From: Frank Geurts (geurts_at_rice.edu)
Date: Tue Feb 11 2003 - 23:30:24 EST
Hi Bill --
I forgot to tell you how the event file will look like.
It is always single line of numbers in a file that will be updated every
minute:
1045013461 61.9 0.15 0.15 0.04 0.04
the first integer number is the unix time stamp. Essentially it counts
the seconds since January 1st, 1970. Solaris appears to support the same
standard. Use this number to check whether the file updates at all ...
it's the least thing that should change after every update. The next
three floating-point numbers are the gasflow in ccm (cubic centimeter
per minute) for freon, isobutane and SF6 respectively. No limits or
alarms apply for those three numbers. The following two floating points
are the ratios isobutane/freon and SF6/isobutane. The ratios are defined
in a rather funny way but the bottom line is this:
- 1.0 good
- any>1.2 *yellow* alarm -- shiftcrew should make a note in the logbook
- any>1.6 *red* alarm -- shiftcrew should take action by physically
shutting down valves and notifying a TOF detector expert.
- if timestamp does not update: *white* alarm, if persistant call TOF
detector expert and/or follow procedures to (re-)start gas-monitoring
software.
- there are no lower limits that require alarms.
We are not sure about the fluctuations in our readout system so we might
want to change the limits on yellow and red alarms.
we can look into this tomorrow morning if you want
cheers,
frank
This archive was generated by hypermail 2.1.4 : Thu Jul 24 2003 - 00:39:34 EDT