The LHCb VErtex LOcator Olaf Behrendt on behalf of the VELO group Novosibirsk, 29.02.08

The LHCb VErtex LOcator

Olaf Behrendt

on behalf of the VELO group

Novosibirsk, 29.02.08

The LHCb VErtex LOcator INSTR08Olaf Behrendt

• LHC experiment dedicated to heavy flavour physics• CP-violation and rare decays of beauty and charm hadrons• constrain unitarity triangles and search for new physics

• forward spectrometer: 1.9 < < 4.9 [bb-pairs produced correlated in space]

• luminosity: 2 · 1032 cm-2s-1 [1012 bb-pairs per year]

Detector requirements:

• reconstruction in harsh radiation environment• selective trigger system [bb-production cross-section small]

• good vertex resolution [proper time]

• good particle identification [K/ separation]

The LHCb Experiment

Bs

Btag

K-

K+

+

+,K+

p p

PV

~1 cm

SV


Role of the VELO in LHCb

Reconstruction Trigger

• precise tracking

• primary/secondaryvertices [lifetime]

• B-decay selection

• suppression of multiple interactions

• absoluteluminosity measurement

4 Tm


Outline

• Design

• Module Production and Assembly

• Installation

• Control and DAQ

• Expected Performance

• Summary

• VELO alignment: see talk of M. Gersabeck


Layout

• 2 retractable detector halves• 21 stations per half with an R and sensor• 2 Pile-Up stations per half [trigger]

injection

stablebeams

RF box

module

Pile-Up module

s

base

LHC vacuum

zx

y

x

y

6 cm

p

p partly overlappingsensors

IP


Sensors

• divided into quadrants

• pitch: 40 - 100 m

• divided into short and long strip region

• pitch: 35 - 100 m

• stereo angle

r = 0.8 cm

r = 4.2 cm

2048strips

-sensor R-sensor

typ

e

invert

ed

not

typ

e in

vert

ed after 1 year of irradiation

[2 fb-1]

R [cm]

Udep [V]

n+ in n-bulk sensors [300 m]

• inherently radiation tolerant

• strip isolation via p-spray

• expected radiation dose:- 1.3 · 1014 neq/cm2/year at r = 0.8 cm- 5 · 1012 neq/cm2/year at r = 4.2 cm

after 3-4 years [~8 fb-1]: run partially depleted


Modules

• double-sided hybrid to balance stresses due to “bi-metallic” effects

• typical non-planarities: 250 m

• sensor-sensor accuracy: < 10 m

• analogue front-end read-out:2 x 16 Beetle chips

• cooling: 2-phase CO2 [silicon @ -7° C]

carbon fibrepaddle

cooling block

carbonfibre

-sensor

R-side circuit

-side

circuit

R-sensor

thermal pyrolytic graphite

x

yz


Module Production

receive and validate components

hybrid assembly

wire bonding on the hybrid

glue sensors on hybrid

wire bonding of sensors to hybrid

glue to carbon-fibre paddle

electrical check

module metrology

visual re-inspection

module burn-in- electrical test in vacuum- thermal stress- electronics burn-in

I/I (

%)

# module

good stability during burn-in

rejected

bonding failure rate:

• Backend: 0.000 %• Frontend: 0.010

%• Sensorend: 0.002

%

bad channels:

• R-sensors: 0.7 %-sensors: 0.5 %


Fully Assembled VELO Half


Vacuum System

LEFT DETECTOR BOX

• silicon detectors operated in vacuum

• RF shield of 300 m Al [3% Mg]

- constitutes beampipe in VELO region

- shape allows for overlapping sensors [alignment]

• avoid deformations:P = Pbeam - Pdetector < 10 mbar

• requirements beam vacuum: Pbeam = 10-8 mbar

• expected VELO vacuum:Pdetector < 10-4 mbar

This is what the LHC beams see …


Installation

Lowering into the LHCb cavern …

… ready for installation …

… insertion of one half

October 2007:

• installation of both VELO halves

• transport and installation without any damage [IV curves]

installation rail

RF box


Pumpdown [

mb

ar]

+2 mbar

-5 mbar

after 100 hoursdetector: 9 · 10-5 mbar

beam @ VELO: 2 · 10-9 mbar [without ion

pumps]

absolute pressure

differential pressure


DAQ and Control System


ARX

ARX

ARX

ARX GBE

TTCRx

CC

PC

• originally developed for the VELO, now used for (almost) all of LHCb

• analog input from 64 links• 10 bit analog to digital

conversion @ 40 MHz [A-Rx]• pre-processing FPGAs

[cross-talk, common mode, pedestals, clustering]

• data sent out via GigabitEthernet [GBE]

• max. L0 output rate: 1.1 MHz• control via credit-card

sized PC [CCPC]

Readout Board (TELL1)


Commissioning

installation of cables and boards

Tests:

• ADC cards [A-Rx]

• Readout Boards [TELL1]

• Control Boards

• each readout slice– timing– cable check– noise level

Control and DAQ system (almost) completely commissioned

adjustment of timing to 0.65

nsproblematic A-Rx

card


Testbeam

• 10 modules• 180 GeV beam

• small scale CO2 cooling system

• 6 modules readout simultaneously[full readout chain with final electronics boards]

• software: DAQ, ECS, tracking, vertexing, online monitoring

x = 0 mm, d = 2 mm

x = 15 mm, d = 5 mm

• data for 0, 4 and 8 degrees• interaction data

- 4 layers with 2 Pb targets [300 m]- test vertex reconstruction- emulate open and closed VELO

• > 50 million events to disk

November 2006

y

z

x


binaryresolution

linear fitto data

Testbeam Results

-sensors

R-sensors

• S/N: 20 - 24[decreases with r due to increasing strip length]

• resolution: 9 - 25 m [pitch: 40 - 100 m]

• S/N: 24 - 29

• resolution: 9 - 20 m [pitch: 35 - 100 m]

Landau Gaussian

active region

R-sensors

expected resolution improvements: cross-talk and corrections

R-sensors

full sensorsimulation


Expected Tracking Performance

reconstructed track - hits in 3 stations

LHCb acceptance: 1.9 < <

4.9

= -ln tan(/2)

For typical B decay modes:

• primary vertex resolution:- x,y: ~10 m, z: ~60 m

• proper time res.: ~40 fs• B mass res.: 12 - 25 MeV

B decays


Status

• module production and burn-in: completed March 2007

• module assembly on detector halves: completed March 2007

• vacuum/detector positioning system: installed

• installation of detector halves:October 2007

• cooling system: installed

• HV-/LV-system: installed

• Electronic boards/cables: installed

• system checkout: ongoing

• system test foreseen within the next two months

CO2 cooling system

positioning system

LV system


• VErtex LOcator is small but complex detector

• precision tracking very close to the interaction region

• radiation tolerant design

• results from the testbeam in November 2006

– S/N: 24 - 29 (), 20 - 24 (R)

– resolution: 9 - 20 m (), 9 - 25 m (R)

• on schedule for first beams in 2008

Summary

Backup Slides


Luminosity Measurement

10-7 mbar Xe1011 p/bunch

30

Hz

per

bu

nch

Method:

• injection of a tiny bit of gas into VELO region

• reconstruction of beam-gas interaction vertices- beam angles, profiles & relative positions- calculate luminosity

• simultaneous reconstruction of beam-beam interaction vertices- calibrate ‘reference’ cross-section

test setup for gas injection

• expected beam size: x ≈ y ≈ 70 m

• expected vertex resolution: pv ≈ 30 m

expected accuracy: ~1%

LHCbmachine

Documents

The LHCb VErtex LOcator Olaf Behrendt on behalf of the VELO group Novosibirsk, 29.02.08