Assembly Station

1
The Making of the BOL-0 _at_NIKHEF
NIKHEF ATLAS MUON group H. vdGraaf, F.Linde, G.
Massaro, M. Vreeswijk, P. Werneke

• Introduction
• Assembly Station
• Sag adjustment
• Gluing BOL-0
• Temperature studies
• Summary

2
ATLAS-Muon

• Muon system with air core toroid aims at 50um
  spatial (sagitta) precision in bending plane.
• Achieved by
• excellent alignment 20um
• precise drift tube chambers 20um (RMS)

West side has expansion length o 0.1m in Xras
s
3
ATLAS-Muon

• Alignment based on RASNIK

Relative measurement of three points

• Alignment scheme

In-Plane
NIKHEF will produce the Barrel Outer Large
chambers, 2mx5m, BOL (100) This talk BOL-0
Projective alignment
Chambers consist of spacer 2x3 drift tube
layers In-Plane system monitors internal chamber
deformations Projective alignment monitors
chamber to chamber movements
West side has expansion length o 0.1m in Xras
s
4
Assembly Station

• Granite table 6m x 2.5m

Situation mid 99
East Z
Up Y
North X

• Three bare Cross-Plates are positioned on the
  station, carried by sphere holders
• The combs which will support the tubes are also
  visible
• In the background wiring station, Quality
  Control setup(s)

West side has expansion length o 0.1m in Xras
s
5
Precision Mechanics

• intrinsic accuracy better than 10um

• Positioning combs and sphere holders
• Tools
• Laser optics (straightness combs),
• silicon sensor (to line up combs),
• Slof (height-meter),
• tilt-meters (check Torque),
• Balmonitor (dZ between multilayers)

The quantities which affect the wire positions
are checked and found to be precise within
10um Granite table has sag of 25um 10um light
on/off
West side has expansion length o 0.1m in Xras
s
6
Spacer

• Spacer monitored by RASAS

Situation mid 2000
No Flexo
In-plane lens
NIKHEF Comb
longbeam
RASAS tower
Xplate
In-plane mask
Frascati comb
Ear with sphere and RASAS-MASK
Sphere tower with changeable block to adapt to
tube layers
West side has expansion length o 0.1m in Xras
s
7
Sag Compensation

• Sag compensation has been used manually.
• Three pressures 2 x outer (practically equal)
  and inner Xplates.

Sagcompen-sation tower
Additional weight for safety
After consistent results we adjusted the sag to
obtain (almost) symmetric weight distribution
s
8
Sag Compensation

• Xplate sag before adjustment

Chamber up
Chamber (upside) down
Asymmetric sag Xplates, due to asymmetric weight
distribution during gluing spacer -gt Adjust
s
9
Sag Adjustment

• Xplate sag after adjustment

Chamber up
Chamber (upside) down
After sag adjustment we observe (almost)
symmetric sags for chamber up and down positions
s
10
Sag AdjustmentCompensation

Chamber before adjustement and compensation
Chamber after compensation
s
11
Assembly Station Spacer

• Summary
• Assembly Station alignment good for wire position
  to the 10um level.
• Asymmetric distribution long-beams adjusted
  successfully.
• Sag (compensation) of cross plates understood
  within 20um.
• Observed sag of granite table of about 30um

s
12
Tube wiring QC

• Wiring Machine (NIKHEF)

13
QC-Wire position
Drift Tube
Coil
EndPlug
Wire position Y,Zlt25um
Rejection 2
1 tube was found with Rgt100um. A destructive
check revealed no twister in endplug
s
14
QC-Wire tension

• Quality Control on drift tubes, specs
• Wire tension 350gr-5 (batch 275gr)

Rejection 5 (3). gt1 month after production,
creep of the order of 1gr over gt1 month
s
15
QC-Leak Rate

• Quality Control on drift tubes, specs
• Leak rate He (4 bar) 2.5 x 10-8 bl/s
• (official Ar(3 bar) 1.0 x
  10-8 )

10-1
Rejection 4
s
16
QC HV Check

• Quality Control on drift tubes, specs

• HV 3000V , current lt2.0nA

Overcurrent tubes
Rejection 4
s
17
Gluing BOL-0

• In the BOL-0 all tubes passed the QC. Different
  wire tensions are used.
• Wire tension for layer 1-4 350 gr (nominal)
• For layer 5 275 gr
• layer 6 275gr 5 tubes with 350gr (interesting
  check X-ray tomograph)

s
18
Gluing BOL-0

• Problems
• Nozzle mounting very critical
• Central glue ropes between tubes not
  uniform/symmetric

glue
Bigger central glue cilinder

• Solutions
• Reduce Glue machine X speed by two (now 10m/s)

• Double amount of glue 103 (instead of 106, 2011)
  for central ropes which is more fluid
• (keep nominal amount of glue for side ropes, 106)

Glue stop for tape
s
19
Gluing BOL-0

• Stability Xplates during gluing

Layer 1
Yrasnik2 x Sag
Layer 1 dSag7um
Layer 2
Start glue
Layer 3
Layer 2 dSag10um

• The sag of the xplates in layer 1 and 2 change in
  time. Glue crimp? Temperature?
• Layer 3,4,5,6 Sag stable

s
20
Gluing BOL-0

• RASAS monitors

Global Z

• Relatively large Z movement layer 6 (spacer
  floats on glue?)

Global Y (up)
Lever arm in stacking block

• Stability in Y appears very good ( for some
  layers the RASAS indicated problems which could
  be online corrected)

s
21
BOL-0
Thickness Multi-layers BOL0 has tape (50um)
between tubes at one side
Much work..
2 layers

• Tubes press glue away?
• Tubes deform?
• Affects precision praxial platforms!

3 layers
Tape seems to improve tube distance
s
22
BOL-0
Heat Studies (No flexoos in BOL-0)
The BOL-0 was covered with heat blankets,
producing 50W/m2
Inplane
Yrasnik 2 x sag
Gradient between MLs 2.5 to 4 oC Sag of Chamber
200um ? 50 to 80 um/oC acceptable Sag of
xplate 25um ? 6 tp 10 um/oC acceptable
s
23
Summary

• We have constructed BOL-0 at NIKHEF

• NEXT
• Chamber will be scanned at CERN in Xray tomograph
• Automise QC setups. Most notably leak rate HV
  (tubbies)
• If OK, Production starts April 1th, 2001
• Produce chamber/2weeks

s
24
Production

• Production of 100 BOL chambers will start April
  1th
• How many people are involved?

• People who do the work

• People who do the visual inspections

• And stuurlui

25
Assembly Station

• Accuracy combs and sphere holders

• The intrinsic accuracy of mechanics is good
• The granite table has a relatively large sag of
  about 25um. Additional effects of 10um with light
  on and off (temperature gradient)

• Positioning combs and sphere holders
• Tools
• Laser optics (straightness combs),
• silicon sensor (to line up combs),
• Slof (height-meter),
• tilt-meters (check Torque),
• Balmonitor (dZ between multilayers)

West side has expansion length o 0.1m in Xras
26
Assembly Station

• The Balmonitor is a Cross-Plate equipped with
  lenses. Each lens is combined with a fork on
  the combs to form a RASNIK system

• The Balmonitor measures the Z and Y difference
  between Sphere tower and Comb between North and
  South. (Z difference between Multilayers)
• Additionally the Balmonitor is used to check for
  height difference between sphere towers at the
  reference and non-reference side (West/East)

West side has expansion length o 0.1m in Xras
s
27
Assembly Station
West side has expansion length o 0.1m in Xras
28
Assembly Station
West side has expansion length o 0.1m in Xras
29
Gluing BOL-0
Layer 1
Layer 2
30
RASAS

• RASAS- stacking

The stacking has been checked using the RASAS
monitors. The bare spacer has been put into A,
A-, C and C- orientations for the 3 stacking
blocks The nominal steps are dY26.011 um and
dZ15.017 um After a fit of the Mask-orientations
(4), the residuals are shown below.
31
Assembly Station

• RASAS- block deviations

The fitted block deviations are within 10um as
expected from measurements with a 3D coordinate
machine.

• The expectation values of the masks are extracted
  from
• CCD angles from dedicated calibration, 0.1 mrad
• RASNIK measurements of relative angle from the
  MASK and CCD (0.1 mrad).

In mrad
32
Assembly Station

• RASAS -temperature stability

The Xplates are made off aluminum and expand
about 24um/C/m.
East side has expansion length of 2.4m in Xras
33
XPlate RASNIKs

• The calibration (zero level) of the Xplate
  rasniks can be extracted from

1) The RASNIK reading for the up and down
position of the bare Xplates before gluing the
spacer. Problem old data, components touched. 2)
The response for the up and down position (A,A-)
of the spacer. Problem large effects due to
large gravitational force, leading to possibly
large second order effects 3) The RASNIK response
for the up and down position of the airborne
spacer. Problemhysteres due to bad design of
crane beams.
West side has expansion length o 0.1m in Xras
34
XPlate RASNIKs

• calibration Xplate rasniks, by rotation of
  airborne (horizontal) spacer

This method yields consistent results compared to
up/down positions of the bare Xplates
West side has expansion length o 0.1m in Xras
35
In-Plane RASNIKs

• The calibration of the In-Plane (Y) is extracted
  from data in A, A-, C, C- positions.
• A model to account for asymmetric weight
  distribution over the XPs in the and --
  orientations has been used to fit the data.

Data of the four in-plane systems
Data
Although steps appear large and uncontrolled,
the data is reasonably well described by the
model (10 um)
Model
West side has expansion length o 0.1m in Xras
36
In-Plane RASNIKs

• Model Parameters

Uncertainties 10
Implications 1) Sag of Xplates as function of
additional weigth consistent with data.
FEA predicts slope of 0.56!!!!!
2) Sphere block in centre are off in height by
appr. 30um as confirmed by laserSi-sensor and
also qualitatively by optical alignment
tool!!!!! 3) In A position, when the spacer was
glued, longbeam weight on outer Xplates!!!!
West side has expansion length o 0.1m in Xras
37
Sag Compensation
38
Sag Compensation
0.4 bar
39
Sag Compensation

• Shape without sag compensation

40
Sag Compensation

• Shape with sag compensation

41
The road to BOL-0
42
Gluing BOL-0

• Layer 3

43
In-Plane RASNIKs

• The calibration of the In-Plane (Y) is extracted
  from data in A, A-, C, C- positions..

West side has expansion length o 0.1m in Xras
44
Gluing BOL-0

• Result for side (106) and central (103) ropes

Central rope
Central rope
Central rope
Central rope
Central rope, sometimes bad (stability glue unit?)
45
Hunting the Higgsby online B-tagging

• The Standard Model
• The Higgs particle recent results!!!!
• The D0 experiment
• The proposal
• Costs
• Conclusions

46
The Standard Model

• The Standard ModelRelativistic Quantum Field
  theory with Local Gauge SymmetrySU(2) x U(1)
• Higgs mechanism endows particles with mass, while
  maintaining Gauge Symmetry.
• The Higgs mechanism gives rise to a particle
  called the Higgs Boson, which has been elusive so
  far.

• Theoretical Mass constraints
• An upper limit is related to unitary and set by
  the Higgs self coupling.
• The stability of the Higgs potential provides a
  lower limit.
• Otherwise new physics has to become manifest at
  some scale L!!!!

Not allowed
Not allowed
Allowed
47
The Higgs Particle
The mass of the Higgs can be predicted by the
combination of precision measurements (from LEP)
which are sensitive to higher order Quantum
Corrections that involve a Higgs Boson.
b
b
Z
Z
Higgs
t
b
b
MHiggslt245 GeV
The yellow band is excluded by the direct
searches at LEP
48
The Higgs Particle
The LEP programme just has been extended by one
month (until November 2th). The LEP experiments
have reported an excess of events above other
standard model processes, compatible with a Higgs
boson of about 114 GeV!
Direct searches at LEP
Signature 4 jets (2 with B-meson)
Z -gt qq H -gt bb
Invariant mass of B quark jets. (in red
theoretical contributions for Mhiggs114GeV)
Deviation from other SM processes. (a handful
events) However, observed Xsection to high.
49
The Higgs Particle
Direct searches at LEP (ALEPH)
SignalBackground hypothesis Likelihood Ratio
(signal/background)

• The Higgs might be around the corner.
• LEP will most certainly not be granted another
  year (the future of CERN is chosen to be LHC from
  2005, which can do the job better . but,
  TEVATRON in 2001 gets the first shot)
• Optimistically, if LEP claims a 3-4 std
  discovery, it will be shared with TEVATRON. The
  TEVATRON (and later LHC) will have the great task
  to study what was discovered.

50
The Higgs Particle

• The Higgs might be around the corner.
• LEP will most certainly not be granted another
  year (the future of CERN is chosen to be LHC from
  2005, which can do the job better . but,
  TEVATRON in 2001 gets the first shot)
• Optimistically, if LEP claims a 3-4 std
  discovery, it will be shared with TEVATRON. The
  TEVATRON (and later LHC) will have the great task
  to study what was discovered.

This talk concentrate on the search for a low
mass Higgs (Mlt140 GeV) at D0 (TEVATRON)
51
Hunting the Higgsby online B-tagging

• The Standard Model
• The Higgs particle recent results!!!!
• The D0 experiment
• The proposal
• Costs
• Conclusions

52
Higgs Production

• Higgs production Cross Sections

The strong process gg --gt Higgs is the main
production channel. Electroweak (Associated)
Higgs production of Z or W important, because of
signature.
53
Higgs Decay

• Branching Ratio

• Up to a mass of 140 GeV the Higgs will
  predominantly decay into b quarks.
• Other decay modes require high luminosity (LHC)

54
Higgs Recent Results
The LEP programme just has been extended by one
month (until November 2th). The LEP experiments
have reported an excess of events above other
standard model processes, compatible with a Higgs
boson of about 114 GeV!
Direct searches at LEP
b
Signature 4 jets (2 with B-meson)
b
q
q
Invariant mass of two b quark jets. (in red
theoretical contributions for Mhiggs114GeV)
Deviation from other SM processes. (a handful
events) However, observed Xsection too high.
55
Higgs Selection
B meson travel several mm before they decay,
leading to a secondary vertex. The main enemy is
the enormous QCD ( a standard process)
background.
What if we can trigger Higgs events by the decay
into b-quarks?

• Conservative When a Higgs event is identified by
  a secondary vertex ? allows lower (missing)
  energy cuts ? more events survive for offline
  analysis. (also increases potential top-quark
  analysis)
• But if the algorithm is successfully it may even
  be possible to study the single Higgs channel.
  (only 1 efficiency gives already 10 Higgs/year)
• Furthermore, studying online b quark tagging
  provides excellent starting point for the final
  offline analysis. (also for TOP quark analysis)

56
Higgs Online Selection
Associated production of Higgs and Z or W is
relatively easy to trigger. For example when the
Z decay to neutrinos, the event can be recognized
by large missing energy (but low event rate and
efficiency).
Single Higgs production only accessible with
powerful b trigger, using tracking
information. The main enemy is the enormous QCD
( a standard process) background.
B meson
In the offline analysis all (low mass) Higgs
studies need to require B tagged jets. We
propose to introduce the B tagging already in the
online event selection.
57
Higgs rates

• What if we can trigger Higgs events by the decay
  in to b-quarks?
• Conservative When a Higgs event is identified by
  a secondary vertex ? lower (missing) energy cuts
  ? more event survive for offline analysis. (also
  TOP)
• But if the algorithm is successfully it may even
  be possible to study the single Higgs channel.
  (only 1 efficiency gives already 10 Higgs/year)
• Furthermore, studying online b quark tagging
  provides excellent starting point for the final
  offline analysis. (also for TOP quark analysis)

• Present expectation

58
Higgs rates

• Single Higgs production is usually not considered
  in performance studies (expected to be too
  difficult)
• Associated Production Need to trigger and cut on
  (missing) transverse energy and offline B
  tagging, leading to only a few surviving events

59
Higgs rates

• Single Higgs production is usually not considered
  in performance studies (expected to be to
  difficult)
• Studies of associated Higgs production, with Z or
  W decaying leptonically (34) or hadronically
  (66), show only a few surviving Higgs events,
  after trigger and cuts.
• All channels use cut on (missing) transverse
  energy.

What if we can trigger Higgs events by the decay
in to b-quarks?
60
The proposed research programme

• Summary
• The selection of top and Higgs events by B
  tagging is of crucial importance
• Proposal
• Develop an algorithm to select online the events
  with a displaced secondary vertex from a B meson
• Such an algorithm most likely uses an Artificial
  Neural Network technique.
• The algorithm can be implemented at the level 2
  global processing stage. The present hardware
  allows to add the needed processor.
• The analysis work can be performed at NIKHEF
  (UVA) in close collaboration with the already
  existing (dutch) D0 group.
• If successful (online secondary vertex tagging
  has not been attempted before), this approach
  will be transferred to the ATLAS experiment.

61
Costs
The trigger system of D0 is a multimillion dollar
project and has been developed over the last
years. The Netherlands (NIKHEF) is paying
yearly contributions to D0. Additionally, NIKHEF
made some hardware contributions. The project
proposed here, aims at an extension of the
already existing trigger system and therefore
requires only a moderate hardware investment.
62
Conclusions

• Driven by the exciting recent results from LEP,
  we concentrated on the low mass Higgs search
  using the D0 detector at the TEVATRON.
• The proposed research programme also aims at top
  quark studies.
• To achieve our goals, we propose to tag secondary
  secondary vertices to identify B mesons in the
  online event selection.
• The trigger system of D0 needs to be extended and
  a state of the art selection algorithm has to be
  developed.
• At a somewhat larger time-scale the approach will
  be transferred to the ATLAS experiment at the
  LHC.

63
FERMILAB (USA)
High Energy Physics in the Wild West!
Fermilab is situated near Chicago (USA). The
TEVATRON is four miles in circumference.
64
TEVATRON
Run IB
Run II RunIIB START PHYSICS
March 2001 2003


Energy 900
1000 1000 GeV
Bunches 6
36 36
Bunch length (rms) 0.60
0.43 0.18 m

Typical
Luminosity 1.6 1031 8.3 1031
2.0 1032 cm-2 sec-1 Integrated Luminosity
3.2 16.7 41.0
pb-1/week

Bunch Spacing 3500
396 132 nsec
Interactions/crossing 2.5
2.2 5.3 (_at_ 45 mb)

65
D0
From inside out Vertex Detector, Central
Tracker, Calorimeter Solenoid Muon chambers,
Toroid
66
ATLAS
67
D0
Vertex Detector (SMT) four barrel layers, 50um
pitch. L1L2 single sided, L2L4 double sided
(2deg stereo). Disk modules have 12 wedges,
double sided (30deg stereo). Central Fiber
Tracker (CFT) 8 concentric layers with fibers.
All doublets with 2 deg stereo angles. VLPC
readout. 72k channels. Solenoid 2.7 m long,
R60cm, E5MJ, B2T. 0.9Xo dPt/Pt0.002 Pt
(combined SMTCFT) Preshower detecor for em or
had particle id. Calorimeter Liquid argon, 15
inter. Lengths. Resolution for jets
80/Sqrt(E). Muons 3 layers. Toroid (iron) in
between L1,L2. Onne and Paul work on muon reco.
68
Tracking
Vertex Detector (SMT) four barrel layers, 50um
pitch. L1L2 single sided, L2L4 double sided
(2deg stereo). Disk modules have 12 wedges,
double sided (30deg stereo). Central Fiber
Tracker (CFT) 8 concentric layers with fibers.
All doublets with 2 deg stereo angles. VLPC
readout. 72k channels. dPt/Pt0.002 Pt
(combined SMTCFT)
69
D0
Once upon a time it all worked
RUN I, event display

• Top event from Run I

70
D0-Top
Top physics!
71
D0-Trigger
The colliding 1 TeV proton and antiproton beam
from TEVATRON lead to an enormous event rate
(millions/sec), leading to the need of fast
(efficient) online event selectiontrigger.
Detector. Collision rate 7MHz
Storage
L2 100ms
L1 4 ms
L3 48 ms
10 kHz
1000 Hz
lt50 Hz
128 terms
128 terms
48 nodes
Detector Preprocessors ( Silicon Track Trigger)
Global Processor
L2


The Global Processor is implemented as a fast
Alpha processor on a VME card
72
D0-trigger
Detector
L1 Trigger
L2 Trigger
7 MHz
5-10 kHz
L2Cal
CAL
FPS CPS
L2PS
L1 CTT
Global L2
L2CFT
CFT
L2STT
SMT
L2 Muon
L1 Muon
Muon
L2FWCombined objects (e, m, j)
L1FPD
FPD
L1FW towers, tracks, correlations
73
The Muon System and Higgs discovery in ATLAS
s x B 300fb ATLAS MH115GeV for 100fb-1( e or
m)
Difficult final state 6 jets Background tt
light quarks Need good B tagging ATLAS eb50
ec10 euds1
74
The Muon System and Higgs discovery in ATLAS
For higher masses MHgt120GeV,
H--gtZZ--gt4leptons becomes important. At MH120
Significance4 for 100fb-1 (need excellent mass
reconstruction) For MHgt2MZ, the 4 lepton channel
becomes absolutely gold-plated (60 events,
background free) PLOT???? For the high mass
Higgs gt150 GeV, also H--gtWW--gtlnln or associated
WH--gtWWW--gt lnln lnln or llnnjj (like sign
leptons) with Significance5 for
100fb-1 DIAGRAM
75
Higgs discovery in ATLAS
ATLAS for 100fb-1 (1100 events)
Expected to be the discovery channel for a low
mass Higgs (lt150 GeV) Background light quarks to
g g and fakes Another kind of interesting
events associated Higgs Production (ttH, WH, ZH
, qqH), with H--gt g g gives for 100fb-1
additional
These channels have smaller Xsection, but smaller
background, especially WH and ZH
Associated production WH and ZH with H--gtbb is
hopeless in ATLAS because of background. These
are the most promising channels at TEVATRON.
76
Higgs discovery at TEVATRON (D0)
Promising associated production
l or n
W, Z
l or n
b
H
b
Expected to be the discovery channel for a low
mass Higgs.
Need good B tagging eb57 ec20 euds1 Need
good Mass resolution, simulation predicts 15
(with 10 mass resolution background reduced by
two!!!)
77
Higgs discovery at TEVATRON (D0)
Associoated production in the mass range 130-190
GeV
l or n or q
W, Z
W,Z
H
W,Z
At higher mass the decay H--gtW,Z becomes important
78
LHC vs TEVATRON
_at_LHC decay channel H--gt g g and associated ttH
production promising at low mass. These channels
are out of reach for TEVATRON, simply because of
the low Xsection
_at_TEVATRON the associated production channels WH
and ZH are most promising. These channels are
out of reach for LHC because of overwhelming
(QCD) backgrounds. NOTE Studies by ATLAS
suggest that these channels may be out of reach
for TEVATRON as well. QCD background LO/NLO
predictions Mass reconstruction B tagging
79
Proposed Improvements for D0
Silicon Vertex Detector dies after 4fb-1 ---gt
replace it in shutdown 2003. Research Proposal
submitted NWO by MV includes the upgrade of the
level 2 trigger with an additional global
processor. This will allow to include vertex
information for B tagging, leading to higher
Higgs discovery potentials. Presently, we are
involved in offline SVTX finding using full
simulation and reconstruction packages.
80
Higgs Production and Decay
The strong process gg--gtHiggs is the main
production channel. Electroweak (Associated)
production of Z or W important, because of
signature.

• Up to a mass of 140 GeV the Higgs will
  predominantly decay into b quarks.
• Other decay modes require high luminosity (LHC)

81
Higgs Production and Decay
82
Higgs Production and Decay
83
Higgs Online Selection
Associated production of Higgs and Z or W is
relatively easy to trigger. For example when the
Z decay to neutrinos, the event can be recognized
by large missing energy (but low event rate and
efficiency).
Single Higgs production only accessible with
powerful b trigger, using tracking
information. The main enemy is the enormous QCD
( a standard process) background.
B meson
In the offline analysis all (low mass) Higgs
studies need to require B tagged jets. We
propose to introduce the B tagging already in the
online event selection.
84
Event rates
85
The vertex fit
The Principle
86
The vertex fit
A Closer Look
Where VVoe Assumption
Then
87
Study events

• SVTX finding algorithm

Start with (free) selected tracks Significancegt3
Make all possible 2-track vertices (vertex fits)
Keep/Kill vertices with shared tracks (based on
probability cuts)
Try to add tracks to vertices (based on
probability cuts)
Select (kill) good (bad) vertices (based on
probability cuts)
88
Study events

• Tracking

Significance wrt pv d/s
New propagator
Private X0 fix
89
Study events

• Track Selection

Track cuts remove PV tracks and demand
Significance wrt pv d/s gt3
Private X0 fix
New propagator
Private X0 fix
90
Study events

• Probability and Cuts

Start with selected tracks and make all possible
two-track vertices. Reject when 1ltDirectionlt3.
91
Study events

• Probability functions.

When required (for fine tuning) more variables
can be included in a simple way
92
Study events

• Resolution

Bad resolution on lifetime.
93
Results (Feb/2000)
Results for standard efficiency definition (for a
optimized cut on Significance)
Where dr is distance PV - SV
Results for efficiency definition 2 Including
all SV-track requirement.
94
Results
A tru SV is considered tagged as
assigned tracks are from tru SV. For gt2 track
vertex one wrongly assigned tru PV is allowed.
Files all gtr-mc-find by Maria Roco
Where dr is distance PV - SV
95
Study events
RECO
TRU
96
Study events
RECO
TRU