Main INDEX
Monthly INDEX
PREV
NEXT
User name R. Michaels
Log entry time 17:38:09 on May 4,2003
Entry number 99855
keyword=helicity errors as indicator of deadtime burps
Once again the helicity is an indicator of a problem. In runs 1577 and 1578 I see
some "deadtime burps" which are periods of ~0.2 to 0.5 sec when the DAQ does
not take data. Don't know why it happens, but would suggest you keep the rates
at 4 kHz or below, and the deadtime at 10% or less. Also keep to a MINIMUM
the number of processes including GUI's that run on adaql2, even if you are logged
in somewhere else and just using the terminal. For the time being, one can assume
the complaints from ESPACE (about "inconsistent" or "impossible" helicity) are
probably an indicator of these burps, and monitor if they happen frequently.
With the present helicity algorithms, the consequence of these burps is a loss of
potentially 3 seconds each time, to reload the helicity. It probably doesn't have to
be that bad, since : 1) The helicity data are ok; 2) The DAQ sub-components are
in good synchronization; and 3) We have several redundant clocks in the data.