Developments of micromegas detector at CERN/Saclay

Shuoxing Wu

08-03-2010

Outline:• Introduction to Micromegas detector 1. properties 2. main problem with micromegas• November test beam 1. test beam set up 2. current-voltage monitoring 3. offline analysis 4. spark topology

Part 1 Introduction to micromegas

Working principle

1 ) Conversion stage:

Ionization of gas by the charged particle which produces electron-ion pairs.Electrons directed by E1 (0.1 to 1 kV / cm) to the micro-grid, and cross the mesh holes.

2 ) Amplification stage:

E2 (≈ 50 kV/cm) >> E1 avalanche, electrons mulitplication.

High voltage 2

cathode

Micro-grille

High voltage 1

Anode strip

Anode plane

Incident charged particle

Avalanche

e-

Conversion gap

Amplification gap

Gas

E1

E2

Widely used in particle physics experiments:

Prototype for ILC-TPC:

Compass:

Prototype for ILC-DHCAL:

CAST@CERN:

The Upgrade of ATLAS muon spectrometer LHC upgrade

LSLHC~ 10 LLHC bunch crossing time:

50ns (25 ns) Critical regions in ATLA

S Muon Spectrometer: EI layers:

CSC (27 m2) EIS/L1 (54 m2) EIS/L2 (68 m2)

EM >2: EMS/L1 (85 m2)

6

Micromegas for ATLAS Muon upgrade

• Combine triggering and tracking functions• Matches required performances:

– Spatial resolution ~ 100 m – Good double track resolution– Time resolution ~ few ns– Efficiency > 98%– Rate capability > 5 kHz/cm2

• Potential for going to large areas (1 m x 2 m) with industrial processes – Cost effective– Robustness

7

Spark in micromegas and resistive coating solution:

-8

-6

-4

-2

0

2

4

6

8

0 0. 1 0. 210. 310. 41 0. 510. 610. 720. 820. 92 1. 021. 12Ti me/ us

Volt

age

drop

/V

A point charge being deposited at t=0, r=0, the charge density at (r,t) is a solution of the 2D telegraph equation. Only one parameter, RC (time per unit surface), links spread in space with time. R~1 M/ and C~1pF per pad area matches µs signal duration:

e tr RC

ttr

RC4

2

2),(

Micro-grid

Resistive film (kapton) or ink(1k-500MΩ/ )

Insulator (75 µm) AnodesMicro-grille

Resistive strip(few hundreds of kΩ/ )

Anodes

Anodes

Resistive pad(few tens of kΩ/ )

Micro-grille

Vacrel

Part2 November test beam

10

Test beam set up:• Telescope:

3 X-Y detectors(10 x 10 cm2) manufactured at Saclay

• Aim: Test different resistive films detectors manufactured

by Rui De Oliveira at CERN and compare behaviour to non-

resistive detectors

• Electronics: GASSIPLEX

• DAQ: realised by Demokritos

• Gas: 95%Ar + 3% CF4 + 2% isobutane

120 Gev π+

Y

ResistiveN

on-Resistive

X Y X X

Beam

1 mm 0.25 mm 1 mm

Detectors in test

YTested detectors:Standard bulk detectors; Resistive coating detectors; Segmented mesh detector

SPS-H6

Summary of tested detectors:Type Name Properties:

Standard bulk: SLHC3 1mm pitch

SLHC2 2mm pitch

Resistive coating:

R3&R4 2mm,2MΩ/ ,kapton+insulator

R5 2mm,250MΩ/ ,resistive paste

R6 1mm,400KΩ/ , resistive strip

R7 0.5mm, tens of KΩ/ , resistive pad

Segmented mesh

S1 1mm pitch,8 segmentations

12

Readout and online monitoring:

Data acquistion based on Labview: Monitoring through raw data by C++/Root code:

Maximum strip ID:

Charge spectrum:

13

Spark counting device:

MM detector

C=50 pF

R1=5,6 kΩ R2=5,6 kΩ

C=470 pF

HVmesh

Pre-amplifer

C=50 pF Voltage-Current monitoring PC

for standard ones for resistive ones

Different sparking behaviors of standard and resistive detector:

Standard SLHC2 (2mm) (@10KHz): R6(1mm,400kΩ/ resistive strip) (@10KHz):

SLHC2: HV=400 V (Gain ~3000): current when sparking < 0.4 A voltage drop< 5%R6: HV=390 V (Gain ~3000): current when sparking < 0.08 A voltage drop<0.5%

350

360

370

380

390

400

410

420

430

440

1 542 1083162421652706324737884329487054115952Spark number

Mesh voltage/V

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1Mesh current/ua

250

270

290

310

330

350

370

390

410

430

1 77 153 229 305 381 457 533 609 685 761 837 913 989 10651141121712931369144515211597Spark number

V

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1 uAMesh Vol tageMesh Current

Sparking behavior of resistive detector R3:

Standard SLHC2 (2mm) (@10KHz): Resistive R3(2mm,2MΩ/ ) (@10KHz):

SLHC2: HV=400 V (Gain ~3000): current when sparking < 0.4 A voltage drop< 5%R3: HV=410 V (Gain ~3000): current when sparking < 0.2 A voltage drop<2%

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 301 601 901 120115011801210124012701300133013601Spark number

Mesh current

340

360

380

400

420

440

460

Mesh voltage

250

270

290

310

330

350

370

390

410

430

1 77 153 229 305 381 457 533 609 685 761 837 913 989 10651141121712931369144515211597Spark number

V

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1 uAMesh Vol tageMesh Current

250

270

290

310

330

350

370

390

410

1 92 183 274 365 456 547 638 729 820 911 1002109311841275136614571548

V

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Spark number

uAS1 mesh vol tageS1 mesh current

Sparking behavior of S1:

Seven,eight sparking

six sparking

five sparking

four sparking

three sparking

two sparking

one sparking

S1:Standard bulk: SLHC2

16

250

270

290

310

330

350

370

390

410

430

1 77 153 229 305 381 457 533 609 685 761 837 913 989 10651141121712931369144515211597Spark number

V

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1 uAMesh Vol tageMesh Current

Charge vs. channel ID:

17

1mm

1mm 1mm

1mm

250m

250m

R3

R6

Pedestal shift in standard MM (telescope) due to sparks.

No pedestal shift in resistive detectors (R3&R6).

Spark rate vs mesh voltage and beam intensity:

R3(2mm, 2MΩ/ ):

18

0. 00E+00

5. 00E-07

1. 00E-06

1. 50E-06

2. 00E-06

2. 50E-06

3. 00E-06

3. 50E-06

4. 00E-06

4. 50E-06

0 10 20 30 40 50Beam I ntensi ty/ KHz*cm-2

Spark rate

Spark rate= #sparks/#incident hadron

19

Detector type Spark rate Spark current /A

Voltage drop

SLHC2 Standard bulk 7*E-5 0.4 5%

R3 2 M Ω/

resistive kapton9.6*E-6 0.2 2%

R6 400K Ω/

resistive strip6.4*E-6 0.08 0.5%

R5 250 M Ω/

resistive paste1.6*E-4 0.1 1.5%

R7 tens of K Ω/

resistive pad5.9*E-4 0.35 4.5%

Detector performance at same gas gain (~3000):

Process to analysis the data:

1. Covert raw data into root tree

1. Decoding

2. Pedestal calculation and subtraction

3. Event filtering

4. Clusterization

5. Track reconstruction

6. Efficiency and spatial resolution studies based on the track

Binary data format:

1.run header (run ID, time, type, …..)

2. event loop:

Loop over 8 detectors: .number of active strips Strip loop: Word 1: 31 30 … 2 1 0

Word 2: 31 30 … 2 1 0...

Word NAS[det]:

3. run footer (event number, run end time….)

31 30 … 2 1 0

….

31 30 29 … 12 11 … 0

Ovf Vld strip data strip ID

Conversion from raw data into root tree:

‘Detector’ class:NofStrips; stripID;stripCharge;

Channel-pin relation

Decoding matrix:

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 3132 30 28 26 24 22 20 18 16 14 12 10 8 6 4 233 35 37 39 41 43 45 47 49 51 53 55 57 59 61 6364 62 60 58 56 54 52 50 48 46 44 42 40 38 36 3465 67 69 71 73 75 77 79 81 83 85 87 89 91 93 9596 94 92 90 88 86 84 82 80 78 76 74 72 70 68 66

1. Decoding:

Before decoding: After decoding:

Check decoding:

2.Pedestal calculation and subtractionCharge-ID distribution: Strip out of event: Strip within event:

Case of ‘noisy’ strip: Charge-ID distribution: Channel noise level:

3. Filter event: Based on the cut on pedestal sigma, only strips with charge larger t

han the threshold are kept.

threshold

4. Clusterization:

Strips with charge above the threshold form a ‘Cluster’( 1 gap is allowed). The cluster center is the charge-weighted barycentre of the strips, the cluster charge is the sum of strips charge within the cluster, cluster size is the number of strips within the cluster.

Number of clusters in one event: Cluster size:

1mm

1mm

1mm

1mm

1mm

1mm 1mm

1mm0.25mm

0.25mm 0.25mm

0.25mm

R3

R6

R3

R6

5. Track fitting:

y=a*x+b, a and b are determined by minimizing the chi-square.

Track angle X: Track angle Y:

R6 (1mm, 400KΩ/ ) efficiency :

>98%

drop<3%

29

@beam intensity of 11kHz/cm2: @voltage of 400V:

Resolution:

beam

MM

MM under test

MMMM

1mm 0.25 mm 1mm

•Residuals of MM cluster position and extrapolated track from MM telescope: •Convolution of: - Intrinsic MM resolution - Track resolution (extrapolated) ~68m

R6(1mm pitch, 400kΩ/ ) R3(2mm pitch, 2mΩ/ )

δ=112mδmm= 90±0.8m

δ=199mδmm= 187±1.9m

30

Resolution vs mesh voltage:

31

R6(1mm, 400kΩ/ resistive strip): R3(2mm, 2MΩ/ resistive kapton):

Pillars:Track position( with hit in test detector): Track position( without hit in test

detector):

32

Combination:

Cluster size:

33

R3 : 2mm , 2 M Ω/ Resistive kapton +insulator

R6 : 1mm , 400 k Ω/ Resistive stripSLHC2: 2mm standard bulk

Number of missing strips in cluster:R6( 1mm, 400 KΩ/ ): R3( 2mm, 2 mΩ/ ):

34

Probability to have missing strips in cluster and number of missing strips:

0

0. 2

0. 4

0. 6

0. 8

1

1. 2

1. 4

1. 6

1. 8

2

0 2 4 6 8 10Cl uster si ze/ stri p

Si ze

R6

R3

00. 10. 20. 30. 40. 50. 60. 70. 80. 9

1

0 5 10 15 20Cl uster si ze/ mm

Probabi l i ty

R6R3

35

Sparks in R7(few tens of kΩ/ Resistive pads):

36

All the ‘sparks’ in R7: Three ‘spark’ type:

‘Spark’ amplitude distribution:

Sparks in R5(250 MΩ/ Resistive kapton):

37

All the ‘sparks’ in R5:

‘Spark’ amplitude distribution:

Four ‘spark’ type:

Sparks in SLHC2(standard bulk):

38

All the ‘sparks’ in SLHC2: One ‘spark’ type:

‘Sparks’ amplitude distribution:

Sparks in R3(2 MΩ/ resistive kapton):

39

All ‘sparks’ in R3:

‘sparks’ amplitude distribution:

One ‘spark’ type:

Sparks in R6(400 kΩ/ resistive strip):

40

All ‘sparks’ in R6:

‘sparks’ amplitude distribution:

One ‘spark’ type:

Spark amplitude vs voltage and beam intensity:

0

0. 2

0. 4

0. 6

0. 8

1

1. 2

1. 4

1. 6

1. 8

2

375 380 385 390 395 400 405 410 415

Vol tage/ V

Spar

k am

plit

ude

mean

/V

0. 5

0. 52

0. 54

0. 56

0. 58

0. 6

0. 62

385 390 395 400 405 410 415

Vol tage/ V

Spar

k am

plit

ude

mean

/V5. 15

5. 2

5. 25

5. 3

5. 35

5. 4

5. 45

5. 5

5. 55

5. 6

370 380 390 400 410

R5 voltage: R7 voltage:

0

0. 1

0. 2

0. 3

0. 4

0. 5

0. 6

0. 7

380 390 400 410 420 430 440 450

Vol tage/ V

Spar

k am

plit

ude

mean

/V

0

0. 1

0. 2

0. 3

0. 4

0. 5

0. 6

0. 7

0. 8

360 370 380 390 400 410 420 430

Vol tage/ V

Spar

k am

plit

ude

mean

/V

0

0. 1

0. 2

0. 3

0. 4

0. 5

0. 6

0. 7

0 5 10 15 20 25 30 35 40 45Beam I ntensi ty

Spar

k am

plit

ude

mean

/V

SLHC2 voltage:

R3 intensity:R6 voltage:R3 voltage:

41

Conclusion:

• Today’s micromegas detector is facing the spark problem, resistive coating is a successful way to solve it.

• High efficiency achieved with a resistive strip detector R6.

• Good spatial resolution.