Difference between revisions of "PID Day 1 Notes"

From GlueXWiki
Jump to: navigation, search
Line 15: Line 15:
 
It's clear that the cosmic data, photon data, and also TRIUMF data need to present a clear picture of the performance of the BCAL.
 
It's clear that the cosmic data, photon data, and also TRIUMF data need to present a clear picture of the performance of the BCAL.
  
The goal is to have all of the above test beam analysis summarized in a report by July 1.
+
The goal is to have all of the above test beam analysis summarized in a report by July 1.  This report should also include what work has already been completed on analysis of data with the converter in front of the BCAL.
  
 
=== BCAL Standalone MC ===
 
=== BCAL Standalone MC ===

Revision as of 17:14, 14 May 2007

The whiteboard at the end of the day laid out tasks to do in the near term. These were grouped into three categories:

Test Beam Analysis

We wanted to decouple PID from photon position and energy resolution. We decided the best thing to do is to focus on the position and energy resolution of photons since this only involves energy and time difference. In theory it a technically easier analysis since it does not depend on understanding tagger systematics. The items to probe are:

  • z dependence of energy resolution and time difference resolution (Alex/Blake)
  • angular dependence of energy resolution and time difference resolution

Regarding charged particle ID, the key quantity from the BCAL is the mean time resolution. We decided there is some significant extrapolation to take information on TOF from photon and muon data and extrapolate to hadron data. In some sense cosmics and photons provide two extremes for the response we expect from hadrons. We need to extract data from test beam on these items:

  • resolution in mean time for low energy photons (Blake)
  • resolution in mean time for cosmics (Christina)

It's clear that the cosmic data, photon data, and also TRIUMF data need to present a clear picture of the performance of the BCAL.

The goal is to have all of the above test beam analysis summarized in a report by July 1. This report should also include what work has already been completed on analysis of data with the converter in front of the BCAL.

BCAL Standalone MC

The goal for the July workshop is construct a simulation that provides accurate simulation of:

  • converter runs
  • energy, mean time, and time difference resolution

Zisis is working on this.

After this MC has been created and validated additional studies can be done. (These will likely not be done before the July workshop.)

  • develop parametrization of resolution that can be put into GlueX GEANT MC
  • explore hadron showers in the BCAL
    • does path length of grazing showers contribute to timing resolution?
    • pi, p, k, n interactions
    • what is the impact of the photosensor spec on performance

Full GlueX MC

  • single photon reconstruction (Mihajlo/Matt/Zisis/Andrei)
    • adjust MC thresholds
      • The threshold now in the MC is set at 30 MeV of total energy deposition in the cell. The simulation models a homogenous mix of lead and fiber and is therefore looking at total energy deposition in the cell. George notes that 26 keV of energy deposited in the fibers gives roughly one photoelectron in the photodetector. To go from deposition in fibers to deposition in lead + fibers divide by 12%, this means that 1 p.e. is roughly 220 MeV total energy deposited in the cell. Therefore if one assumes a sensitivity of around 5 p.e. then a suitable threshold for the GlueX MC might be more like 1 MeV than 30 MeV.
      • Having the threshold set too high will cause one to loose lower energy cells at the edge of a cluster and significantly affect energy resolution. This could be the dominant effect in Mihajlo's studies.
      • The MC code needs some speed optimization so that it is useable for high statistics studies -- Matt will work on this.
      • We need to use the GlueX MC to map out the response of the BCAL near the downstream end with high statistics. The response will vary dramatically as one increases angle.
      • The sampling error of 5% of sqrt(E) needs to be incorporated into the BCAL MC
      • All hard coded parameters in the MC need to be checked. These include fiber properties, thresholds, etc.
    • π0 reconstruction (Matt/Mihajlo)
      • lower thresholds are expected to enhance resolution
      • combinatoric backgrounds due to beam related noise (which increase with lower thresholds) need to be explored as they will affect the purity of reconstructed π0's
    • Material (Richard/Mihajlo/Matt)
    • What impact does alternate FDC geometry have on efficiency and resolution of MC.