photomultiplier signal processing Characterisation of the ...ndip.in2p3.fr/ndip11/AGENDA/AGENDA-by-DAY/... · Data loss

Multi-time-over-threshold technique for photomultiplier signal processing

Characterisation of the SCOTT ASIC

S. Ferry1, F. Guilloux2

CEA Saclay, Irfu

and E. Delagnes2, F. Louis2, E. Monmarthe2, J-P. Schuller1, Th.Stolarczyk1, B. Vallage1, on behalf of the KM3NeT consortium3

1IRFU/SPP, 91191 Gif/Yvette, France2 IRFU/SEDI, 91191 Gif/Yvette, France3 Supported by the European Commission through FP6 and FP7

KM3NeT

• Now in preparatory phase (FP7)

• A European deep-sea research infrastructure for a neutrino telescope – Mediterranean Sea– Instrumented volume

5 km3

– 300 detection units– 12000 photo-detectors

• Cherenkov light detection– Upward – Downward cosmic – Optical background (100 Hz / cm-2)

40K decay in water

Bioluminescence

Mainly single photon signals

• All data to shore– On shore computer farm

2Sophie Ferry - NDIP11 - Characterisation of SCOTT

Front end requirements (from CDR)

• Time resolution (PMT+FE) < 2 ns at 1 pe (RMS)

• Charge dynamic range : 100 pe / 25 ns– Charge estimation

• Two-hit time separation < 25ns

SCOTT ASIC


SCOTT concept

• Time and Amplitude Sampling– “wave-form”-like image of

the pulse

• Master Clock 50 MHz– 16 bit Time stamp

• 1.25 ns sampling

• 16 independent channels– Dynamic range : 0- 3V– 10 bit adjustable thresholds

• Auto-triggered– Each threshold can be a

trigger– Handling several PMTs e.g.:

2 PMTs x 8 thresholds

16 PMTs x 1 threshold

Sophie Ferry - NDIP11 - Characterisation of SCOTT 4

Analogue input

Analogue switchesInternal sharing of

signals

10 bit DACAMS BiCMos

0.35mProgrammable

thresholds LSB<3mV

DiscriminatorAnalogue to digital

convertor

SCOTT architecture

Sophie Ferry - NDIP11 - Characterisation of SCOTT

5x16 channels

SOC

Circular sampling memory

Delay locked loop (DLL)

All channels are synchronous

Circular digital memory

2 memory banks (MB)1MB = 1 time slice =20 ns

No dead time

16 bit coarse time counter

LSB = 20 ns

Time step TCK

/16

Ts

= 1.25ns (~ 800MHz)

SCOTT architecture


6x16 channels

SOC

SCOTT architecture


7x16 channels

SOC

FIFOAuto-triggered

Logical or channel triggered

Size32 time slices

640 ns continuous wave-

form

Data loss <1‰ for 200 kHits/s

Double portFck

sample ≠Fck

readoutReadout driven by System-on-Chip

Selective readoutData reduction

Scott output example


8

11280

0000000000000111000000000000001100000000000000010000000000000001

11281

11111111000000001111111000000000111111100000000011111100000000001111110000000000111111000000000011111000000000000111000000000000

Am

plitu

de

Time

20 ns

1.25 ns

Coarse Time stamp11283

0100000001111100

The SCOTT Opera

•

Act I : Determination of thresholds– Which values for how many thresholds ?

•

Act II : Charge

•

Act III: Timing– Correction of the walk effect– SCOTT Intrinsic timing accuracy


ACT I THRESHOLD DETERMINATION

ActorsPMT : Antares 10’’ Gain=3·107

1 pe

50mV 5pC

LED : flashes at 1pe up to 70peOscilloscope : F=2.5GHz Bandwidth=500MHz


10

11

Tpeak


Performances verified with SCOTT

ACT II CHARGE RESOLUTION

ActorsPMT : Antares 10’’ gain=3.107

1pe

50mV 5pC

LED : flashes at 1pe and 11peSCOTT : 5 thresholds


12

• Integration of FRED : needs CPU time• Alternative based on Time-Over-Thresholds (TOT)

– Upper (Q+ ) and lower(Q- ) area under the thresholds

• Q

(Q+ + Q- ) / FQ– FQ depends on

the last threshold reached

the threshold values

requires calibration

Q+ Q-


Charge resolution from TOTs

14

Charge 1pe

Mean=5pCRMS =2pC = 40%(PMT charge resolution

35%)

SCOTT + TOT enable charge reconstruction with good accuracy


ACT III TIMING RESOLUTION Scene 1 : Pulse time resolution


15

ActorsPMT : Antares 10’’ gain=3.107

1pe

50mV 5pC

LED : flashes at 1pe and 11peSCOTT : PMT 5 thresholds

LED trigger 1 threshold

Pulse time reconstruction

Walk effect- Threshold crossing time

depends on the charge- Peak time Tpeak unchanged


residuals

1st

thres.

TOT[i] (ns)

Tpea

k-T 0

[i]

All thresholds

Tpeak and Walk correction with SCOTT

• T1/3 -TLEDT1/3 =crossing time of the 1st

threshold

• Tpeak - TLED


(11+1)pe Nentr 1852RMS 1.5

1pe11pe

11+1pe

(11+1)pe Nentr 1852RMS 2.5

1pe11pe

11+1pe

TTSPMT

1.3 ns

Time resolution = 1.5 nsNo charge estimate needed

ACT III TIMING RESOLUTION Scene 2 : SCOTT Intrinsic timing accuracy

ActorsGenerator: HP 81110A 330MHz

SCOTT


18

SCOTT intrinsic timing

• Generated Pulse– 120mV height– 1kHz

• SCOTT: 6 thresholds15% 35% 50% 70% 85% 90%


00000000111111110000000000000111

11111111111111111111111111111111000111111111111100000000111111110000000000000001

111111111111111111111111111111111111111111111111111111111111111111111111111111110111111111111111

• Consider all possible time between crossing times for every couple of thresholds

(ns)

SCOTT intrinsic time resolution 0.8ns

Conclusion


The End


21

PM

T

SCOTT architecture

• Analogue input

Analogue switchesInternal sharing of signals

10 bit DACAMS BiCMos 0.35m

Programmable thresholds LSB<3mV

DiscriminatorAnalogue to digital

convertor

• FIFO and zero suppressAuto-triggeredLogical or channel

triggered

Size32 time slices

640 ns continuous wave- form

Data loss <1‰ for 200 kHits/s

Double portFck sample ≠Fck readout

Readout driven by System-on-Chip

Selective readoutData reductionSophie Ferry - NDIP11 - Characterisation of SCOTT 22

• Fast memory samplingDelay locked loop

(DLL)All channels are synchronous

Circular digital memory2 memory banks (MB)

1MB = 1 time slice =20 nsNo dead time

16 bit coarse time counter

LSB = 20 ns

Time step TCK /16

Ts = 1.25ns (~ 800MHz)

1channel

FRED

23

Apeak

Tpeak

tR


Tpeak residuals

Sophie Ferry - NDIP11 - Characterisation 24

Apeak

Tpeak

tR

Test bench


25

PMT

HTAnode

Diff

LED

Black Box

PMT

HTAnode

oscilloscope

Diff

LED

Black Box

SCOTT

Generator

PMT

HTAnode

Diff

LED

Black Box

Generated pulse viewed with SCOTT


26

30164

00000000111111110000000000000111

30165

11111111111111111111111111111111000111111111111100000000111111110000000000000001

30166

111111111111111111111111111111111111111111111111111111111111111111111111111111110111111111111111

30167

111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111000

30168

11111111111111111111111111111111111111111111100011111110000000001000000000000000

30169

1111111110000011100000000000

Am

plitude

Time

20 ns

1.25 ns

PMT characterisation

• PMT + Oscilloscope


PMT

HTAnode

oscilloscope

Diff

Black Box

Define the 1PE 50 mV 5.4 pC

Amplitude max.

Charge

Amplitude with FRED & SCOTT


T1/3 versus charge

• T1/3PE versus charge


1PE and 11PE

Tpeak versus charge

• Tpeak versus charge


1PE and 11PE

Sophie Ferry - NDIP11 - Characterisation of SCOTT31

Charge 11peCharge 1pe

Charge resolution with FRED + SCOTT

Mean=70pC RMS=22pC: 32%

Mean=5pC RMS=2pC: 37%

SCOTT + Pulse fit enable charge reconstruction without introducing further error

SCOTT data set


• 1 + 11 pe data sets

0 1/3 2/3 1 4/3 5/3 2 3 8 pe

Analogue to digital@ SCOTT


Front End

Asic

System on

Chip

Analogue

Signal

Digital

data

Ethernet TCP/IP

data link

Shore

Documents

photomultiplier signal processing Characterisation of the ...ndip.in2p3.fr/ndip11/AGENDA/AGENDA-by-DAY/... · Data loss