SJVAQS91: SONICS Messages: 16 Entries..

sonic problems - 1d and UW 
 
I have had to replace 2 transducers today to get sonics going.  The first was on 
the conditional sampler's 1d ati, which wasn't sending data at all due to no sonic 
signal.  I replaced one of the transducers with one of my spares and got it going. 
The zero looks okay (w's about 0), so I guess it works.  Even though the transducer 
is a different shape in the back, it should have the same dimension, so I think the 
gain is okay.  Incidentally, this sonic should be LEVELED before really trusting the 
data. 
 
About 0130 tonight, I noticed that B on uw.ed had died and traced it to the B top 
transducer.  (This was in addition to the other problems this sonic had on path A.) 
I put in another one of my spares and got this going too.  Note that this sonic will 
have to be completely REZEROED, since I changed both gains and thresholds on both 
B receive cards.  At 0238, it finally is working again.  (Prior to this time, it was 
running but either had bad data or zeros - a failure mode fixed by cycling power.) 
I also raised thresholds on both A transducers. 
 
This raises the question as to the reliability of ATI transducers in the heat.  (ARM 
tested them in cold!)  I think a call to Herb and/or Ski is in order.  Thus: 
 
DO LIST FOR TOMORROW: 
- level 1d tower 
- zero uw.5m.ed 
- call Ski 
 
 

uw.5m.ed zeroed 
 
I've just replaced a B transducer (top) again, and have now computed zeros. 
They are: 
A -115   B -89   C -3  S 36877 (with T=27, Tw=18, p=1006mb) 
 
I did this by: 
 
ashow marigold 200 > A 
cut -10-80 A > uw1_zeroA.193 
Splus 
zeroa <- scan("uw1_zeroA.193",uwlist) 
hist(zeroa$a) 
mean(zeroa$a) 
q() 
 
and also for B,b and C,c to get all the means.  Note that 
s is from zeroa$s.  This seems like a nice way to do it. 
 
Am going to raise the 5m.ed mast now (it was dropped about 9:15 local). 
 
 

sonic "painting" 
 
Both uw sonics have been bouncing over the full-scale range recently.  I suspect the 
problem is overheating of the electronics boxes.  I'm working on sorting this out. 
 

ati 1d zeroed 
 
Since I had changed the transducer, and since any w bias is critical for the conditional 
sampler, I checked the zero on the 1d ati today.  After a lot of pain (I loosened one 
of the transducers by mistake and took a long time to find it), I captured the file 
/home/aster/projects/SQVAQS91/cals/ati1d.195 
The mean w from this file is -0.006 m/s, so the zero apparently is pretty good! 
 
 

UW sonics woes 
 
I have just traced the speed of sound (and maybe more?) problem to the 
74HC74 above the 32 MHz oscillator.  I have moved it to sonics #2, which  
has just started having the same problem.  I'll try to get a replacement  
tomorrow - at worst, I'll ask Steve to bring one.  I will have to trace through 
the schematic to see what effect this had on the data up till now.  All the 
wiring in UW #1 should be redone when we get back. 
 
 

uw sonic spiking AGAIN 
 
It has to be the same connector, but I don't know what more I can to to prevent 
this.  I guess I'll have to remake it for the 4th time. 
 
 

wet towel now on 5m.uw.eddy 
 
This sonic started painting, so it called for desperate action.  Problems 
started at an air temp of about 32C.  I'll think about airconditioner ductwork 
for a more permanent solution. 
 
 

uw eddy moved to shelter 
 
This was done to allow the electronics to be cooled, but requireed 
longer cables, which introduced noise.  We are about to take another 
set of zeros on it now. 
 
 

Very heavy spiking on all channels of  
5m.uw.ed. 
 
 

uw sonic on scalar flux tower is currently  
spiking very heavily. 
 
 

Some spiking on 5m uw sonic on scalar 
flux tower, but better than previously. 
 
 

Created new sonic calibration routine, sonic_uw_4a 
to account for new zero wind values for 5m uw sonic 
on scalar flux tower that were obtained yesterday. 
Edited prep.config files in ~aster/config and  
~aster/...results/covar.  The prep.config file in 
~aster/config had calibration routine _1, while 
the prep.config file in ~aster/...results/covar had 
calibration routine _3a. 
 
 

EDDY SONIC: 
 
	From 23:30 to 00:30 GMT, the eddy sonic was done. 
The electronic box was moved from the cond. sampler shelter 
to the ozone shelter. The signal cables were reduced to the 
short length. Initial cockpit look at the data shows good 
signals. 
 

uw sonics appeared to have run fairly cleanly 
last night, including the 5m scalar flux tower 
sonic that was moved (electronics) yesterday. 
 
 

EDDY SONIC: 
 
	The uw sonic on the eddy tower is 
still spiking. It doesn't appear to be as bad 
as before; however, there are more spikes than 
one would expect. Will try playing with the 
threshold levels. 
 

EDDY UW SONIC: 
 
	The sonic was moved from Dick's shelter 
to the outside so Dick could put his calibration 
instrument in the shelter. This took place around 
19:50 GMT.