SHEBA: Sonic Messages: 43 Entries..

Station 1 was setup with the sonic aligned parallel to the electronic compass
north, with the north position facing 'down the TRH xarm toward the tripod' and
90 CCW viewed from the top from the sonic xarm.

This configuration will cause excess wind obstruction from the sonic's mounting
arm and needs to be changed so that the north faces in the parallel direction
with the sonic xarm.

Station 3 was setup with the sonic aligned parallel to the electronic compass
north, with the north position facing 'down the TRH xarm toward the tripod' and
90 CCW viewed from the top from the sonic xarm.

This configuration will cause excess wind obstruction from the sonic's mounting
arm and needs to be changed so that the north faces in the parallel direction
with the sonic xarm.

Station 1 visited today:
Asymmetric Gill sonic realigned to have 'north' pointing outward from the
boom arm per the standard installation technique.

3/31/99
This appears to have occurred at 22:00 GMT, Oct 15 - TWH

Station 3 visited today:
Symmetric Gill sonic realigned to have 'north' pointing outward from the
boom arm per the standard installation technique.

3/31/99
This appears to have occurred at 19:00 GMT, Oct 15 - TWH

Sonic height measurements:

Date:		Sta-1		Sta-2		Sta-3		Sta-4
10/15 (setup)	346cm		346cm		345cm		
10/23 (moved)							348cm

1/25/99 John recalls that these heights were to the center of the
        sonic arrary or "sampling volume".

  Sonic 123 (?) was mounted at site 4 yesterday. The angle
of orientation was not taken and the sonic was not mounted
to its correct height. To date this site is being used
as a test site. The screws which hold the bottom plate of the sonic
to the mounting bracket were missing.

  The sonic flag counts started increasing at about
0:00 GMT. There was a sharp decrease in temperature,
from -10 C to -17 C, over a 3 hour period before the counts
went up.

  The problem with the sonics appears to be related to icing.
Station 4 transducers were covered. After cleaning the signals
were ok. 

  Sonics are stilled iced up at stations 1 thru 3. We cleaned
off station 4 and it came back to normal operation.

  The sonics continue to be a problem. Even after cleaning with
a broom signals do not look good. There maybe a fine film of ice
covering the transducers. Maybe a spray bottle of isopropol alchol
would help.

  Teh sonic scontinue to ice up. Even after we clean them the
data looks bad. At station 3 today we tryed using an alcohol swap
to help remove ice in the transducers. Based on the time series
plots there was some improvement but it still looked bad. If this is
a continuing process we need to give some serious thought on how
to prevent it. A heat source would probable be the best thing.
Something we turn on every few hours for a short time.

  We ran a test last night with the sonic at station 4. We
replaced the sonic with the spare to see how long a clean sonic
would last and if the iced up sonic was still functional. The
clean sonic lasted about 1 hour! The old sonci worked fine. We
may try to setup a test bead on ship wit the sonic outdoors to
try and solve the icing problem.

  The sonic response at stations 2 and 4 has improved to the point
where the data is believable. The temperatures have risen so the
icing may be shedding off. These sites have had their sonics cleaned
over the last few days while 1 and 3 have not. We will get to 1 and 3
today and clean the sonics. 

  The sonics were well behaved until about 9:00 local time.
Florida and Atlanta started coming back to life at 22:00 local.

> Some uncertainty has arisen about the exact method that you use to
> access the sonics when cleaning them, particularly how you are able to
> warm the transducers with your fingers.  Do you climb onto the tripod
> and then reach out to the sonic?  Do you remove the sonic from the
> boom?  Do you lower the mast to bring the sonic down to ground (snow)
> level?
        The very first time I cleaned a sonic beyond brushing it with the
feather duster, I climbed onto the tripod, and removed the one screw
holding the sonic base support to the horizontal arm.  With the sonic
temporarily upside down, I cleaned the heads.  This was rather cumbersome,
and didn't bode well for the future of the cable connector.
        Ever since then, we have taken a ladder along with us on the 
emergency gear sled.  I punch the ladder legs into the snow, climb up the 
ladder, and melt the frost with my fingers.

	Florida, being right by the ship, gets more frequent cleaning.  On 
some cleanings, we just step up on the tripod and clean the sonic frost off 
with the feather duster.

	If conditions are ripe for riming, no matter the method of cleaning, 
the sonics rime back up usually within 2 hours.

  The sonic at station 4 has been modified with the heaters. In
order to make the change, the TRH was modified and the new config
was downloaded. This took place at about 9:00 pm local.
The heater was set for 5 minutes on and 60 minutes off.
If everything goes ok we may get to Atlanta or Baltimore tomorrow.

The TRH instrument used was Florida. The instrument on Florida
before the change was the unit from Baltimore (refer to entry 190).

The heater turned on at 22:05 GMT. The compass showed a
~0.8 degree shift. The current load went up about .4 amps.
There was no noticeable change in T ro RH.

  The heater worked throughout the night. Looking at the response
of the sonic I would say it was a positive test. However, the level
of icing is low so further evaluation is needed. The change in the
TRH sensor, entry 199, may be related to the heaters.

  The sonic temperature changes ~2 degrees when the heaters are on.
Jeff and I went out and visual examined the sonic transducers. Eventhough
there is not major riming going on, there was some buildup on the sonic
support arms. The transducer heads were clean.

  At around 5:00 am local the error count flag on the sonic started.

UPDATE:  While installing the 4-comp, a visual inspection of the
GILL sonic showed alittle bit of icing on 2 of the transducers. The
heads were clean compared to the rest of the sonic.

1/21 17:54
	Changed heaton to 600.  (10 minutes)
	Changed heatoff to 6600.  (110 minutes)

1/21 17:54
	Changed heaton to 600.  (10 minutes)
	Changed heatoff to 6600.  (110 minutes)

1/22 18:55gmt
	Turned Atlanta's sonic heaters off (heaton=0) due to drop in battery 
voltage.

1/22 18:55gmt
	Turned Baltimore's sonic heaters off (heaton=0) due to drop in battery 
voltage.

1/24 00:05gmt
	Having trouble keeping sonic heaters turned off at Atlanta.  Set 
heaton=6 and heatoff=60000 at both Atlanta and Baltimore.

1/24 19:42gmt
	Set heaton=1 and heatoff=32000 at Atlanta.

2/9 ~23:15gmt
	Turned Florida's sonic heating cycle back on.  Heaton=600, Heatoff=6600.

2/26 16:47gmt
	Changed heaton=5, heatoff=3600 in Florida's TRH for sonic heater 
control.

At 1015 AKST downloaded a new configuration to Florida. The new configuration
utilizes sonic tc flag count rather than sonic w flag count as the critrium
for activating the ATI heaters. 

This is a repeat of the Gill orientation comment from the 'Florida Moved'
logbook entry:

The Gill sonic head was rotated so N arrow is now due north by removing two of 
the three set screws, since the N arrow was originally about 10-15 deg off
(CCW viewed from the top) from the sonic arm.

Installed ATI #3 on Cleveland for check against Fla.

Tower sited immediately to W of Fla tower.  Both sonics pointing N.
ATI boom: 333 + 23 variation
TRH boom:  66 + 23 variation

Notes:
- ATI is approximately .5-m lower than Gill.  Have not measured vertical
distance to snow surface (of today)
- TRH boom sent up was a full length unit, need to cut-off and reinstall
later.
- Clev/Fla sleds to S of towers.

From:	Steve Oncley, (some edits/additions from John M.)
Subject: sonic serial numbers


Enclosed is a summary I've just made of logbook comments pertaining to moving
the sonics.   According to my reading, the current status is:
Atlanta: ATI 0202
Cleveland: ATI 0303
Baltimore: Gill 123 (Kurt thinks that this has the "weird" mount)
Florida: Gill 87
Spares: Gill 104 (original spare);
	Gill  67 (old Cleveland, no heaters);
	Gill 112 heater installed (last used on Florida)

Notes on Rotated Mounts (North arrow with respect to boom):
	s/n 87	13   deg CCW viewed from top measured on site by compass.
	s/n 112 34.2 deg CCW viewed from top measured by machinist on ship
	s/n 67  47.3 deg  CW viewed from top measured by machinist on ship

s/n-87
I note that comments 33 and 349 should pertain to the same sensor (87), and 
indicate that the 13 degree problem developed during handling, rather than
being a problem during set-up.  Furthermore, this Gill should have had a 
"good" mount, with the screws installed, so slippage would not be expected.
[JWM note, this mount was not 'good' and as noted it did not slip.  In
order to install all 3 screws, the mis-machined mount required the sonic
head to be rotated.  When realigned 2 screws were removed to get it right]

s/n-112
On Oct 16th this sonic alignment was corrected for N outward from boom.
Noting that on Jan 15th it was removed from Atlanta, the heater installed and
put on Florida.  On feb 26th is was removed and replaced at Florida with an ATI.
It is unclear when the mount was rotated by 34.2 degrees.

s/n-67
Was installed at Cleveland on Oct 14th.  The next entry refers to retrieval
after the crunch of Clev. on Feb 13.  It is unclear, but possible that
the mount was incorrect at setup.  Possibly notes buried in Daily_Status
may shed light on these questions.

-----------------------------------------------------------------------
Logbook	Date	Site	Sonic	Comment
Comment			s/n

26	oct 11	all		Gill: aligned with 'u' north to parallel the 
				north on the TRH compass so that they are 
				reporting similarly even though the xarms are 
				90-deg off.
31	oct 12	1	112	Carl Friehe's R2 sonisc s/n 112? installed
33	oct 13	3	87	NCAR asymmetric R2A sonic s/n 119? installed
37	oct 14	2	67	ETL asymmetric R2A sonic s/n
38/41	oct 16	1	112	[A]symmetric Gill sonic realigned to have 
				'north' pointing outward from the boom arm
39/42	oct 16	3	87	[a]Symmetric Gill sonic realigned to have 
				'north' pointing outward from the boom arm
212	jan 13	4	123	Removed Florida's sonic and TRH for 
				installation at Baltimore.
		3	87	Installed Baltimore's sonic and TRH, with 
				sonic heater modifications made to them, at 
				Florida.
216	jan 15	4	87	Removed sonic and TRH from Florida to install 
				at Atlanta.
217	jan 15	1	87	Installed modified TRH and sonic with head 
				heaters.
221	jan 17	4	112	Installed Atlanta's old TRH and sonic at 
				Florida.  Sonic now has head heaters on it.
258	feb 13	2	67	Retrieved in apparently operational condition 
				... the sonic unit...
267	feb 20	4	112	Inspected Florida in order to decide how to 
				replace Gill sonic with ATI sonic 
282	feb 26	4	0202	install ATI and to try to get ATi to operate.
328	mar 21	1	0202	Replaced Gill sonic with Fla ATI
329	mar 21	4	87	Gill sonic from Atl put back on
348	apr 1	4	87	Florida moved.  The orientation of the sonic 
				is at 360 deg true and the T/RH is at 86 deg 
				true.
349	apr 1	4	87	The Gill sonic head was rotated so N arrow is 
				now due north by removing two of the three set
				screws, since the N arrow was originally about
				10-15 deg off (CCW viewed from the top) from 
				the sonic arm.
362	apr 6	2	0303	Installed ATI #3 on Cleveland for check 
				against Fla. Both sonics [Fla] pointing N.
				ATI boom: 333 + 23 variation
				TRH boom:  66 + 23 variation
				ATI is approximately .5-m lower than Gill.  
				Have not measured vertical distance to snow 
				surface (of today)

4/12 2:20Z
	Reinserted the GPS cable into EVE which had been out.
	With the GPS declanation, and updated orientation.
	the Splus plots should now come out right.

	Confirmed ATI sonic was reporting consistent values on EVE
	for the observed wind.

Sonic height above the snow were measured today:
	Atl: ATI	89"
	Cle: ATI	111"
	Fla: gill	131"

4/13 18:25Z
Anticipating a trip to Balt when the weather is better, removed ATI
from Clev.

Site Visit to Atl to 'plywood' protect pentapod:

~21Z	The ATI sonic had not shown 'flags' or heating cycles
	overnight despite that Fla/Sea (Gills) had and there
	was a pretty good freezing mist / super cooled fog.

	The ATI V transducer facing the wind was pretty well
	covered in rime and ice which had to be melted off
	by hand.  There was some other ice on the U component.
	Otherwise the 'W' looked OK.

TWH changed site number from 2 to 1, 1/22/99

4/22 ~18:30
	Seattle: 
		ATI #980303 installed
		Gill #123 removed
		ATI boom height = 114"

	NOTE: the s/n labeled on this sonic conflicts with the electronic
		serial number.
		The number above is the electronic id.

during unpacking, the ati sonic v vector was discovered to be bent.
it is not known at this time how long it has been bent. v vector is
reporting 2 M/S in the lab, and v samples is low (15 instead of 20).
Three possibilities are: 
1.) it happened during a bear visit (last one at maui was about
    one month before teardown
2.) it happened during teardown
3.) it was man handled in Customs

I don't think that the ATI array was bent during the data collection for SHEBA.
As the attached plot shows, the wind speeds generally agreed between Maui and 
Florida before and after the bear visit and just prior to Maui teardown.  
However, Kurt has noticed that v on this sonic now has a 2.5 m/s offset, which
should show up on this plot if it were present.

Furthermore, Kurt noted that the vsamples are now low, but they were still
solidly 20 after the bear visit.  (They were down to 0 the day before
teardown, but this probably is due to riming since usamples and wsamples
also are low - though not 0 - during this period.  Also, open leads close
to the station were noted during this period which would have supplied
moisture for riming.)  (vsamples is shown as the lowest line on this plot,
normalized by 20 to have it fit on scale.)

...Steve Oncley

11-Feb-1999

Kurt looked at the ATI sonics returned from SHEBA and found the 
following:

Atlanta ATI was tagged S/N 980202 (ATI #1?) and the internal digital
serial number is also 980202.  This sonic was removed from Atlanta in
August because of a failed v axis.

Maui ATI was tagged S/N 980302 (ATI #2) but the internal digital serial
number is 980303.  This sonic was found to have a misaligned v axis.

Baltimore ATI was tagged S/N 980303 (ATI #3) but the internal digital
serial number is 980302.


From oncley@ale.atd.ucar.edu Tue Feb 16 16:15:13 1999
Subject: Re: ATI S/N
To: horst@atd.ucar.edu (Tom Horst)

> Steve:
> 
> I found a note with the SHEBA ATIs stating that S/Ns 980203 and 980202
> had been swapped.  Did you do this?  Did you swap the external labels
> or change the internal serial numbers?
> 
> Tom
> 
As far as I know, we received the ATIs with the serial numbers in ROM
not matching the labels on the boom for these 2 units.  I recall noticing
this before they were sent to SHEBA, but apparently forgot to do anything
about it.  Kurt noticed that they were still mismatched upon return,
so I (yesterday) changed the ROM values to correspond with the boom label.

Since these serial numbers do not appear in the data stream, I assume that
the labels on the boom were used by folks in SHEBA to identify the sensors.

...Steve

I have been reviewing and modifying the sonic orientation file
/net/pam/projects/SHEBA/sonics/orientation.dat.

I tested the orientations for Oct through early Jan by plotting wind
direction differences with Florida as the reference.  Data were used
only when sonic.samples exceeded 99.5% and I also clipped the differences
at +/- 50 deg.  A summary of the data follows:

Another Time Series (ats):
Observations: 25941  on 3 channels

       dir                  dir                   dir            
 Min.   :   -50.000   Min.   :-5.0000e+01   Min.   :   -50.000  
 1st Qu.:    -1.556   1st Qu.: 8.7790e-02   1st Qu.:    -1.861  
 Median :     2.209   Median : 3.3890e+00   Median :     3.980  
 Mean   :     1.464   Mean   : 2.7720e+00   Mean   :     2.330  
 3rd Qu.:     5.507   3rd Qu.: 6.0470e+00   3rd Qu.:     7.176  
 Max.   :    50.000   Max.   : 5.0000e+01   Max.   :    45.350  
 NA's   : 21287.000   NA's   : 2.1429e+04   NA's   : 22541.000  

Time Parameters :
                          start                            end   units deltat1 
 [1997 10 15 jd:288 000230.000] [1998 01 12 jd:012 235730.000] seconds 300.000
 
Looking at the absolute value of the median differences, 

Atl - Fla = 2.2 deg
Cle - Fla = 3.4 deg
Bal - Fla = 4.0 deg
Atl - Cle = 1.2 deg
Alt - Bal = 1.8 deg
Cle - Bal = 0.6 deg

Many of the Gill sonic "height" measurements in the logbook are actually
the height of the sonic boom.  According to logbook entry #659, the
center of the Gill sonic array is 60 cm above the center of the boom.

Kurt measured this distance in the laboratory and found it equal to 58 cm.

Sonic height measurements

   Station 1         Station 2         Station 3         Station 4
Date       ht(m)  Date       ht(m)  Date       ht(m)  Date       ht(m)		


97 10 16   3.46   97 10 16   3.46   97 10 16   3.45   97 10 23   3.48 
98 04 11   2.26   98 04 11   2.82   98 04 22   2.46   98 04 11   3.33 
98 04 17   2.26   98 04 18   3.45   98 05 15   2.34   98 04 21   3.20 
98 05 22   2.22   98 05 17   2.87   98 06 18   2.63   98 05 18   3.42 
98 06 12   2.39   98 06 12   3.02   98 06 30   2.68   98 06 08   3.47 
98 06 22   2.51   98 06 23   3.03   98 07 08   2.69   98 06 13   3.48 
98 07 06   2.77   98 07 07   3.12   98 07 30   2.97   98 07 29   3.51 
98 07 21   2.87   98 07 23   3.24   98 08 10   2.97   98 08 26   3.53 
98 08 03   2.92   98 07 30   3.23   98 08 24   2.93   98 09 08   3.51 
98 08 29   3.57   98 08 26   3.28   98 09 20   2.92   98 09 20   3.50 
98 09 16   3.48   98 09 16   3.26                     98 09 30   3.52 
98 09 30   3.31 

Added 0.58 m to heights of center of boom for Gill sonics