On 12apr21 12meter xband spectra were dumped at
a 1millisecond rate to check on the dropouts we had seen on
the oscilloscope (more info). We later
found that these dropouts were caused by the sband
portion of the signal saturating the system (probably
the Fiber tx ?). On 26apr21 we retook some fast
dump spectra with the sband signal terminated (at the input to
the postamp chassis in 50 ohms).
The setup we used was:
- Telescope sitting at az=120, el=41.285 . It was a little
after noon so the sun had transited and was not close to the
- The sband signal was terminated at the postamp chassis
input so the system would not go into compression
- take 7 172 MHz bands covering xband,.
- take 512 channel spectra across 172MHz every .238
milliseconds for 180 seconds.
Processing the data
- For each .238 millisecond spectra compute the total
power (2pols by 7 freqbands)
- there were 756000 spectral/totalpower samples in the
180 seconds of data.
- When computing the spectra of the total power
- spcTp= abs(fft(totalPwrTmseries - meanvalue))
Displaying the results
power time series for each freq band (.ps) (.pdf)
- Each frame is as separate 172 MHz freq band
- Black is polA, red is polB
- The .238 millisecond samples have been smoothed/decimated
to 7.38 millisecs (so the file doesn't get too big)
- The timeseries has been divided by it's mean value (to put
it in units of Tsys).
- offsets have been added for display: +.01 for polA, -.01
- The slow variation is similar for all 7 freq bands. It is
a about 1% of Tsys..
power spectra: polA (.ps) (.pdf)
power spectra: polB (.ps) (.pdf)
- The frequency of each band is plotted in a different color
with an offset for display.
- The spectra was smoothed/decimated to .056 Hz frequency
- You can see birdies at 60hz, 120hz, and around 1Khz.
- The 1Khz spike is seen in polA and polB
of the total power spectra (.ps) (.pdf)
- These spectra have the full freq resolution: .0056 Hz
- Page 1: blowup 0 to 2 hz
- top is polA, bottom is polB
- each color is the spectra from a different 172 MHz band
- Nothing stands out.
- Page 2: blowup around 1020 Hz.
- top polA
- The birdie is centered around 1018.71Hz
- Bottom: polB
- the birdie is centered around 1021.589
- it is a little stronger and wider than polA
- The 1KHz birdie is probably coming from the fiber optic
- It is dithering the output at 1Khz.
- Each fiber optic transmitter has a slightly different
freq (differs by about 3hz).
- the total power varied by about 1%.
- We should probably try this at night to make sure
the sun is not coming in through a sidelobe.
- There is a 1018.7 Hz birdie in polA
- there is a 1021.589 birdie in polB
- These may be coming from the dithering of the fiber