NPS Frame and Infrastructure Meeting (JLab, 6/25-6/26 2019)
Participants: V. Berdnikov, G. Hull, T. Nguyen Trung, S. Lassiter, M. Fowler, P. Medeiro, C. Munoz-Camacho, E. Rindel, T. Horn, J. Segal, B. Sawatzky, A. Camsonne, J. Crafft
[ PMT module prototype photos]
- PMT module assembly: design a special tool to pull out the PMT?
- Some concern about what to pull out (initial thought by hand/finger and sturdy enough for that) - after settling perhaps parts are more stuck together, would connector or base come off?
- Calorimeter: consider including a special transportation frame when entire assembly may be tilted, e.g. during tranportation down truck ramp to experimental hall
- Fiber side PCB: Question: if fiber in place and remove crystal, can crystal move and break fiber?
- Detector box:
- exchanging PMTs on sides: no line of sight, unscrew by feel good enough?
- bottom of box: leg room? Have only about 18inches, box sits on a platform, bottom plate removable; how to get more space? - modified platform? region between rails with more room? Needs to be checked
- Chiller for cooling crystals: where to put chiller? - length of cooling lines? size of box? - needs to move with and be protected from radiation - send specifications to S. Lassiter, M. Fowler, P. Medeiro
- Assembly space - check with Walt Akers: requirements: space with crane acces, electrical power, high voltage, DAQ, table space, quasi clean room/tent, quiet, temperature controlled (AC), how long? summer 2020 (June?) start, ask NPS collaborators to come Sept 2020 for assembly tasks
- Cables - external HV: request to include option to bypass safety system for case if just test one channel or one column/row and want to bypass all others - maybe use switch or a jumper?
- Cables - LED: length needs to be defined
- Cables - LEMO connectors: request to JLab to purchase connectors for top of NPS box (Carlos can send part numbers), connector can be mounted on board, cable connected to connector, plug inside, mounted on top, when mounted cannot remove stuff from top
- Cables - routing: estimate size of cable bundle - fit on spectrometer and movement with it; need 1080 signal cables strain relief scheme (not to damage LEMO conn); need to work out LEMO cables - RG58 comes off, goes to controller in SHMS, BNC at patch; how much clearance on top of box; idea is to run all cables together (plus LED cables) to SHMS hut - relatively close; slow controls: temperature sensors (number?); need a plan for plug/unplug connectors? two spare cables to have ability to connect one column to another; timeline
- Presentation mechanical design
- new system for fibers
- no change to cables
- PMT assembly: design final, one long captive screw, some modifications for support of fiber - metal support
- Wrapping: horizontal thickness (260um (2x70 + 2x60)), assume no air between crystal and reflective sheet, vertical: 260um (), front face: reflector on plastic frame/connector
- Pitch aveoli: gap aveoli and crystal, assume 260um +50um wrap, SICCAS: 20.85, CRYTUR: 20.77, so pitch 21mm inner - gap between two crytals 1 mm, gap 0.15mm (SICCAS), 0.23mm (CRYTUR)
- Shearing on one tower with no gap underneath:
- simulation assuming ~0.7kg/crystal, 1.4MPa/each stage, shear resistance carbon plate 90MPa, security coeff 64 - very safe
- simulation (ANSYS) assuming all load supported at bottom plate: 125N down, 36P/2, Max stress 48MPa, Shear res 90MPa, security coeff 2 ->safe but not secure -> need to reduce gap under vertical plates
- tests in lab:
- load on tower from top on 3x3 prototype with crystals in it, 76kg (3 columns weight), gap under vertical carbon plates -> no damage
- alveoli deformation with one crystal: simulation (ANSYS), 4 measurement points, no significant deformation
- buckling: 1.35 (>1, so ok)
- remove lower row of crystals -> not a good idea, get shearing since load is not perfectly vertical from top - note in actual, full design crystals share the load - fine in principle since not concentrated load, but distributed (also note HYcal had no problems with stacking)
- Fiber side crystal: metal ferrule screwed on alumin support at PMT/crystal junction joined to silicon fiber sleeve for fiber, fiber: quartz (stiff)
- Fiber side PCB: between 2 Al plates, protected with plastic tube, inserted NOT glued in plastic ferrule, plastic to not damage fiber, fiber length: 51cm; fibers can be dismounted, PCB can be dismounted if necessary
- Interface PCB: have 40 PCBs, cables now routed along top rather than side as before; have done mock up with real dimensions, PCB shifted 15cm vertical of PMT column
- Electronics input/output:
- Cooling: crystals, external, heating zones; use chiller + two fans (top/bottom like CCal)
- Discussion of tasks list:
- Orsay: provides components and assemblies, e.g. cutting reflective sheets, prefolding, wrapping plastic front part, glue mu metal with plastic support, block PMT assembly; cutting/polishing fibers vs purchase (~21eu/meter) - use bundle of ~100, cut in insert tube then polish all same time
- JLab/NPS collaboration: assemble remaining components, e.g. wrapping crystals (initial estimate 12min/crystal -> 216 hrs, can likely be reduced), insert column PMT assembly and PCB; Mounting: insert cooling, finalize mounting, mounting (Hall C techs); Cabling: Brad, J. Beaufait et al.
- Presentation high voltage
- PCB feasibility connectors - small sizes not possible, HV=1100, best are oval pads for soldering and grid setup
- 2 layers, 1.6mm thick, 780x214mm2, solder
- Safety considerations for HV: use safety loop - pin42 connect to ground (if open switch off), requires modification of PCB and connect pin to external switch at detector door; CCAL uses additional light sensor tied in with humidity control; make a note
- Presentation: Cables and connectors
- between PMT base and interface board: length chosen based on prototype, 295mm length cable HT, 290mm length cable anode - have prototype cable with connectors (Jack has it now!); discussion about crimp vs solder - detailed instructions available - these cables will be made at JLab
- external HV: length to be defined, depends where power supply is located, scheme from distribution board - present channel schematic (channel to channel or ground), connectors=Samtec
- Cable for LED: inside box by flat cable, present channel/pin schematic - in this layout even pins=ground, cable length to be defined