Birdie pixel 0b of alfa receiver
resolved 08sep04 (see below).
A birdie has been seen in Pixel 0b of the alfa receiver
near 1380.7 Mhz. It is .5*tsys in amplitude and has a fwhm of .7
Mhz (900 second average). Over a 15 minute integration it can wander up
to a few Mhz. The problem has been there since at least 21aug04 (the earliest
data set i looked at). On 7sep04 the first LO (high side) was moved up
by 1 Mhz and the position of the birdie moved down by 9 Mhz. The LO was
moved down 2 Mhz and the birdie in the rf band moved down by 18 Mhz.
If this is an intermod, then it is the 9th harmonic of the 1st LO beating
with a hiFreq birdie in the system. The hiFreq birdie would be at:
HiFreqBirdie= (lo1*9) + (if1 - (rfCfr-birdieInRf))= 14960.7 Mhz
On 03sep04 a 15 minute drift scan using 100 MHz
centered at 1385 Mhz was taken. The 900 records were averaged together.
show the birdie in pixel 0B. The red linear are polA and the
green lines are polB. The plots are labeled B1(pix0) thru B7(pix6).
Fig 1. This is the 15 minute average of the data. The birdie is in pixel
0b (B1 green line). pix0a has no power since that amplifier is dead.
Fig 2. A baseline has been removed and then the data was scaled to units
of Tsys. Top Plot pix0: The green birdie is .5*Tsys and has a fwhm
of .7 Mhz. Bottom plot pix1: The bottom plot is pixel 1. There is
small birdie at 1380 Mhz in all of the other bands.
Fig3. The top figure plots the frequency of the birdie in the rf
band versus the center frequency of the rf band (assuming 250 IF center
frequency). The red portion of the line marks the range of rf center frequency
that would see the birdie in a 100 Mhz band centered at the RF center frequencies.
The bottom figure is a plot of the center frequencies for all of
the scheduled alfa observations from 20aug04 thru 09sep04. Only the observations
with a center frequency of 1385 Mhz (A1946) would have seen the birdie
in their band. The other observations would have the birdie outside of
the observing band.
Resolution of the problem.
On 08sep04 the module that converts from RF to
IF and then to fiber was replaced for pixel 0 of alfa. After doing
this, the birdie went away. When this same module was probed in the lab,
the birdie was present.