Observations of the Queanbeyan BPL Trial Nov 2004
Broadband over Power Lines is the latest incarnation of attempts to utilise power transmission and distribution infrastructure for electronic communications. Earlier efforts included PLC which was principally used for protection, supervision, telemetry and control, and PLT which included extension to more generalised telecommunications (eg voice channels).
Although BPL offers the opportunity for power industry telemetry and control (eg remote meter reading, dynamic load management etc), it depends on delivering commercial Internet access to subscribers for its economics.
Though the receiver test setup was relatively insensitive (using a -50dBi antenna), interference from the BPL systems was readily observable to medium strengths on the receiver, more than 35dB above the receiver internal noise.
Here is a short mp3 sound clip (10s / 40KB) of the SSB receiver being tuned across about 20kHz of the 40m band over 10 seconds. The beat notes are from the BPL carriers spaced about 1.1kHz apart, and the noise that is sometimes heard is the effects of data modulation of the carriers.
These carriers were measured at about 3.5µV at both precincts with a -50.8dBi gain small loop antenna system. With a half wave dipole mounted at about 10m and feeder (gain about 6.5dBi), one would expect about 2573µV, or about 34dB above S9 (assuming s9 equivalent to 50µV).
The electric field strength in 9kHz measurement bandwidth would be 10.8dBµV (3.5µV) + 38.6dB/m (Antenna Factor) + 4.77dB (bandwidth factor) or 54.3dBµV/m measured at a distance of 15 metres from the nearest power line. The electric field strength normalised to 1Hz bandwidth is 14.8dBµV/m measured at a distance of 15 metres from the nearest power line
At the time of observation, the system in Hinksman Avenue seemed to radiate frequencies from 5.5MHz to 8.0MHz, and the system in Campbell Street seemed to radiate frequencies from 5.5MHz to 8.0MHz, and 9.0MHz to 12.0MHz. It is not known whether they use other frequencies, eg in response to demand.
This study measured interference over a narrow band of frequencies around 7MHz, although the equipment radiated interference of similar intensity over a much wider range. The impact will vary with frequency.
The observed interference levels at 7MHz would be extremely disruptive of radio communications for hundreds of metres from the deployment, and would exceed the background noise level at distances over 1Km from the deployment. If this infrastructure was deployed on all LV distribution in a geographic area, it would render radio communications unusable at this frequency and probably all frequencies where it radiated such interference.
The infrastructure used in the Queanbeyan trial includes injection equipment that is mounted in plain view on power poles in the street. The LV distribution is 240/415V 50Hz three phase with neutral. Figure 1 shows the injector equipment mounted on poles at two locations in Queanbeyan.
Note the aluminium power line conductors at the Hinksman Avenue site. A recent report commissioned by government stated:
Clearly this contains errors of regarding the resistivity and permeability characteristics of the materials (copper, steel and aluminium), skin effect, inductance, capacitance, reactance, and understanding of theory and practice of transmission lines at radio frequencies.
The receiving antenna was a square loop with sides of 0.5m, and fed with a 1:1 transformer balun (RAK type BL) and 8m of RG-58C/U coax.
Choices for a measurement antenna fell mainly into two categories, antennas responding to the E field and those responding to the H field. Short dipoles respond to E field and have a very small equivalent series resistance and very high capacitive reactance. Small loops have a very small equivalent series resistance and moderate inductive reactance. It was felt that the impedance of a small loop would not be unsuited to the 1:1 balun that was available, whereas a short dipole would not be as suited to the balun. So, a square loop with 0.5 metre sides fed in one corner was constructed, the antenna is pictured in Figure 2. The loop is readily rotated and can be used to identify the source of radiation.
Measurement antenna system gain estimates are set out in Table 1.
Table 2 shows the Antenna Factor (E/V (electric field strength / receiver terminal voltage)) calibration estimates.
The measurement receiver was a standard IC-706IIG with pre-amp on.
The receiver S-meter calibration was measured about two hours after the field observations, and there is no reason to expect that they would have been different during the observations. Note that the reported receive signal strength in µV does not depend on any assumption of relationship between S meter reading and terminal voltage, the relationship was measured for the test receiver and the results are set out in Table 3.
The plan was to make measurements in the vicinity of the injector equipment, but not so close to the injector equipment to be subject to leakage if any from the boxes themselves. The criteria for a measurement location was a place on a public street 50 to 100 metres from the injector, and as far as possible from aerial power lines (either the lines down the street or branches to customer premises). This strategy was intended to avoid hot spots that would misrepresent the interference level that would be generally experienced. At both locations small changes in the antenna position did not make measurable changes in received signal strength. The loop antenna was always rotated for maximum signal as the antenna gain calibration was calculated on the major lobe of the loop.
The BPL Interference Evaluation Tool calculates the impact of radiation at the maximum level permitted under the draft CENELEC standard.
I am not sure of the equipment in use in Queanbeyan, either brand or model. I have searched the web site of the company that is rumoured to supply the equipment, but they are unusually shy about their BPL product range. I have no knowledge of its compliance with any standards.
Nevertheless, it is interesting to compare the measured interference level with the predicted interference from the maximum permitted radiation under the draft CENELEC standard, as predicted by the BPL Interference Evaluation Tool.
The measured value is +12dB +/- 3dB compared to the maximum impact predicted by the BPL Interference Evaluation Tool.
A Broadband over Power Lines (BPL) installation is operating in at least two separate precincts in Queanbeyan.
The BPL installation does radiate electromagnetic energy on a wide range of radio frequencies and the observed radiated signal levels are much higher in intensity than the background noise level.
The measured interference levels at 7MHz would be extremely disruptive of radio communications for hundreds of metres from the deployment, and would exceed the background noise level at distances over 1Km from the deployment. If this infrastructure was deployed on all LV distribution in a geographic area, it would render radio communications unusable at this frequency and probably all frequencies where it radiated such interference.
Use at your own risk, not warranted for any purpose. Do not depend on any results without independent verification.
V1.06. Last changed: 20 February 2009 19:25
© Copyright: Owen Duffy 1995, 2021. All rights reserved. Disclaimer.