CACHE94: PSYC Messages: 40 Entries..

Aspirated Humitter changes: 
 
 Due to the special 10hz with .5 gain 
filter problem ( refer to message under 
A2D), the gain settings of the humitter 
probe was changed to 1 with 0 offset. 
Below are the calibration information. 
 
channel		gain/offset	cal 
 
106		1/-		0.015258789, -40.0 
 
Humitter T:   t = v*100 - 40 
 
t = (5/32768) * 100 - 40 
 

AIR humidity bad! 
 
 The AIR humidity sensor has gone 
bad already! At some time we will 
pull it down and check it out. 
 

AIR RH looks good! 
 
 The AIR rh is looking good now. 
It may have been due to liquid water 
on the sensor. 
 

!2.5m psychrometer removed 
 
At 19:30 the 12.5m psych was removed from the wind tower. 
It had failed due to water in the electronics. 
Cloud water was drivn up the butt end of the boom. 
 
Silicone caulk was used to fill the butt end 
in an attempt to correct the promblem 
 

Comparision RH HMP35C, ASTER PSYC 
 
From 1994 Aug 6 (218) 03:00 til 1994 Aug 7 (219) 16:00  
GMT a Vaisaila HMP35C Rh/Temperatur probe was operated  
next to PSYC.2.5m.  
 
Data files: 
 Aug  6 08:00 1994 psyc/psyc218025031 
 Aug  6 16:00 1994 psyc/psyc218080000 
 Aug  7 00:00 1994 psyc/psyc218160000 
 Aug  7 08:00 1994 psyc/psyc219000000 
 Aug  7 16:00 1994 psyc/psyc219080000 
  
 psyc/psyc218025031 psyc/psyc218080000 psyc/psyc218160000  
 psyc/psyc219000000 psyc/psyc219080000 
 
 

Remote site RH probe-PSYC comparison 
 
Aug 15. 1994 pa: 
  Reference temperatur was logged with -+5V, but excitation  
  was >5 Volt, ca. 5.84 Volt, 1/2 Voltage division of 
  ASTER voltage supply. 
  Data might not be recoverable, even with usage of 5.84 
  Volt as excitation, because a small change in excitation 
  of about 0.1 Volt results in a differenz of ca. 0.8 C in 
  the Temperatur. 
	 
Start: Aug 13 (225) 19:10  
End :  Aug 14 (226) 16:20 
 
 
Previous: comparsion see JDays 218,219 
 
RH/Temp probe HMP35C/No. 606433 is setup next to psyc.2.5m. 
RH, temp, und tempref voltages are logged on a 1Hz basis. 
 
calibration: 
  
	RH[%] <- RH[V] * 100 
 
	x <- temp[V]/tempref[V] * 8000 
	T[C] <- -53.4601 + 9.08067 * x - 0.832569 * x^2 +  
		0.0522829 * x^3 - 0.00167234 * x^4 + 2.21098e-05 * x^5 
 
 
 

Aspirated humitter drying out! 
 

15-m Psyc Replaced / RH Profile Test 
 
Because psyc #51 was continuing to consistently read approximately 
4% RH lower than what was expected given what the others were doing, 
we replaced it with a freshly calibrated unit, #52. 
However, sensor 52 was also reading about 4-6% low similar to what 
51 did.  The reason we expected the readings to be higher is a 
typical profile curve which shows approx. 1-2% increase in RH between 
sensors at each step up the tower.  However, the 15-m sensor had 
values more consistent with the 2.5-m readings. 
 
RH profile test: 
After observing that the new sensor behaived similarly to the earlier 
sensor, we tried a silly test by putting sensor #51 in a bag and running 
it up the tower on the rope, stopping at each location, and recording 
the data on the toshiba.  These data have been moved to  
$ASTER/projects/CACHE94/results/toshiba 
The times for each file are: 
	(local time, ~23Z) 
	start   stop  file  Gortex filter used? 
 jd 229	161320 161550 rh2_5    no 
	161800 162030 rh5_0    no 
	162320 162550 rh7_5    no 
	163400 163600 rh10_0f  yes 
	163830 164030 rh5_0f   yes 
	164900 165100 rh12_5f  yes 
 
Results: 
Terrible, as expected.  Forget these data. 
Test needs to be rerun with a better 'mounting' 
and aspiration/rad-shielding arrangement. 
 

psyc.2.5m replaced 
 
One of the new units (#67) I brought from Boulder has just been 
installed at 2.5m.  The previous one apparently had electronics 
which died in moisture, though it was not obvious where the 
water might have gotten to it (the outer boom connector?). 
 
The old unit, which is bad, is #76. 
 

Aspirated 10-m Humitter removed/corroded 
 
Earlier today, the vaisala 50Y humitter was removed from 
the 10-m flux level. 
Upon examination of the sensor, the temperature probe 
was seriously corroded and the sensor itself was 'dead'. 
Attempted to clean the oxidation with alcohol and water. 
 
Both readings appear to be out; sensor is done. 
 

Status of Psyc swap.  Ref Note 131. 
 
For the past two days or so, the humidity readings have 
taken on a more expected 'characteristic profile,' which 
leads to the conclusion that the swap that was done on 
Day 230 was unnecessary and that all of the units have 
been working within tolerance. 
 

RH profile 
 
Since we are now nearly in cloud, I would expect all of the RH's to 
be about 100%.  According to "xrh" the profile is: 
15m	100 
12.5m	101 
10m	99 
7.5m	98 
5m	98.5 
2.5m	100 
 
We have suspected that 12.5m is a bit high.  I don't know if the 
5-10m levels are low or not.  We'll work up our "roving probe" 
(hauling line) profile data today. 
 

Psyc's Replaced at 7.5-m, 15-m 
 
The dead psyc at 15-m was replaced with the old unit 
that had been up there, #51.  Reason: #52 died last night 
 
The psyc reporting low temps at 7.5-m was replaced with 
a freshly calibrated unit brought from Boulder. 
 
	was	now 
15-m	#52	#51 
7.5-m	#35	#40 
 

Swapped 2.5-m Psyc 
 
Unit #52, installed yesterday after having been removed 
from 15-m and being 'fixed' definitely has a problem 
and goes down shortly after being repowered. 
 
Put #67 back into the 2.5-m location. 
 

2.5 Meter Psychrometer 
 
The humidity sensor was reading about 20% high. 
Cleaned with alcohol and now is back to normal. 
 

2.5 Meter Psychrometer 
 
The 2.5 meter humidity sensor saturated at 105% during 
a high humidity peak at jd 241, 22:00 gmt. 
The other sensors returned to normal, but the 2.5 sensor 
stayed high. I noticed this in the morning of jd242 and 
tried drying it. When that did not work, I cleaned it 
(see prev. log entry) and that brought it back to normal 
at 10:30 gmt. 
 

Humitters drying out 
 
The psychrometers and Vaisala RH sensor are all finally 
drying out. This seems to have become a daily event, 
considering the moist nights. 
 

Humitter failure 
 
The 10 meter humidity sensor just went from 100% 
down to -1.8%..."obviously a major malfunction". 
 

Removed the 10m psyc around 2200 
 
Removed the 10m psyc since the humidty probe 
is bad. Found very large deposits on it. 
 
This is serial number 34 
 
 

Replacing gortex shields 
 
We are reinstalling the gortex shields 
inside the humitter caps, since the 
sensors seem to be corroding quickly. 
 
We plan to install external gortex covers when 
the material arrives here from Boulder. 
 

Reinstalled 10m psyc 
 
Serial number 16. 
 

It is a raging storm outside; winds up to
20 ms/s. Perhaps water in a connector
or the electronics box

Charlie retrieved the psych, and it was DOA. No Serial data output.
I replaced the ua78S40 voltage regulator chip plus a small choke
which looked toasted. Seems to run fine, so we reinstalled it
on the tower about an hour ago, and it's still ticking...

The 10 meter unit died again today. The problem appears to be
the power supply. Since we do not have a schematic, it's time
to replace it with another winner from the boneyard.
Outgoing serial # 16.
Ingoing #35 which seems to read ~ 1 C low.

Unit #35 at 10 m. was reading about 1C low consistently, so I
removed it and put # 67 which was at 2.5 m. in its place.
Recall that #35 was used to replace #16 just yesterday @ 22:00,
after # 16 had died.
Meanwhile, I replaced the power transformer on 16, which is suspect,
tested, and reinstalled # 16 at 2.5 m for ease of servicing, in
case it dies again.

The 5m psychrometer just decided to stop sending
data. I cycled the power on it, and it restarted okay.

The 2.5 m unit (#16) just latched up again.
Cycling power did the trick.

another power supply latch-up and restart,
after it started raining.
Up again.

10m psyc stopped around 06 GMT. This unit 
had a lot of water in it; circuit board was
very wet, with slime deposits

2.5m psyc stopped around 09 GMT

All the rain yesterday and last night killed 2 more units.
s/n 67 at 10 m. was full of water and drying it out did'nt
help - there was a lot of corrosion. It has been replaced
with s/n 75 which arrived today.

Unit s/n 16 at 2.5 m had the usual power supply problem.
I replaced it with s/n 20.

I sealed both new units with RTV Silicone so hopefully
they will survive the next storms.

T and RH look good now.

Restarting it did the trick, fortunately.

The 5m hygrothermometer temperature has been occasionally spiking on the 
past two days, 252-253.  The relative humidity has also been constant at
105% today, 254.  At ~0930 the RH appeared to dry out and afterwards 
generally matched the other hygrothermometers.

psyc.5m is not transmitting.

psyc.10m seems to be transmitting intermittent garbage which is causing
arnica (and then marigold) to crash.  I've disconnected the cable for
psyc.10m at the breakout box.

Marigold rebooted.

Retrieved the 5m (#01) and 10m (#75) hygrothermometers from the psyc
tower.  Opened the electronics cans and found that #01 (5m) was dry
inside with only damp dessicant; #75 (10m) had liquid water in the
can, despite the fact that it appeared to have been sealed with some
type of caulking.  Left them open to dry out and will try them out
tomorrow.

Checked out the two hygrothermometers taken off the tower yesterday.
#75 did not transmit, but #01 appeared to be working.  Replaced #75
with newly calibrated unit #70 and sealed both #01 and #70 with silicone
caulking.  Put #01 back at 5m and put #70 at 10m.  Data look reasonable,
particularly RH which is on the order of 90%RH at 5 and 10m but is
much closer to 100%RH at other levels, suggesting that the humitter
filters are still saturated.

15 meter psyc stopped transmitting somewhere around 9:00 am PDT this morning.
Cycled power and it came back to life.

Looking back over the temperature, RH, and specific humidity profiles
for the JD 261 and 262, the 12.5 m RH appears to be in the most
disagreement with the other levels.  However, it still may be within
the stated 2%RH accuracy.

The psyc at 15 meters (ID 51) had stopped transmitting earlier this morning.

It was taken down, dismantled, and assembled, using silicon caulk to seal,
and 2 dessicant packs.  This didn't help much.  It runs for ~5 minutes,
looking OK, though sometimes a character is dropped,
and then starts generating garbage, which tends to crash mariold.

The psyc was brought down, and the same behaviour was observed in the lab
connected to the Toshiba PC.

So, we'll leave it off the tower, waiting for replacements which should
arrive within a day from Boulder.

Removed psyc #51 from 15 meters, where it had been misbehaving
(generating gibberish now and then).  Replaced it with #3 which
is fresh from Cathy's lab.  Sealed it with RTV.
JD 263, 0900PDT.

Moved 12.5m (#30) and 15m (#73) hygrothermometers to 10m level on psych
tower, and moved 2.5m (#20) and 5m (#17) hygrothermometers to 7.5m level
for intercomparison.  Completed move about 3:45 pm PDT.  Also moved 
Rick's remote site hygrothermometer to 7.5m level.