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

    User name R. Michaels

    Log entry time 10:09:09 on April23,2009

    Entry number 267496

    This entry is a followup to: 267287

    Followups:

    keyword=R-HRS DAQ configs redone

    The R-HRS DAQ configs have been "finalized" and are
    similar to the L-HRS (see halog 267287). This is
    on adev account on adaql1.

    RightHrs -- Default config for production. Runs in buffered mode
    and includes HAND.

    RightUnbuff -- Same as above, but unbuffered.

    RightOnly -- Excludes the HAND crate. Unbuffered at the moment.
    (easy to make it buffered if that's what we want)

    pedrun -- The usual pedestal run, generates pedestal files.
    Note: this only generates ped1.dat and ped2.dat for ROC1 and ROC2.
    It does not generate them for ROC8. One may generate those by
    hand or we may modify this config to do that at some point.

    Additional notes:

    1. ROC8 (HAND) does not have pedestal suppression. It may therefore
    dominate the deadtime. There was a similar note in halog 267287
    about L-HRS where the BB crates are not ped suppressed. For some
    reason I anticipate that people will ignore this and ask "Why is
    the DAQ deadtime so high ?". :) :)

    2. The RightHrs config does not actually work yet because apparently
    the FrontEnd busy signals are not all plugged in. You get messages
    from Event Builder like below. Deja vu to L-HRS and halog 267287.
    Use RightUnbuff if you are afraid. It has no such errors.

    3. All SPLITMB set to 4 Gbytes, on both HRS.

    -----------------------------------
    Error message in buffered mode: (probably will be cured
    with proper FB, see note 2 and halog 267287):

    daLogMsg: Event (Num 224 type 8) SYNC mismatch -- ROC mask = 0x00000506 , SYNC mask = 0x00000100
    Event (Num 68 type 8) SYNC mismatch -- ROC mask = 0x00000506 , SYNC mask = 0x00000100