CACHE94: CO2/O3/H2O2 Messages: 24 Entries..

Output signal of O3: 
 
 The output signal from the o3 sensor 
can jump 800 mvolts due to a faulty contact 
in the sensor. When this happens, it will 
appear on the cockpit display as a negative 
value, 800 - Vout.  This is a signal to 
the OSU people to look at the sensor. 
This is the reason for the calibration routine 
doing funny things if the output voltage is 
above 800 mvolts.  
 
 

O3 Calibration and Zero 
 
From 18:50 to 19:25 on August 6 (GMT) the ozone meter was 
run through a zero and calibration run. 
 
 

Licor CO2 fast sensor installed 
 
The Licor had been set up several days ago but was not 
connected to the zero power timer until today. 
The span was left at 5.0, the zero was left at 5.0. 
A mass flow meter connected to the house vacuum system  
monitored the flow. The best available flow was 6.2 slpm. 
This remained reasonable constant. (5.9 - 6.2 slpm). 
The sensor was located inside the seatainer on a shelf 
immediately inside the high north cable port. 
The intake tube of high density polyethylene was  
run throught the port and to the boom. The inlet was on the ATI 
at the base of the transducer supports. 
The system was checked out and by noon was fully operational. 
The cross correlation lag was done both against the ATI and the Solent. 
The lag against the ATI was 0.7 seconds. This is the tube flow time. 
The lag against the Solent was 0.3 seconds. Apparently a -0.4 second 
delay occured between the two sonics. This delay should be wind direction 
and speed dependant. I will check this again when there is more data. 
 
Added: As the cross corellation between the w.atik.10m.walk and 
co2.10m.walk gave a delay of 0.7 seconds this should be put in the calibration . 
 
 

OZONE INLET CHANGE 
 
During calibration of ozone meter, Pit noticed loose connection at 
inlet.  We ran the generator/calib. prior to and after tightening 
the inlet connection at the back, and got higher readings afterward. 
 
Generator 20 results:    29/33 
Generator 30 results:    133/140 
                         (before/after) 
 
 

AIRBLAST on.signal shows up as mode.licor 
  
airblast on signal has an output of ca. 5.4 Volt 
I know that Gordon changed the calibartion for  
the mode.licor. I don't know why i does not show 
up in the cockpit correctly. 
 

Gain setting on licorco2mode 
 
Between 1805 1nd 1815 gmt the half gain setting on the  
licorco2mode was examined and found to be  
inoperative.(the jumper was not making good contact). 
The half gain setting was made operative. 
 

H2O2 status: 
 
 The H2O2 sensor will require more 
then 1 analog channel as planned. There 
is the possibility to collect data from it 
as a serial channel. The sensor has not 
arrived yet. However, they will use a PC 
to collect the data. Since it is a slow sensor 
it may be possible to pump data out of the PC 
as a serial message. This would interface to 
marigold. 
 

Added channels for CO2.licor 
 
Added chanels are flow rate of flow meter and 
temperatur of CO2 measurement cells. 
 
flow rate= Volt* 4.213 
(Check by varying low rate and measure volt output: 
 data: 1.40lpm=0.329V,2.94lpm=0.672V,4.26lmp=0.984V,5.4lpm=1.28V) 
t.cell = Volt * 10 (or manual mV * 0.01) 
 
flow rate is going to be hook up on ref.peter.2.5m did:114 chanC.peter 
temp cell on t.peter.2.5m did:112 chanA.peter 
t.licor stuck into the wrong channel, should be in 113 
changed 229, 05:54 GMT 
 
 

changed co2 zero check from 20s to 30s 
 
Start: jday 228 23:20 GMT 
New configuration 1730s operational, 30s CO2 zero check. 
 

mode.licor changed to 20 Hz 
 
It appeared from prep that the mode information was coming in late 
and "zero" data were getting into the co2.licor channel.  (This  
check "synchronized" to the 1Hz data.)  Having the mode switch 
come in late seems to me to be a problem since it is hard for the 
co2 calibration routine to look into the future (maybe modify timetag 
or use lag times?)  I wondered if the low sampling rate was the  
cause of this problem, so I've just changed channel_config to sample 
marigold:105 at 20 Hz.  We rebooted marigold (deliberately, this time!). 
We still haven't changed ops. 
 

 
aOzone meter calibration Sat 8-20 9:50 am 
The OSU ozone was calibrated between 9:50 am and about 10:11 am. 
Signals include 130 ppb generator, 32 ppb generator, control freq, 
and sample freq. as well as zero and span. See manual for voltage 
spec. during control req. mode 
 

CO2 zero mode signal delayed (slow rising) 
 
It takes about a second for the mode.licor.10m.walk 
signal to go from 0 Volts to 1.5 Volts. 
 
 

CO2 mean volt output is fairly low 
 
Mean output of ca. 1.4 Volt gives a CO2 conc 
of ca. 240 ppm. (according to calibration sheet, 
25 Jan 1993 for IRG1-256) 
A check to a known span gas has to be performed 
as soon as possible. 
Expected voltage output would be in the range 
of 1.9 volt at 350 ppm. 
 
 

CO2 licor check 
 
Start: 238, 20:00 
End  : 238, 22:15 
 
Flow rate has an effect on the CO2 volt 
output. 
Flow set to 5.08 lpm. Zero check reading has higher 
flow rate, since it is through a shorter tube. 
 

CO2 changes 
 
 
- zero check tubing extended along sample tubing 
  to keep same flow rate (flow rate has an effect on 
  co2 reading). 
- time delay check with filled C02-ballon: time delay = 1.4secs 
- time delay check with cross-correlation of 
   u.atik:co2,fastT:co2 give time delay of ca. -1.5 sec 
   w.atik:co2 has a broad flat peak between: -0.7 and 1.5 sec 
  
ops1/prep.config phase values for co2.licor10m.walk and 
mode.licor.10m.walk got set to 1400 ms. 
 

O3 still high 
 
O3 values have been high all day (>40ppb); 
may be related to the fire, but  
Andy said that they started going up 
yesterday. 
 

CO2 licor calibration test 
 

The CO2 licor was drawing air water bubbles through the tube.
It is a rainy day with clouds at Cheeka Peak.
Co2 Licor got turned off at 21:25 GMT.
It is going to be turned on again, by Rick Vong after
cloudy period.

the CO2 flow was restarted after Peter had it down to blow out possible H2O.
I turned on the flow and the LiCOR analyzer at 15:50
Time down ca. 2 hr, see OSu written notes elsewhere

The OSU ozone meter was calib from 12:15 until 12:45 (noon) and appeared
OK in general. Normal routine.

Dasibi O3 calibration/check ca. noon Wed 9-7-94. Result= OK.

The ozone calibration/checkout was performed from  1440 until 1510 today.
all results were `in-control' (passed). 

On JDay 249 @ 14:20..15:20 the CO2 output jumped up by
ca 0.51 volt. Might be caused by cloud water entering the
sample cell or the internal filters. So that either the 
internal filters or the optics needs to be cleaned (see
section 8.5 in the manual)

The offset is persistent until the end of the experiment.

Changed cal_files for chanB.peter and chanC.peter

chanB.peter (Temp of CO2 sample cell)
  edited comment line #1994 229 ...

chanC.peter (flowmeter)
  changed calibration to get only voltage output for the time
  the flowmeter was hooked up to chanC.peter