Parkes RFI Studies - Multibeam Ripple
Reference Fault Report 1358, 1670, & 1779:Wideband RFI
The reported RFI, affecting P372, P410, P431 manifested itself as solar-like (ripple) interference, commonly seen on Livedata display, when daytime observations occur, and the sun's disk (broad-band RF noise) falls in a position such that, there is a direct reflection of the sun's energy from the telescope's tripod legs, to the underside of the focus cabin. This reflection causes a coherent, reflected signal of RF broad-band energy, to be coupled into the Multibeam receiver feedhorn array. The reported wideband RFI showed identical effects as the solar interference, but occurred in the middle of the night (moonless nights). At first, there was some correspondance of the RFI with farming activities that were happening in the night, ie, ploughing and/or sowing of crops, which utilisd the services of a tractor with ground speed monitoring equipment (K-Band radar), but this was later discounted, as it became obvious, that the solar-like interference was still evident, when the farming activities were finished. There was also some discussions as to whether the interference was being caused by satellites in the geostationary arc, which, seemed to correlate to some of the observations, in terms of their scans, being close to the geostationary arc. In particular, a number of geostationary satellites offer services at L-Band (around 1535-1559-1610MHz), which would provide a similar response as solar noise, ie, a transponder radiating the thermal energy of the earth, amplified, and over much of the spectrum, even with output filters. Again, this theory was later discounted, as the solar-like RFI was still present, even when the observations had later moved to regions of the sky, where the scans were well away from the geostationary arc, and from the satellites of concern. The nature of this RFI, was such, that it seemed to occur at odd times during the previous 2-3 years, with no real correlation to any hardware signatures, site black-out conditions (apart from the telescope), and time of appearance during the course of the night, apart from specific observing angles, ie, for the same observing conditons, in terms of telescope pointing parameters, the RFI could be there one night, and not the next. The only signature that could be recorded, was a large ripple, seen in the differenced spectral response, from the correlator display, which seemed to wander through the differenced spectrum, like it was drifting, or related to some multipath effect, and moved in frequency as the telescope position slowly moved in relation to the source. A differenced spectrum is obtained by taking a snapshot of the spectrum, and then subtracting subsequent spectrum displays from the reference. This technique of looking at differenced spectrum was the only way of looking for the RFI in real time, as the Livedata display, was sometimes, many seconds, or minutes behind, in displaying the current spectrum, as a waterfall plot of spectrum vs time.
In 2003, a staff member had some contact with one of the local farmers, a few km's away from the observatory, about some other non-observatory business, and the farmer mentioned, that, near one of the 11kV feeder poles, which runs across his property, that his am radio made quite a lot of static, and hiss. Subsequent investigations showed that at this pole, the insulator on one of the phases was well dusted, and depending as to whether there was any moisture at the time, the insulator was arcing over, causing significant sparking. In addition, the cable was well and truly damaged, requiring immediate replacement. The local authority for power distribution was alerted to this, and they replaced the faulty insulator on the offending pole, and cable. They have also checked on a number of poles in the district, for this particular type of insulator, as they have been known to cause problems with arc-over, and replaced any that they have found.
Unfortunately, since the discovery and repair of this pole, the Multibeam receiver has been removed from the focus cabin, for repair and maitenance, so it has been hard to ascertain as to whether the effect is still present, or not. Only with the return of the Multibeam receiver system, scheduled some time in the 2nd half of 2004, that an opportunity to ascertain the mitigation of the solar-like RFI has been successful.