Wide band look at AC birdies in 327 rcvr.
oct05
Plots/images:
dynamic
spectra for the 170 seconds of data (.gif):
average
spectra when the Ac was on and off (.ps) (.pdf)
The 2 air conditioners in the dome (18K Btu, 33K
Btu) generated a comb with a 1 Mhz spacing. It was seen in the
327
and 430 Mhz receivers. The comb was coming from the digital portion of
the AC unit (used for the thermostat and the hand held communications).
A new 33K Btu unit was purchased. It as well as the old 33K Btu unit
were
fitted with an analog thermostat and the digital control was removed.
The
smaller 18K Btu unit was not modified.
On 22feb06 a test was done to see if any 1 Mhz combs
remained in the 327 receiver. The test consisted of:
-
Run the two new AC units continuously.
-
Turn the older AC unit on and then off to generate the comb.
-
A 25 Mhz band centered at 320 MHz was used with 2048 channels.
-
170 1 second records were taken. At second 84 the old Ac unit was
turned
on. At second 148 it was turned off.
The images show the dynamic
spectra for the 170 seconds of data (.gif):
The top plot is polA. The bottom plot is polB. Horizontal dashed lines
have been drawn where the old Ac unit was turned on and off.
The next plots show the average
spectra when the Ac was on and off (.ps) (.pdf):
-
The top plot is polA, the bottom plot is polB. The red lines are when
then
old Ac was on. The black lines are when the old Ac was off (they have
been
offset for display). The dashed red lines line up with the 1 Mhz
comb when the old Ac is on.
Conclusions:
The combs from different Ac units had the same comb
spacing but there was a frequency offset between combs. There is lots
of
rfi left in the band when the old Ac was off, but not of it looked like
it was part of a 1 Mhz comb.
Another test would be to try and turn off the new
Ac units (if we can find their breakers) and see if and birdies go
away...
processing: x101/060222/327ac.pro
home_~phil