Ripple in pixel 4A seen in alfalfa (wapp) data

    Alfa pixel 4 polA has a ripple in the bandpass with a period of 174 kHz. It was discovered while debugging the galfa spectrometer back in sep04 (see pixel4aripple sep04).

    Project a2010 (alfalfa) does 600 second  drift scans dumping once a second using 100 Mhz centered at 1385 Mhz. A single drift from 21may05 was examined and shows that the ripple is still present in pixel4a. The band passes for pixel4 with and without bandpass correction were examined to see the strength of the ripple and how stable it was over time.   The plots show the band passes and the ripple for the 600 second drift (.ps) (.pdf) :

• fig 1 No bandpass  correction:
• Top: The bandpass averaged over the 600 seconds. The edges of the bandpass have been removed to exclude the 1350 radar. Black is polA, red is polB. The black trace looks noisier than the red one.
• Center: A 10 Mhz blowup of the average band passes. You can clearly see the ripple in polA.
• Bottom: Each spectra was fft'd and then the magnitude computed. The average was then computed over all of these spectra. The spike at 5.78 useconds is the ripple. It has an amplitude of .0033 Tsys or about .09K. The corresponding ripple frequency is 173 Khz.
• Fig 2 With bandpass correction: The median bandpass was computed and then divided into each 1 second spectra.
• Top: This is the average of the bandpass corrected spectra. The spike at 1380 Mhz is a harmonic of the 1350 radar folding n.
• Middle: A 10Mhz blowup of the average bandpass. You do not see the ripple here.
• Bottom: The spectra were fft'd and then then magnitude were computed and then averaged over. The ripple spike is still present at 5.78 Usecs although it is not much weaker : .0002 Tsys or .007K.

Summary:

• The 174Khz spike is still present in alfa pixel4a. It is not seen in any other pixel.
• The amplitude of the ripple is about .09 Kelvins.
• The strength of the ripple can be reduced by about 11 db if you divide each spectra by the median. This was computed over a 600 second time period.
• 5.78 usecs is a distance of 150m*5.78=867M in air. If the reflection is in the fiber optic cable then the distance would be about 580 meters.