TPC Design

1 HW

TPC Design

Howard Wieman

CCAST Student Lecture

2 HW

Topics

TPC what it does TPC How it works

» Electron drift in gas» Wire chamber/pad readout

3 HW

TPC

I will cover TPC technology, but examples will be for the STAR TPC at RHIC

The TPC is the main detector in the STAR experiment

STAR

BRAHMS

PHENIX

PHOBOS

4 HW

TPC what it does

Measures many of the charged particles radiating from the collision point» Measures momentum

vectors of the particles. This is done by electronically recording high resolution space points along the particle path as it curves through a magnetic field

» Determines the particle species from the density of ionization left by the particle as they travel through gas

5 HW

TPC – How it works – 3D from 2D The TPC is basically an empty volume of gas

with an electric field High velocity particles leave a trail of

ionization (electrons and positive ions) as they pass through the gas

The particles are detected using the released electrons they leave in their path

The electrons drift in through the gas driven by the electric field (typical drift velocities – centimeters per s)

The drifting electrons are detected at the end cap with a 2D array of detectors

Two dimensions of the path are obtained by the 2D pattern traced on the end cap

The third dimension is obtained from the arrival time, the time it takes the electrons to drift from the track creation point to the end cap detectors

The start time is know from the accelerator bunch timing or from fast detectors such as scintillators which detect the fast particles thus identifying the time of interaction

The end time is determined by the end cap array of detectors

The positive ions drift much slower ( ½ sec) than the electrons (10s of sec)

» The positive ions are not detected but they do affect TPC performance – a subject to be addressed later in this talk

Path of ionization left by a particle radiating from the collisionElectrons greenPositive ions red

E

6 HW

Topics to be covered xxxxx

overview» 3D tracking with dE/dx for particle ID

generating the signal track point reconstruction limits on momentum resolution

» multiple scattering » accuracy of track reconstruction

– readout resolution– distortions in the electron drift path

7 HW

More topicsxxxxxx

particle identification with dE/dx» limitations due to ionization fluctuations

front-end low noise electronics

8 HW

The best way to learn about TPCs and other gas based detectors

9 HW

An excelent reference on signal generation and readout elelectronics

Ann. Rev. Nucl. Part. Sci. 1988. 38: 217

10 HW

TPC Capability

Tracking, momentum reconstruction for 4000 charged particles in 0<||<1.8p/p = 1.5% low p, p/p = 3% at 10 GeV/c

Particle ID by dE/dx, 6.7%

11 HW

Why we need a TPC

to get particle momentum and particle identification in a high track density environment one needs to over sample, i.e. lots of 3D pixels

the TPC provides by far the lowest cost per pixel of any detector

it does this by recording 3D space with 2D hardware

12 HW

STAR TPC

13 HW

Outer and Inner sector

14 HW

Sector Wire Geometry

15 HW

gas gain amplification at the anode wire on the sector

avalanche in the high fieldregion near theanode wire

Er

2

1

0

r m10

E = ~200 kV/cm

16 HW

current induced in the wire as the avalanche positive ions drift away from

the wire

I tq

a b t t

ln

1

0

I(t) q(t)

the “one over t tail”

17 HW

charge induced on the pad plane surface by a line charge at the wire

location

a good measureof the true surfacecharge generated by the avalanchepositive ions

x

D x D

2

1

cosh

+

x

D

18 HW

pad signal as a function of distance between avalance and pad center

(pad response function)

0

x

ex

2

22

19 HW

extracting track position

the avalanche position is extracted assuming a gaussian pad response function if the cluster is narrow (~3 pads wide)» i.e. the track is close to perpendicular to

the wires ( = ~0) a weighted mean is used for wider

clusters where is larger

20 HW

sources of error in the hit position determination

electron cloud width due to diffusion while drifting

tan() effect due to clustering or non uniform deposition of charge along the track

EXB term in the drift velocity as the electrons approach the anode wire

electronic noise of the amplifier Total r direction error in STAR

TPC:

z zm

cmT

215

210 31cm mm .

20:1 signal/noise center pad

~ 500 m

21 HW

gating grid

reduces drift distorting space charge by preventing avalanche positive ions from reaching the drift volume

reduces wire aging by preventing electrons from non-trigger events from reaching the MWPC

symmetric + - on alternate wires largely prevents induced signals on the wires and pads

22 HW

measuring momentum

tracking in a magnetic field

sources of error» multiple coulomb

scattering» errors in hit

reconstruction» global distortion in

the electron drift path

23 HW

Less fun