TGC LVL1 Commissioning

1
TGC LVL1 Commissioning

• 27 Nov 2006
• Hironori Kiyamura
• Kobe University

2
Contents

• Introduction
• Data sample Software
• Method
• How to calculate trigger efficiency for TGC.
• Analysis
• How to choose muon samples.
• How to decide which muon is triggered.
• Results
• Pt vs Trigger efficiency
• validation
• Position dependency of trigger efficiency
• Luminosity vs Sigma/Efficiency
• Summary Outlook

3
Introduction

• It is important for commissioning how to measure
  LVL1 muon trigger efficiency using real data.
• Analysis method has to be fixed before ATLAS
  experiment starts.
• Key points,
• Collect LVL1-unbiased samples.
• Select muon tracks without knowing MC truth.

• For this study,
• Muons in Z -gt mm process can be a good probe.
• Using simulation data,
• We have developed an analysis tool.
• Validate this method by comparing with the result
  of single muon samples.

4
Data sample Software

• Data sample (No background is included so far)
  csc11.005151.McAtNloZmumu.simul.HITS.v11004103
• generator cut (selection efficiency 89)
• Mmm gt 60GeV, at least 1 mhlt2.8, ptgt5GeV
• cross section 1662pb(after above cut)
• In this study, about 40pb-1 data are used.
• Software
• Athena 11.0.42
• TGCcabling-00-00-31 TrigT1TGC-00-01-15
• With these packages, digitization and
  reconstruction are performed and AOD files are
  analyzed.

5
Muon kinematics (MC truth)
Fig.1
Fig.2

• This sample is appropriate for study for high
  Pt(40GeV) region.
• There is enough data in endcap region.
• Two muons from Z are well-isolated from each
  other.

Fig.3
6
Method Analysis
7
Method
Leg technique

• The events triggered by MU20 are used.
• Look for a triggered muon based on LVL1 ROI.
• Look for another muon of decay products from Z.
  (named probe muon)
• Trigger efficiency for Endcap is calculated using
  the probe muon.
• If both muons from Z are MU20 ones, both of them
  can be used to the calculation.
• Use only well-measured muons (i.e. combined muon
  track).

8
Step.1 (triggered muon selection)

• Matching between reconstructed muon and LVL1 ROI
  is based on dR.
• Parameters are defined as follows
• dh hrec hROI, df frec fROI
• h and f of reconstructed muon are at IP.
• h and f of LVL1 ROI are at the center of ROI.
• dR (dh2 df2)1/2

• Look for the nearest ROI to reconstructed tracks
  within dR of 0.1.
• If it is found, the muon is decided to be
  triggered.

9
step.2 (probe muon selection)

• Reconstruct invariant mass(Minv) of triggered
  muon and another muon.
• When Minv Mz(91.19GeV) is less than 10GeV,
  the muon pair is regarded to be from Z. -gt probe
  muon is found.

• When more than one pair are found,
• the pair with Minv nearest to Mz is chosen.

Minv of two muons

• Selection efficiency is about 50 .
• Fraction of events with more than 1 pair is less
  than 0.01.

10
Step.3(trigger judgment)

• Matching scheme is the same as step.1.
• Performed only for endcap region. (1.05lthlt2.4)
• For 20GeV muons, 99.9 of triggered muons are
  included within 0.1 of dR.

Selection efficiency
(40pb-1)
11
Results
12
Pt vs Trigger efficiency
Fitting results
(40pb-1)
13
Validation
Comparison with the results using single muon
samples

• In trigger efficiency calculation using single
  muon samples, there is no matching ambiguity and
  reconstruction bias.
• By comparing with the results of single muon
  samples, it can be seen if the method is without
  bias or not.
• Comparison is performed at pt6GeV, 20GeV, and
  40GeV.
• Eta distribution of muon is different in Z(mm)
  and single muon samples, and trigger efficiency
  has position dependency.
• To be compared with the results of single muon
  samples, the results of Z(mm) samples are scaled
  to have flat eta distribution.

14
Results of Comparison
dR0.1 is not sufficient for low pt muons. -gt
other matching scheme is needed.
Pt vs Trigger efficiency for MU06
Pt vs Trigger efficiency for MU20
Results of single muon
Results of single muon
15
Summary table of the comparison

• The table shows that the results of Z(mm) and
  single muon samples are consistent with lt 1 at
  20GeV and 40GeV.
• -gt It comes to be sure that the method used in
  this study is unbiased and promising for high Pt
  region.

16
Position dependencyof trigger efficiency
17
Eta dependency of trigger efficiency
Only muons with pt gt 30GeV(at plateau) are used
for both results. Calculate only for endcap
region(1.05lthlt2.4)
Eta vs trigger efficiency
Eta vs trigger efficiency

• MU06
• MU20

• charge h gt 0
• charge h lt 0

There is position dependency for trigger
efficiency.
Covered by RPC.
18
Trigger efficiency for each sector
19
TGC geometry
Geometry acceptance of TGC is 100.

• In simulation geometry, muon trigger system is
  divided to 8 octants.
• Octants are constructed completely same.
• Each octant is divided to 2 regions named Forward
  and Endcap.
• Forward/Endcap is divided to 3/6 sectors,
  respectively.
• For each sector, trigger efficiency is calculated.

octant
7
6
4
8
3
5
1
2
0
Endcap the red region 1.05ltetalt1.92 Forward
the green resion 1.92ltetalt2.4
20
Trigger efficiency at plateau for each sector
Use only muons with pt gt 30GeV
0.9
It will be recovered with improved coincidence
window.
0.8
Sector number
21
Luminosity vs Sigma/Efficiency

• MU06
• MU20

• Sigma/Efficiency is calculated
• for trigger efficiency at plateau.
• at luminosity of 10pb-1, 20pb-1, 50pb-1, 100pb-1
  by scaling the results to each point.

For both MU06 and MU20, trigger efficiency at
plateau can be measured with an error of about
0.1 with data of 100 pb-1.
22
Summary Outlook

• Evaluate LVL1 endcap muon trigger efficiency
  using Z(mm) samples and the method looks
  promising for high pt region.
• Result is consistent with those of single muon
  samples.
• Take into account HLT for event selection.
• Improve matching scheme rather than using dR
  matching at IP.
• For more detailed position dependence of trigger
  efficiency
• use track extrapolator and use dR on TGC
  chambers.
• use segments (hits) of LVL1 and reconstructed
  muons.(need ESD)
• Using J/Y(mm) for trigger efficiency in low pt
  region.
• Perform also for Barrel.

23
Backup
24
dR of several Pt
For high pt muons, more than 99.9 of muons are
included within 0.1 of dR. For low pt muons,
larger range of dR has to be applied.
dR
25
Trigger efficiency for single muon

• Results of these three ways are consistent.

1
2
3