Performance of L1 CSC TriggerPerformance of L1 CSC Trigger(and strange geometrical effects)(and strange geometrical effects)
D. AcostaUniversity of Florida
PRS/Mu Meeting, July 2, 2002 Darin Acosta2
Eta Eta = 1.6 Region= 1.6 RegionAs reported by Hannes, a large number of high quality
muons reported by L1 CSC trigger originate from PT1 sample Contribute ~1 kHz to L1 rate
1 1.2 1.4 1.6 1.8 2 2.2 2.40
5
10
15
20
25
30
htempNent = 63 Mean = 1.671RMS = 0.3191
abs(Etac[0]) {Ncsc>0&&Qualc[0]>2&&Ptc[0]>10} htempNent = 63 Mean = 1.671RMS = 0.3191 50Kevt
PRS/Mu Meeting, July 2, 2002 Darin Acosta3
ηη Distribution of ME SegmentsDistribution of ME Segments
55 60 65 70 75 80 85 90 95 1000
20
40
60
80
100
120
140
160
180
200
220
htempEntries 5937Mean 76.73RMS 11.34
stubeta[0] {nstub>0&&station[0]==1&&Cscid[0]<3} htempEntries 5937Mean 76.73RMS 11.34
25 30 35 40 45 50 55 600
20
40
60
80
100
120
140
160
180
htempEntries 3920Mean 41.49RMS 9.393
stubeta[0] {nstub>0&&station[0]==1&&Cscid[0]<6&&Cscid[0]>2} htempEntries 3920Mean 41.49RMS 9.393
ME1/1
ME1/2ME1/1 and ME 1/2 overlap in ηsomewhat: 1.5875–1.6625
Creates ambiguity as to which station to use in PT assignment
Also region of detector cracks
PRS/Mu Meeting, July 2, 2002 Darin Acosta4
ηη Distribution after Boundary CreationDistribution after Boundary Creation
55 60 65 70 75 80 85 90 950
20
40
60
80
100
120
140
160
180
200
220
htempEntries 5937
Mean 76.76
RMS 11.3
stubeta[0] {nstub>0&&station[0]==1&&Cscid[0]<3} htempEntries 5937
Mean 76.76
RMS 11.3
25 30 35 40 45 50 550
20
40
60
80
100
120
140
160
180
200
220
htempEntries 3920
Mean 41.42
RMS 9.279
stubeta[0] {nstub>0&&station[0]==1&&Cscid[0]>2&&Cscid[0]<6} htempEntries 3920
Mean 41.42
RMS 9.279
ME1/1
ME1/2
undfl
ovflw
One way to solve ambiguity is to define the η bins so that there is no overlap (even though there is). Define an upper η bin for the ME1/2 chamber and a lower η bin for the ME1/1 chamber (overflow and underflow bins). The deviation from the true η is only 2 bins (0.025 units), which does not affect the L1 CSC Track-Finder efficiency.
Avoids adding more memory chips on board
PRS/Mu Meeting, July 2, 2002 Darin Acosta5
1/P1/PTT ResolutionResolution
-1 -0.5 0 0.5 10
2
4
6
8
10
12
14
16
18
20
22
hPtResEntries 310Mean -0.07004RMS 0.3146
/ ndf 2χ 54.32 / 37Prob 0.03293Constant 1.544± 15.63 Mean 0.01664± -0.0325 Sigma 0.01923± 0.2495
Ptgen/Pt-1 hPtResEntries 310Mean -0.07004RMS 0.3146
/ ndf 2χ 54.32 / 37Prob 0.03293Constant 1.544± 15.63 Mean 0.01664± -0.0325 Sigma 0.01923± 0.2495
-1 -0.5 0 0.5 10
5
10
15
20
25
hPtResNent = 310 Mean = -0.008793RMS = 0.2597Chi2 / ndf = 34.3 / 34Prob = 0.4533
1.637 ±Constant = 19.75 0.01439 ±Mean = -0.01955
0.01258 ±Sigma = 0.2169
Ptgen/Pt-1 hPtResNent = 310 Mean = -0.008793RMS = 0.2597Chi2 / ndf = 34.3 / 34Prob = 0.4533
1.637 ±Constant = 19.75 0.01439 ±Mean = -0.01955
0.01258 ±Sigma = 0.2169
Before:PT often overestimated when wrong ME1 chamber was assumed
(less bending between ME1/2 and ME2, so looks like high pTif you assume ME1/1 to ME2)
After:With clean η boundary, tail removed and PTresolution improved
PRS/Mu Meeting, July 2, 2002 Darin Acosta6
PT1 CandidatesPT1 Candidates
1 1.2 1.4 1.6 1.8 20
5
10
15
20
25
30
htempNent = 53 Mean = 1.544RMS = 0.216
abs(tracketavalue[0]) {ntrack>0&&trackquality[0]>2&&ptvalue[0]>10} htempNent = 53 Mean = 1.544RMS = 0.216
1 1.2 1.4 1.6 1.8 20
1
2
3
4
5
6
htempNent = 13 Mean = 1.235RMS = 0.3243
abs(tracketavalue[0]) {ntrack>0&&trackquality[0]>2&&ptvalue[0]>10} htempNent = 13 Mean = 1.235RMS = 0.3243
Before
After:low PT muons removed !
PRS/Mu Meeting, July 2, 2002 Darin Acosta7
Why ?Why ?A little strange that we see a large impact from this effect now, because the ambiguity has been there all along in previous productions
More punch-through in latest production ?Larger cracks in geometry ?
PRS/Mu Meeting, July 2, 2002 Darin Acosta8
0 1 2 3 4 5 60
2
4
6
8
10
12
14
16
htempNent = 37 Mean = 3.412RMS = 1.564
abs(trackphivalue[0]) {ntrack>0&&trackquality[0]>2&&ptvalue[0]>10&&tracketavalue[0]>=1.6&&tracketavalue[0]<1.7} htempNent = 37 Mean = 3.412RMS = 1.564
0 1 2 3 4 5 60
2
4
6
8
10
12
14
16
18
20
htempNent = 53 Mean = 3.41RMS = 1.546
abs(trackphivalue[0]) {ntrack>0&&trackquality[0]>2&&ptvalue[0]>10&&tracketavalue[0]>-1.7&&tracketavalue[0]<=-1.6} htempNent = 53 Mean = 3.41RMS = 1.546
Why spikes in φ for the η=1.6 triggers before fix? Or why the asymmetry for that matter ?
Not aligned on any sector boundary
About 1/3 are associated with generated muons with PT ~ 3 GeV (lower limit of ntuple)
About 2/3 of these CSC triggers have RPC confirmation
+η Endcap
+η Endcap
φφ Distribution of Distribution of ηη=1.6 Triggers=1.6 Triggers
PRS/Mu Meeting, July 2, 2002 Darin Acosta9
Distribution of RPC TriggersDistribution of RPC Triggers
0 1 2 3 4 5 60
1
2
3
4
5
6
7
8
htempEntries 68Mean 2.858RMS 1.711
Phir[0] {Nrpc>0&&Ptr[0]>5&&(Etar[0])>=1.5&&(Etar[0])<1.7} htempEntries 68Mean 2.858RMS 1.711
0 1 2 3 4 5 60
1
2
3
4
5
htempEntries 64Mean 2.993RMS 1.831
Phir[0] {Nrpc>0&&Ptr[0]>5&&(Etar[0])<=-1.5&&(Etar[0])>-1.7} htempEntries 64Mean 2.993RMS 1.831
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.20
10
20
30
40
50
60
70
80
htempEntries 467Mean 1.344RMS 0.5665
abs(Etar[0]) {Nrpc>0&&Ptr[0]>5} htempEntries 467Mean 1.344RMS 0.5665
Maximum triggers for η=1.6
φ Distribution shows spikes
PRS/Mu Meeting, July 2, 2002 Darin Acosta10
Correlation with E/Gamma ?Correlation with E/Gamma ?Chris Seez:
“We deliberately remove trigger tower 18 (1.479 < eta < 1.566)from our fiducial region for precision e/gamma work. This region is strongly shadowed by tracker cables exiting through the barrel/endcap 'corner' (at eta~=1.479). On the high side of the corner there are a number of "missing" crystals - but the location of these small "holes" in the material is/should be 4-fold symmetric (each quadrant of theendcap is identical).”
Possible correlation, or just accidental coincidence ?
Anyway, latest CSC simulation removes low pT triggers (but need to keep an eye on this region if punch-through is large)
Barrel-Endcap crack & tracker cables
η
Trig. Tower 18
PRS/Mu Meeting, July 2, 2002 Darin Acosta11
Other Changes to L1 CSC SimulationOther Changes to L1 CSC SimulationThe addition of the η boundary for the CSC trigger is implemented in the head of the L1CSCTrigger and L1CSCTrackFinder packages
The ORCA_6_1_1 tagged release is several months old
Many other changes have been made to these packages to bring them up-to-date with prototypes currently being built and tested
New anode trigger logic New Sector Receiver look-up table schemeAdditional features in Sector Processor logic
But a few more changes to the default settings need to go in before general release
Also to L1DTTrackFinder because of an interface change
PRS/Mu Meeting, July 2, 2002 Darin Acosta12
CSC Efficiency with Latest CodeCSC Efficiency with Latest Code
Pt (GeV/c)0 20 40 60 80 100
Eff
icie
ncy
0
0.2
0.4
0.6
0.8
1
hEffPt0Entries 27
Mean 51.66
RMS 28.03
Efficiency vs. Pt hEffPt0Entries 27
Mean 51.66
RMS 28.03
1 1.2 1.4 1.6 1.8 2 2.2 2.40
0.2
0.4
0.6
0.8
1
Efficiency vs. Eta
Basically, as good as it ever has been
Need to check for fake di-muon triggers, however