Minutes of the Muon Software Meeting 19 February, reported by Victoria. ----------------------------------------------------------------------- 1. Camille Ginsburg: Update on Z's using BMU muons. Camille is using CMUP-BMU and CMX-BMU muon pairs to reconstruct Z's. The BMU muons were reconstructed using either the "standard" COT+SVX tracking or silicon only tracks. The dX cut used is +/- 10cm. 333 Z candidates are found in 90pb^-1. Only around 10 of the 333 events actually have silicon-only tracks. Camille has also looked at the turn-on curve for BMU muons, which seems to be fully efficient for pT's of 2.4 GeV/c and above. Lee Pondrom noticed that the z0 distribution of the muons from these Z events is skewed wrt a typical z0 distribution. He said that this is because the BMU is more likely to detect muons from that originate from high-z0. 2. David Dagenhart: CMX reconstruction efficiency from J/psi events. David has calculated the CMX reconstruction efficiency using J/psi->mu+mu- events. He requires a first CMUP muon leg, plus a second minimum-ionizing track pointing at a CSX hit. (In simulation the efficiency for finding a CMX muon when a CSX hit is present is 0.999, so this requirement should not bias the measurement. The second track must also be fiducial in the CMX with a fiducial distance of less than 0 cm. Looking for a CMX stub to be associated with the second leg he finds an efficiency of (0.978+/-0.009). If the fiducial distance is varied, the calculated efficiency varies from over 99% to 96%. Phil Schlabach said that this was probably due to the fact that the CMX chambers are not as efficient at the ends, the efficiency probably drops sharply in the last last few cm's at each end of the chamber. The same is true for the CMU and the CMP chambers. This inefficiency is not modeled in the MC. We discussed what could be done to account for this effect. Phil told us that in run 1 fiducial cuts were applied to avoid this area of the chambers. Phil said that the muon fiducial tool could be used to make this cut in run 2. The number of muons that scatter into the acceptance of this cut, will be equal to the number of muons that scatter out of the acceptance of this cut. David also find one bad run for the CMX (152581) which is still included on lepton good run list. 3. Natasha Miladinovic: CMX drift velocity. Natsha is using CMU-CMX J/psi events to investigate the CMX drift velocity. In order not to be influenced by potential miscalibrations in the database, she used hardcoded values for the CMX calibration constants. The delta-Z (dZ) between the reconstructed stub position and the projected position of the track in the CMX is plotted. If the drift velocity is correctly tuned then dZ should be centered on 0. The dZ distribution is fitted to a Gaussian. The mean of this Gaussian is plotted against wedge number, and against week that the data was taken. Two main observations are made: (a) for about 4 weeks last October (runs 152942 - 153739) the mean of dZ is offset by about 6cm. For the rest of the time the mean of dZ is within 2 cm of 0. (b) there is a significant dependence of dZ with respect to the wedge number. The overall shape of this dependence does not change over time, but may be shifted up or down. We discussed what could cause these two effects. Liz SK and Andreas Korn said there had been many changes in the online software and the trigger conditions at around the time when the drift velocity seems to change. Phil said that another (remote) possibility is a change in the CMX drift gas which may also occurred around that time. David said the variation of the drift velocity with respect to the wedge number could be due to misalignment of the chambers in Z. Phil said this could also be due to the HV or the gas flow around the wedges, although this last effect is not very likely. Natasha plans to calculate a new set of drift velocity constants. There will be one overall constant, which can change on a weekly basis, plus a set of drift velocity factors for each of the CMX wedges. Everyone discussed if there would be enough data every week to make a accurate calculation of the drift velocity, but it the end we all agreed Natsha's plan was the most sensible thing to do.