Muon Offline Meeting -- Minutes ================================= 18 - Sept - 2002 ---------------------------------------------------------------------- Anyes Taffard -- Cosmic Ray Tagger ================= Anyes has made some major revisions to the cosmic-ray tagger. The main purpose has been to integrate Ashutosh's work on refitting cosmic tracks and looking for their partners on the other side of the detector, but this created an opportunity to rearrange the code into more easily-digestable pieces. The tagger still only tags events that have at least one muon with a stub in them. Now Ashutosh's code is run, along with all the algorithms that the tagger had before, including looking for TOF and hadron TDC hits, and flipping tracks over and looking for stubs to identify new "muons." There have been improvements to some of the algorithms, especially the TOF, and the performance of just about everything is improved by refitting the tracks first, i.e. the helix coming out of the Ashutosh refit is more likely to point to the correct outside-COT parts of the detector than its predecessor was. All of this information gets written out into an upgraded CosmicRayInfo object. Anyes has been running the tagger on various control samples, including W and Z to electrons (which shouldn't be identified as cosmics), but she has yet to set the default cuts for cosmic identification. She was encouraged to consider just taking an "or" of the cuts used this summer with the new cuts from Ashutosh based on track refitting as a first pass as soon as possible. Henry pointed out that it would be useful if parts of the tagger could easily be run on non-electrons. ---------------------------------------------------------------------- Un-Ki Yang -- MC Study of Kaon Decays in Flight ================================= Un-Ki has continued his investigations of kaon decays in flight. He simulated a large number of single kaons and pions. The simulation does reproduce the main features of the data. Taking the kaon pT to be 6 GeV, the probability to obtain a CMUP muon is 1.6 x 10**-4. (No additional quality cuts, such as a tighter cut on dX, are applied at this stage.) Un-Ki used production version 4.7.1. The trigger rate of single isolated tracks is 120 mb, for pT > 5 GeV. Combining this with the results of the simulation, the trigger rate for kaon decays in flight is about 5.8 nb, 1.6 nb for pions. This is in fair agreement with the data. Un-Ki showed distributions of the characteristic quantities, such as the calorimeter energy distribution, the angle wrt the closest track, Z(muon)-Z(vtx), etc. About 1/3 of the muons have zero calorimeter energy. Un-Ki plans to study signatures based on the pattern of hit residuals. ---------------------------------------------------------------------- Jim Bellinger -- BMU Stubs ========= Jim has implemented new alignment and calibration constants which result in a slight increase in the number of muons per events. There are some spikes at the arc boundaries, caused by slight errors in the geometry description. Jim will be able to fix these errors quickly. There was a bug in the calculation of the Z position which Jim fixed recently. A plot of the (X,Y) position shows about 10% of stubs have nonsensical values. These are certainly junk stubs and will be easy to eliminate. The chi-squared distribution for the stub fits now looks reasonable. The corrections for the Z position, etc., are ready to be tagged. The elimination of the junk stubs will be ready soon, in time for the 4.8.3 test reprocessing. ---------------------------------------------------------------------- Dave Dagenhart -- Progress on CMX Drift Velocities ================================ Dave has completed the measurement of the CMX drift velocities for the `Spring' data set (Feb - June). For this he has to consider each wedge individually, where in this context `wedge' is based on the way the HV is distributed in the CMX. Dave's plots show a large improvement in the dZ distribution before and after his calibration. The rms of these dZ distributions is very close to those of the August data, for which the HV was uniform across the CMX. Dave noted that wedges 7&8 west appear to have some residual time dependence. Presumably there were frequent adjustments to the HV for these wedges. There is no such time dependence for the August data. Wedges 23&24 east apparently have an HV jump in April. It was agreed that the pre-Feb data should not be calibrated as it will never be used for physics. The data from June to 7-August are chaotic as regards CMX HV. It was agreed to declare these data bad and that no calibration was warranted. Dave checked the alignment of the CMX wedges and found very good alignment. These constants have been given to Victoria to put in the data base. Dave will investigate some residual time dependence, and he is writing a CDF note. He pointed out the need to monitor the CMX drift velocity, perhaps on a weekly basis. It was suggested that Dave contact Michael Gold to ask him to tune the simulation to reproduce the data, as he has done with the CMU. ---------------------------------------------------------------------- Victoria Martin -- Status of Calibration Constants & DB ==================================== There are two data bases: the online and the offline ones. Constants are put into the online data base, and then are automatically propagated to the offline. Recently Victoria has deposited many new calibration constants, including the new CMU Z constants from Lucio, the CMP drift velocity and T0 from Lucio, and the CMX drift velocities from Dave (including the Feb-Jun period). Victoria is waiting for updated constants for the CMU and BMU. We expect new CMU constants from Hyunsoo Kim next week. This will be too late for the 4.8.3 test reprocessing, but not for the `final' reprocessing for the Winter conferences. Victoria plans to make a new L3 Valid Set today, and a new Production Used Set tomorrow. She wants to check the CMP wall positions, too. ---------------------------------------------------------------------- K.Bloom & M.Schmitt