1. Test boxes software and first tests V. Kulikovskiy

2. DarkBox Setup CLB II WRS CLB I LASER 20 kHz 31 PMT Equalised distance optical fiber splitter (C. Mollo) equal distance PMT base cables (M.Circela/Napoli) PC

3. Software - DarkBox Detector Manager (C. Bozza) High-level control via scripts, web-based technology, DarkBox Control JAVA-App (V. Kulikovskiy) launch scripts for DM, operate DataBase, PMT tests and analysis with 1 button click RemoteControl JAVA-App (V. van Beveren) low level CLB control RemoteControl JAVA-App (V. van Beveren) low level CLB control USB shells - LM32-2 nd, WR V. van Beveren, CLBv2 group USB shells - LM32-2 nd, WR V. van Beveren, CLBv2 group DEBUG STABLE OPERATION DarkBoxAnalyzer C++,ROOT offline analysis (TT, ToT..) DarkBoxAnalyzer C++,ROOT offline analysis (TT, ToT..) CLBSwissKnife C++, BOOST C. Pellegrino raw data acquisition tool CLBSwissKnife C++, BOOST C. Pellegrino raw data acquisition tool ROYWebServer T. Gal rate visualizer ROYWebServer T. Gal rate visualizer Database PROMIS – Default HV requests (wget) XML database output CLB firmware/software (CLBv2 group) modifications for PPS (VK) CLB firmware/software (CLBv2 group) modifications for PPS (VK) Intensive testing of all components together! Heavy debugging of the whole system. A lot of support (usage, bug fixing): NIKHEF (V. van Beveren, P. Jansweier…) IFIC (D. Calvo, D. Real C. Bozza, C. Pellegrino, T. Chiarusi, A. Orzelli…

4. Dark Box Setup “CLB I” with PPS signal modified (output is LEMO, signal is 20 KHz trigger, synchronized with PPS). V2.2 “CLB I”I V2.2 no modifications Laser Pilas Advanced Laser diodes White rabbit switch, SFP avago & comm fibers Extended cables for the PMT bases (Napoli) Calibrated Optical fibers to each PMT (equalized distance). During this week only one PMT + 1 fiber was used.

5. First DarkBox Calibration and Tests 3 tough and intense days in Napoli: Setup finalization (DarkBox is so dark!) Software installation (SL6, 1.5 hard days) ~100 different data acquisitions (PMT base cable reattached ~50 times) Excellent pizza + and lost flight to Catania External help of C. Pellegrino & C. Bozza! C. Mollo, G. Riccobene, V. Kulikovskiy 24-26 Nov., Napoli

6. ToT vs TT afterpulses

7. ToT

8. 1 CLB drives laser, and reads PMTs The same channel evening/morning next day Laser + PMT to the same CLB delta T ~130+-25 ps difference between the same measurements

9. 2 CLBs + WRS syncro CLB drives Laser and detects light with PMT at channel 0 One CLB drives Laser, second one detects light with PMT at channel 0 Hit time peak value difference is 120+-35 ps (from fit) delta t(ns)

10. Fixed latency: CLB I drivers Laser, CLB II reads PMT (ch 28) CLBs restart CLBs + WRS restart mean value gauss fit

11. delta T for all Octopus channels (same PMT and setup) Large-Small mean difference is 780 ps Max difference 1.4 ns different path length? LARGE SMALL 111.157+-0.004 110.374+-0.004

12. Added measurements with other CLB + other octopuses (no laser attached to it) Scope – test CLB + Octopus channels (to be completed) The difference in delta t are smaller than 200 ps synchronization (for each channel measurement CLBs are switched OFF)? or real difference ?  negligible LARGE SMALL PMT+Laser+CLB I Laser+CLB I, PMT + CLB II

13. HV tuning Hamamtsu provides the Nominal High Voltage obtained in Current mode operation ad for 2:1:1 dynode ratio. KM3Net works in pulsed mode and at 3:1:1 Tune HV : TOT 26.4 (reference by Oleg K. – average value og 600 PMT) Scan HV from Hamamatsu NHV -100 V; +50 V with a step 25 V Base Threshold 1094 mV (2F)  0.3 p.e.

14. HV tuning Hamamtsu provides the Nominal High Voltage obtained in Current mode operation ad for 2:1:1 dynode ratio. KM3Net works in pulsed mode and at 3:1:1 Tune HV : TOT 26.4 (reference by Oleg K. – average value og 600 PMT) Scan HV from Hamamatsu NHV -100 V; +50 V with a step 25 V Base Threshold 1094 mV (2F)  0.3 p.e. Hamatsu Nominal HV HV tuned error is gaus sigma

15. TT dependence from HV 70 ns (Carlos’ previous measuremetns) 110 ns ->71.3 ns measured TT peak – -87.9 ns optical path delay – +49 ns trigger anticipation before PPS

16. Software – GreenBox (to develop) Detector Manager (C. Bozza) High-level control via scripts, web-based technology, GreenBox Control JAVA-App (V. Kulikovskiy) Rotation of the DOM AHRS data output Piezo control LED beacon contrlol RemoteControl JAVA-App (V. van Beveren) low level CLB control RemoteControl JAVA-App (V. van Beveren) low level CLB control USB shells - LM32-2 nd, WR V. van Beveren, CLBv2 group USB shells - LM32-2 nd, WR V. van Beveren, CLBv2 group DEBUG STABLE OPERATION GreenBoxAnalyzer C++,ROOT offline analysis GreenBoxAnalyzer C++,ROOT offline analysis CLBSwissKnife C++, BOOST C. Pellegrino raw data acquisition tool CLBSwissKnife C++, BOOST C. Pellegrino raw data acquisition tool ROYWebServer T. Gal rate visualizer ROYWebServer T. Gal rate visualizer Setup creation (Bologna?) Setup creation (Bologna?) XML database output CLB firmware/software (CLBv2 group) modifications for AHRS operations (VK) CLB firmware/software (CLBv2 group) modifications for AHRS operations (VK)

17. Conclusions DarkBox has a great mechanical realization – darkrate << 1 kHz – easy to operate – fast test of 62 PMTs is proven to be possible and fast – it already unique measurements of the channel delays already GreenBox – Mechanical realization of 2 boxes is ready – Almost all the software can be reused from the official ones for the future detector