SHEBA: Gill_Sonic_Tests Messages: 21 Entries..

From: "Gill Electronic R&D" 
Subject: Solent HS Anemometer
Date: Fri, 15 Nov 1996 19:02:39 -0000

John

Sorry for the delay in responding further to your email.

The temperature qualification of the HS anemometer has been carried out in
our environmental chamber here at Gill.  This operates down to -40°C.

Over the full temperature range we have monitored receive waveforms,
transducer delays, speed of sound, analogue inputs and the inclinometer
readings.

The performance of the transducers is reduced below -35°C.  We are
currently performing tests to obtain more data on this.  

The test time at low temperatures is currently hours rather than days or
weeks.  However, we have not seen any degradation of transducers in other
models that have been operated at low temperatures.  This includes a Solent
R2 operated at the Antarctic by Phil Anderson of British Antarctic Survey. 
His telephone number is 44 1223 251400.

It is worth noting that the HS and R3 carry out transducer delay correction
with temperature which the R2 does not.

I would expect the power consumption of the HS to remain <4 watts at low
temperatures.  We will measure it to confirm this.

Regarding cables, I have a note in my logbook that a cable supplier that I
contacted a while ago does not recommend polyethelene for operation below
-20°C.  A plastics book here at Gill gives a glass transition temperature
for polyethelene of -20°C which may support this view.

I hope the above is helpful.

Angus

----------------------------------------------------------------------------
------------------------
Gill Electronic Research & Development   
Solent House, Cannon Street, Lymington, HAMPSHIRE. SO41 9BR.
Telephone :  +44 (1590) 679955   Fax :   +44 (1590) 676409
E-Mail :  gill@gill.co.uk     Website :  http://www.gill.co.uk/index.html

From horst@stout.atd.ucar.edu Thu Jun 12 12:16 MDT 1997
Date: Thu, 12 Jun 1997 12:16:20 -0600 (MDT)
Subject: Solent cold temperature performance

Dear Mr. Clark:

I spoke to you earlier this morning regarding a question on the
performance of our 1-D Solent for shorter path lengths.  Following is
another, more urgent question regarding the performance of Solent
anemometers in cold temperatures.

We are hoping that we can use R2 Solent sonic anemometers for a 13
month field program to study the surface heat budget of the arctic
(SHEBA).  We expect average minimum temperatures to reach -30 degC and
extreme lows to approach -50 degC.  Could you tell me the expected cold
temperature performance of our R2 sonics?  The serial numbers of our
Solents are:

1012R2 0067; 1012R2A 0129; 1012R2A 0087

There is also a fourth R2 Solent for which I do not yet have the serial
number.

I spoke with Phil Anderson of the British Antarctic Survey.  He used an
R1 Solent in the Anarctic in 1992 and was quite disappointed in the
performance below -5 degC.  Angus Raby provided the attached
information on the R3, but I find in your sales brochure a quoted
performance down to -20 degC on early model R2s and -40 degC on later
model R2s (corresponding storage temperatures are -40 degC and -50
degC).

Could you also tell me the nature of the tests used to establish these
temperature specifications.  If our sonics are only rated for performance
to -20 degC, is there a simple upgrade that could be done to improve
their low temperature performance?

We are on an extremely tight schedule to prepare for a field deployment
in September, so I would very much appreciate a quick response.

Best regards,
Tom Horst


>=46rom gill@gill.co.uk Fri Nov 15 12:25 MST 1996
>Message-Id: <199611151907.TAA22638@zeus.tcp.net.uk>
>From: "Gill Electronic R&D" 
>To: 
>Subject: Solent HS Anemometer
>Date: Fri, 15 Nov 1996 19:02:39 -0000
>
>John
>
>Sorry for the delay in responding further to your email.
>
>The temperature qualification of the HS anemometer has been carried out in
>our environmental chamber here at Gill.  This operates down to -40=B0C.
>
>Over the full temperature range we have monitored receive waveforms,
>transducer delays, speed of sound, analogue inputs and the inclinometer
>readings.
>
>The performance of the transducers is reduced below -35=B0C.  We are
>currently performing tests to obtain more data on this. =20
>
>The test time at low temperatures is currently hours rather than days or
>weeks.  However, we have not seen any degradation of transducers in other
>models that have been operated at low temperatures.  This includes a Solent
>R2 operated at the Antarctic by Phil Anderson of British Antarctic Survey.=
>=20
>His telephone number is 44 1223 251400.
>
>It is worth noting that the HS and R3 carry out transducer delay correction
>with temperature which the R2 does not.
>
>I would expect the power consumption of the HS to remain <4 watts at low
>temperatures.  We will measure it to confirm this.
>
>Regarding cables, I have a note in my logbook that a cable supplier that I
>contacted a while ago does not recommend polyethelene for operation below
>-20=B0C.  A plastics book here at Gill gives a glass transition temperature
>for polyethelene of -20=B0C which may support this view.
>
>I hope the above is helpful.
>
>Angus
>
>---------------------------------------------------------------------------=
>-
>------------------------
>Gill Electronic Research & Development  =20
>Solent House, Cannon Street, Lymington, HAMPSHIRE. SO41 9BR.
>Telephone :  +44 (1590) 679955   Fax :   +44 (1590) 676409
>E-Mail :  gill@gill.co.uk     Website :  http://www.gill.co.uk/index.html
>
>

From angus.raby@virgin.net Thu Jun 19 12:09 MDT 1997
To: horst@stout.atd.ucar.edu
Subject: Solent cold temperature performance and path length

Dear Tom

This is largely in response to your email of 12th June concerning low
temperature performance of the R2.

The results that Phil Anderson obtained at Halley in the Antarctic in
1994 were disappointing.  The limited data that I have shows noise on
the speed of sound and the UVW outputs.  We were unable to identify the
source of the problem from the data available, but it does not appear
to be entirely due to low temperatures.  A plot of sonic temperature
against absolute temperature is much cleaner for absolute temperatures
from -12 to 15 degrees C (the coldest in the data set)than the majority
of the data between -3 and -12 degrees C.  It was suggested at the time
that the presence of high power RF transmitters at Halley could have
contributed to the problems.  Phil Anderson obtained more consistent
data from two Kaijo Denki sonics.  However, a proposal for the STABLE
experiment reports that the sonic proposed (presumably from KD) gives
increasingly erroneous readings above 12m/s.

On a more positive note, Andy Black from the University of British
Columbia, Vancouver recently reported successful deployment of an R2
for twelve months at temperatures down to "almost -40 C".  I have
emailed him to ask if he can supply any further information.

Other than these I do not have any data from the field on low
temperature performance.  I view this as encouraging since we are more
likely to hear of problems than good results!  It may be worth asking
RISO if thay have any data at low temperatures.

The tests used to establish the low temperature performance of the R2
have been carried out in environmental test chambers, observing receive
ultrasonic waveforms and measuring transducer delays for a number of
transducers down to -40 C.  In the early days we only had access to a
temperature chamber that would go down to -20 C, hence the original
specification.  A few components required upgrading to -40 C, but from
what I remember the original parts worked OK at -40 C anyway.  All the
anemometers you mention contain the latest temperature specification
parts.  There are no further upgrades that improve the low temperature
performance of the R2.

From our tests to date, the components that appear most affected at low
temperatures are the transducers.  If you have time you could consider
returning your R2s to Gill Instruments for recalibration.  Part of this
process is to check the transducer waveforms.  It appears from our
records that R2 serial number 0067 has not been calibrated since it was
built in 1992 and is worth checking.  The other two anemometers you
mention were recalibrated on 23rd Nov 95 (R2A No 87) and 14th Dec 95
(R2A No.  129).  These units therefore checked and calibrated less than
two years ago and should be OK.

As I mentioned in my email to John Militzer on 15th November the
transducer performance starts to deteriorate below -35 C.  Our chamber
will not go below -40 C so I cannot be certain, but I would not expect
satisfactory performance much below this.

Regarding operation at shorter path lengths I am not aware of any
compensation for path length in the software.  There is however a
receive time window.  I am working from memory here but I think that
the start of the receive time window is timed to work at temperatures
up to 60 C and wind speeds up to 60m/s.  This gives a minimum allowable
transit time of 350uS.  The minimum path length that should work at say
25 C and a wind speed of 10m/s is 125mm.  This may well be the
problem.  What path length did you try?  It is possible to move the
time window if the software is changed, but the receive timing is
interleaved with analogue inputs and other things that make it a
significant task.
    
I hope the above information is helpful.

Best regards

Angus Raby

From ablack@unixg.ubc.ca Tue Jun 24 15:19 MDT 1997
X-Sender: ablack@interchg.ubc.ca
To: Tom Horst 
Subject: Re: Solent cold temperature performance 
MIME-Version: 1.0
X-Lines: 30


Dear Tom,

We have operated the R2 during two winters at Prince Albert, Sakatchewan.
Air temperatures got down to about -35 C. The R2 worked fine. No
missing data except when the coating of hoar frost or snow got too thick.
We're talking about a remotely controlled electrical wire heater, but
haven't done it yet. Maybe that will not be a problem for you being so 
dry up there. Mean cup wind speed agreed well with that from our R.M.
Young propeller-vane. The slope and intercept of the mean sonic air
temperature vs. mean air temperature (thermocouple) changed as a function
of air temperature as it got colder. You just need to keep your eye on
this relationship if you're calculating sensible heat fluxes or sigma Ts
using the sonic air temperature. In summary, we've been very pleased with
the performance of the R2.

Best regards,
Andy Black

------------------------------------------------------------------------------
T.A.Black
Department of Soil Science
139-2357 Main Mall
University of British Columbia
Vancouver, B.C.
V6T 1Z4 CANADA
Phone (604) 822-2730
FAX   (604) 822-8639     

97/06/25


NCAR Sonic:	1012R2A - 0087
Test Purpose:	Cold Temp. data quality assessment

Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	 25	00:10	soak		176/13:30	 7:30, Wed 6/25
 2	-45	12:00	ramp-down	    13:40	 7:40
 3	-45	 1:00	hold		177/ 1:40	19:40
 4	25	12:00	ramp-up		     2:40	20:40
 5			End-of-Pgm	    14:40	 8:40, Th  6/26	


Configuration:	Gill is oriented diagonally in chamber with asymmetric
	head toward the upper, rear corner away from the circulation
	system.   There is about >= 1" between the Gill framing tubes
	and the chamber wall.  	Styrafoam boards are holding up the
	TRH in the middle of the chamber.
	Cosmos  202: Gill
		200: TRH

Notes:	Ingest from Cosmos seemed to be down even though rserial
	showed data coming in.  Restarted Cosmos.

97/06/26
 
 
ETL Sonic:      1012R2A - 0129 (Asymmetric)
Test Purpose:   Cold Temp. data quality assessment
 
Chamber Program
Step    Temp    Time    Type            Begin-Z(approx) Local-Time
----    ----    ----    ----            ---------       ----------
 1       25     00:10   soak            177/15:30        9:30, Th. 6/26
 2      -45     12:00   ramp-down           15:40        9:40
 3      -45      1:00   hold            178/ 3:40       21:40
 4      25      12:00   ramp-up              4:40       22:40
 5                      End-of-Pgm          16:40       10:40, Fr. 6/27
 
 
Configuration:  Gill is oriented diagonally in chamber with asymmetric
        head toward the upper, rear corner away from the circulation
        system.   There is about >= 1" between the Gill framing tubes
        and the chamber wall.   Styrafoam boards are holding up the
        TRH in the middle of the chamber.
        Cosmos  202: Gill
                200: TRH
 
Notes:  Cosmos up.

97/06/27, Friday
 
 
ETL Sonic:      1012R2 - 0067 (Symmetric)
Test Purpose:   Cold Temp. data quality assessment
 
Chamber Program
Step    Temp    Time    Type            Begin-Z(approx) Local-Time
----    ----    ----    ----            ---------       ----------
 1       25     00:10   soak            178/17:30       11:30, Fr. 6/27
 2      -45     12:00   ramp-down           17:40       11:40
 3      -45      1:00   hold            179/ 5:40       23:40
 4      25      12:00   ramp-up              6:40       00:40
 5                      End-of-Pgm          18:40       12:40, Sa. 6/28
 
 
Configuration:  Gill is oriented diagonally in chamber with Symmetric
        head toward the upper, rear corner away from the circulation
        system.   There is about >= 2" between the Gill framing tubes
        and the chamber wall.   Styrafoam boards are holding up the
        TRH in the middle of the chamber. 
 
        Note that on the previous test, the sensor head assy was very
        close to the chamber walls, possibly touching, and that there
        was a PAM electronics box also inside the chamber which may
        have disrupted the results.
 
        Cosmos  202: Gill
                200: TRH

Notes: 	Cosmos up

97/06/23


ETL Sonic:	1012R2 - 0067	(symmetric)
NCAR EVE Chassis:

Test Purpose:
	- Survivability / Cold Temp. data quality assessment of Gill

	- Survivability of EVE:
	  This is the second Industrial Solutions VME chassis to be
	  tested.  The first one failed so we'll see if it's
	  symptomatic of the species.
	  Using the same VME boards as the first series of tests (they
	  work), but a different PTB220 and front panel board.


Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	 25	00:10	soak		174/18:25	12:25, Mon 6/23
 2	  0	 1:00	ramp-down	174/19:35	12:35
 3	-25	 4:00	ramp-down	174/20:35	13:35
 4	-35	 4:00	ramp-down	174/23:35	17:35
 5	-35	12:00	hold		175/03:35	21:35
see Note:
 6	+25	12:30	ramp-up		175/15:30	 9:30-ish
 7			End-of-Pgm	176/03:30	22:00-ish


Configuration:  Chamber is crammed full.
	EVE box is diagonally (sort-of) placed on its side with its back
	to the Gill, which is oriented diagonally in chamber with its
	head all the way toward the upper, rear corner away from the
	circulation system.   There is about <= 1" between the Gill
	framing tubes and the chamber wall.

	Cosmos  202: Gill
		200: TRH

Note:	It's easy to see the effect of the internal air circulation
	and possbile vibration in the Gill Cockpit display.  It
	appears to be as much as .5m/s


Note:	6:40 local Tues 6/24, 
	EVE at -35C beginning to show 'spurious interrupts' and will
	likely die soon.

	DECISION TO END TEST EARLY: reason was to speed the process and
	perform quicker/shorter tests on the Gills to screen them
	first.   Then the chamber would be warm and ready for another
	gill in the morning.

jm

97/06/30


Campbell Sonic:	....s/n later....
Test Purpose:	- Campbell Cold Temp. data quality assessment
		- Also to verify that the periodic waveform seen in the
		Gill sonic's covars which begins near room temp. and
		continues through low temps is a dynamic of the sensor
		itself and not a manifestation of something the
		chamber is actually doing.
		- Also to try and break those Qualimetrics quad
		pressure port with the cold....ie will they survive?


Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	 25	00:10	soak		181/22:40	16:40, Mon 6/30
 2	-45	12:00	ramp-down	    22:50	16:50
 3	-45	 1:00	hold		181/10:50	 4:50  Tue 7/1
 4	25	12:00	ramp-up		    11:50	 5:50
 5			End-of-Pgm	    23:50	17:50, Tue 7/1


Configuration:	The sonic head is "horizontal" in the middle of the
	chamber oriented "N/S" such that the w is perpendicular to the
	chamber's aspiration system grill and the probe assemblies
	are both about >=2.5" below the chamber ceiling. 
	The electronics package is resting against the chamber wall
	immediately below the cable access hole.

	This probe arrangement seems better than what the Gills had.

	The TRH is on some styrafoam at the bottom of the chamber with
	the transducer near the door.
	Two Qualimetrics quad disk pressure ports are laying in the
	back of the chamber.

	Cosmos  202: 
		200: TRH
		203: Campbell

Notes:	Cosmos up


jm

97/06/25

NCAR Sonic:	1012R2A - 0087
Test Purpose:	Cold Temp. data quality assessment.
		All of the Gills tested so far have had a periodic
		waveform modulated onto the data over a 6-deg. cycle.
		Tests using the Campbell sonic revealed this is not
		an effect of the chamber.
		Gill suggested that the effect is related to a single 
		wavelength and it might be minimized by 'quieting' the
		chamber by lining it with felt.

Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	 25	00:10	soak		183/00:00	18:00, Tu  7/1
 2	-45	12:00	ramp-down	    00:10	18:10
 3	-45	 1:00	hold		    12:10	 6:10
 4	25	12:00	ramp-up		    13:10	 7:10
 5			End-of-Pgm	184/01:10	19:10, We  7/2


Configuration:	Gill is oriented diagonally in chamber with asymmetric
	head toward the upper, rear corner away from the circulation
	system.   There is about >= 1.5" between the Gill framing tubes
	and the chamber wall.   Black felt lines the top,back,bottom,port-
	access side and door of the chamber.  The circulation system is
	partially covered below the exhaust vent.
	1 Styrafoam board is holding up the TRH roughly in the middle
	of the chamber.
	Will the 2-sided tape holding the felt last through the low temp
	cycle and vibration of the chamber?....Only the shadow knows.

	Cosmos  202: Gill
		200: TRH

Notes:	Cosmos up

Results:	The felt stayed in place and did not impact the sonic.
	The test was ended at 183/23:00 (5pm Wed) when the chamber was
	at ~15C.

jm

97/07/07

Friehe Sonic:	1012R2A - 0104
NCAR TRH with Electronic Compass and Beacon

Test Purpose:	Cold Temp. data quality assessment.

Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	 25	00:10	soak		189/22:10	16:10, Tu 7/8
 2	-45	12:00	ramp-down	    22:20	16:20
 3	-45	18:00	hold		190/10:20	04:20  We 7/9
 4	25	12:00	ramp-up		191/04:20	22:20
 5			End-of-Pgm	    16:20	10:20, Th 7/10


Configuration:	Gill is oriented diagonally in chamber with asymmetric
	head toward the upper, rear corner away from the circulation
	system.   There is about >= 3" between the Gill framing tubes
	and the chamber wall.
	The TRH is resting on a foam pad about 1" above the chamber floor.
	The Beacon and Electronic compass are on the floor of the chamber.

	Cosmos  202: Gill
		200: TRH

Notes:	Cosmos up
	"Noise" on the sonic began immediately when the chamber started up,
	it was quiet before that.

Results:Sonic:
	The sonic had problems symptomatic of the others beginning at 
	-30.....Tom....

	Compass/Beacon
	The TRH with compass and beacon stayed up and looked OK.
	The beacon did turn on/off at the -45 cold soak.
	The attempt to rotate the compass to confirm it wasn't frozen
	partially worked: the wire pull broke but the compass reading
	did change by about 4-degrees.
jm

97/07/07

Friehe Sonic:	1012R2A - 0112  Symmetric
NCAR TRH with Electronic Compass and Beacon
AGI Electronic Level

Test Purpose:	Cold Temp. data quality assessment.

Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	 25	00:10	soak		192/01:00	19:00, Th 7/10
 2	-45	12:00	ramp-down	    01:10	19:10
 3	-45	18:00	hold		    13:10	07:10  Fr 7/11
 4	25	12:00	ramp-up		    21:10	15:10
 5			End-of-Pgm	193/09:10	03:10, Sa 7/12


Configuration:	Gill is oriented diagonally in chamber with symmetric
	head toward the upper, rear corner away from the circulation
	system.   There is about >= 3" between the Gill framing tubes
	and the chamber wall.
	The TRH is resting on a foam pad about 1" above the chamber floor.
	The Beacon and Electronic compass are on the floor of the chamber.
	The Electronic Level is taped to a PVC pipe that protrudes through
	the access hole by about 4"

	Cosmos  202: Gill
		200: TRH

	EVE electronics box interfaces the second TRH with compass/beacon
	as well as the electronic level.
	Format of asciidat file:

	d/t  box-t  temp  rh  Compass  FET-compass FET-beacon xlevel ylevel


Notes:	Cosmos up


Results:

jm

97/07/07

Peter Guest Sonic:	1210R3 - 19  Symmetric
NCAR TRH with Electronic Compass and Beacon

Test Purpose:	Cold Temp. data quality assessment.

Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	 25	00:10	soak		193/21:20	15:20, Sa 7/12
 2	-45	12:00	ramp-down	    21:30	15:30
 3	-45	18:00	hold		194/08:30	03:30  Su 7/13
 4	25	12:00	ramp-up		    16:30	11:30
 5			End-of-Pgm	195/04:30	23:30, Su 7/13


Configuration:	Gill is oriented diagonally in chamber with symmetric
	head toward the upper, rear corner away from the circulation
	system.   There is about >= 3" between the Gill framing tubes
	and the chamber wall.
	The TRH is resting on a foam pad about 1" above the chamber floor.
	The Beacon and Electronic compass are on the floor of the chamber.
	The Electronic Level is taped to a PVC pipe that protrudes through
	the access hole by about 4"

	Cosmos  202: Gill
		201: Gill.r3
		200: TRH

	EVE electronics box interfaces the second TRH with compass/beacon
	as well as the electronic level.
	Format of asciidat file:

	d/t  box-t  temp  rh  Compass  FET-compass FET-beacon xlevel ylevel


Notes:	Cosmos up

Results:

jm

Summary of temperature chamber tests of Solent sonics at low temperature
------------------------------------------------------------------------

We tested 5 R2 and one R3 Solent sonic anemometers in a laboratory
temperature chamber to investigate their performance at low
temperatures.  The chamber is just large enough for a Solent to fit
diagonally between opposite corners, with the measurement array within
a few inches of two of the chamber walls and the top of the chamber.
The chamber temperature was ramped from 25 C to -45 C in a 12 hour
period, held at -45 C for one hour or more, and then ramped back up to
25 C over another 12 hours.  The sonic data were sampled at 21 Hz and
statistics (means, variances, etc.) were calculated in 5-minute
blocks.  The chamber temperature was monitered with a slow-response
sensor whose low temperature limit is -40 C.

The principle measures of sonic performance that we looked at were the
5-minute variances of the three orthogonal wind components and sonic
temperature.  A more detailed look at a limited subset of the data used
10-minute calculations of the power spectra and cospectra.

1012R2A 0087 (NCAR) Solent: (97/6/25 13:30 - 97/6/26 14:30)
-----------------------------------------------------------

The performance of this sonic is acceptable to -35 C and somewhat
degraded at lower temperatures. 

The velocity variances gradually increased with decreasing temperature
from 0.05-0.10 (m/s)^2 at 25 C to 0.05-0.13 (m/s)^2 at -30 C.  Below
-30 C the variances increased more rapidly to 0.08-0.2 (m/s)^2 at -40
C.  Below -30 C there was evident noise contamination of the power
spectra in the inertial subrange, particularly for sonic temperature.
The contamination generally increased with decreasing temperature.

There was a noticeable and repeatable modulation of the variance
amplitudes at all temperatures, with a period of 6-7 C.  This
corresponds to one period of the 180 kHz sonic pulse.  Angus Raby of
Gill Instruments suspects that this is caused by reflections of the
pulses from the walls of the chamber.  This modulation was present for
all R2 Solents, but was not present for the R3 Solent or a Campbell
sonic.  However we were able to locate the Campbell sonic more
centrally in the chamber (further from the walls) and it operates at
400 kHz, which would cause more attenuation of the reflected signals.
Angus also suggested that the more sophisticated 32-bit signal
processing of the R3 and Campbell sonics is better able to discriminate
against reflected signals.  At the suggestion of Angus, we tried,
unsuccessfully, to attenuate the reflections with a thin layer of felt
lining in the chamber.  Thus we have shown that the modulation is not
an artifact of the chamber ventillation system, but we have not shown
conclusively that it is caused by reflections from the chamber walls.

1012R2 0067 (ETL) Solent: (97/6/27 17:30 - 97/6/28 18:30) 
---------------------------------------------------------

The performance of this sonic is acceptable to -35 C and somewhat
degraded at lower temperatures.

The velocity variances were 0.05-0.09 (m/s)^2 and roughly independent
of temperature from 25 C to -30 C.  Below -30 C the variances increased
to 0.06-0.1 (m/s)^2 at -40 C and to 0.06-0.13 (m/s)^2 below -40 C.  
The power spectra show various evidence of noise, particularly for
sonic temperature at temperatures of -40 C and below.

1012R2 0112 (UCI) Solent: (97/7/11 01:30 - 97/7/12 09:00)
---------------------------------------------------------

The performance of this sonic is acceptable to -35 C and somewhat
degraded down to -40C.  The performance is poor at -40C.

Note that this sonic was held at -45 C for 8 hours, rather than 1
hour, to simultaneously test other equipment.

The velocity variances increased with decreasing temperature from
0.03-0.05 (m/s)^2 at 25 C to 0.03-0.07 (m/s)^2 at -30 C.  Below -30 C
the variances increased more rapidly to 0.09-0.2 (m/s)^2 at -40 C.
From about -38 C to -44 C the variances increased very rapidly to
1-3 (m/s)^2 just below -40 C and then decreased again to 0.05-0.12
(m/s)^2 at -45 C.  Note that this local maximum around -40 C has a
width near the 6-7 deg C period corresponding to one wavelength of the
sonic pulse.  Without this local maximum, this sonic performed
similarly to the previous two sonics.  Note that this `resonance'
occurred as the temperature passed through -40 C on the decreasing
ramp, but not on the increasing ramp.

1012R2A 0129 (ETL) Solent: (97/6/26 15:30 - 97/6/27 16:30) 
----------------------------------------------------------

The performance of this sonic is not acceptable below -32 C.

The velocity variances increased with decreasing temperature from
0.03-0.15 (m/s)^2 at 25 C to 0.05-0.2 (m/s)^2 at -30 C.  Below -30 C
the variances increased very rapidly, with a local maximum between -32 C
and -37 C that had peak variances of 4-10 (m/s)^2.  Below a local
variance minimum at -37 C, the variances increased rapidly again to
levels as high as 50-150 (m/s)^2.  However around -45 C the variances
were down to 5-20 (m/s)^2.  Below -30 C there was evident noise
contamination of the power spectra in the inertial subrange,
particularly for sonic temperature.  At -35 C and below, there is very
obvious noise on the 21 Hz time series for all 4 output variables.

1012R2A 0104 (UCI) Solent: (97/7/8 23:00 - 97/7/10 16:00)
---------------------------------------------------------

The performance of this sonic is not acceptable below -35 C.

Note that this sonic was held at -45 C for 18 hours, rather than 1
hour, to simultaneously test other equipment.

The velocity variances increased with decreasing temperature from
0.03-0.05 (m/s)^2 at 25 C to 0.05-0.06 (m/s)^2 at -30 C.  Below -30 C
the variances increased very rapidly, with a local maximum between -38
C and -42 C that had peak variances of 50-200 (m/s)^2.  The sonic
stopped reporting for about 15 minutes just below -40 C.  However, at
-45 C the variances again returned to amplitudes of 0.04-0.10 (m/s)^2.

The power spectra show a small amount of noise in sonic temperature
at -30 C, very obvious noise in v and w and a small amount in
temperature at -35 C, and then a small amount of noise in u, w, and T
at -45 C.  At -40C there is very obvious noise in the time series for
all components.

1210R3 0019 (NPS) Solent:  (97/7/12 21:30 - 97/7/14 05:30)
----------------------------------------------------------

The performance of this sonic is not acceptable below -30 C.

Note that this sonic was held at -45 C for 8 hours, rather than 1
hour, to simultaneously test other equipment.

The u and w variances increased gradually with decreasing temperature
from 0.04-0.05 (m/s)^2 at 25 C to 0.04-0.06 (ms/)^2 at - 30 C.  In the
same interval the v variance increased rather sporadically from 0.11
(m/s)^2 to 0.2 (m/s)^2.  The periodic modulation of the variances seen
on the R2 sonics is not evident on this R3 sonic.

Below -30 C the variances increased very rapidly, with a local maximum
between -33 C and -40 C that had peak variances of 40-150 (m/s)^2.
However around -45 C the variances were down to 0.1-0.2 (m/s)^2 for u
and w and 1.2 (m/s)^2 for v.

97/07/14

Friehe Sonic:	1012R2A - 0112  Symmetric

Test Purpose:	Cold Temp. data quality assessment.
		The earlier test had reasonable results except for a
		bit of a problem when the -45 cold soak began. 
		This test is intended to determine whether or not
		the problem was a quirk or not.
 
Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	 25	00:10	soak		195/23:00	17:00, Mo 7/14
 2	-45	12:00	ramp-down	    23:10	17:10
 3	-45	18:00	hold		196/11:10	05:10, Tu 7/15
 4	25	12:00	ramp-up		    12:10	06:10
 5			End-of-Pgm	197/00:10	18:10


Configuration:	Gill is oriented diagonally in chamber with symmetric
	head toward the upper, rear corner away from the circulation
	system.   There is about >= 3" between the Gill framing tubes
	and the chamber wall.
	The TRH is resting on a foam pad about 1" above the chamber floor.

	Cosmos  202: Gill
		200: TRH

Notes:	Cosmos up

Results:

jm

97/07/29

NCAR Sonic:	1012R2A - 0087
Test Purpose:	Cold Temp. data quality assessment.
		

Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	-45	06:00	ramp-down	210/21:00	15:00, Tu  7/29
 2	-45	18:00	hold		211/03:00	21:00
 3	-35	 1:00	ramp-up		    21:00	15:00  We  7/30
 4	-35	 6:00	hold		    22:00	16:00
 5	 25	 6:00	ramp-up		212/04:00	22:00
 6			End-of-Pgm	    10:00	04:00, Th  7/31


Configuration:	Gill is oriented diagonally in chamber with asymmetric
	head toward the upper, rear corner away from the circulation
	system.   There is about >= 1.5" between the Gill framing tubes
	and the chamber wall.

	Cosmos  202: Gill
		200: TRH

Notes:	Cosmos was down for the first ~1.5 hours of the chamber run and
	archiving didn't begin until about 16:30 local on Tuesday afternoon.
	The temp. at that point was about 8-deg.

Results:


jm

97/07/29

ETL Sonic:	1012R2A - 0067
Test Purpose:	Cold Temp. data quality assessment.
		

Chamber Program
Step	Temp	Time	Type		Begin-Z(approx)	Local-Time
----	----	----	----		---------	----------
 1	-45	06:00	ramp-down	214/16:45	10:45, Sa  8/2
 2	-45	18:00	hold		    22:45	16:45
 3	-35	 1:00	ramp-up		215/16:45	10:45  Su  8/3
 4	-35	 6:00	hold		    17:45	11:45
 5	 25	 6:00	ramp-up		    23:45	17:45
 6			End-of-Pgm	216/05:45	23:45, Su  8/3


Configuration:	Gill is oriented diagonally in chamber with asymmetric
	head toward the upper, rear corner away from the circulation
	system.   There is about >= 1.5" between the Gill framing tubes
	and the chamber wall.

	Cosmos  202: Gill


Notes:	Cosmos up.   No TRH available for this test.

Results:


jm

The recent cold soak tests of NCAR R2A 0087 and ETL R2 0067 Solent
sonics produced results that appear to be entirely consistent with the
previous, shorter duration tests in the temperature chamber.

On Wednesday,25 Feb Jeff opened the ATI boom to check on the internal circuitry.
The flat connector on the internal circuit board was loose. The conector was 
reaffixed and we took the ATI out and remounted it on the Florida sonic boom.
Again no response.
Today, Thurs, 26 Feb we brought the ATI and eve into the lab. Again it was
discovered that the connector was loose. This time a cable tie was used to cinch
the connector tight. Also several test points were definately pushed out of 
accidental contact range. At ~1330 - 1400 AKST the ATI was reinstalled and it 
worked!!!!!!!!!!
Iplot of winds, u,v and w indicate that it is behaving.

The spare Gill was installed ( 8/24/98 )  1315 hours.  Config 52 was 
downloaded but how do I check to make sure it has the new and improved
Gill software, like the Fla Gill?  
The Gill is mounted very close to the mast and I need to bring out the Gill
size boom extention. Does anybody know where one was last seen?  I checked out
the PAM room and a quick search on deck, but no luck yet. Also, it 
was mounted with the North arrow 30 degrees CCW to the boom.
The "SHEBA PAM Operators Guide" said the usual mounting 
orientation of the Gill sonic on the PAM station is with the "N" arrow 
parallel to the sonic boom.  I looked at the mounting bracket at Fla today
and saw that only one screw is necessary to hold the bottom of the Gill
in place.  The data needs a 30 degree correction until I can get back 
out there and rotate it. 
The ponds where covered with ice and I moved the sled about a foot higher on
the ablation shield.

D. Costa 

I was able to stop by Alt on the way back from Maui and align the N arrow
with the boom.  The N arrow now points straight out from the boom. I haven't
found any of the Gill sonic booms.  The screws didn't line up on the bottom
bracket so the top clamp and the edge of the bottom bracket are
the only things holding it on.  This Gill is "Symetric" and the one at Fla
is "Asymetric".

Silly Gilly