Preparation for 2021 Run

From GlueXWiki
Revision as of 15:48, 6 May 2021 by Somov (Talk | contribs)

Jump to: navigation, search

Below is a list of action items which we would like accomplish before the PrimEx Phase II run in Spring/Summer of 2021

Aligning FCAL in the Hall -- Coordinated by Tim Whitlatch
Moving FCAL platform by about 5 mm in X-direction to make sure that the beam goes through the middle of the FCAL hole (3x3 modules). The exact position of the platform should be confirmed by the alignment group

Preventative maintenance of FCAL PMT bases -- Coordinated by Mark D, with support from Chris S., Nick S., Fernando B., Matt S., Ben R., Paul S.
  1. Change out bases which draw excessive current (> 13 mA) when they are on or which show other signs of problems. High current bases are associated with increased temperature, increased deviation from setpoint and future high voltage failure. There are about 200 high current bases. They are repaired using the standard procedure for cleaning and replacing resistors on the base.
  2. Diagnose bases which draw excessive current (> 13 mA) when they are on and off. This curious symptom is not understood. [1]
    • Status:
      • 8 bases have been remove from the FCAL for special diagnosis. These plus 2 more have arrived with Paul at IU with details of the problem symptoms.
      • Report from Paul. Base 23286 drew 14 mA when powered up. The COMM board by itself draws 14 mA, so the problem is there. It was not immediately obvious what is wrong with the COMM board. The other 7 had current draws > 20 mA when powered up. In all cases the COMM board by itself had most of the excess current draw. In all 7 chip U9 on the COMM board was warm or hot to the touch. This chip drives the test pulser on the SOCK board. If there is a spark at the PMT socket the test pulser transistor is destroyed and U9 is damaged.
  3. Improve robustness of firmware for handling communication failures. When the beam is on we regularly have about 2-3 bases that lose communication per day. The HV drops to zero and the only known way to recover is to power cycle all the bases at the end of the run. This means that we permanently have some holes in the FCAL from perfectly working bases. We will try to modify the firmware to fix this issues. Ben R., Mark D.
    • Plan
    1. compile the existing code and load onto a test unit to make sure we can reproduce what you currently have
    2. have it communicate over the CAN interface like normal
    3. try to crash the micro-controller with a radiation source
    4. make a small firmware change to enable the watchdog timer and try to crash the micro-controller again with source (we should plan to track/observe if a watchdog reset occurs which should match the rate of lockups from step 3 if conditions are the same, but in this case it should self recover)
    5. look into remote/boot loader firmware update code & power to try and optimize.
    • Status
      • 2020-10-08 Ordered a RLINK-STD STM8 series Debugger, Programmer (In-Circuit/In-System) to program the base firmware.
        • 2020-10-15 The RLINK has arrived and has been delivered to Ben.
      • 2020-10-16 Adam Hartburger toured EEL118 and began the paperwork to give us a test source.
      • 2020-10-22 Source testing has begun.
        • 2020-11-06 More than a week of study have not been able to reproduce the problem.
      • 2020-11-11 No drops observed, source returned.
      • 2020-12-03 The firmware changes were made (v1.5) and Nick will be testing in the PMT base setup he has in EEL. We were able to load firmware and see the new version reported. Some bench top test shows the watchdog timer works and resets the unit when the CPU is forcefully halted (using a debugger to trash the CPU) state).

Now Nick will add new commands to the backend software (to get status and be able to test some various forced reset conditions). Once we see the various features working it will be good to port/test on whatever software is used in the Hall D environment (and look into remote firmware updates if needed to lower power to make that work).

  1. Evaluation of bases in inner rings for Compton events
    • 25 bases have already been replaced in this area. Status of FCAL bases in 3 inner rings Colin
    • A possible 47 more could be replaced, we should evaluate which of these are showing any issues. Mark D.

CCAL active bases
- Refurbish CCAL active bases: -- Chris S., Fernando B.

- Chris has already prepared a new PCB design for the active base
- ( replace bases: A.Somov, J.Crafts, V. Berdnikov )

Increase LED light pulse on the CCAL LMS system
- replace (use multiple LEDs) on the LMS system (A.Somov, J.Crafts, V. Berdnikov)
- Driver(s) for multiple LEDs Chris S., Nick S., and Fernando B.