Radphi Video Conference 11/18/2003

November 18, 2003
University of Connecticut
Storrs, CT

Meeting Agenda:

1. Latest draft of NIM paper - S. Teige
2. Progress on kinematic fit - D. Krop
3. Progress on BGV calibration - D. Steiner
4. December meeting at JLab - R. Jones

• http://dustbunny.physics.indiana.edu/~teige/nim
• http://zeus.phys.uconn.edu/radphi/call-notes
• http://dustbunny.physics.indiana.edu/~dkrop/kinefit
• http://cabibbo.physics.wm.edu/~steiner/radphi

Present: R. Jones, D. Steiner, D. Armstrong, S. Teige, D. Krop

At the above link on Scott's web site, he has posted the latest draft of the NIM article. There has been no progress on the article since the last call. David has yet to complete the section on radiation damage to the lead glass blocks. Richard offered to draft a new section on calibration and resolution of the LGD.

• Last time Dan described a problem he saw when he applied his kinematic fitter to the omega 3gamma sample, that the residuals returned by the fit for some of the kinematic quantities showed narrow spikes at zero. This can be due to a failure of the fit to converge, but Dan said that most of the fits returned with convergence. In particular, his theta_recoil residual distribution looks weird, with a very asymmetric tail toward large positive differences. However he did observe that if he cut on the chi-squared from the fit that it did select the true omega region of the mass plot. With some reservations he continued on to look at the four-gamma sample and to try to use the fit to distinguish pi0,pi0 from eta,pi0 events. He determined the rate of misidentifying pi0,pi0 events as eta,pi0 to be 30% based on the chi-squared only, based on Monte Carlo. Adding a cut of 50 MeV around the eta and pi0 masses reduced the mis- identification rate to 4% and still kept 50% of the pi0,pi0 reconstructions.
• The spikes at zero are still present in the distribution of the residuals. He showed this effect in the azimuthal angle of the recoil proton. He tried reducing the Fermi smearing to zero of the initial proton in the fit, and it seemed to reduce the spike, but it is clearly not physical to turn off Fermi smearing.

• Dan looked closer at his mass plots for the 1barrel-1forward events to see if the peak is really coming from the eta. To check this, he moved the forward Egamma threshold around and looked to see if the mass peak moved with it. He also tried the same thing with the barrel gamma threshold. In the plots he showed, the shape of the edge at low energy did change but the position and width of the peak at about 500 MeV did not shift. This seems to indicate that it is the eta. A discussion followed regarding whether we should proceed to calibrate on the pi0 peak that we feel sure is there, or to try to calibrate on this broad bump we think might be the eta. If the broad bump is not the eta then we expect that trying to tune calibration constants will not make it any narrower. If it does sharpen up with calibration then this might be be the best handle we have. Otherwise we might tune on the pi0 and use the eta as a diagnostic. Either choice should work. Dan will take the lgd_tune code and refit it for doing the barrel calibration, keeping both possibilities open.

• We agreed to hold a face-to-face meeting at Jefferson Lab on Monday December 15. That date immediately follows the GlueX meeting and so is a convenient time and place from the point of view of travel. Richard will see about booking a room and distributing an agenda. There will be no more conference calls before that meeting.