power levels in lband wide while doing
The power levels at the input to the lband wide
receiver were measured on 22feb19. This data can be used to estimate
the dynamic range needed
for the new ALPACA (ao40) phased array feed. Data across all of
lband wide was recorded while doing 2 azimuth swings (with the dome
at 18 degrees za). The
cal was used to convert the data to degrees Kelvin (and then dbm).
The average spectra , total power vs azimuth ,
and total power vs azimuth for some known rfi sources is plotted for
each azimuth swing.
The levels shown are for the 50 millisecond averaged spectra. Radar
pulses with around 10% duty cycles will have peak values at least 10
db higher than the values plotted.
The aerostat radar balloon was not up during this test.
- The gregorian dome was set to za 18 degrees. The azimuth swung
from -90 to +270 degrees (and back) at .35 degrees/second.
- Data was recorded with the mock spectrometer
- 6 x 160 MHz bands were recorded covering 1112 to 1760
- spectra were dumped at 50 millisecond with 21 KHz
- A 10 second calOn,calOff was done before and after the first
azimuth swing to convert from spectrometer counts to kelvins.
Processing the data:
- The calon,off was used to compute the scaling factor from
spectrometer counts to kelvins.
- The average spectra was computed for the 10 second
calOn and calOff spectra.
- an attempt was made to exclude frequency channels with rfi
by fitting to calOn/Caloff and excluding freq channels
residuals greater than 3 sigma.
- the spccntsToKelvins was computed by using the cal value for
the center of each band and the mean value of (calon-caloff).
- outliers (from step 2) and the bandpass edges were
excluded in the average.
- the same channels were used in the actual spectra when
converting spc counts to kelvins.
- A band pass correction (to remove the IF bandpass) was
computed using the calon-caloff spectra
- a robust 13th order harmonic fit was done to calOn-caloff.
- let bpIF(frq) be the IF bandpass.
- assume Tsky,Tcal is constant in freq over the 160 MHz
- CalOn-calOff = (Tsky + Tcal)*bpIF - (Tsky)*bpIF =
- the band pass correction is then
Tcal*bpIF(freq)/mean(Tcal*bpIF) .. (where the average is
over the non rfi channels).
- Each 50 millisecond spectra was then converted to
kelvins by multiplying by spcCntsToKelvins/bpc
- PolA and polB were then averaged together.
- To convert from kelvins to dbm:
- dbm= alog10(degK * resolutionBandWidth*Kboltz)*10 + 30 ..
- The 6 x 160MHz spectra were interpolated to a single spectra
covering the entire band.
The average and peak hold spectra:
- The average and peak hold
spectra in degrees Kelvin (.ps) (.pdf):
- The vertical axis is degrees Kelvin.
- The black trace is the first azimuth swing, the red trace is
- top Frame: the average spectra over each azswing.
- The rfi above 1700 Mhz did not repeat in the 2nd swing.
Looks to be time variable (which is why it didn't show up in
the median spectra).
- 2nd Frame: the median spectra over each azswing.
- 3rd Frame: the peak hold spectra for the azimuth swings.
- This kept the highest value in each frequency channel
(20Khz resolution) for a swing.
- Bottom Frame: rms/Mean for each frequency channel over
each az swing.
- If we had just noise, the radiometer equation would give
sqrt(1./(21Khz*2pol*.05 secs) = .021..
- the second red swing had a higher rms/mean than the first.
I think this was caused by crossing more of the galactic
anti center (ra=6hrs) during the second swing.
- The average and peak hold
spectra in dbm (.ps) (.pdf)
- The average spectra in degK was converted to dbm using the
21Khz channel resolution.
- Black is the first 1st azswing, red is the 2nd.
- top Frame: The average spectra
- 2nd Frame: The median spectra over each az swing.
- bottom Frame: The peak hold spectra.
The total power vs azimuth
total power vs azimuth (.ps) (.pdf):
- The total power (1113-1758 MHz) was computed for each 50
- Black is the first az swing, red is the 2nd azswing.
- Top frame: the average temperature in deg K vs azimuth
- Bottom frame: The average power in dbm vs azimuth.
- This power is averaged over 50 milliseconds.
- The rapid changes are the various radars with their 12 second
- When they point at AO, the signal goes up.
- The radars have a duty cycle of around 10%.
- Their peak radar power (within the rf pulse) will be
at least 10db higher than the plotted values.
- The strong rfi around 130 deg az in the first swing looks to
have moved to around 110 deg az in the second.
- The peak value is around -89.5dbm
- the dynamic range is about 6.5 db (averaged over 50
Plotting some known rfi vs azimuth.
The power in some known rfi bands was computed
and plotted vs azimuth. This showed which frequencies were causing
the changes in the total power.
The radar bands total
power vs azimuth (.ps) (.pdf)
The top frame is the average total power for a swing. The other
frames show the power in a particular frequency band.
- Page 1 is az swing 1. Page 2 is az swing 2
- 2nd frame: the 4 frequencies used by the FAA carsr radar (more
- The 4 frequencies are plotted in different colors.
- The spikes are from the 12 second rotation of the radar.
- The carsr radars have around a 10% duty cycle.
- 3rd frame: puntaBorinquen radar (more info)
- This is the same type of radar (carsr) as the faa. it too
has a 10% duty cycle.
- bottom frame: the punta salinas frequency agile radar (more info)
- It was running in ModeA.. using only 4 frequencies.
- I only plotted 3 frequencies since one of them overlapped
- punta salinas has a short pulse/ipp: 51/1500usecs
=3.5% and a long pulse: 409/2600usec 15%.
- We are probably seeing the long pulses sticking out in the
The radar bands total
power vs azimuth (blowup in azimuth) (.ps) (.pdf)
- The same radar plots with the azimuth range limited to 70 to
100 degrees azimuth.
- You can see the individual 12 second rotations.
- You'll notice that the punta salinas has a large blanking
period as it points at AO.
- The faa and punta salinas have shorter blanking periods when
they point at AO.
Total power vs azimuth
for gps, iridium, and 1710-1750 Mhz (.ps) (.pdf)
Total power vs azimuth for
gps, iridium, and 1710-1750 Mhz (blowup in azimuth) (.ps) (.pdf)
- Page 1 is azswing 1, page 2 is azswing 2.
- 2nd Frame: total power for the 3 gps frequency bands (more info).
- The bump in total power at az=150 degrees was from a gps
- it was stronger in az swing 1, and weaker in azswing 2.
- The satellites are at 1/2 geostationary orbit, so their
movement on the sky is much slower than the LEO satellites.
- I didn't check how close to the beam the satellite came (i
could if there is interest).
- 3rd Frame: iridium satellite band (more
- the power was computed over 1617-1627 MHz. The official
iridium band is 1616 to 1626.5 Mhz.
- The 1st azswing has a peak around az=150 degrees
(similar to gps).
- The 2nd azwing has a large bump at 90 to 120 degrees
- The iridium satellites are about 780 Km above the earth.
They only remain above the horizon for about 15 minutes (more
- the time between the measurement of the two bumps was at
least 14 minutes, so the two bumps are probably coming from
different iridium satellites.
- The iridium satellites are making some of the largest
change in the total power for the 2 azimuth swings (in the 50
- Bottom Frame: power in the 1710 -1750 Mhz band
- I plotted 4 separate frequency ranges in the 1710 to 1750
- 1710-1720 and 1730-1735 were very strong in az swing 1.
- They were not present in the 2nd azimuth swing.
- These are the same gps, iridium, 1710=1750Mhz plots with the
azimuth range limited to az=100 to 180 degrees.
- In the 1st az swing, you can see the iridium satellite going
through the sidelobes of the telescope.
processing: x101/190222/azswinglbw.pro, docalonoff.pro
- the total power vs azimuth was computed for the lband wide
- the power varies by about 6 db in the 50 millisecond averaged
- The largest signals came from
- the 1710-1720 rfi. This only occurred once during the
- The iridium satellites. These occurred on each swing (from
- the gps satellites.
- The strength of a satellite signal will depend on how close
the satellite comes to the AO beam and sidelobes (so these
values are not the largest possible)
- The faa radars have a 10% duty cycle.
- The values show for the 50 millisecond integrations probably
need to be increased by at least 10db to get the peak power
during the radar pulses.
- The aerostat radar balloon was not up during this test. When
running it will have a value up to 10-15 db higher than the
- The values shown here depend on the geometry of the AO
beam/sidelobes and the location/timing of the rfi