CRRES LIVE TIMES
There are 16 rate ACCUMULATORS; these guys count 1 by 1 and
collect counts for each rate during almost all of a 4.096 second
long interval.
Every 4.096 seconds these 16 ACCUMULATORS are all log-compressed
into 16 LOG-REGISTERS. Then these LOG-REGISTERS are readout out 2 at a
time in each 360-bit readout. Since a 360-bit readout is telemetered
down to the ground once every .512 seconds, and it takes 8 readouts to
telemeter all 16 LOG-REGISTERS down to the ground, then it takes
(8 x .512 =) 4.096 seconds to telemeter all 16 LOG-REGISTERS down to the
ground.
Immediately after all the LOG-REGISTERS are telemetered, then all
the ACCUMULATOS are again log-compressed into these LOG-REGISTERS,
and so goes the cycle.
Only during log-compression, are the ACCUMULATORS dead to incoming
counts. So, once every 4.096 seconds we get this 5 milli-seconds
of compression time.
Here's a table of CRRES Rate & PHA dead times that I prepared
in Dec of 1992.
CRRES EHIC-LT
Rates (once per count) 12 micros 6.1 micros
Rates (once per 8 Readouts) 7.0 milli 7.0 milli
PHA (once per Pulse Height Analysis) 2.088 milli 2.088 milli
Rates & PHA (once per Readout) 24.5 milli 3.91 milli
In addition, when the RATES are disabled during compression, Pulse Height
Analysis is NOT inhibited. So, it is possible to get a PHA Event for which
there is not a rate count--but this occurs only 0.17 % of the time.
The BOTH is due to the transfer of the 360-bit buffer to the buffer that will
then dribble it out into the Telemetry at only 16 KHz on CRRES.