EMU Trigger Simulations
D.Acosta, S.M.Wang
University of Florida
EMU Meeting
Feb 1999
Goal: Study reconstruction of Pt of muons in the End Cap
chambers at the trigger level, to help in the design of
the End Cap Track-Finder.
Simulation:
use CMSIM 114
{ mf.tz has wrong strip staggering !
It was corrected for this study
{ For CMSIM 115,
staggering was not xed
ME4 is not in the geometry le
MB chambers have similar eciency as in
CMSIM 114
Produced single muon events at various Pt, in
0:9 < < 2:4 range with no backgrounds
Cathode strips and anode wires that carry hits caused
by traversing muon are used to reconstruct the trigger primitives (high Pt patterns only) (refer to CMS
TN/96-69)
Study the dependence of on and Pt
: 12 = 1 , 2
13 = 1 , 3
23 = 2 , 3
vs. gen at dierent Pt
Extrapolation
5
2.5
0
ME2-ME3
Φ2-Φ3 (degree)
7.5
7.5
5
2.5
0
(GeV)
1
3
0
-1
-2.5
1
2
1
ηgen
Φ1-Φ3 (degree)
3
2
1
2
1
ηgen
3
Φ2-Φ3 (degree)
-2.5
Φ1-Φ2 (degree)
ME1-ME3
Φ1-Φ3 (degree)
Φ1-Φ2 (degree)
ME1-ME2
Pt
2
1
2
ηgen
0.4
10
0.2
0
2
0
ηgen
0.4
0.3
0.2
0.1
0
1
2
0.4
0.3
0.2
0.1
0
1
2
ηgen
1
ηgen
Φ2-Φ3 (degree)
1
Φ1-Φ3 (degree)
Φ1-Φ2 (degree)
0
2
ηgen
0.4
0.3
100
0.2
0.1
0
1
2
ηgen
1
\Error" bars are the RMS of the spread
Max < 9 5 + 2 , 2.
< 15 ) drop 2 MSB from 60 range in
2
ηgen
Track-Finder
Jump in at gen 1:6 for ME1-ME2 and ME1ME3 is due to ME1/1 and ME1/A being closer to IP
compare to ME1/2 and ME1/3
Parameterize vs. Pt and Fit to = A()=Pt
Invert relation to obtain Pt
1:80 < < 1:95
1:95 < < 2:10
2:10 < < 2:25
2:25 < < 2:40
1 , 1
Ptrec Ptgen Distributions
150
2.35-2.40
Events
1.70-1.75
Events
Events
1.25-1.30
Pt
150
(GeV)
100
75
100
100
10
50
50
0
50
-0.2
0
0
0.2
25
-0.2
-1
-0.2
100
75
100
200
0.2
1/Ptrec-1/Ptgen (GeV )
Events
150
0
-1
1/Ptrec-1/Ptgen (GeV )
Events
Events
0
0.2
-1
1/Ptrec-1/Ptgen (GeV )
300
0
50
50
50
100
0
-0.1
0
0
0.1
25
-0.1
-1
200
150
100
100
50
0
0.1
-1
1/Ptrec-1/Ptgen (GeV )
Events
250
200
0
-0.1
1/Ptrec-1/Ptgen (GeV )
Events
Events
300
0
0.1
-1
1/Ptrec-1/Ptgen (GeV )
400
0
150
100
100
50
-0.1
0
0.1
-1
1/Ptrec-1/Ptgen (GeV )
0
-0.1
0
0.1
-1
1/Ptrec-1/Ptgen (GeV )
0
-0.1
0
0.1
-1
1/Ptrec-1/Ptgen (GeV )
Pt1rec , Pt1gen for single muon events, with no background
Distributions are Gaussian. No signicant tails
σ( 1/Ptrec - 1/Ptgen )/( 1/Ptgen )
Resolution of 1/Pt as function of 1
0.9
Pt = 10 GeV
Pt = 50 GeV
Pt = 100 GeV
0.8
0.7
ME1/3
ME1/2
ME1/1
MB1
0.6
0.5
0.4
0.3
0.2
0.1
MB1 / ME1 / ME2
0
0.8
1
1.2
1.4
1.6
1.8
2
2.2
2.4
ηrec
Ptrec obtained from measured between MB1-ME1
(0:9 < < 1:2), and ME1-ME2 (1:2 < < 2:4).
Resolution 30% at low Pt
Resolution in 2:2 < < 2:4 may be poorer since in
CMSIM 114 there are 65 cathode strips in ME1/A
instead of 48
Expected to be improved as Track-Finder design evolves
σ( 1/Ptrec - 1/Ptgen )/( 1/Ptgen )
1/Pt Resolution with and without
Measurements in MB1 or ME1
1.4
ME2 - ME3
ME1/3 - ME2/2
ME1 - ME2
MB1 - ME1/3
Pt = 10 GeV
1.2
1
0.8
0.6
0.4
0.2
0
0.8
1
1.2
1.4
1.6
1.8
2
2.2
2.4
ηrec
1/Pt resolution 70% without MB1 in overlap
region, or without ME1 in endcap
Cannot satisfy single muon rate requirement without
ME1 in the End Cap region, or without MB1 in the
overlap region
Study of Single Muon Rate in CSC
Muon Flux dN/dPtdηdt (Hz/GeV)
Estimated Muon Flux
10 6
All µ
10 5
µ from π/K
10 4
µ from b/c
10 3
10 2
10
1
10
10
10
-1
-2
-3
1
10
10
2
PT (GeV/c)
Estimate muon ux at L = 1034 from QCD events
{ Parameterization based on Pythia (CMS Note 1997/096)
{ Includes =K which decay before calorimeter
Fold in probability to punch through calorimeter in
endcap region (determined from CMSIM)
Assume 100% chamber eciency
Trigger Eciency Curves
E.
10% resolution
HH
HHHj
6
Threshold
YHHH
HH
50% resolution
True Pt (GeV/c)
Assume Gaussian errors for 1/Pt resolution
Convolute muon ux with trigger eciency curve to
determine trigger rate
Muon Rate dN/dηdt (Hz)
Single Muon Trigger Rates in CSC
10 7
L = 1034 cm-2s-1
10 6
10 5
50%
10 4
40%
10 3
10 2
30%
10
1
10
10
20%
CSC resolution from CMSIM
-1
10% resolution
-2
1
PYTHIA6
10
10
2
Effective PT Threshold (GeV/c)
Single rate from Pythia, convoluted with eciency
curve
Threshold set for 90% eciency
Require rates < 1 kHz per unit rapidity
Not satised for Pt resolution worse than 30%
σ( 1/Ptrec - 1/Ptgen )/( 1/Ptgen )
Parameterize Resolution as Function of 1.4
ME1 - ME2
Pt = 10 GeV
1.2
1
0.8
0.6
0.4
0.2
0
0.8
1
1.2
1.4
1.6
1.8
2
For Pt = 10 GeV (end cap region only)
2.2
2.4
ηrec
23 vs. 12
∆Φ23 (deg)
Pt=3 GeV
Pt=5 GeV
Pt=7 GeV
Pt=10 GeV
5
4
3
2
1
0
-1
-2
-3
-4
-5
-2
1.6 < η < 2.0
0
2
4
∆Φ23 vs ∆Φ12
6
8
10
∆Φ12 (deg)
Using the relation between 23 and 12 to improve
resolution at low Pt
Eect of Misalignment of CSC
on Track-Finder
Estimate overall accuracy of station placement
3 mm (from Dick Loveless)
) relative misalignment between 2 stations < 6 mm
Simulate eect by misaligning ME1 by x = +3 , +8
mm
sin(jj) = sin(,jj)
x
R
j j x
R sin()
Dierent values of 1 at dierent locations in ME1
1 ! 01 = 1 + 1
δΦ1 (degree)
1
vs. 1 for x = +8 mm in ME1
0.5
0.4
ME 1/A
ME 1/1
ME 1/2
ME 1/3
0.3
0.2
0.1
0
-0.1
-0.2
∆x= +8 mm for ME1
-0.3
-0.4
-0.5
0
50
100
150
200
250
300
350
Φ1 (degree)
j 1 j is larger for the trigger primitives located in
the chambers closer to the beam pipe
1 has opposite signs for trigger primitives in
opposite halves of the station
Eect on the Reconstruction of 1/Pt
1/Pt obtained from measured between MB1-ME1
(0:9 < < 1:2), and ME1-ME2 (1:2 < < 2:4)
,1=Ptaligned
Dierence in 1/Pt = 1=Ptmisaligned
1=Ptaligned
200
0
-200
0
200
0
-200
200
ΦME1 (degree)
200
0
-200
0
200
0
-200
0
0
-200
200
ΦME1 (degree)
200
ΦME1 (degree)
200
0
-200
0
ΦME1 (degree)
200
0
200
ΦME1 (degree)
200
ΦME1 (degree)
200
0
0
Difference in 1/Ptrec (%)
200
-200
ΦME1 (degree)
Difference in 1/Ptrec (%)
0
0
Difference in 1/Ptrec (%)
-200
200
Difference in 1/Ptrec (%)
0
Difference in 1/Ptrec (%)
200
Difference in 1/Ptrec (%)
Difference in 1/Ptrec (%)
Difference in 1/Ptrec (%)
Difference in 1/Ptrec (%)
MB1-ME1-ME2 +8mm ME1 Pt=50 GeV
200
ΦME1 (degree)
200
0
-200
0
200
ΦME1 (degree)
0
200
-200
0
ΦME1 (degree)
200
0
-200
0
0
200
ΦME1 (degree)
200
200
0
-200
0
ΦME1 (degree)
200
0
-200
0
Difference in 1/Ptrec (%)
-200
200
200
ΦME1 (degree)
Difference in 1/Ptrec (%)
0
Difference in 1/Ptrec (%)
200
Difference in 1/Ptrec (%)
Difference in 1/Ptrec (%)
Difference in 1/Ptrec (%)
MB1-ME1-ME2 +8mm ME1 Pt=50 GeV
200
ΦME1 (degree)
200
0
-200
0
200
ΦME1 (degree)
For +3 mm :
j Dierence in 1/Pt j < 20% (for Pt = 10 GeV)
< 100% (for Pt = 50 GeV)
For +8 mm :
j Dierence in 1/Pt j < 50% (for Pt = 10 GeV)
< 200% (for Pt = 50 GeV)
Correction on due to Misalignment
Perform an average correction in each detector (e.g.
δΦ1 (degree)
ME1/A, ME1/1, ME1/2, ME1/3)
Perform corrections in 15 bins of (0:9 < < 2:4)
0.5
0.4
ME 1/A
ME 1/1
ME 1/2
ME 1/3
0.3
0.2
0.1
0
-0.1
-0.2
∆x= +8 mm for ME1
-0.3
Ave corr for each detector
-0.4
-0.5
0
50
100
150
200
250
300
350
Φ1 (degree)
-20
-40
0
200
20
0
-20
-40
0
20
0
-20
-40
0
ΦME1 (degree)
40
200
ΦME1 (degree)
Difference in 1/Ptrec (%)
0
40
200
40
20
0
-20
-40
0
40
20
0
-20
-40
0
ΦME1 (degree)
Difference in 1/Ptrec (%)
20
Difference in 1/Ptrec (%)
40
Difference in 1/Ptrec (%)
Difference in 1/Ptrec (%)
Difference in 1/Ptrec (%)
MB1-ME1-ME2 +8mm ME1 Pt=50 GeV Ave Corr/det
200
ΦME1 (degree)
200
ΦME1 (degree)
40
20
0
-20
-40
0
Average correction per detector
< 2%
{ For +3 mm j Di in 1/Pt j 200
ΦME1 (degree)
(for Pt = 10 GeV
resolution 30%)
< 10% (for Pt = 50 GeV
resolution 30 , 40%)
{ For +8 mm
j Di in 1/Pt j < 5% (for Pt = 10 GeV)
< 20% (for Pt = 50 GeV)
Corrections in 15 bins of (for +8 mm)
< 10% (for Pt = 50 GeV)
{ j Di in 1/Pt j Summary
Preliminary studies show that :
1/Pt resolution 30% for low Pt
Single muon trigger rate will be too high if resolution
worse than 30%, or without station ME1, and MB1
(for overlap)
Corrections of of the LCTs in small number of bins
is sucient to reduce the eect on the reconstruction
of 1/Pt due to misalignment (for MB1-ME1-ME2)
Further studies
Improve on the 1/Pt resolution for low Pt muons using
the relation between 12 and 23
Further studies on the station misalignments for ME2ME3 (... expect greater eect due to smaller 23),
and its eect on trigger rates and eciencies
Write Track-Finder simulation program to test the
Track Finding algorithm, its eciency and rejection
power on background
Use more realistic LCT simulation (from UCLA) in
CSC and DT (e.g. CMSIM 116)