FOOTPRINT92: SF6 Messages: 15 Entries..

SF6  Analysers 
 
The four (4) fast SF6 analysers each have two (2) outputs. 
 
There is a data output corresponding to the analog signal 
of the SF6 detection system and there is a mode output which 
indicates the operational mode of the analyser. There are three (3) 
different modes in which the system can be  operated. 
There is the sample mode where the analyser is allowed to sample 
the atmosphere, there is the zero mode where the analyser is made to  
sample zero air containing no SF6, and there is the span mode 
where the analyser is made to sample calibration gas containing 
a fixed amount of SF6. 
 
The data output of -5 to +5 vdc is to be recorded at 20 hz. 
The calibration factor is given in the form: 
	       [SF6] = c = a(delta v)**b 
            where delta v = signal voltage - baseline voltage  
                          =     v - v(o)   
  where signal voltage correspond to the atmospheric sample. 
  where baseline voltage corresponds to zero gas signal. 
      The default value of a may be taken as 1 pptv.(mv)**-1 
      The default value of b may be taken as 1   
      The calibration is then approximately linear 
		   1mv = 1pptv 
 
The value of v(o) is to be determined every 30 minutes and applied for the 
following 30 minute period. 
The values of a and b will be given by Brian and their constancy checked 
every 30 minutes by keeping track of the output during the span period. 
 
The mode output of -5 to +5 vdc is to be recorded at 1 hz. 
Three voltage levels (possibly -3, 0, +3 vdc) will be set by Brian. 
 
 
NOTE ADDED 23 April 92: 
Jim Rydock FAX'd to say that: 
the delay time for the SF6 sensors is approximately 2.6 seconds 
the response time for the SF6 sensors is approximately  0.4 seconds. 
 
 

Analyzer D had a rip in the pump tubing; probably for the 
 last several days.  It was replaced between 
 1522 and 1529 during Test 2.  
  
 

SF6 CALIBRATION  ROUTINE: 
 
	The calibration routine for the sf6 
sensor was tested earlier today. Everything 
worked. Tomorrow Brian will provide real calibration 
coefficients. For now, the calibration coefficients 
being used are -3000 mv offset, a= 1.0, and b = 1.0. 
 

SF6 @ COSMOS: 
 
	I'm playing with the sf6 at cosmos 
to checkout the calibration program. 
 

SF6 DELAY TESTS: 
 
	Below is an outline of the procedure 
to do delay tests for the sf6. To make the job 
easier the data rate of the sf6_mode was set to 
20hz. 
 
	In directory /home/aster/raw_data/flow is 
a special archive script which will collect data from 
the sf6 signals. This archiver should be run only 
during the test. If you don't kill it you will be wasting 
disk space. 
 
	After data has been collected and the archiver 
has been shut down, you can run the script "flow.plot". 
This plot will process the data and create an xgraph of 
the data. At that point you can interactively expand the 
scale and determine the time delay. 
 

SF6 COVAR STARTED: 
 
	The sf6 covar routine was started 
about 5 minutes ago. 
 

SF6 1hz COVAR STOPPED: 
 
	The 1hz covars for sf6 were killed 
at 21:55Z. 
 

SF6 MODE @ 250M BAD: 
 
	Since the change over to the old 
datel code the mode signal for sf6 at 250m 
has not worked right. The problem is in the 
amplifier/filter board. Replacement of the 
zero switching chip and the first stage amplifier 
did not help. The board will be replaced. 
 

SF6 MODE @ 250m FIXED: 
 
	The problem with the sf6 mode 
at the 250m site was a bad solder connection 
at the sf6 unit. Time of fix was 00:30Z. 
 

SF6 CODE WORKING: 
 
	The sf6 calibration code seems 
to be working now. The fault was operator 
error! Will monitor. 
 

SF6 TRACER TEST PERIODS 
 
Line releases of SF6 were made under good wind conditions 
 
		    TRACER TEST PERIODS 
 
   Test    Date      J D    Start    Stop     Duration         Comments 
 ----------------------------------------------------------------------------- 
    1     2 Jun 92   154    14:06  ~19:50      5:44        
 
    2     4 Jun 92   156    10:54   18:15      7:21 
 
    3     5 Jun 92   157    06:27   08:46      2:19   
 
    4     6 Jun 92   158    13:48   16:25      2:37     No mass flow data 
 
    5   7-8 Jun 92   159    13:21   00:59     10:38 
 
    6     8 Jun 92   160    16:50   19:13      2:23 
 
    7     9 Jun 92   161    13:01   16:29      3:28     No mass flow data 
 
    8  9-10 Jun 92   161    16:42   09:23     16:41 
 
    9    11 Jun 92   163    08:54   11:12      2:18 
					    --------- 
					      53:29 
 

Cosmos crashed at 22:00 jday 171 
 
  Probably due to too much testing of xpsyc which eventually 
was not disconnecting from the socket successfully. 
 
  Restarted cosmos and archiving.  Restarted covar.  This new 
covar is using phase lags for the sf6 channel.  It will apparently 
need 1.3 Mbyte of main memory to synchronize all channels at cosmos 
with a 8.1 sec sf6 phaselag.  Wish it luck! 
 
  Restarted all by 22:40. 
 
 
 

covars at 50, 100 and 175 now use sf6 phaselags 
 
 
  The above 3 covars were restarted at 19:56, jd=173 
and now have the following phaselags for the sf6 channels: 
 
	250m	8.113 sec	<- since jd=171 22:00 gmt, see comment 127 
	175m	6.458 sec 
	100m	7.164 sec 
	 50m	8.448 sec 
 

  Added cal.sf6 and zero.sf6 average groups to all covar jobs. 
 
 
 

sf6 mode cycle lengthened to 200 minutes at 50 & 100m 
 
This was changed beginning at 20:00 PDT 6/26. 
The sf6 instruments at 175 and 250m are still on 30 minute 
cycles.