endcap modules for the atlas semiconductor trackersct/welcomeaux/papers/florenz... ·...

Endcap Modules for the ATLASSemiConductor Tracker

RD03, Firenze, September 29 th, 2003

Richard Nisius (MPI Munich)

[email protected]

(For the ATLAS-SCT Collaboration)

http://atlas.web.cern.ch/Atlas/GROUPS/INNER_DETECTOR/SCT/

http://mppmu.mpg.de/~nisius

http://www.mppmu.mpg.de/

mailto:[email protected]

The plan of this presentation

Introduction

The forward modules

Mechanical performance

Electrical performance

Test beam results

Status of module production

Conclusions and Outlook

Endcap Modules for the ATLAS SemiConductor Tracker Firenze September 29th, 2003 Richard Nisius

The ATLAS inner detector

»»»»»»»»»:

»»»»»»»»»9

AAK

AAU

2.3 m

7 m

The Pixel Detector− Radius 4.8 − 16 cm.− 3 layers, 10 disks.− 1.4 · 108 read-out channels.− σ: 12 µm (RΦ) and ≈ 70 µm (z/R).

The Semi Conductor Tracker− Radius 27 − 52 cm.− 4 layers, 18 disks− 6.3 · 106 read-out channels.− 4088 modules, 61 m2 silicon− σ: 16 µm (RΦ) and 580 µm (z/R).

The Transition Radiation Tracker− Radius 56 − 107 cm.− 100 k / 320 k straws in barrel / end cap.− 420 k read-out channels.− Xe radiator for electron-detection.− σ: 170 µm / per straw.

In this presentation, only the SemiConductor Tracker will be discussed.


http://atlas.web.cern.ch/Atlas/GROUPS/INNER_DETECTOR/inner_detector.html

The SCT forward part - how to build a module

Middle module

Support-structure (Spine)Thermal Pyrolytic Graphitethickness 500 µm, 1700 W/m/K

@@

@@I

¾¡

¡¡

¡ª

FixationsPrecision 20 µm

³³³³³³³1

@@

@@

@@

@@R

Si-wafersca. 6x6x0.285 cm3

768 single-sided p-typestrips on n-type substrate with50-90 µm pitch

»»»»»»»»»»:

XXXXXXXXXXz

HybridA 6-layer copper-polyimidflex on a carbon substrate

@@RRead-out chips6 ABCD3T chips per sidewith binary read-out

©©©©©¼Fan-ins4/module, glass with Al-stripsof 300 µm thickness

¾

Outer moduleInner module


SCT forward modules - some design values

¾ -16 cm

6

?

8 cm

− Bias voltage: < 500 V.

− Produced heat: < 7 W per module.

− Gain: 50 mV/fC.

− Linearity: better than 5%.

− Peaking time: 20 ns.

− Signal charge: 3.3 fC, S/N > 10.

− Noise: < 1500(1900) e− ENC before (after) irradiation.

− Noise occupancy: < 5 · 10−4 .

− Hit efficiency: > 99%.

− Time walk: ≤ 16 ns, and bunch crossing resolution better than 99%.

− σ = 16(580) µm ⊥ (‖) to the strips.

− Two-track resolution: 200 µm.

The SCT modules are very delicate objects that have to be built with great care.


http://www.mppmu.mpg.de/~sct

The mechanical reproducibility

Results from 28 pre-series modules ( - - - tolerances)

0

2

4

6

8

10

12

14

-0.02 -0.01 0 0.01 0.02

∆detector distance / mm

N

0

2

4

6

8

10

-0.2 0 0.2

∆stereo angle / mrad

N

02.5

57.510

12.515

17.520

22.5

-0.02 -0.01 0 0.01 0.02

∆X-centre / mm

N

0

1

2

3

4

5

6

7

-0.01 -0.005 0 0.005 0.01

∆y-centre / mm

N

-

6

y

x

The modules can be built within specifications with high yield.


An example of the electrical performance

Channel Number0 100 200 300 400 500 600 700

Vt5

0 (m

V

0

50

100

150

200

250Vt50 at 2.00fC Nent = 762 Vt50 at 2.00fC Nent = 762

Channel Number0 100 200 300 400 500 600 700

Gai

n (

mV

/fC

)

20

25

30

35

40

45

50

55

60

65

70Gain at 2.00fC Nent = 762 Gain at 2.00fC Nent = 762

Channel Number0 100 200 300 400 500 600 700

Ext

rap

ola

ted

Off

set

(mV

)

0

20

40

60

80

100

120

140

Extrapolated Offset Nent = 762 Extrapolated Offset Nent = 762

Channel Number0 100 200 300 400 500 600 700

Inp

ut

No

ise

(EN

C)

500

1000

1500

2000

2500

3000

Input Noise at 2.00fC Nent = 762 Input Noise at 2.00fC Nent = 762

0 . . . . . . . . . 768

− Test of a non perfect hybrid (dead

channels are visible as black lines).

Shown are the results for one side of

the hybrid (6 × 128 = 768 channels),

tested at 2 fC injected charge.

− The results are obtained by analyzing

the S-curves from threshold scans.

AA

AA

AA

AAAK

The output signal.

150 mV

@@

@@

@@I

The calculated gain.

55 mV/fC

XXXXyThe extrapolated offset for 0 fC.

40 mV

´´

´´́+

The noise in e−ENC.1500 e−ENC

The channel to channel varia-tions within a module are small.


Some properties of non-irradiated modules

Chip number0 2 4 6 8 10

No

ise

[e-

EN

C]

0

500

1000

1500

2000

2500

3000

〈Noise〉 = 1450 e− (Outer)1350 e− (Middle)1000 e− (Inner)

Chip number0 2 4 6 8 10

Gai

n [

mV

/fC

]

0

20

40

60

80

100

〈Gain〉 = 50 mV/fC

Efficiency vs. Noise Occupancy

Threshold (fC)0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5

Eff

icie

ncy

0.9

0.91

0.92

0.93

0.94

0.95

0.96

0.97

0.98

0.99

1

No

ise

Occ

up

ancy

10-6

10-5

10-4

10-3

10-2

10-1

1

1 fC =̂ 6250 e−

Noise =̂ 0.24 fC

Signal =̂ 3.3 fCSignal =̂ 3.3 fC

6 6 6

? ?design ?

The non-irradiated modules are well within the design limits.


Performance degradation for irradiated modules

¡¡µ3.3 · 1014 p/cm2

Threshold (fC)0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5

Eff

icie

ncy

0.9

0.91

0.92

0.93

0.94

0.95

0.96

0.97

0.98

0.99

1

No

ise

Occ

up

ancy

10-6

10-5

10-4

10-3

10-2

10-1

1A module after full dose

− Gain: 50 → 30 mV/fC ⇒ Smaller signals.

− Noise: 1500 → 2100 − 2400 e−ENC ⇒ Worse S/N ratio.

− Temperature dependence: 6 → 20 e−ENC/◦C ⇒ Higher temperature sensitivity.

− Threshold spread: 160 → 600 e−ENC ⇒ Increased noise.

The operation margin for irradiated modules is rather narrow.


Lets do the time walk

¾ time

− Efficiency vs. timing of clock and trigger signal at

1 fC threshold. The width corresponds to 25 ns.

¾ti

me

− Timing per chip vs. injected charge. The timewalk

is defined as the maximum difference in timing at

1 fC threshold for 1.25 to 10 fC injected charges.

− The chips get slower after irradiation.

Peaking time: 20 → 30 ns.

− The threshold will be crossed later after irradiation.

Timewalk: 10 → more than 16 ns

− By increasing the Voltage for the analog part of the

chip from 3.5 to 3.8 V, the timewalk can be brought

within specifications for most of the channels.

The chips get slower and the timewalk increases.


Tracking performance from test beam measurements

� � � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � ��

� � � � � � � � ��

� � � � � � � ��

� � � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � ��

� � � � � � � � ��

� � � � � � � � ��

� � � � � � � ��

� � � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � ��

� � � � � � � � ��

� � � � � � � ��

� � � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � ��

� � � � � � � � ��

� � � � � � � ��

� � � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � � ��

� � � � � � � � � � ��

! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !

" " " " " " " " " " " "" " " " " " " " " " " "" " " " " " " " " " " "" " " " " " " " " " " "" " " " " " " " " " " "" " " " " " " " " " " "" " " " " " " " " " " "

# # # # # # # # # # ## # # # # # # # # # ## # # # # # # # # # ## # # # # # # # # # ## # # # # # # # # # ## # # # # # # # # # ## # # # # # # # # # #

$ $ $ $ $ $ $ $ $$ $ $ $ $ $ $ $ $$ $ $ $ $ $ $ $ $$ $ $ $ $ $ $ $ $$ $ $ $ $ $ $ $ $$ $ $ $ $ $ $ $ $$ $ $ $ $ $ $ $ $$ $ $ $ $ $ $ $ $$ $ $ $ $ $ $ $ $

% % % % % % % %% % % % % % % %% % % % % % % %% % % % % % % %% % % % % % % %% % % % % % % %% % % % % % % %% % % % % % % %% % % % % % % %

& & & & & & & & && & & & & & & & && & & & & & & & && & & & & & & & && & & & & & & & && & & & & & & & && & & & & & & & && & & & & & & & && & & & & & & & &

' ' ' ' ' ' ' '' ' ' ' ' ' ' '' ' ' ' ' ' ' '' ' ' ' ' ' ' '' ' ' ' ' ' ' '' ' ' ' ' ' ' '' ' ' ' ' ' ' '' ' ' ' ' ' ' '' ' ' ' ' ' ' '

( ( ( ( ( ( ( ( (( ( ( ( ( ( ( ( (( ( ( ( ( ( ( ( (( ( ( ( ( ( ( ( (( ( ( ( ( ( ( ( (( ( ( ( ( ( ( ( (( ( ( ( ( ( ( ( (( ( ( ( ( ( ( ( (( ( ( ( ( ( ( ( (

) ) ) ) ) ) ) )) ) ) ) ) ) ) )) ) ) ) ) ) ) )) ) ) ) ) ) ) )) ) ) ) ) ) ) )) ) ) ) ) ) ) )) ) ) ) ) ) ) )) ) ) ) ) ) ) )) ) ) ) ) ) ) )

* * * * * * * * ** * * * * * * * ** * * * * * * * ** * * * * * * * ** * * * * * * * ** * * * * * * * ** * * * * * * * ** * * * * * * * ** * * * * * * * *

+ + + + + + + ++ + + + + + + ++ + + + + + + ++ + + + + + + ++ + + + + + + ++ + + + + + + ++ + + + + + + ++ + + + + + + ++ + + + + + + +

, , , , , , , , ,, , , , , , , , ,, , , , , , , , ,, , , , , , , , ,, , , , , , , , ,, , , , , , , , ,, , , , , , , , ,, , , , , , , , ,, , , , , , , , ,

- - - - - - - -- - - - - - - -- - - - - - - -- - - - - - - -- - - - - - - -- - - - - - - -- - - - - - - -- - - - - - - -- - - - - - - -

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

/ / / / / / / // / / / / / / // / / / / / / // / / / / / / // / / / / / / // / / / / / / // / / / / / / // / / / / / / // / / / / / / /

Light tight, thermally insulating box

Scintillator Trigger

180 GeV pions

TelescopesTelescopes

0044

*

KB

105

0026

0034

*

0097

0007

*

K5

310*

0047

*

K5

303

K5

308*

KB

104

K5

312*

− Several fully irradiated modules, are placed at about the distances they will have in ATLAS.

− The tracking performance is studied, varying the noise occupancy via the threshold.

− The observed residuals of the space points per module are according to expectation

from the geometry, about 17 (800) µm perpendicular (parallel) to the strips.

− For a track reconstructed from three modules, and at 1.2 fC threshold, an efficiency of

more than 97% at about 10−3 fake rate is achieved. For four modules the efficiency is

still larger than 97%, however at lower fake rate, which is compatible with zero.

The fully irradiated modules still allow for tracking with high efficiency and low fake rate.


Status of the production - the barrel part

− Module production is distributed over the world.

Production centers are at various institutes in

Japan, Scandinavia, UK and the US. This

means a complex logistics is needed.

− The module production is about 1/3 completed.

− Module to barrel mounting is done

in the UK.

− Lots of services (cooling, cables,

sensors for monitoring, . . .) have

to be installed.

There is still much work ahead.


Status of the production - the forward part

− The module production is distributed

over various institutes in Europe

and Australia. The existence of three

different module types makes the

logistics even more complex.

− Module production is about to start.

− Module to disk mounting is done

at NIKHEF and Liverpool.

− Lots of services (cooling, cables,

sensors for monitoring, . . .) have

to be installed.

The time schedule to complete the forward SCT is tight.


Conclusions ...

− A part of the ATLAS inner detector will be equipped with silicon microstripdetectors, the SCT.

− A number of prototype forward modules demonstrate that the mechanicalrequirements can be met with sufficient yield.

− The electrical performance of non-irradiated modules is according to the design.

− The electrical performance of irradiated modules only allows for a marginaloperating flexibility after receiving the full LHC dose, i.e. in the year 2017.

− The series production is underway for barrel modules and about to start for forwardmodules.

... and Outlook− In december 2004 the barrel part is expected to be ready for integration.

− In may 2005 the forward part is expected to be ready for integration.


endcap modules for the atlas semiconductor trackersct/welcomeaux/papers/florenz... ·...

Documents