1. Omega Users 2/07/2009 Short Status Report ANR-06-BLAN-0186 (LAL, IPNO, LAPP and Photonis till March 09) J.E Campagne project leader

2. 65m 60m Liquid Scintillator (→ 50 kton) 100m 30m LENA MEMPHYS Liquid Argon (≈10→100 kton) 70m 20m GLACIER New large Underground Detectors Similar projects in USA (DUSEL) and Japan/Korea (T2HK) In EU (FP7): LAGUNA & Euro

3. J.E Campagne Omega Users 2/07/2009 About 170 γ/cm in 350 < λ < 500 nm With 40% PMT coverage, Q.E.≈20% Relativistic particle produces  ≈14 photoelectrons / cm  ≈7 p.e. per MeV Up to 400 kton fiducial mass 1 shaft ≈ 4xSK ≈ 215 kton H 2 O MEgaton Mass Physics: MEMPHYS

4. J.E Campagne Omega Users 2/07/2009 PMm 2 ANR (2007-2010) LAL: front end electronics + water tight box LAL: front end electronics + water tight box IPNO: photodetector tests + mechanics + integrated electronic board IPNO: photodetector tests + mechanics + integrated electronic board LAPP: Data network (electronics + cables + protocol) LAPP: Data network (electronics + cables + protocol) Photonis: PMTs provider Photonis: PMTs provider Funded 500k€/3yrs (1 post- doc included) designed to involve 5FTE Engineers And in fact ~20p participate to our “monthly” meeting. k€ http://pmm2.in2p3.fr Recently ULB Bruxells (K. Hanson) has joined for DAQ

5. J.E Campagne Omega Users 2/07/2009 Concept: digitize everything! Concept: digitize everything! (supported by low energy physics) Triggerless Front-End from preamp. to Q & T digitization Common HV FPGA Integration Board 16 2m x 2m 12’’ PMs FPGA  Controler Power + Decoding Board 100ml submarine cable power + data TCP/IP 10bars water pressure IPNO & Photonis LAL & IPNO LAPP ULB Size: manageable 12’’ PMs: no manual operation for large production, better ratio cost/QExCE

6. J.E Campagne Omega Users 2/07/2009 Photodetector XP1812 12’’ 10b IPNO shape and glass thickness computation to subtend 10 bars Photonis Electrostatic and realisation of 9 pieces (Sept. 08) First measurements: Photocathode: uniformity quite good Skcb=10.5µA/lmF (with corning blue filter) Skb=77µA/lm (with white source) ~24% at peak => reasonable for first attempt (nb: 8’’ XP1886 new process bialkali super² gives 36% at peak) Gain: 10 5 @ 2kV => low value compared to expectation 10 6 (8 stages) Redesign of the first dynode PHOTONIS had halt the PM activity March 09 Potting tested

7. J.E Campagne Omega Users 2/07/2009 High pressure tests BNL tank (150b)Mid-Feb 09 3’’2’’1,5’’ 3,2 atm 6,5 atm 16,3 atm Preliminary comparison simulation/measurements indicates Beaking limit of glass : 40 - 50 MPa In agreement with glass properties given by Manufacturer. This is a good news to design a 12’’ shape and choose glass thickness to subtend 10 atm. (first test with 12’’ end march 09’) 8’’ XP18060 for Daya Bay (reinforced) does not break at 10 atm IPNO

8. J.E Campagne Omega Users 2/07/2009 PARISROC  Target features:  16 independent channels  Variable gain to allow single HV for all PMs  1/3pe 100% efficiency  Triggerless  1ns time resolution  Low cost LAL

9. J.E Campagne Omega Users 2/07/2009 PArISROC v1 architecture CRRC2 Slow Shaper (50, 100, 200 ns) Fast Shaper (15ns) 16 charge inputs 1 mux. charge output Track & hold Discri Variable Gain Amplifier (1-8) Gain Correction (8bits) DAC 10 bits variable delay Bandgap Vref SSH Vref FSH Vref SSH Channel 1 Channel 16 Read Hold ADC discri 24 bits counter 10MHz Threshold DAC 4 bits 16 Trigger outputs OR 1 OR output Track & hold ADC discri 12 bit TDC ramp 12 bit ADC ramp ADC conversion 40MHz Read-out 10MHz registers Serialized dataout (digitizedcharge and time) Digital part SCA management LAL

10. J.E Campagne Omega Users 2/07/2009 PARISROC v1 Technology : AMS SiGe 0.35  m Size : 5mm X 3.4mm Package : CQFP160 ADC ramp 16 analogue channels Digital part TDC ramp BandgapDual DAC Bias + supply LAL Submitted July 08

11. J.E Campagne Omega Users 2/07/2009 20ns 5ns  input signal : Ii= 0 to 5mA  0 to 300 pe  0 to 50 pC  PM’s Gain =10 6 (1pe=160 fC) Measurements LAL Start Dec 08’ default value to present the results

12. J.E Campagne Omega Users 2/07/2009 PARISROC test board (S/N #3) PARISROC test board (S/N #3) Analog signal read on high impedance on the oscilloscope, the adaptation being achieved by the PARISROC input Analog signal read on high impedance on the oscilloscope, the adaptation being achieved by the PARISROC input Probe signal: can be the fast or the slow shaper. Software controlled Probe signal: can be the fast or the slow shaper. Software controlled Discriminator output: chip output, measured with an oscilloscope probe Discriminator output: chip output, measured with an oscilloscope probe 10-inch Ø PMT for « standard » measurements, 1-inch Ø PMT for pressure / time measurement 10-inch Ø PMT for « standard » measurements, 1-inch Ø PMT for pressure / time measurement PARISROC Oscilloscope Probe Discriminator output 1 M  10-inch Ø PMT Photonis XP1804 S/N 1086 IPNO+LAL PARISROC v1 (1st tests) 1-inch Ø PMT

13. J.E Campagne Omega Users 2/07/2009 Slow shaper test Nearly the same SER * shape obtained with the oscilloscope and the slow shaper signal fit. 10-inch Ø PMT Photonis XP1804 gain = 2.4×10 7 Noise : around 1 kHz at room temperature IPNO+LAL *:Single Electron Response

14. J.E Campagne Omega Users 2/07/2009 Discri. test PMT signal and discri. Signal leading-edge fitted PMT signal and discri. Signal leading-edge fitted Resolution = 600 ps RMS (> 0.3 pe) with a low threshold on PARISROC discriminator (220 DAC channels). Resolution = 600 ps RMS (> 0.3 pe) with a low threshold on PARISROC discriminator (220 DAC channels). Threshold = 220 DAC channels 0.3 pe 1 pe IPNO+LAL 1-inch Ø PMT Photonis XP3102 (10 stages) special design 2.4×10 7 at 1200 V Small PMT  much faster than the 10-inch Ø PMT + lower noise (≈ 100 cps) = better for time resolution measurement

15. J.E Campagne Omega Users 2/07/2009 FPGA To Power & Decoding Board PARISROC USB LOCAL TEST IPNO High Voltage generator ±0,2% Residual The strong support of the IPNO Electronic Service

16. J.E Campagne Omega Users 2/07/2009 16 195mm 250 mm Water tight GLENAIR overmodling DAQ sub-marine cable 16 HT/signal cables GND CARTE Carte fille PARISROC GND cable DAQ RJ45 48V/12V 12V/2kV RJ45 Bandes de contact De masse Implementation in the water tight box (10bars) LAL Integration board implementation Very crude « mecano » design

17. J.E Campagne Omega Users 2/07/2009 Cables and Data Transmission LAPP 48V Power & Decoding box 100ml

18. J.E Campagne Omega Users 2/07/2009 Power & Decoding board LAPP 48V DC/DC 48V/3.3V Transfo POE Cable to water tight box FPGA Mezzanine board µC Ethernet RS485 2 twisted pairs Clock Data bi-direct. 5Mbits/s 5kHz/PM dark current POE: Power (30W/48V) over Ethernet

19. J.E Campagne Omega Users 2/07/2009 Cables Bi-directional Cat. 5 type of data network cables Bi-directional Cat. 5 type of data network cables At least 2 twisted pairs (may be revisited) At least 2 twisted pairs (may be revisited) 100 ml 100 ml The only connector we accept to ease the mounting and maintenance. The only connector we accept to ease the mounting and maintenance. LAPP Hydrocable with Souriau connectors Prelim. price large prod. 400-500€ (cable + connectors) EUROCEANIQUE with GISMA connectors HYDROCABLE with GLENAIR connectors

20. J.E Campagne Omega Users 2/07/2009 Cable tests with connectors LAPP Some x-talk inside the connector No change w/ or w/o connector Attenuation loss Better w/ connector Return loss Geometry of the cable impedance ≠100 

21. J.E Campagne Omega Users 2/07/2009 Injection up to 16 channels ( test input) Pulse period 200µs Transmission tests during several hours Ok for 1 complet set. Next steps: 1)Use 1 PMs as « noise pulser » 2)Use several PMs 3)Use the TCP/IP network for the DAQ 4)… Integration Board (PARISROC) PMm² Surface_Board DAQ Data + Power Clock PM/pulse Generator USB 100m Submarine cable 48v PMm² DAQ tests IPNO+LAPP+LAL

22. J.E Campagne Omega Users 2/07/2009 2m Water tight box Black panel for reflective insulation Mounting unit Compatible with MEMPHYNO water vessel (join LAL- IPNO-APC and ULB) Demonstrator Design Ok Construction Q4 2009 16 1’’ PM demonstrator in the 10bars BNL-IPNO vessel

23. J.E Campagne Omega Users 2/07/2009 Coming next… PARISROC PARISROC –Next version submission scheduled Nov 09’ Electronics Demonstrator @ 10bars (BNL vessel) Electronics Demonstrator @ 10bars (BNL vessel) –16 x 1’’ Photonis PMTs –All the present versions of the PARISROC, submarine & surface boards, TCP/IP DAQ Large Photodetectors: Large Photodetectors: –We have bought Hamamatsu 8’’ PMTs for the demonstrator  July 09’ –To transfer to the how-know to use these new PMTs will take some times –Finalize the glass & PMT shape breakdown studies Final Demonstrator Final Demonstrator –Design will be finished July 09’ –Start building Sept 09’ –Will wait for Large PMTs delivery –Tests with all the electronics + DAQ will start early 2010. ANR has accepted 6 months continuation of the R&D due to Photonis halt  end July 2010. ANR has accepted 6 months continuation of the R&D due to Photonis halt  end July 2010.

24. Omega Users 2/07/2009 Backup

25. J.E Campagne Omega Users 2/07/2009 @ source ~ 670 bars !! @ 40 cm ~ 50 bars !! Pressure @ 40 cm on the neighbor PMT ∆ 50 bars !! 47,5cm 12’’:31cm Implosion simulation IPNO CONSOL software + Semi-analytical approach Preliminary validated with exp. data Need decoupling Not investigated by PMm 2

26. J.E Campagne Omega Users 2/07/2009 Cf Input: 10pe G_pa=8 10mV/div 10ns/divPreampmeassimVout(1pe) 5 mV 5.2 mV Rms noise 1 mV 316  V Noise in pe 0.20.06 SNR516 Some clock noise and low frequency noise under investigation Gain adjustment linearity : Vout/vin = Cin/Cf 8 bits range 2% non linearity Vmax_pa vs 1/Cf Preamplifier measurements LAL

27. J.E Campagne Omega Users 2/07/2009 Fast shaper & discriminator Fast shaper meassim Vout (1pe) 30 mV 37 mV Rms noise 2.5 mV 2.5 mV 2 mV SNR1216 Rise time 7 ns 5 ns 10 pe input Spread 0.5 pe Noise 0.1 pe Overlapped curves Trigger efficiency : Pedestal Spread < 0.1 pe LAL

28. J.E Campagne Omega Users 2/07/2009 Slow shaper G_pa = 8 Slow shaper meassim Vout (1pe) 12 mV 18 mV Rms noise 3.5 mV 3.5 mV 1 mV SNR3.518 Rise time 28 ns 27 ns Low frequency to be understood and removed (power supply noise ?) LAL slow shaper held slow shaper T&H

29. J.E Campagne Omega Users 2/07/2009 Internal 12bit Wilkinson ADC DAC=500=1.4523V Number of acquisitions: 10000 ADC_UNITS =1961 ΔADC_units = 4 0,2% accuracy ! LAL

30. J.E Campagne Omega Users 2/07/2009 Internal Wilkinson ADC Wilkinson ADC well suited to multichannel conversion Wilkinson ADC well suited to multichannel conversion Very good uniformity and linearity Very good uniformity and linearity Uniformity of 10bit Wilkinson ADC 16 channels superimposed ! Linearity of 12bit Wilkinson ADC LAL

31. J.E Campagne Omega Users 2/07/2009 Overall behaviour Complete chain : autotrigger + T&H + internal ADC Complete chain : autotrigger + T&H + internal ADC LAL

32. J.E Campagne Omega Users 2/07/2009 PARISROC v2 LAL PM noise # buffers per channel ADC bits number Readout frequency loss (%) 5 kHz 212 10 MHz 16  12 0.1 kHz 212 10 MHz 0.008  0.6 5 kHz 28* 40 MHz 0.24  3.1 5 kHz 49* 40 MHz 0.007  0.7 V1 V2 After discrimination p.e can be lost by the digital part of PARISROC depending on the dark current rate of the PMs and readout parameters. simulation

33. J.E Campagne Omega Users 2/07/2009 Ramp TDC, Ramp ADC and 10-bit DAC common to all channels 50  ext PA 4-bit DAC Threshold 10-bit DAC Trigger Output Discri. FSH SSH VARIABLE DELAY OR Auto-trig Adjustable gain amplifier Adjust = 1-4 PA SSH SCA high gain SSH SCA low gain Ramp ADC Ramp TDC Time output Charge output SCA fine time ext. Hold.Hold 10-bit DAC Threshold gain Output MUX Added blocs 2 gains per channel 0-10pe & 0-300pe 8-9-10 bits ADC/TDC PARISROC v2 LAL

34. J.E Campagne Omega Users 2/07/2009 Digital improvments Management of 2 gains for charge (16 inputs to add) Management of 2 gains for charge (16 inputs to add) Multiplex SCA depth 1 and SCA depth 2 gain (analog part) Multiplex SCA depth 1 and SCA depth 2 gain (analog part) Store Gain in 1-bit register Store Gain in 1-bit register Conversion 8-9-10 bits by selecting ramp slope Conversion 8-9-10 bits by selecting ramp slope Internal counter 10-bit (Parisroc1: 12-bit) Internal counter 10-bit (Parisroc1: 12-bit) Max conversion time : 25 us @ 10 bits (Parisroc1: 100us) Max conversion time : 25 us @ 10 bits (Parisroc1: 100us) Readout @ 40 MHz Readout @ 40 MHz Max readout time 25 us (Parisroc1: 100us @ 10 MHz) Max readout time 25 us (Parisroc1: 100us @ 10 MHz) Readout format (50 bits / hit channel unchanged) : Readout format (50 bits / hit channel unchanged) : (MSB) 4 bits channel # (MSB) 4 bits channel # 24 bits timestamp 24 bits timestamp 1 bit timestamp counter 1 bit timestamp counter 1 bit gain 1 bit gain 10 bits charge 10 bits charge (1 bit “TDC ramp 1 or 2 converted” if needed) (1 bit “TDC ramp 1 or 2 converted” if needed) 10 bits time (LSB) 10 bits time (LSB) LAL Chip submission Q4 2009 Design will end July 09

35. J.E Campagne Omega Users 2/07/2009 Power & Decoding board LAPP FPGA synchronisation and dialogue with submarine electronics (RS485) MOD570 Connexion with external networks (TCP/IP) GPS 10MHz 10MHz Internal clock External power supply 48V/50W The dialog between this board and the Integration Board is home made (simple serial protocol with 32bits words) transmitted coded in Manchester (Phase Encoding) on RS485 diff. pair. Data integrity test @ 8Mbits/s during 2h: no failure

36. J.E Campagne Omega Users 2/07/2009 Cables –All the contractors even if they have very good references (space qualifications, nuclear plants…) they have made unconceivable mistakes! –This implies a close follow up of the quality all along the production, and by consequences this would be included in the price… –But they can improve also their products as we have experienced too with Hydrocable for instance.