(1)Emanuele Fiandrini EPS 2017 Emanuele Fiandrini University of Perugia - I.N.F.N

(1)

Emanuele Fiandrini EPS 2017

Emanuele Fiandrini University of Perugia - I.N.F.N. Perugia (IT)

Development and characteriza2on of near-‐UV sensi2ve Silicon Photomul2pliers for the

Schwarzschild-‐Couder Telescope prototype for the CTA collabora2on

www.cta-‐observatory.org

1

(2)

CTA Cherenkov Telescope Array observatory SiPM sensors for Cherenkov radiation detection

Packaging, assembly & performances

Development and characteriza2on of near-‐UV SiPM for the Schwarzschild-‐

Couder Telescope prototype for the CTA collabora2on

2 Emanuele Fiandrini University of Perugia - I.N.F.N. Perugia (IT), G.Ambrosi, M.

Ambrosio, A. Aramo, E.Bissaldi, B. Bertucci, A.Boiano, C.Bonavolontà, M. Caprai, N.GiglieJo, F.Giordano, M.Ionica, C. de Lisio, L.Di Venere, V.Masone, R.PaoleO, V.Postolache, D.Simone,

L. Tos2, V.Vagelli, M.V alen2no

(3)

Imaging Air Cherenkov Telescopes!Gamma Ray

CTA Telescopes ( km² )

3

(4)

Physics of TeV γ-‐ray Telescopes!

Dark matter Supernova

remnants

AGNs Binaries

Pulsars GRBs

Starburst

galaxies …?

EBL Axions, …

Spacetime Cosmic

rays

3033 GeV sources 176 TeV sources

The gamma-ray sky as of 2016

Physics*with*TeV*gammaPray*telescopes*

ICHEP*2016* Jus:n*Vandenbroucke:*CTA* 3*

4

(5)

Emanuele Fiandrini EPS 2017 5

1,350 members

from 210 insBtutes in 32 countries

The CTA Project!

Headquarters Science Data

Management Center

Array Sites

(6)

The CTA Project!

CTA Key Science Projects

Adapted from W. Hofmann!Adapted from W. Hofmann!6

(7)

CTA performances!

LST MST SST

DIFFERENTIAL SENSITIVITY

5 bins per decade in energy Improvement of a factor 5-10 in sensitivity w.r.t. the current IACTs in the core energy range

Extension of the accessible

energy range below 100 GeV and above 50 TeV

1. Introduction to CTA Science 1.1 Key Characteristics & Capabilities

(TeV) Energy ER

−2

10 10⁻¹ 1 10 10²

)°Angular Resolution (

0 0.05 0.1 0.15 0.2 0.25

www.cta-observatory.org (2015-05-11)

CTA South

HAWC

Fermi LAT Pass 8 MAGIC

VERITAS

Figure 1.1 – Comparisons of the performance of CTA with selected existing gamma-ray instruments. Top: dif- ferential flux sensitivity for five independent five standard deviation detections per decade in energy. Additional criteria applied are to require at least ten detected gamma-rays per energy bin and a signal/background ratio of at least 1/20. Bottom: angular resolution expressed as the 80% containment radius of reconstructed gamma rays (the resolution for CTA North is similar).

Cherenkov Telescope Array

Science Case Page 8 of199 v.2.1 | 26 September 2016

ANGULAR

RESOLUTION ENERGY

RESOLUTION

EFFECTIVE AREA

7

(8)

Camera Sensors!

The current IACT generation cameras are equipped with PMTs.

CTA is evaluating the possibility to equip the focal plane camera of Small and Medium size telescopes with SiPMs

SiPM features for Cherenkov light detection:

√ Smaller areas (<1cm²), hence higher pixel angular resolution

√ Higher photo-detection efficiency at UV wavelengths (~ 50@ 350 nm)

√ Fast response O(1-10) ns

√ Not damaged by moonlight, can be operated during bright Moon nights, enhancing the DAQ duty cycles

√ Can be operated with bias voltages <100V

√ Low power consumption (µW)

√ Light-weight

X Noisy, dark count rates O(10-100) KHz/mm² at room temperature, but below the expected average night sky background.

8

(9)

FBK NUV-‐HD SiPM sensors! (a) FBK NUV-HD (b) Hamamatsu S13360-3050CS (c) SensL J-series 30035 Figure 1: Full scale pictures of the three tested SiPMs.

(a) FBK NUV-HD (b) Hamamatsu S13360-3050CS (c) SensL J-series 30035 Figure 2: Close-up pictures of the cells of the three tested SiPMs. The scale in the pictures shows 20 µm.

more, all cells are connected to one common output. For a review of the history of SiPMs and their basic functionality the reader is referred to [? ] and references therein.

55

The three tested devices are

• a NUV-HD SiPM from FBK,

• a S13360-3050CS MPPC from Hamamatsu,

• and a MicroFJ-SMTPA-30035-E46 from SensL.

A picture of all SiPMs is shown in Figure 1 All three de-

60

vices are p-on-n, which means that the avalanche structure consists of a p-implant in an n-doped substrate. In this configuration the electric field directs electrons produced by blue photons just below the surface into the avalanche region, which is why all three devices have a sensitivity

65

that peaks in the blue or near UV.

2.1. FBK NUV-HD

The FBK device is fabricated in NUV-HD technology, which is discussed in more detail in [? ]. The device we investigated is a special development for the Cherenkov

70

Telescope Array (CTA) [? ]. Unlike the other two devices this one does not have an epoxy, silicon resin, or similar protective coating, which is one of the reasons why it has better UV sensitivity than the other two SiPMs. The dimensions of the FBK SiPM are (6.3 × 6.8) mm² with a

75

micro-cell pitch of 30 µm. One SiPM has a total of 40,394 cells. The chip came glued onto a PCB carrier and is wire bonded. Figure 2a shows a microscope picture of four cells.

Clearly visible are the quench resistors (red) and the metal lines that connect all cells.

80

2.2. Hamamatsu LCT5

The SiPM from Hamamatsu is a S13360-3050CS MPPC.

It is fabricated using their latest technology, which is also called LCT5 because it is the fifth iteration of a low-crosstalk development. The dimensions of the tested device are

85

(3 × 3) mm² with a cell pitch of 50 µm (s. Figure 2b) resulting in a total of 3,600 cells. The device is mounted onto a standard ceramic chip carrier and coated with UV- transparent silicon raisin. Electrical contacts between the chip and the pins of the carrier are made with wire bonds.

90

We note that Hamamatsu also produces SiPMs using the 2

(a) FBK NUV-HD (b) Hamamatsu S13360-3050CS (c) SensL J-series 30035 Figure 1: Full scale pictures of the three tested SiPMs.

(a) FBK NUV-HD (b) Hamamatsu S13360-3050CS (c) SensL J-series 30035 Figure 2: Close-up pictures of the cells of the three tested SiPMs. The scale in the pictures shows 20 µm.

more, all cells are connected to one common output. For a review of the history of SiPMs and their basic functionality the reader is referred to [? ] and references therein.

55

The three tested devices are

• a NUV-HD SiPM from FBK,

• a S13360-3050CS MPPC from Hamamatsu,

• and a MicroFJ-SMTPA-30035-E46 from SensL.

A picture of all SiPMs is shown in Figure 1 All three de-

60

vices are p-on-n, which means that the avalanche structure consists of a p-implant in an n-doped substrate. In this configuration the electric field directs electrons produced by blue photons just below the surface into the avalanche region, which is why all three devices have a sensitivity

65

that peaks in the blue or near UV.

2.1. FBK NUV-HD

The FBK device is fabricated in NUV-HD technology, which is discussed in more detail in [? ]. The device we investigated is a special development for the Cherenkov

70

Telescope Array (CTA) [? ]. Unlike the other two devices this one does not have an epoxy, silicon resin, or similar protective coating, which is one of the reasons why it has better UV sensitivity than the other two SiPMs. The dimensions of the FBK SiPM are (6.3 × 6.8) mm² with a

75

micro-cell pitch of 30 µm. One SiPM has a total of 40,394 cells. The chip came glued onto a PCB carrier and is wire bonded. Figure 2a shows a microscope picture of four cells.

Clearly visible are the quench resistors (red) and the metal lines that connect all cells.

80

2.2. Hamamatsu LCT5

The SiPM from Hamamatsu is a S13360-3050CS MPPC.

It is fabricated using their latest technology, which is also called LCT5 because it is the fifth iteration of a low-crosstalk development. The dimensions of the tested device are

85

(3 × 3) mm² with a cell pitch of 50 µm (s. Figure 2b) resulting in a total of 3,600 cells. The device is mounted onto a standard ceramic chip carrier and coated with UV- transparent silicon raisin. Electrical contacts between the chip and the pins of the carrier are made with wire bonds.

90

We note that Hamamatsu also produces SiPMs using the 2

6 x 6 mm² 30 x 30 µm² cell

•  Produced at FBK(Trento, IT)

•  p-n SiPM

•  Active area: 6.03 x 6.06 mm²

•  Microcell size: 30 x 30 µm²

•  Fill Factor: 76%

•  High PDE (50%) for UV photons

bonding pads

SiPMs mounted on single sensor test PCB

Time (ns) 0 50 100 150 200 250 300 350 400

Amplitude (mV)

−30

−25

−20

−15

−10

−5 0 5 10

1 p.e.

3 p.e.

5 p.e.

SiPM signal amplified in transimpendence G=500Ω

•  NUV-HD technology successful,

development of further improvements are ongoing

9

(10)

Samples of sensors are thoroughly studied in diﬀerent temperature ranges in a climatic chamber to extract sensor performance informations (gain, dark count rates, crosstalk, …)

Inverse Bias (V)

24 25 26 27 28 29 30

Current (nA)

−1

10 1 10 102

103

104

Temperature (C)

−20 −10 0 10 20 30 40

Breakdown Voltage (V)

26.5 27 27.5 28 28.5

p0 27.23 ± 0.008073 p1 0.03185 ± 0.000424 p0 27.23 ± 0.008073 p1 0.03185 ± 0.000424

FBK NUV-‐HD SiPM sensors!

•  Breakdown Voltage < 30V

•  Small temperature dependence

10 +35°

-10°

@V_b

@T =⇡ 32 mV/^oC

(11)

Emanuele Fiandrini EPS 2017 Wavelenght (nm)

250 300 350 400 450 500 550 600 650 700

PDE (%)

0 10 20 30 40 50 60

70 Cherenkov)Light)[a.u])

(La)Palma,)2200)m))

Night)Sky) Background)[a.u])

(La)Palma,)2200)m))

NUVCHD)SiPM)

Photo)DetecGon)Eﬃciency)

OverVoltage (V)

5 6 7 8 9 10 11

)6 10× Gain (

1.6 1.8 2 2.2 2.4 2.6 2.8 3 3.2

3.4 _{+ 30 deg}

+ 20 deg + 10 deg + 0 deg - 10 deg

Photo Electron Threshold 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 )2 Rate (kHz/mm

1 10 102

+ 10.5 OV + 9.5 OV + 8.5 OV + 7.5 OV + 6.5 OV + 5.5 OV

Dark Count Rate, T = +10 deg

OverVoltage (V)

5 6 7 8 9 10 11

)2 Rate (kHz/mm

10 102

Dark Count Rate, 0.5 P.E. Threshold

1 10 102

+ 10.5 OV + 9.5 OV + 8.5 OV + 7.5 OV + 6.5 OV + 5.5 OV

1 10 102

+ 10.5 OV + 9.5 OV + 8.5 OV + 7.5 OV + 6.5 OV + 5.5 OV

NUV-HD SiPM sensitivity peaks towards NUV (Cherenkov signals) with maximum PDE ≈ 50%

Wavelenght (nm) 250 300 350 400 450 500 550 600 650 700

PDE (%)

0 10 20 30 40 50 60

70 Cherenkov Light [a.u.]

(La Palma, 2200m) Night Sky [a.u]

La Palma, 2200m)

NUV-HD SiPM

Photo Detection Eﬃciency

•  Wide dynamic range

•  gain G=O(10⁶)

11

(12)

Wavelenght (nm) 250 300 350 400 450 500 550 600 650 700

PDE (%)

0 10 20 30 40 50 60

70 Cherenkov)Light)[a.u])

(La)Palma,)2200)m))

Night)Sky) Background)[a.u])

(La)Palma,)2200)m))

NUVCHD)SiPM)

Photo)DetecGon)Eﬃciency)

OverVoltage (V)

5 6 7 8 9 10 11

)6 10× Gain (

1.6 1.8 2 2.2 2.4 2.6 2.8 3 3.2

3.4 + 30 deg

+ 20 deg + 10 deg + 0 deg - 10 deg

1 10 102

+ 10.5 OV + 9.5 OV + 8.5 OV + 7.5 OV + 6.5 OV + 5.5 OV

OverVoltage (V)

5 6 7 8 9 10 11

)2 Rate (kHz/mm

10 102

Dark Count Rate, 0.5 P.E. Threshold

1 10 102

+ 10.5 OV + 9.5 OV + 8.5 OV + 7.5 OV + 6.5 OV + 5.5 OV

1 10 102

+ 10.5 OV + 9.5 OV + 8.5 OV + 7.5 OV + 6.5 OV + 5.5 OV

•  NUV-HD dark count rate DCR << 100kHz/mm² up to 20 deg

•  DCR doubles every 7.0°

NUV-HD performances are compliant with minimum requirements specified to equip the focal planes of CTA telescopes

12

+30°

+20°

+10°

+0°

-10°

(13)

SCT Telescope!

Schwarzschild-Couder dual mirror optics Medium Size Telescope

Primary mirror (9.7m diam) Secondary mirror (5.4m diam)

Focal plane camera

Dual mirror optics designed to cancel aberration and de-magnify images, to be compatible with compact high-resolution SiPM camera and resulting in a smaller point spread function (PSF) and improved angular resolution compared to the classical single mirror Cherenkov Telescope.

Mechanical stability and mirror alignment are the main challenges.

VERITAS site, Arizona

13

(14)

SCT Telescope!

Schwarzschild-Couder dual mirror optics Medium Size Telescope

Primary mirror (9.7m diam) Secondary mirror (5.4m diam)

Focal plane camera

VERITAS site, Arizona

14

22 institutes, universities and observatories 5 institutes, universities and observatories 3 institutes and universities

1 university 1 university

•  Optical Support Structure and Positioners installed in September 2016

•  Complete assembly planned by end of 2017

SCT is the unique proposal of the innovative SC optics for the CTA Medium Size Telescope

(15)

pSCT Telescope!

Camera design for Schwarzschild-Couder telescopes

ICHEP*2016* Jus:n*Vandenbroucke:*CTA* 40*

•  Excellent*op:cal*resolu:on,*small*

plate*scale*of*dualPmirror*telescope*

well*matched*to*ﬁne*pixela:on*

supported*by*silicon*photomul:pliers*

and*TARGET*readout*electronics*

•  11,328*6x6*mm²*pixels*(temperatureP stabilized*silicon*photomul:pliers)*

•  Pixel*size*0.067°*(highP*resolu:on*

imaging)*

•  Readout*directly*behind*focal*plane*

•  1*GSa/s,*10*bits*eﬀec:ve*(TARGET*7)**

•  3*kW*power*budget*

•  Shares*many*common*components*

with*the*Compact*High*Energy*

Camera*for*CTA*Small*Size*Telescopes*

8° field of view, 81 cm diameter

0.4m² active area per telescope

Prototype demonstrator for the Medium Size SCT solution

15 Prototype main goals:

•  Demonstrate the performances of the optical system

•  Gain experience with the optical alignment and operation of the SiPM camera

(16)

pSCT Telescope!

Hamamatsu INFN prototype

pSCT camera mechanics in Univ. of Wisconsin

FBK 6x6mm² SiPM intented to replace the original Hamamatsu MPPC solution and equip a possible upgrade of the pSCT camera

pSCT camera currently equipped with Hamamatsu MPPC S12642-0404PA-50(X)

16

(17)

pSCT focal plane modules!

4 Board gluedwithonea layerof glue

Board 3

PCB bonding pad

500 μm SiPM bonding pad

27x27mm² PCBs are equipped with 16 SiPMs to cover uniformly the exposed area

36 modules made of 16 SiPMs will be coupled to the electronic readout and the pSCT camera in the next months, to be tested in situ and to prepare for the next massive production of modules.

17

(18)

pSCT focal plane modules!

Module(assembly(

Die5Bonder(manual(machine(@(FBK,(1(HD(+(4(HD2(modules(assembled(

PCB modules are assembled with SIPM sensors in the laboratories of INFN.

SiPMs are positioned on the PCBs using a die-bonder machine.

18

(19)

Module assembly and packaging!

300 350 400 450 500 550 600 650 700

µm) L37 SiPM border X distance (

Pixel (X)

−0.5 0 0.5 1 1.5 2 2.5 3 3.5

Pixel (Y)

−0.5 0 0.5 1 1.5

2 2.5 3 3.5

µm) L37 SiPM border X distance (

300 350 400 450 500 550 600 650 700

µm) L37 SiPM border Y distance (

Pixel (X)

−0.5 0 0.5 1 1.5 2 2.5 3 3.5

Pixel (Y)

−0.5 0 0.5 1 1.5 2 2.5 3 3.5

µm) L37 SiPM border Y distance (

µm) SiPM border X distance ( 300 400 500 600 700

Occurrence

0 1 2 3 4 5 6 7

µm : 28.25 m -- σ

: 506.64µ

µ: 506.64µm -- σ: 28.25µm µ

µm) SiPM border Y distance (

300 400 500 600 700

Occurrence

0 1 2 3 4 5 6 7 8 9

µm : 27.01 m -- σ

: 499.49µ

µ: 499.49µm -- σ: 27.01µm µ

−40

−20 0 20 40 µm)

∆ X ( L37 SiPM Center

Pixel (X)

Pixel (Y)

−0.5 0 0.5 1 1.5 2 2.5 3 3.5

µm)

∆ X ( L37 SiPM Center

−60

−40

−20 0 20 40 60 µm)

∆ Y ( L37 SiPM Center

Pixel (X)

Pixel (Y)

−0.5 0 0.5 1 1.5 2 2.5 3 3.5

µm)

∆ Y ( L37 SiPM Center

−0.4

−0.3

−0.2

−0.1 0 0.1 0.2 0.3 0.4

(deg) L37 RotationXY

Pixel (X)

Pixel (Y)

−0.5 0 0.5 1 1.5 2 2.5 3

3.5 L37 Rotation_XY (deg)

µm)

∆ X ( SiPM Center

−50 0 50

Occurrence

0 0.5 1 1.5 2 2.5 3 3.5 4

µm : 17.34 m -- σ

: -8.79µ

µ: -8.79µm -- σ: 17.34µm µ

µm)

∆ Y ( SiPM Center

−50 0 50

Occurrence

0 0.5 1 1.5 2 2.5 3

µm : 26.21 m -- σ

: 7.66µ

µ: 7.66µm -- σ: 26.21µm µ

(deg) SlopeXY

−0.4 −0.2 0 0.2 0.4

Occurrence

0 0.5 1 1.5 2 2.5 3

: 0.17 deg : -0.05 deg -- σ

µ: -0.05 deg -- σ: 0.17 deg µ

Inspection with ruby-head touch probe and an optical metrology machine to verify the quality of the sensor alignment.

Sensor alignment better than 30/40µm

position along X axis

diﬀerence w.r.t. nominal position position along Y axis

diﬀerence w.r.t. nominal position sensor XY rotation

diﬀerence w.r.t. nominal position

19

µm

µm deg

deg

(20)

SiPM module assembly tests!

Bias (V)

5 10 15 20 25 30

Current (nA)

10 102

103

[01]

[02]

[03]

[04]

[05]

[06]

[07]

[08]

[09]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

Entries 16 Mean 27.77 RMS 0.04108

27.5 27.6 27.7 27.8 27.9 28

Occurrences

0 1 2 3 4 5 6 7 8

9 Entries 16

Mean 27.77 RMS 0.04108

pSCT(module(assembly(

Bias (V)

5 10 15 20 25 30

Current (nA)

10 102

103

[01]

[02]

[03]

[04]

[05]

[06]

[07]

[08]

[09]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

Entries 16 Mean 27.77 RMS 0.04108

27.5 27.6 27.7 27.8 27.9 28

Occurrences

0 1 2 3 4 5 6 7 8

9 Entries 16

Mean 27.77 RMS 0.04108

Matrix(sensors(are(tested(before(dispensing(of(protec5on(epoxy.(

Any(defec5ve(sensor(is(replaced(

Placement(in(

bonding&transport(jig( Bonding((approx.(15(mins/matrix)( Bonding(with(20μm(Al/Si(wire( Dispensing(of(UV$transparent(

protec5ng(epoxy(

CTA$HD(prototype(module(

After quality checks, SiPMs are wire-bonded and the PCBs are protected with UV-transparent epoxy layer

Tests of SiPM homogeneity and assembly quality

20

(21)

SiPM module readout!

Module signal readout using “TeV Array Readout with GS/s sampling and Event Trigger” (TARGET7) board

•  16 input channels

•  Analogue ring buffer of 16384 capacitors

•  Switched Capacitors Array

•  Storage of analogue waveforms in a limited period of time @ 1GSa/s sampling frequency

•  Compact chip for high density channel camera

4x SiPM module connectors 4x TARGET7 ASICS

Pre-amplifier stage pulse shaping

pole zero cancellation network two-stage AD8014

21

(22)

Backplane of the pSCT camera hosting 2 TARGET7 modules

@ Univ of Wisconsin (US)

Experimental setup for SiPM module tests and characterization

@ INFN Bari (IT)

After assembly, SiPM modules are tested simulating the whole pSCT readout chain

380 nm laser 16 SiPM module

TARGET7 module

22

(23)

Zero network R229 and C161 were replaced with R_pz = 5.6 k⌦ and C_pz = 18pF to perform a better shaping of the signal. With these modifications, performed on channel 0 of the used ASIC, 3000 waveforms were acquired. The output of the pre-amplifier of the TARGET 7 board was connected via LEMO cable to the oscilloscope where waveforms where visualized, as shown in Figure 5.2.

Figure 5.2: SiPM signal visualized on the oscilloscope at 2.5 GSa/s. On the left, the charge (blue histogram) and the amplitude (pink histogram) distribution performed online on the waveforms. The quantization is clearly visible on waveforms, too.

The charge spectrum obtained by integrating each waveform presents a clearer resolution with respect to results previously obtained, as reported in Figure 5.3.

It was therefore possible to perform a fit of the spectrum. The spectrum follows a Poissonian distribution which reflects the probability to observe pulses equal to 0, 1, . . . n fired pixels. Each peak has a Gaussian distribution due to excess noise of charge multiplication, structural di↵erences between cells in the single SiPM and noise introduced by the readout electronics.

Superimposed on the histogram there is a multi-Gaussian function to perform a fit of the data. Each peak was fitted with a Gaussian function in the form:

f_i(x) = ai e

(x µi)2

2 2i . (5.1)

Parameters from the multi-Gaussian fit where extracted for each photoelectron (p.e.); in particular, µi represents the mean value of the Gaussian corresponding to

73

Integrated signal (charge) signal amplitude

Preamplifier signal at 2.5 GS/s oscilloscope

4.3.2 Comparison between external and FPGA generated trigger mode

First measurement were performed with external trigger mode, triggering with signals provided from an ArbStudio Waveform Generator both the laser driver and the TARGET ASIC, in order to synchronize the digitization process with the pulse.

In this operational mode, the waveform is typically stored in a di↵erent frame of the 16 µs storage array at each event; as we previously mentioned, the acquired data file contains information about the first digitized cell, which is coincident with trigger arrival. Moreover, in order to correctly subtract the pedestal, it is necessary to perform a pedestal run for all the 16k capacitors before proceed with analysis.

What immediately appeared clear was an evident jitter of about 10 ns in acquired waveforms. An example of waveform acquisition is reported in Figure 4.12.

The reason for this jitter in the signal may be due to the intrinsic coupling between trigger and sampling circuit, which introduces a phase displacement between the trigger arrival and the sampling process.

Time (ns)

160 180 200 220 240 260 280 300

Amplitude (mV)

-100 0 100 200 300 400

waveform_50events_ch10_HV3700_ped_sub

(a) (b)

Figure 4.12: Waveforms of acquisitions made with external trigger. The 10 ns jitter is even more visible in saturated pulses.

When operating in FPGA generated trigger mode the signal generated from the

68

channel [10]

TARGET7 readout 37V

Modules are being characterized in terms of gain, dark rate, crosstalk…. at the end of the TARGET7 readout chain

Charge spectrum

Gain (fC)

23

(24)

Overview!

•  FBK NUV-HD SiPM technology has been tested and its performances are compatible with the requirements to equip the focal planes of CTA telescopes.

•  Multi-SiPM modules have been developed to equip a possible upgrade of the Medium Size Schwarzschild-Couder telescope prototype pSCT

•  Assembly, packaging and tests of multi-SiPM modules is ongoing, and 36 modules are planned to be installed on the pSCT camera by September 2017.

24