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

    User name R. Michaels

    Log entry time 14:50:02 on November 5,2010

    Entry number 338669

    Followups:

    keyword=helicity issues, cont.

    First note, the gate-veto appears to have been not plugged
    in since yesterday. Julie will try to restore this during access.
    (Talking about gate on scalers)

    Looking at the pattern of failure in the ring buffer,
    I think the following occurs. I don't know why it occurs.
    But if this is a robust pattern it may suggest a recovery.

    "Bad" data is defined by a clock value that's not close to
    normal (normal = ~830 or 865 depending on if gate veto plugged in).

    The first bad data that appears to have correct values for
    the QRT and helicity. Sometimes there are 2 or more bad
    data in sequence; the ones after the first have bogus QRT/helicity.
    I think they are all zero.

    The first subsequent good data has bogus QRT/helicity, but might
    have good data; that's hard to tell but at least data looks normal.
    After that, the data are all ok until the next episode.

    So, the recovery might be: for the 1st bad data, take QRT/hel
    and save it. For the first following good data, assign the
    QRT/hel that you saved to this.

    I was starting to try to implement this for the online packaging
    but there are too many other things going on with the DAQ this
    afternoon -- Alex and Dave Abbott doing stuff -- so I gave up.