• Main INDEX
  • Monthly INDEX
  • PREV
  • NEXT
    Make New Entry, Make Followup Entry

    User name R. Michaels

    Log entry time 02:10:36 on February 12, 2009

    Entry number 260569

    This entry is a followup to: 260563

    keyword=re: 1875 missing gate

    Since unbuffered seems to suffer no missing gates, it may be wise to run like that
    for the time being on L-HRS.

    Two possibilities come to mind about this 1875 which Vince found is the culprit:

    1. Maybe it is programmed wrong. But looking from home this doesn't seem to be the case.
    One may compare it carefully to another 1875, see if it's CSR is ok, it's counter incremented,
    etc. Like I said, I think it's ok ...

    2. More likely, it's busy signal is not communicated correctly to the Trig. Super. This is
    the slowest module in the system (frontend busy 50 usec, compared to 1877 and 1881M
    which both have 10 usec). Meanwhile, there is a "safety" margin of 20 usec, see the line
    ts->timer[3] = 500 in ts_scaler.crl. But this only corresponds to 20 usec, so one may try
    (at least as a test) to bump this up to ~60 usec and see if the problem goes away (I'd bet
    $5 it will). Next, one may check the busy signal gathering at next access. I thought I checked
    this but maybe something went wrong there.

    A copy of this log entry has been emailed to: brads@jlab.org,vasulk@jlab.org