�� . �� : http://www.naic.edu/~phil/usrproj/s3039/s3039.html
�� : Fri Feb 19 17:03:45 2016
�� : Mon Apr 11 01:51:39 2016
��:
�� : � � � � � � � � � � � � � � � � � �

T2874 meteors on moon.

S3039 meteors on the moon with alfa

may14

Setup
observing notes
spectra
telescope position

Experiment setup

Veena and john mathews looked for meteor flashes on the moon using alfa (also see t2874 meteors on moon at 430). The experiment used the alfa receiver (lband) at arecibo tracking the moon.

The AO setup was:

alfa receiver (rotation angle fixed at 0deg)
Alfa notes:

beam 3b was turned off .. it was dead
beam 6b has been bad.. but it think it had recovered (check cal deflection).

jpl horizon ephemeris file was used to track the moon.
one 6.4 MHz band centered at 1415 Mhz
mock spectrometers used in voltage sampling mode

the 6.4 MHz band was complex sampled and recorded 16 bits (both pols)
a +10Mhz analog offset followed by a -10 Mhz digital offset was used to get rid of a/d DC and analog 60 hz.
output was done using the pdev file format
see pdev file format for the output file format.

Observing notes

14feb16:

Start file: /share/pdata[1..7]/pdev/s3039.20160214.b[0..6]s1g0.00000.pdev
there were 232 files for each alfa beam
All times below are approximate.. except the start time is exact
wait till on moon.
adjust power levels for 1 sigma = 15 counts

most were higher since all gain was removed and still about 15 counts.

16:49:54 (ast) start taking data (exact)
16:51:31 high cal turned on (manually)
16:5200 high cal off (manually)
19:16:00 high cal on .. still on moon
19:16:30 high cal off
xxxx off the moon
19:29:30 high cal on (off the moon) (rms probably now has too few counts)
19:30:30 high cal off
19:30:xx done with observation

06may14

start file: /share/pdata[1..7]/pdev/s3039.20160215.b[0..6].s1g0.000000.pdev

had 3 separate scans

000 .. 021
100 ..172
200 .. 342
Total of 238 files/beam

Waited till on moon.
adjust mock power levels. 1 sigma=15 counts
17:40:55 started scan file .00000.
17:32:00 high cal on for 30 secs
17:32:30 high cal off
17:05:00 cal on 10 seconds
17:56:41 started scan file 00100
18:48:34 start scan file 00200
20:22:00 high cal on for 30 secs (still on moon)
20:22:30 high cal off
20:27:00 high on for 30 secs (off the moon) rms counts probably now too low
20:27:30 high cal off
20:27:xx done with observation

Quick look at the spectra:

The first 50 files of run (300-349 05may14, 100-149 06may14) were used to examine the spectral density.

The processing:

2^24 points from the start of each file were processed for the first 50 files of scan 300.
the 16 million points is about 1.6 seconds of data
8k fft's were done (1.2 Khz resolution), power computed and then averaged for the 1.6 seconds
Each 1.6 sec spectra were normalized to their median value (Tsys).

The plots below show the 1.6 second average spectra from the start of the first 50 files:

date	433 MHz band		438 MHz band
	polA	polB	polA	polB
05may14	(.pdf)	(.pdf)	(.pdf)	(.pdf)
06may14	(.pdf)	(.pdf)	(.pdf)	(.pdf)

Notes on the spectra:

Both days:

rfi in polA looks a lot stronger than that in polB
rms's on moon were 30 to 40 counts for 1 sigma in the 8 bit digitizer.
The cal temps are 95.5, 102.6 for polA, polB
On the oscilloscope we saw rfi with 12 second periods reflected off the moon.

land base radars (in PR) have 12 second rotation period.
I've seen ship based radars with 4 second rotation periods.

05may14:

moon come into beam during the 12th file : .00011.pdev
Tsys was about 130 K off the moon.

06may14

Tsys was about 80 K off the moon.
file 121, 431.8MHz birdies starts drifting to higher frequency.

at file 149 it has moved about 1 MHz.
This may be rfi from the sbn postamp chassis. in the 327 receiver we saw it drift by this amount (more info)

The rfi off the moon looks much stronger in polA than polB (pols are circular).
this may be real or just a difference in Tsys for the 2 pols (when the dewar is warm.. need to check the cals).

processing: x101/14050[5,6]/moonjm.pro

Telescope position.

The telescope position for the two runs was taken from the telescope logfiles. These files have the encoder position information dumped once a second.

Processing:

model correction was removed from the az,za encoder values
the az,za was backcomputed to ra,dec of date.

The plots show the telescope and moon ephemeris position for the 2 days (.ps) (.pdf):

Page 1: 05may14:

Top: RA (of date) vs hour of day (UTC)

black AO telescope
red moon ephemeris file

2nd: DEC(of date) vs hour of day (UTC)

black AO telescope
red moon ephemeris file

bottom: position error Asecond (great circle)

Black RaEphm - RaTelescope

the error was computed as: (radif)*15*3600*cos(declination)

red : decEphm - DecTel

The error was computed at: (decDif)*3600

at 21.6 the telescope tracking was stopped so the moon drifted out of the beam

. a few minutes later we drove back onto the moon.

Page 2: 06may14 run.

at 22.9 the telescope tracking was stopped so the moon drifted out of the beam

. a few minutes later we drove back onto the moon.

processing: x101/140506/moonpos.pro

 home_~phil