Transcript Resolution study of the GE1/1

GE1/1-III GEM Cluster Size and Resolution
Studies with the FNAL Beam Test Data
Aiwu Zhang, Vallary Bhopatkar, Marcus Hohlmann
Florida Institute of Technology
CMS GEM workshop IX
14-18/07/2014
Outline
• Current data analysis status from the FNAL
beam test
• Cluster size (number of strips in a cluster)
studies
• Resolution results
• Summary and next plans
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
FNAL beam test data analysis status
•
Finished basic characteristics: cluster charge distributions, gain curve,
detection efficiency, etc.
•
Implemented spatial resolution study method in polar coordinates.
•
This talk presents the cluster size studies and resolution results, with HV
scan data in middle-sector 5 and position scan data in middle position of
each sector. Note analog pulse information are decoded from APV chips.
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Mean cluster sizes
with different thresholds
HV scan in middle sector 5
7/15/2014
Position scan in middle sectors at 3250V
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Fraction of different cluster sizes
– HV scan in middle-sector 5
3σ threshold
4σ threshold
5σ threshold
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Fraction of different cluster sizes
– HV scan in middle-sector 5
N=1
N=4
7/15/2014
N=2
N>=5
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
N=3
Fraction of different cluster sizes –
middle-sector position scan (3250V)
3σ threshold
4σ threshold
5σ threshold
• Majority of the events have cluster
size 2 and 3.
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Fraction for different cluster sizes –
middle-sector position scan (3250V)
7/15/2014
N=1
N=2
N=4
N>=5
N=3
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Spatial resolution study -- method
•
Y offset
vertex
9.94°
tracker
X offset
Step 1. Fix an X offset, loop over
Y offset with a step 0.2mm; we
get a pair of (X,Y) by requiring
residual mean equals to 0.
Eta5
We study resolution in polar coordinates (φ)
due to the trapezoidal shape and radial strips.
We use the vertex as origin and align the
trackers by finding the X, Y offsets. (rotation of
the GEM is also taken into account, not shown
here.)
Step 2. Fix a Y offset, loop over X
offset with a step 1mm; we find
another pair of (X,Y) from the
minimal of the parabola of
residual sigma.
We assume that any misalignment will shift peak of residual distribution
away from zero and give larger residual width (or make distribution wider)
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
GE1/1-III GEM Residual distributions
σ=138μrad
Exclusive
residual
σ=110μrad
Inclusive
residual
Exclusive (inclusive) residuals at 3250V in sector 5 (from HV scan data)
Cluster size in GEM and trackers all >0.
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Resolution results – behavior in detail
Resolution vs. HV in middle-sector 5
7/15/2014
Resolution vs. sector at 3250V
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Strip non-linear response study -- motivation
• Centroid position distribution from COG method (in middle-sector 5).
N=2
N=3
• From the centroid position distributions of fixed cluster sizes N=2 and 3, we
observe that they have apparent bumps around each strip. For N=2, the
positions in the middle of two neighboring strips; for N=3 the positions in the
strip. (note we define the position of a strip in its center.)
• This brings us to study the non-linear strip response of charge distribution on
position reconstruction, and hence to flat these distributions.
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Strip non-linear response study -- method
h(η2) distribution
Correction function
for 2-strip events
7/15/2014
h(η3)
distribution
Correction function
for 3-strip events
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Cluster position distributions
before and after correction
HV scan data @ 3250V
N=2
Before correction
N=3
7/15/2014
After correction
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Strip response functions
HV scan on
middle sector 5
N=3
N=4
N=3
N=4
N=2
Position scan
at 3250V
N=2
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Resolution results (HV scan)
before and after position correction
All valid events with
cluster sizes 2,3,4
N=3
N=2
N=4
Resolution is not improved after position correction with HV scan data!
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Resolution results (position scan)
before and after position correction
N=2
All valid events with
cluster sizes 2,3 and 4
N=3
N=4
Resolution is not improved after position correction with position scan data!
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX
Summary
 Most of the clusters have sizes no more than 5.
 Spatial resolution is on the order of 120μrad.
 We don’t observe improvement on resolution after applying cluster
position correction.
• What else can be done with these data (HV scan, position scan in
middle sectors):
 Study details of cluster size distributions using VFAT-like strip hits.
 Apply position correction a second time to flatten position
distributions even more and to see if resolution could be improved.
 Study resolution with different thresholds.
• What needs to be done with other data:
 We also performed two position scans with the detector tilted by
7degrees from normal incidence. The effect of the inclination needs to
be studied.
7/15/2014
A. Zhang et al., GE1/1-III GEM Studies with FNAL Beam Test Data, GEM workshop IX