CASES97: Logbook Entries

CASES97: CHEM Messages: 22 Entries..

Return to Logbook Contents Page
Entry Date Title Site Author #Graphics
262 Wed 04-Jun-1997Plumbing delay times for O3 Meyers and CO2 Licor7Delany, Tony
261 Wed 28-May-1997Post-CASES97 inspection of chem and rad7Delany, Tony
244 Tue 20-May-1997Licor CO2/Chem sensor problems7Militzer, John
237 Mon 19-May-1997Seclod zeroed and Target swap7Militzer, John
221 Sat 17-May-1997Serviced o3.meyers and co2.licor7Horst, Tom
211 Wed 14-May-1997Seclod Zero and Target7Jirak, Cathy
196 Sun 11-May-1997Changed SECLOD target7Horst, Tom
194 Sun 11-May-1997Seclod zero7Militzer, John
169 Mon 05-May-1997Zeroed SECLOD7Michaelis, Matt
139 Wed 30-Apr-1997Restarted CO2 pump yesterday afternoon.7Horst, Tom
134 Wed 30-Apr-1997ozone targets replaced7Oncley, Steve
130 Mon 28-Apr-1997ozone sensors have old targets7Oncley, Steve
129 Mon 28-Apr-1997licor pump off again/soda lime refreshed7Oncley, Steve
124 Mon 28-Apr-1997licor replumbed/seclod zero7Oncley, Steve
89 Wed 23-Apr-1997Zero.chem programming7Oncley, Steve
88 Wed 23-Apr-1997Care and Feeding of Meyers Ozone Sensor7Delany, Tony
87 Wed 23-Apr-1997Care and Feeding of the Licor CO2 Sensor.7Delany, Tony
85 Wed 23-Apr-1997Care and Feeding of SECL7Delany, Tony
84 Wed 23-Apr-1997Care and Feeding of the ThermoElectron Ozone detector7Delany, Tony
69 Mon 21-Apr-1997Installation and Testing of SECLOD7Delany, Tony
67 Mon 21-Apr-1997Zero.chem programming7Oncley, Steve
45 Fri 18-Apr-1997Set-up Chemical Sensors7Delany, Tony


262: CHEM, Site 7, Wed 04-Jun-1997 20:04:15 GMT, Plumbing delay times for O3 Meyers and CO2 Licor
In order to calculate fluxes the delay times for the chemical sensors
needed to be calculated.
For CASES97 the slight variation due to wind direction was not considered.

SECLOD was mounted immediatelty behind the sonic anemometer and a
delay time of zero was used.

For both Meyers and Licor the delay time were defined during high-wind
period when the flow was down the length of the boom. This condition
gives the tubing + sensor delay when the contribution of the
sonic anemometer to intake is at a minimum.

Cross corellation lag time ("o3.meyers.s7","w.s7") = 2.95 s
Cross corellation lag time ("co2.licor.s7","w.s7") = 1.20 s

These values are in rough agreement with the times calculated
from the flow rates of the sensors and the volume of the 25'tubing
Mass of water in 7.25' of tubing = 27.4 g
Mass of water in 25'   of tubing = 94.5 g
Volume of 25' of tubing = 94.5cc = 0.0945 l 

Delay time through tubing at 3 lpm (0.05 lps) = 0.0945/0.05 = 1.9 s for Meyers
Delay time through tubing at 7 lpm (0.12 lps) = 0.0945/0.12 = 0.8 s for Licor

261: CHEM, Site 7, Wed 28-May-1997 22:42:07 GMT, Post-CASES97 inspection of chem and rad
The equipment returned to NCAR and was unloaded on 27 May''97
I took the Chem Box and the Four component raiometer array
into the lab.

Powered-up the Chem Box:
 Teco: Air flow ~ 2 lpm, internal pump and solenoids could be heard operating
       The front panel came alive and settled down to give a reasonable value
       for inside air ~ 10 ppbv.

 Meyers: Air flow ~3 lpm, Internal pump could be heard. 

 Licor: The pump drew ~ 7 lpm. However the front display was dead.

 Gulp: Connected output to chart recorder. Ran for two hours: no change
       in output. The program is dead.


 Radiometer array: Domes all look clean. Cables disconnected from patch box
                   The 12 VDC fan operated on test

ADDED 6 June''97
    On 3 Jun''97 removed Meyers from chem-box and tested on bench. Functions OK
    On 5 Jun''97 removed Licor from Chem-box and tested on bench. Front panel 
      display active. Zeroing works and sensor responded to hurrs  





244: CHEM, Site 7, Tue 20-May-1997 23:13:47 GMT, Licor CO2/Chem sensor problems
The Licor CO2 sensor has been mostly down since the 17th although there
were some brief periods in which is was running.

It was brought back up: ~15:20 CDT, JD140, 20 May via simple power cycle.

NOTE the Gulp also appears to be down and not firing the solenoids to
zero the seclod, licor and meyers ozone.

Part of the outages were related to the Datel problems.
However, it appears that when the Sodalime and Ozone target change was done
on the 17th, the box A/C power cables may have been moved and then became
intermittent for at least the Licor if not the others.

This status of the chem. sensors needs to be investigated further
and fully reported.
237: CHEM, Site 7, Mon 19-May-1997 22:33:24 GMT, Seclod zeroed and Target swap
Zeroed seclod at 16:17:05 local (21:17:05 GMT) for 2 - minutes.
Replaced Target at 18:00 local:  removed GFAS3, installed ATDD3
221: CHEM, Site 7, Sat 17-May-1997 16:56:13 GMT, Serviced o3.meyers and co2.licor
Changed the soda lime in the co2 scrubber and replaced the o3.meyers
target between 1100 and 1200 CDT today.
211: CHEM, Site 7, Wed 14-May-1997 22:22:46 GMT, Seclod Zero and Target
Zeroed seclod 17:36 local  
(This cannot be the right time, looks like 19:36 GMT, twh)
Removed target ATDD2 replaced with GFAS3 17:00 (CDT, twh)
replaced seclod 18:00





196: CHEM, Site 7, Sun 11-May-1997 21:32:58 GMT, Changed SECLOD target
Removed SECLOD from tower at 3:45 pm CDT
Replaced GFAS2 target with ATDD2
Replaced SECLOD on tower at 4:30 pm CDT

194: CHEM, Site 7, Sun 11-May-1997 15:10:25 GMT, Seclod zero
Zeroed SECLOD with helium balloon technique at 0944-0946 CDT.
169: CHEM, Site 7, Mon 05-May-1997 22:01:57 GMT, Zeroed SECLOD
Used a helium balloon to zero the SECLOD today at 21:49--21:50 GMT.
The zeroing may have been rather poor since a good deal of the helium
escaped in puncturing the balloon.



139: CHEM, Site 7, Wed 30-Apr-1997 20:56:50 GMT, Restarted CO2 pump yesterday afternoon.
Restarted CO2 pump around 1800-1830 CDT Tuesday afternoon.
134: CHEM, Site 7, Wed 30-Apr-1997 17:09:30 GMT, ozone targets replaced
We (finally) have replaced the targets for Meyers and Seclod.  
I deliberately didn't replace them just before the IOP on Monday
(which ended yesterday) in case they needed time to settle.

The Meyers target was yellowed (beige?).  We installed the
#2 target in its place.

Seclod still looked white, however there are bits (1mm^2) of
glass(?) shards which obviously have cracked off of the window
in front of the PMT.  I tried to shake most of the shards out,
and reinstalled the sensor with the new (ATDD#1) target.
130: CHEM, Site 7, Mon 28-Apr-1997 22:18:06 GMT, ozone sensors have old targets
Although Tony's schedule called for the ozone targets to be changed today,
I thought that I shouldn't do this just before an IC, on the theory that
it will take some time for them to equilabrate to outside conditions.
Since tomorrow continues the IOP, the next opportunity will be Wed.

129: CHEM, Site 7, Mon 28-Apr-1997 22:13:47 GMT, licor pump off again/soda lime refreshed
When I checked the pump at 4pm today (after it had been running since 10am),
it was pretty hot (though not as bad as yesterday).  I was going to leave it
going - perhaps outside the shelter to help it cool - but then started having
cosmos crashes.  On the outside chance that these crashes (now during IOP1)
were power related, I decided to leave the pump off.  

The licor is still on.  I also refreshed the soda lime. at this time.
124: CHEM, Site 7, Mon 28-Apr-1997 16:21:33 GMT, licor replumbed/seclod zero
After talking to Tony last night, I've replumbed the licor in a way which
should make the pump happy.  (The pump sucks on a T fitting which is connected
to the licor sample outlet and to an adjustable valve.)  I've set the flow
rate (at the inlet of the chem box) to 7 lpm.

While this was going on, I had "gulp" manually set to turn off the flow
on seclod from about 0950-1010 and printed this out for Tony.  It didn't
do a good zero.


89: CHEM, Site 7, Wed 23-Apr-1997 18:43:39 GMT, Zero.chem programming
The commands for the HC11 to program "gulp" to generate zero.chem are:
(The laptop must be set to 1200 baud and connects to the cable on the bottom,
left of the shelter.)

TS (only if data are scrolling)
C
[RAM] A
[index:] 1
H39 600 (sets 60 seconds with solenoids on)
[index:] 2
H00 17400 (sets 1740 seconds = 29 minutes with solenoids off)
[index:] 3
END
[index:] 

(repeat this for RAM B:)

C
[RAM] B
[index:] 1
H39 600
[index:] 2
H00 17400
[index:] 3
END
[index:] 

E


88: CHEM, Site 7, Wed 23-Apr-1997 18:43:18 GMT, Care and Feeding of Meyers Ozone Sensor
The Meyers O3 sensor uses a chemiluminescence target which reacts with Ozone
to produce light. The efficiency of the target decreases with time.
The target needs to be changed 

To change the target:
Slide the Ozone Sensor slightly foreward and open the lid.
Power down the Ozone Sensor by pulling apart the BLUE 12 VDC disconnect.
Unscrew the Al plug on the right hand end of the metal cylinder
Take the Al plug to the trailer.
Remove the previous target, replace the double sided sticky tape, lay the 
    new target on the plug face, store the old target in the envelope.
Return the Al plug, screw it in finger-tight and repower the sensor.


Once a week install a new target. I suggest:

          Target	 Install           Use

	  #1             18 April 1997     18 April - 28 April
	  #2             28 April 1997     28 April -  8 May
	  #3              8 May   1997      8 May   - 16 May
	  #4             16 May   1997     16 May   - 24 May


The instruction sheet and circuit diagrams for the Meyers Ozone Sensor are
in the chemistry drawer in the middle room of the base trailer. 
The targets are in a plastic bag in the plastic box in the chemistry drawer.





87: CHEM, Site 7, Wed 23-Apr-1997 18:42:49 GMT, Care and Feeding of the Licor CO2 Sensor.
The Licor sensor relies on a carbon vane pump in the Chem box to draw air
through the sensor at a high flow. The pump flow is indicated on the
floating ball guage in the top right of the front of the box. The flow 
should be between 5 - 7 lpm. There is an adjustment valve in the aft part of the box. It can be reached from the front.

The Licor CO2 sensor has a small lucite cylinder containing SodaLime.
This is to scrub CO2 from the sample after it passes through the sample
cell and before it passes through the zero cell. If the capacity of the
SodaLime to scrub CO2 from the air-stream is exceeded, then there is
no good zero for the differential optical measurement.

Once a week 
Power down the Licor
Open the back of the Chem Box
Detach the connector panel 
Remove the swagelok connectors for little lucite cylinder.
Dump out the old SodaLime and refill from the bottle.i
Refit the cylinder and tighten the swagelok connectors
Reassemble Chem Box
Restart Licor

A 9/16 wrench is required to undo and redo the Swagelok nut. Do not
over-tighten.
SodaLime is not toxic nor very corrosive and small amounts can be disposed
of in the garbage.
 
I suggest changing the SodaLime on the same day as changing the Meyers
ozone sensor target discs.

        20 April
        28 April
         8 May
        16 May




85: CHEM, Site 7, Wed 23-Apr-1997 17:24:07 GMT, Care and Feeding of SECL
The SECL ozone sensor uses a chemiluminescence target which reacts with
ozone to produce light. The efficiency of the target decreases with time
The target needs to be changed

Two different versions of target are being tested; the GFAS target
and the ATDD target. GFAS is more efficient but has a water interference.
ATDD is less efficient but may be free of the water effect.

To change the target:
   Select the replacement target from the box in the Chemistry Drawer
   Carry the target holder out to site7 in the little tupperware box
   Climb the tower, unplug the cable, carry SECL down.
   Remove the putty around the base of the intake (light leak prevention)
   Unscrew three screws around back edge
   Withdraw the entire sensor.
   Remove the Al sticky tape securing the target holder.
   (The target holder fits into the front top of the sensor)
   Remove old target holder, implace the new target holder
   Replace the Al sticky tape (maybe use fresh tape)
   Insert the sensor in the can, reapply the putty, reattach the back.
   Climb the tower and reattach to Velcro , recable.

I  would like the following schedule

   GFAS 1  23 Apr - 28 Apr
   ATDD 1  28 Apr -  3 May
   GFAS 2   3 May -  8 May
   ATDD 2   8 May - 13 May
   GFAS 3  13 May - 18 May
   ATDD 3  18 May - 23 May

(The actual change time should be noted in this logbook entry)

Zeroing of SECL.
The GULP microprocessor is set up to no-flow zero for 1 minute every 30 minutes.
(See logbook entry 83 , Zero chem programming)
I would like a zero-ozone zero for each new target.
A day after the new target is installed take a helium party balloon out to site.Bring SECL down, still attached to the cable. Rest SECL on the break-out box.
Make a small slit in the ballon with a razor or knife and insert SECL's intake.
Let SECL inhale Helium for a minute or so.
Return SECL to tower.
The slit can be sealed with tape and one balloon will do for two zeros 





84: CHEM, Site 7, Wed 23-Apr-1997 16:57:43 GMT, Care and Feeding of the ThermoElectron Ozone detector
The TECO operates continuously and rarely requires attention.
It may occasionally go into a state of microprocessor confusion when it
generates a hi-lo-hi-lo output.
Mostly it will cure itself . However if the condition persists for more than 
several hours the sensor should be switched off and after a minute, repowed.
69: CHEM, Site 7, Mon 21-Apr-1997 17:11:31 GMT, Installation and Testing of SECLOD
The Chemical Box was addressed on Friday 18 April 97.
It was hauled into the trailer and put into order. Intakes, power
 data connections were made and Tecod, Licor CO2 and Meyers O3 were operated.

The GULP microprocessor was reprogrammed by Steve O.

On Saturday, 19 Apr,  the other sensors were ready and in the afternoon SECL
was addressed.
Test points for HV input were installed. (1 VDC = -250 VDC)
An ATTD target was installed.
Twenty hopkalite screens were inserted into the killer cartridge.
The reversing fan sucessfully operated and the flow through
the newly installed hopkalite screen tube was checked. The question of flow
obstruction by the screens was left till the sensor was running.
The output remained constant at ~ 1.3 indicating that the electronics
were not operating properly.
A -12VDC supply was wired into the system.

On Sunday, 20 Apr, the sensor was run with the extra -12VDC. Exposing the intake
to direct sun gave a response indicating that the PMT and amplifier
were working.
The HV was increased from -750 VDC to -1150 VDC and the sensor run outside.
Still no good.
The ATTD target was replaced by a GFAS target and the sensor run outside.
Steve O reprogrammed GULP.
At ~ 1500 the sensor began to respond.

The Chemical sensors, in their box were deployed and started up at ~1700.
The outputs were ragged but basically OK.

At 1830 CDT the HV of SECLOD was decreased to -1100 VDC.

On Monday, 21 Apr, the chemical sensor data was examined.
SECLOD had settled down and the spectral responses od SECLOD and Meyers
were compared. 
At 11:48 CDT the GFAS target was replaced with the ATTD target.
At ~17:15-17:45 SECL was brought into the Base and the entire killer cartridge
 was removed. SECL was reinstalled.

 FINDINGS
SECL works more or less the same as during FLATLAND96

The reversing fan system works
Twenty Hopkalite screens in the cartridge doesnt significantly decrease flow
       but there is breakthrough with only ~ 30% ? destruction.
With no killer cartridge the reverse flow is 10% ? low but variance smoothed.

Both ATDD and GFAS targets work but GFAS is ~4X higher luminescent

On Tuesday, 22 Apr, at 9:45 did a bag zero using a garbasge bag of base air into which Hopkalite was added.
At 10:55 SECL removed from tower. 20 screen cartridge removed (had it not been removed the previous eve?),  
  and a 5cm of Hopkalite killer cartridge was added. SECL replaced on tower at 11:50
At 13:30 SECL intake inserted in Hopkalite containing zero bag
	possibility of contamination of sensor
At 15:04 did He zero 

At 15:10 SECL and GULP removed to base. SECL cartridge removed entirely, sensor cleaned. GULP relays
   reconfigured for 59 min of fan sample, 1 min of fan off (rather than 59 min aample 1 min reverse)
At 16:00 Restored on tower, It was confirmed that the fan did stop for 1 min.

















67: CHEM, Site 7, Mon 21-Apr-1997 15:20:02 GMT, Zero.chem programming
The commands for the HC11 to program "gulp" to generate zero.chem are:
(The laptop must be set to 1200 baud and connects to the cable on the bottom,
left of the shelter.)

TS (only if data are scrolling)
C
[RAM] A
[index:] 1
H39 600 (sets 60 seconds with solenoids on)
[index:] 2
H00 17400 (sets 1740 seconds = 29 minutes with solenoids off)
[index:] 3
END
[index:] 

(repeat this for RAM B:)

C
[RAM] B
[index:] 1
H39 600
[index:] 2
H00 17400
[index:] 3
END
[index:] 

E


45: CHEM, Site 7, Fri 18-Apr-1997 16:22:48 GMT, Set-up Chemical Sensors
Brought the chem-box into the trailer and checked out the sensors.
Using the three 8 m lengths of teflon 1/4" tubing drew air in from outside.
With power off inserted fresh target in Meyers fast ozone sensor

Teco ozone: looks OK, showing approximately 55 ppbv

Meyers ozone: output voltage ~ 1.7 VDC with fast variability that showed 
	      a decrease when Kurt blew cigarette breathe over intake.

Licor carbon dioxide: output voltage ~ 1.7 VDC with fast variability that
		      showed increase when Kurt blew tobacco breathe over
		      intake

Zero indicator: the Zero indicator shows a response but Gulp requires
		reprogramming