important for everyone doing tracking analyisis. It is therefore most urgent to implement the results of the work done by the ITEP and CMU groups into the official tracking code as soon as possible. A target date for this to happen was set at May 1, 1998. By this time we should have a consensus on what changes on the DC geometry need to be implemented. Even if this is not the final geometry, it should provide a significant improvement over the current situation. Individual reports: ==================== N. Pivnyuk: Current Status of the Straight Track Analysis (B-field OFF). ----------------------------------------------------------------------- The major result that was achieved in the last couple of weeks, is that for the first time it was demonstrated that Drift Chamber spatial resolution is REALLY very close (exactly the same ?) to the desigh values, namely, sigma ~ 200 - 300 microns in the middle part of the Drift Chamber cell, and rise to the values 500-600 microns approaching the sence wire and the edge of the cell. The time-to-distant functions and spatial resolution vs. drift time were obtaineded for the superlayers 3,5,6. The mutual geometrical misalignment between reg.2 and reg.3 in sector 1 was studied. After the proper geometrical corrections were implemented the SUCCESSFUL (chi^2 ~ 1) semultaneous Time-Based-Tracking in the three (3,5,6) superlayers was achieved. It could be considered as the strong evidence, that mentioned above values of the spatial resolution are CORRECT and REALISTIC. It also can be considered that straight track analysis is a powerfull tool for the investigating the Drift Chamber time-to-distant functions and spatial resolution as well as the geometrical missalignment between different regions. J. McNabb: Drift Chamber Alignment & corrections to the survey data - Work in Progress by Robert Feuerbach - The program being developed takes the RECSIS tracking code, with minor modifications to allow zero field tracking, and does a fit to region 2 only. The resulting tracks are then projected into the other regions. Several iterations are used to do a chi^2 fit of the driftchambers locations. This is a fully automated process which only requires a human to check final results. M. Mestayer: Drift Chamber Tracking Status ------------------------------------------ Work Plans: (NOT including operations or monitoring) General Software Improvements (Klein, Mueller, Manak, Stepanyan, others) Simplify bank structure Simplify I/O - fewer TCL variables Cleanly separate hit-based and time-based tracking Develop utilities - wire->elec. and wire->geom. map, SSLT, etc. Improve Track-Finding (Klein, Mueller, others) Use all hits, not just one per layer Other? Wire Status (Thompson, Coleman, Carman, Yun) Program to find dead wires and load status into MAP: PDU Program to ``kill" GSIM hits for dead wires AFTER generation: GSIMKO Time Calibration (Qin, Niyazov, Skabelin) Program to analyse pulser data: RECCAL Study drift time spectra for ``real" data; shifts of T0 indicate mistake in pulser set-up, time-walk, problem with flight times or signal propogation times Distance vs. Time Calibration (Qin, Klein, Joo, Niyazov, Skabelin) Clean up procedure to read parameters, evaluate function D(t,...) to fill x vs. t tables Develop function which obeys constraint, D(TMAX,...) = DMAX, where DMAX depends on local track angle Build in dependence on B field Develop an automatic calibration program -choose unbiased data sample -fit drift time distribution for TMAX -fit DOCA vs. time for other parameters -do for each run, enter parm's into MAP Alignment (Pivnuk, Mikhailov, Schumacher, Feuerbach, McNabb, others) Enter survey data into MAP Minimize chi-squareds for a sample of straight tracks by adjusting chamber positions and orientations A. Vlassov: Status of the Cerenkov counter efficiency analysis ----------- The status of Cerenkov efficiency was presented. 1.6 GeV data February runs (#8797 - #8815,#8825 - #8857 ; ~ 21,700,000 events total) were analysed . The minitorus current was 6000, torus - 1500 A. Events with negative (TBT) tracks were selected. Next cuts were applied to select the electron candidate : 1. Missing mass : 0.85 - 1.1 2. EC energy deposited / initial momentum : 0.2 - 0.4 3. Difference between TBT matching point in EC and EC hit position: on X : -5.0 up to 15.0 on Y : -15. up to 15. cm The goal was to estimate the mean number of photoelectrons as a function of polar angle Theta & azimuthal angle Phy , Theta and Phy were calculated as a Theta and Phy of the strait line from the (0,0) point to the TBT matching point in CC. The number of events in a bin (0.5 degree on Theta, 1.0 degree on Phy) deviated from ~1000 in small segment numbers up to ~ 5-10 at segment 18. The number of photoelectrons was calculated according to the single photoelectron positions, stored in the database. The plot of the mean photoelectron numbers looks like it was predicted by GEANT estimations, mostly this number is > 6. One region of lower efficiency is near the Sector middle plane (N Phe ~ 3-5 ), which is expectable. The other region was the region of segment 15 in sector 4 - PMT #29, Where this number is ~ 2-4 . But it appead to be a mistake in SPE peak position in the datamap - (480 instead of ~100) so Sector 4 is also not bad. I intend to put the pictures of mean number of photoelectrons for all Sectors to WWW soon ( http://www.cebaf.gov/~vlassov/ ). The questions to investigate are : 1. We need to see the efficiency estimation as a function of detected photoelectrons, to be sure, that this efficiency is consistent with the Poisson distribution, and that we define the number of photoelectrons correctly. Unfortunately, I have to use December runs without CC in the trigger for this goal. ( Better use same runs for both. ) 2. We need to get same plot for the other value of Torus current : it will be a test of Nphe dependence on the angle between the track and normal to CC detector surface. I don't expect it to be large, but it must be checked. B. Asavapibhop: A first look at the photon commissioning data -------------- I've presented some plots from the photon run data taken last week. I've used the time from the SC and ST to get the start time. By using the current calibration constants for other detectors(no ST calibration yet) and an estimated time offset between ST and SC(around 90 ns), I can use RECSIS and modified SEB to analyse data up to particle id witout any correction. I got the pion peak at 139.1 MeV with sigma=40 MeV and proton peak at 989 MeV with sigma= 30 MeV. The reconstructed pi0 mass from EC is 134.6 MeV with sigma=21 MeV.