Home » Groups » STAR Protected

PXL : tracking efficiency in overlapping areas

Updated on Sun, 2013-04-14 15:12. Originally created by bouchet on 2013-04-14 11:06.
PXL Geometry for Y2013  : 3 sectors.
Example of tracks with more than 2 PXL hits due to the overlap between 2 outers ladders is show below.
  1. straight lines in red are for vertex Z = 0 cm
  2. even for a slighly shift in vertex Z, we can never have tracks with 4 PXL hits in that case (should consider curvated tracks)
1) Quick estimation of the number of tracks with 3 PXL hits.

a) simulation setup
  1. 400 events
  2. 50 pions/event
  3. 0.1<Pt <5.0
  4. |η|<1
  5. 0<Φ<2π
 detp geom y2013
 gexec .$STAR_HOST_SYS/lib/xgeometry.so
 gclose all
 user/vxyz 0.001 0.001 0.01
 user/vsig 0.01 0.1
 gfile o debug66.fz
 aguser/gkine 50 8 0.1 5.0 -1.0 1.0 0 6.28
 trig 400
The numbers of TPC and PXL hits used in fit ( per global track ) is counted via StGlobalTrack :
    tTree.pt = gTrack->geometry()->momentum().perp();
    tTree.tpc = gTrack->fitTraits().numberOfFitPoints(kTpcId);
    tTree.pxl = gTrack->fitTraits().numberOfFitPoints(kPxlId);
The following histograms shows the PXL hits distribution for this sample  (21914 global tracks)


The number of tracks with 3 PXL hits (overlap btw ladders) is 226, which makes a ratio of = 226/21914 ~ 1% 
 
 
2) tracking efficiency for tracks having 3 PXL hits

a) simulation setup
  1. 5000 events
  2. 1 muon/event
  3. Pt =1.0 GeV/c
  4. |η|<1
  5. 1.80<Φ<1.82
The azimutal angle has been chosen to increase the number of tracks hitting the overlap btw ladder 1 and ladder 2 of sector 1
 detp geom y2013
 gexec .$STAR_HOST_SYS/lib/xgeometry.so
 gclose all
 user/vxyz 0.001 0.001 0.01
 user/vsig 0.01 0.1
 gfile o singlemuon_overlap.fz
 aguser/gkine 1 5 1.0 1.0 -1.0 1.0 1.80 1.82
 trig 5000



b) results

file : output_Ana.root
# events : 5000
# events with PXL hits =0 or >3  : 104
# events with PXL hits =1  : 114
# events with PXL hits =2  : 2842
# events with PXL hits =3  : 1940
# events with PXL hits used =3 for PXL hits =3  : 1845

 
root4star [2] 1845./1940
(const double)9.51030927835051498e-01
 
comments :
  1. the current version of the tracking (Sti) seems to not complain for overlapping areas (issues could have been that the tracker expects 2 PXL hits and sometimes finds more)
  2. Sti still uses 8microns for the PXL hit resolution (not sure if it's a real issue here given the very low track density)

 
Groups: