NPS Frame and Infrastructure Meeting (JLab, 11/13-11/15 2018)

From Cuawiki
Revision as of 23:27, 19 November 2018 by Hornt (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search


  • Photographs of the IPN-Orsay 3x3 prototype


  • Test radiation hardness of carbon fiber and insulator (e.g. DEPRON) samples - concern with carbon fiber is mainly the epoxy inside
  • Check NPS PMT requirements - question is if CAEN HV cards with 1 mA/channel (vs 3 mA/channel) are sufficient
  • UPDATE: The total resistance of our divider is about 3.6 MOM. Hence, at operating voltages of 1.1-1.5 kV the divider current would be in range of 300-400 uA. --> 1mA/channel is sufficient.
  • Sliding PBC boards: mechanical stability? - completely parallel movement. Check what angles can be reached if assume detector needs 10cm on one side and 20cm on the other. Also, is vertical limit given by the second floor SHMS and/or LWC pipe a problem when detector is on the large angle side?
  • Signal cables: check space needed for BNC vs. LEMO connectors on top/side of NPS. Would BNC work or need to use LEMO? Also, check on flat micro ribbon coax cables - any drawbacks, e.g. signal distortion?
  • UPDATE ON FLAT MICRORIBBON CABLES: the most recent and largest implementations using flat micro-coaxial cable assemblies at JLab is for the CLAS12 MVT detector. These assemblies are 2m long and have drawbacks such as: Cost, specific connection schemes, and almost impossible to re-work or fabricate on site. These attributes are common for most other types of flat cable coaxial cables plus one has to be aware of the signal attenuation characteristics for the given drive signal.
  • High voltage cables: assume CAEN 52 conductor multiwire cable - need to make sure that proper strain relief implemented at detector
  • Crytur crystals are slightly smaller than SICCAS crystals - measure 40 SICCAS and 20 Crytur crystals for tolerances
  • Heat removal - temperature gradient seems large - how to test and control?
  • Would it be feasible to run some cheap poly (1/4"?) tubing off a (chilled?) air manifold into the gaps between PMTs near the central region? It wouldn't take much air flow to dramatically improve the cooling, so I expect you get get away with a relatively small number of tubes. Alignment isn't important, so I would imagine just threading them in where desired and cable-tying them in place after everything else has been installed.
  • Check how HyCal or ComCal cooling system would work with NPS
  • Test silicone cookies, e.g. Sylgard 527 instead of optical grease for PMT-crystal interface
  • Fiber-crystal interface: add optical grease here too?
  • Light monitoring system: radiation hardness of LEDs and fibers? Possible to distinguish possible radiation damage of LED and fibers from radiation damage of crystal?
  • Comments on 3x3 prototype
  • All removable metal components on the posts (washers, spacers, and the nuts) should preferably be non-conductive nylon/delrin equivalents
  • The spring probably needs to remain metal, but it should be mechanically captured on the rod (perhaps crimp the spring or wrap some wire around it at the base so it can't come off)
  • Screws/threads - need to stay on axis
  • Special tools
  • Tape in front of crystal
  • Tedlar in addition to reflector?



  • High voltage cables
  • Inside NPS: Bogdan's special connector
  • Top and outside multi-channel connector - standard CAEN 52-pin cables seems best
  • would need 20-21 multiconductor cables (might want some spares as sometimes conductors don't work)
  • Important to implement proper strain relief at the point where 52 wires from PMTs are coming together
  • Signal cables
  • Inside NPS: short flat micro ribbon coax cables should work - outside, longer does not seem good (see comment in action item list on 2 m long cables)
  • On top of NPS detector: use BNC or LEMO connector depending on space availability
  • A movable patch panel will be located near the NPS that moves along with the detector, from there would route high density cables (BNC)
  • SHMS carriage patch panel is BNC
  • Crates HV - plan to order for Hall C in FY19
  • Discussion about maximum current requirement per channel - needed for decision on 1mA or 3mA CAEN HV card
  • Weight per foot of cables?
  • Cabling cost discussion - LEMO is a factor of 2-3 more expensive, but saves space


  • Expect 540W from heat zone, a relatively large gradient - discussion about heat removal options
  • Insulator under consideration: DEPRON (polystyrene insulation material, temperature range: -60 -- +70 degC, ignition temperature: 355 degC)
  • Check radiation hardness of this material - could place a sample in Hall C
  • Concern mainly about the epoxy in carbon fiber plates - there is no glue used at front of detector
  • Note that have carbon fiber (vendor: Dragonplate) in SHMS aerogel tray SP11


  • Access to NPS normally from back, also access from one side
  • Cables are routed through the top
  • Consider sliding PCB board for access, e.g. change a PMT
  • discussion about mechanical stability, complete parallel movement, and clearance with beamline and SHMS/HMS
  • Need ~10 space for access on one side and ~20cm for access on the other - what angles can one reach with that?
  • on HMS side, need also consider vertical clearance with SHMS second level and/or LWC pipe


  • Based on previous detectors, difficulty with LMS is continuous operation
  • Discussion about radiation hardness of LEDs and fibers
  • might check with fast electronics group on materials that are radiation hard
  • How to disentangle radiation damage of LED and fibers from that of crystals? - can one make an estimate of level of precision needed and see if one can still distinguish?
  • Currently using optical grease for PMT-crystal connection - considering to explore silicone cookies like Sylgard 527 (DOW-Corning, silicone gel, 0.1-0.6mm thickness) or similar


  • Assembly will be from front - requires inserting of carbon structure layer by layer, and then crystals into the compartments
  • 3D printed carbon fiber parts hold assembly in place, springs to tighten
  • crystals wrapped in VM2000, no tedlar due to space limitation - perhaps more space with Crytur crystals
  • Discussion about optimization of components
  • removable parts - nonconducting material to avoid shorts if parts accidentally dropped into detector during access
  • special tools
  • About half to one centimeter aluminized mylar tape wrapped around crystals at front and back end to facilitate sliding into place - any issue with that?
  • Impact of not having tedlar around crystal in addition to VM2000?
  • Estimate tolerances for crystal-to-crystal space