Polarized Tsys using the lbw receiver
In jun02 a polarized
component of tsys at 5 Ghz was measured with the cband receiver.
On 27mar03 and 17apr03 measurements of tsys were done using lbw to see
how the component changed with frequency.
The setup and processing for these days was:
run the correlator with 25 Mhz bandwidth, at 1405/1415 Mhz, sampling
at 1 second intervals. Set the lbw receiver to linear polarization (hybrid
drive the telescope in az at .2 deg/sec (half slew) from az 270 to 630.
Do this at za=2,4.5,7,9.5,12,14.5,17, and 19.5 deg (reversing direction
on each swing).
Fire the cal 4 times during the measurements to convert to kelvins.
For each az swing compute TsysPola- TsysPolB.
Fit a 2 az sine wave to each swing: polDif=TsysA-TsysB=A*sin(2*az + phase)
The data was taken 10:40 to 15:00 on 27mar03 and 10:50 to 15:10 on 17apr03.
The data was also used to measure compression
of the system by the various radars and to watch the sun move in the
sidelobes of the telescope. Since the data was taken through noon, some
of the swings were corrupted by the sun in the sidelobes. Another problem
occurred on 27mar03 with a drift
in sbc1 polB caused by a loose cable in a chassis of the correlator
(this was later corrected).
The plots show the difference TsysA-TsysB
vs az and za for this dataset:
Fig 1. TsysA-TsysB is plotted vs azimuth. Each za strip is coded by color.
The za strips have offsets for plotting purposes. The data was taken between
270 and 630 degrees. Plotting 0 to 360 causes a discontinuity at az=270
The top plot is 27mar03. The strips at za =4.5 and 7 (red,green) have the
sbc1b instability. za=12 and 14.5 have sidelobes of the sun. The narrow
spikes are continuum sources.
The bottom plot is 17apr03. za=7 and 9.5 (green, red) have sidelobes of
Fig 2. 27mar03 TsysA-TsysB vs az. A 2*az sin wave is fit to the dataset.
The black lines are the data. The red line is the fit to the raw data,
the blue line is the fit after cumfiltering the data (see
cumfilter idl routine). The green * were the data points dropped by
the filtering process.
Fig 3. 17apr03 version of figure 2.
Fig 4. The fit amplitude an phase vs za. The really bad fits were not included.
Black is 27mar03, red is 17apr03. It is important to remember that the
phase angle is relative to the orientation of the probes in the receiver,
not the ground or the sky.
lbw 1400 Mhz has a .2 to .3 Kelvin polarized component to Tsys. This compares
to a .5 Kelvin value at 5 Ghz.
The amplitude decreases at the highest za.
The phase is repeatable between the two days (except for the smallest za).
It can not be compared to the cband value (or the ground) until the angle
of the probes is measured (or this could be used to measure the angle of
This really needs to be repeated at night to improve the quality of the
data (and see how much the sun is contributing to the value).