Minutes of the CALCOM meeting, Novermber 7, 1997. ================================================= Let me start with some good news discovered after the meeting. As of Saturday 10pm or so, some problems in the elastic cross section analysis got resolved: (1) some weeks ago the elastic cross section showed a slower than expected falloff in the angular dependence. This was traced back to a missing "sin(theta)" term in the d(Omega) = sin(theta) dtheta dphi. (2) the second problem, the factor ~2.5 in the absolute normalization was found to be largely related to the "prescale factor" in the trigger. A "1" was interpreted to mean prescale factor 1 while it really means skip 1 event, thereby reducing the numberr of events by a factor of 2. These factors seem to explain the bulk of the disprepancy. The remaining missing 20-25% may be due to radiative corrections (5-10%)+ tracking inefficiencies (5-10%) + chamber misalignment + magnetic field + ... but at least we are now in the right ballpark and may begin the fine tuning. I just wanted to let you know about this. Volker Burkert ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Agenda: V. Burkert - CALCOM project list, where are we? L. Murphy - Status of the photon tagger and downstream & A. Longhi devices anaylsis K. Joo - A first look at the electron trigger(s) P. Rossi - Calibration of the LAEC H. Egiyan - TORUS magnetic field analysis status ============================================================================== Individual reports: ================== V. Burkert: --------- The task list was updated to include some new projects some of them suggested by Steve Dytman. Also the list has been amended by tasks that need to be completed before the next data taking period (11/24 cosmics). This list will grow as time goes on, and may develop into a check list for things that must be checked before each run. The good news is that CALCOM work is progressing on many fronts, however time is scarce, and more people need to get involved to enhance the CALCOM effort to the degree necessary for physics data taking in December. This is in the interest of each individual and the entire CLAS collaboration. CALCOM Analysis Tasks for October/November 1997 =============================================== 1. Absolute tracking efficiency, resolution, track reconstruction efficiency vs luminosity (F.J. Klein, B. J. Mueller, B. Niczyporuk, L. Qin) -> work in progress -> most urgent is tracking efficiency versus luminosity (1.6, 2.4, 4 GeV) (relative changes) - hit-based tracking - time-based tracking 2. DC B=0 straight track analysis to identify possible sources of background in beamline ( N. Pivnyuk, K. Mikhailov, A. Skabelin) -> work completed 3. Extract elastic and inclusive ep cross sections as a means of absolute normalization, do this for different luminosities, energies and +/- 10% accuracy. (A. Coleman, J. Manak, K. Joo) -> very urgent, work in progress. (reasonable understanding was obtained after the meeting, see comments above) 4. Trigger studies (S. Barrow, L.C. Smith, K. Joo, R. Thompson, V.B.) - uniformity of trigger response in EC - does the prescaling work as expected? - electron selectivity of EC, EC.CC, EC.TOF CC.TOF (fraction of elastic electron events) - what is the efficiency of each trigger? Are there non-uniformities due to missing channels, or low pmt/electronics gains? -> very urgent, work has started 5. Is our live time measurement correct? (J. Manak) - look at FC, clock info in data stream, is # elastic event/FC(corrected) = constant (throughout a run) and independent of luminosity (for different runs, same energy, trigger, target) -> very urgent 6. DC performance: Hit occupancy versus luminosity, noise versus track related hits, DC layer efficiency (D. Carman, J. Mueller, B. Niczyporuk, U. Pitt.) -> very urgent, hit occupancy has been completed, work on DC layer efficiency has started 7. Gain & time - matching for TOFs (S. Taylor, C. Loukachine) -> gain matching in progress -> time matching for panel 1 in progress 8. Particle id using matched tracks and ToF, SEB (L. Elouadrhiri) -> good progress on dE/dx -> needs extensive work on paddle by paddle time matching automated procedure should be developed 9. EC calibration using min. ion. particles & reconstructed gamma-gamma (from pi-0) (L.C. Smith) -> progress on all fronts, need to extract new attenuation lengths from cosmic runs and/or from min. dE/dx beam runs 10. Checking the drift chamber alignment (N. Pivnyuk, K. Mikhailov, A. Skabelin) - straight track analysis - elastic ep scattering with B = 0, ep angle correlations - reconstruct target walls (empty target cell) -> machinery has been setup, work to proceed within RECSIS framework 11. LAEC calibration (P. Rossi, Marco Battaglieri, INFN) -> cosmic ray calibration well established -> beam calibration procedures started using elastic protons. 12. How well do we understand the Torus magnetic field (H. Egiyan, R. Minehart, K. Joo) - elastic scattering with and w/o magnetic field - study correlation of kinematical invariants (W) in ep elastic, or MM in ep -> eppi0, enpi+ vs theta/phi angles - check that the B-field map has no impact on results by using different grids (rectangular (standard) and trapezoidal (M. Kossovs new grid). Elastic peak should not shift by more than 1-2MeV. Same for miss. mass. of exclusive reactions. 13. Single particle acceptance functions vs px,py,pz,Q (V.Burkert) - use TBT and known DC positions & correct magnetic field to map out charged particle acceptances from minimum biased events. Use these to make GSIM more realistic -> work to start after tracking and detector geometry better understood Checks that must be done before the e1 run: =========================================== (1) Verify the correctness of our live time measurement procedure using electronic pulser (S. Barrow, V. Gyurdjyan, E. Wolin) (2) Verify all tubes fire, give signals in ADC, TDC Detector: resposible: Cherenkov - J. Price TOF - S. Taylor, C. Loukachine EC - L.C. Smith, A. Coleman LAEC - (3) Verify all input lines give a signal to the level 1 trigger input, especially those that may be used in e1 (EC_TOT, CC_OR, TOF_OR_panel_1) Check this by firing only one channel (e.g. with pulser) responsible: S. Barrow ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Luc Murphy: Update on the Photon Tagger Effort ---------- A) Introduction: We have presented the status of the tagger system. This includes the status of the analysis of the T and E-counters separately, as well as in coincidence. We also discussed the status of the downstream devices and the methods we use in order to determine the quality of the beam and the photon flux. Finally we started to present the results obtained by the Saclay group to determine the accurate event time, by using the tagger time in coincidence with the RF signal from the accelerator. B) The T-counters: The T-counter plane is complete and operational except for two counters out of the 61. We have shown the clear coincidence TDC peaks from these counters, when the daq was triggered by the one of the pair counters. It has thus been clearly demonstrated that an event in the tagger can be directly associated with photon events down the beamline. C) The E-counters: The same exercise was performed with the E-counters as had been done with the T-counters. Once again we identified with no difficulty the coincidence peaks between the tagger and the appropriate downstream devices. We remarked on the double timing peak associated with each events. This is due to the fact that the TDCs record both the rising and falling edges of the pulses. We also pointed out good alignements of the TDC peaks from E-counter to E-counter. D) The E-counter response as a function of applied voltage: In our attempt to plateaux the E-counters and determine the optimal high voltage with which to drive them, we have found that the E-counter electronics become extremely sensitive to dark current from the PMTs. We have shown how increasing high voltages adversely affected the TDC spectra. For a more in depth discussion on the subject, please refer to: http://www.cebaf.gov/~murphyl/tagger/analysis_index.html E) The response of the E counter detector plane with respect to the response of the Total Absorption Counters: We have shown on a two dimensional plot the relationship between the energy deposited in the Total absorption counter as a function of which E-counter was hit. Not only did a clear correspondence emerge, but the magnitude of the 'peaks' suggested strongly the shape of the bremsstrahlung spectrum. F) Correspondence of the E and T counters hit patterns: We showed on another two dimensional plot the correspondence between events on the E and the T detector planes. Once again the relationship was striking, and demonstrates clearly that cuts can be applied in the analysis package requiring coincidences between appropriate E and T counters. G) Tagging Efficiency and Photon Flux Monitoring: Events from the T-counters are recorded by three banks of scalers. One bank of scalers is free running and counts T-counter hits. The other two banks of scalers are gated by the downstream devices. The selection of which device gates the scalers is done remotely. In the example we presented, we showed the ratio of scalers gated by the Total Absorption Counter(TASC) by those which are free running. This is essentially the measure of the tagging efficiency. For each T counter, we define tagging efficiency = (tag in coincidence with TASC)/tag. During operation with an unobstructed beam we measured raw tagging efficiencies of approximately 88% (except for the extreme ends of the focal plane). The values determined from the scalers were consistent with those obtained by analysis of the TDC's and TASC ADC. The photon beam for this run traveled through approximately 35 m of air, producing pairs which were swept out by the pair spectrometer magnet field before reaching the TASC. Thus about 9% of the "inefficiency" is due to the air path, resulting in a true tagging efficiency of greater than 95% except at the high-photon-energy end, where Moeller-scattered electrons are known to produce false tags with no photon in the beam. In the same vein we showed the T-counter scalers gated by the pair counter, whose conversion efficiency is of order 1%. The results are very consistent, giving us the confidence to use the pair counter as a monitoring device during high intensity runs. H) Determining Accurately determining the Time of an Event by Using the Beam RF signal: We presented results obtained by the Saclay group to better determine event timing. They showed that the time jitter associated with the tagger trigger is too large to allow good particle discrimination based on the beta of the particle. The Saclay group worked on a method to associate events with discrete beam bunches. The results presented showed their success, and demonstrated that the timing of a tagger event can be reconstructed within a sigma of 80nsec. Gerard Audit and Eric Anciant are to discuss their results further next week. ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ P. Rossi: -------- I presented some results of the LAC analysis performed using tracking information. We analyzed the run 6079. The results were obtained using the following selection: -2 tracks with opposite charges -Hypothesis of negative track being an electron -Calculation of kinematical quantities under this assumption (W, Q^2, et) -Selection of elastic scattering events using cuts in energy and angular correlation -Selection of events with elastic electrons or protons in the LAC I showed the angular distribution of both elastic electrons and protons. The statistics for elastic electron going into the large angle calorimeter is very poor while for elastic proton we have much more numbers. The proton energy calculated in the large angle calorimeter versus the angle theta extracted from tracking shows the expected behaviour of the elastic scattering; also the correlation between the proton energy calculated in the large angle calorimeter versus the proton energy obtained using the tracking is good. All the plots that I presented are in the web page: http://aiacehp.ge.infn.it/~ripani/offline.html Finally I showed the preliminary results of our gain monitoring system. H. Egiyan: ---------- In order to understand the discrepancy between field mapping data and computer simulated (expected) data the survey analysis was done, using least square fitting of Z and X coordinates sepa- rately. Each fit had 6 unknown parameters, like encoder-to-length and encoder-to-angle parameters, X0, Z0 coordinates of the mapper position etc. It was found that these fits describe the survey data with Sigma about 1-2 mm within the same cryostat. But for different cryostats the values of X0 and Y0 values vary significantly and the reason for that is not yet understood. The comparison between simulation and actual data using these values of parameters for each Phi angle hasn't been finished yet. K. Joo: ------ I analyzed the runs, 6079 of 1.6 GeV and 6389 of 2.4 GeV. The run 6079 used high threshold EC trigger and the run 6389 used two triggers, EC and EC*CC triggers. For run 6079, I studied CC efficiecies by looking at the electron EC x and y hits with/without matching the informations for each track between the drift chamber and the CC system for electron identification. For this analysis, I used inclusive and exclusive data for elastic and inelastic proton scattering (e,e'pi+) to study a wide range of electron momentum. For particle identification between proton and pi+, I used the TOF informations, the details of which were presented by Latifa at the calcomm meeting a couple of weeks ago. The results show about 77% overall effieciencies from either inclusive or exclusive data and also local ineffieciencies due to the geometric acceptance difference between the EC and CC. Inelastic scattering data also show similar results. For run 6389, I analyzed two types of triggers separately. One is evtype=1 and the other one is evtype=14. Hadron invariant mass W spectrum from evtype=14 show the loss of the second resonance peak. It shows that we need a better understanding of our trigger types. More studies are underway. -------------------------------- The next meetings will be on Tuesday 11/16 and Friday 11/21. ======= Tentative agenda(s): -> update on trigger studies (threshold, efficiency, hardware and analysis) -> elastic x-section, DAQ live time -> tracking efficiency vs luminosity analysis -> update on TOF calibration and particle id -> update on the DAQ/Online/Run control -> more on photon tagging analysis from Saclay -> ....... -> preparations for e1 ==============================================================================