OwenDuffy.net 


 

An interpretation of the ACMA report on the Aurora Energy BPL trial in Hobart, Tas  on 11-12 January 2007.

This report is an interpretation and analysis of the ACA's report on the Aurora Energy BPL trial in Hobart, Tas on 11-12 January 2007.

Contents

Background

The ACMA published a report on 31 May 2007 entitled "Report on Measurements of BPL Emissions performed 11-12 January 2007". The report is an analysis of the ACMA's measurements of electromagnetic field strengths and receiver voltages resulting from radiation from that BPL trial. The full ACA report is available online from the ACMA's website at http://www.acma.gov.au/webwr/lib284/measurements%20bpl%20emissions%2011_01_2007.pdf .

Analysis

This document is an interpretation of the ACMA's report and an analysis of the impact of the measured emissions on a typical amateur radio station.

ACMA Tests 1-3

Tests 1-3 of the ACMA suite was a scan of HF using an amplified loop antenna and EMC receiver.

Fig 1: ACMA Tests 1-3 graph with added expected ambient noise line

Fig 1 shows the graph of ACMA's Tests 1-3 measurements with the addition to scale of the expected ambient noise level per ITU-R P.372-8. It can be seen that the "Background at Quoin Ridge" is around 20dB higher than the expected ambient noise per ITU-R P.372-8 which suggests the this measurement is limited by and mainly reporting the instrument internal noise.

Though the report identifies a "Rohde & Schwarz HFH2-Z1 Loop Antenna" and shows pics of a loop, the -Z1 is a rod antenna. The Rohde & Schwarz HFH2-Z2 Loop Antenna specs specify a worst case lower limit to measurement range in 200Hz bandwidth of 10dBμV/m to 5dBμV/m from 1 to 30MHz for an average indicator, and 23dB higher for QP indicator. Adjusting for 10kHz bandwidth (+17dB), the specified lower limit would be 27dBμV/m to 22dBμV/m from 1 to 30MHz for an average indicator. The process of S-plane summary of separate readings in x, y and z planes will result in a further increase of 4.8dB in S-plane summary of the instrument noise floor, so a realistic noise floor for this instrument with S-plane summary of measurements with an average indicator is 32dBμV/m to 27dBμV/m from 1 to 30MHz. The 23dB factor for QP indication is a further increase in the effective noise floor should that mode be used for accurate QP recordings in impulse scenarios.

It can be seen from Fig 1 that emissions are up to 55dB above the expected ambient noise level, and even in the presence of the notch in the 14MHz amateur band, emissions at 45dBμV/m are 37dB above the expected ambient noise level at 7.8dBμV/m.

ACMA Tests 4, 9

Test 4 explored the notches in the 14MHz and 21MHz amateur bands. From the supplied graphs:

  • the notch in the 14MHz band appears to notch less than the full band and has an average best depth around 22dB; and
  • the notch in the 21MHz band appears to have an average best depth of at least 18dB, but instrument noise floor as probably limited assessment of the true notch depth.

Test 9 shows the notch in the 18MHz amateur band, and again it appears to notch less than the full band and has an average best depth around 21dB. 

ACMA Test 6

Test 6 was a scan of HF using the ESCI receiver at 10kHz measurement bandwidth and an amateur's two element quad antenna.

Because the antenna is a narrow band antenna designed for 14.2MHz and its performance reasonably predictable at that frequency, the following analysis focuses on just that frequency. The band was subject to BPL emissions and a notch had been applied.

Data for this section is taken from graphs in the ACMA report and is subject to small error in scaling the values from the graphs. The relevant graph Y axis is labelled "Emissions dBµV/m (not corrected)", but discussion with the ACMA reveals that the values assume an antenna factor of 0dBµV/m, and so they actually represent receiver terminal voltage in dBµV.

Table 1: Interpolation of measured receiver voltage to a typical amateur receiver
Frequency (MHz) 14.2
Measurement with R&S ECSI receiver
Bandwidth (Hz) 10000
Rx voltage (dBµV) 25
Interpolation to amateur receiver
Bandwidth (Hz) 2000
Rx level (dBm) -89.0
Rx level (S-meter) S6

Table 1 shows the calculation of received signal level in dBm and S-meter reading in a typical 2kHz bandwidth amateur receiver from the ACMA receiver voltage measurements in a 10kHz wide receiver.

Table 2: Calculation of emission / expected ambient for Quad antenna (Test 6)
Measurement  
Frequency (Hz) 1.42E+07
Bandwidth 10000
Rx voltage (dBµV) 25
 
Expected ambient noise
ITU-R P.372-8 curve B
Label Residential (curve B)
Ambient FS (dBµV/m) 7.8
 
Quad antenna - expected ambient
Directivity (dB) 0.0
Loss (dB) 1.0
Ant gain (dBi) -1.0
AF (dB/m) -5.7
Rx voltage (dBµV) 13.6
 
Increase in noise floor (dB) 11.4

Table 2 shows the calculation of the expected ECSI receiver voltage from ambient noise per ITU-R P.372-8 using some reasonable assumptions for the quad antenna behaviour, and the increase in noise floor suggested by the actual measurement.

Fig 2: Emission / expected ambient for quad antenna (Test 6)

Fig 2 shows the measured emissions relative to the expected ambient noise voltage, noise floor degradation being 11.4dB, or approximately 2 S-units.

ACMA's OPINION’S AND CONCLUSIONS

The report contains quite some significant and arguable material in this section.

The most concerning is the discussion regarding compliance with the FCC standard, presumably Part 15.209. The inference is that if an unintentional radiator creates emissions at a level below that provide for in 15.209, that the emissions are "acceptable" implying not regarded as a cause of interference. The emission limits stipulated in 15.209 are a regulatory upper limit to emissions from non-intentional radiators in a strategy to minimise the risk of interference from such radiators, but in no way does compliance with 15.209 override the right of so called "licensed services" to protection from interference. The ARRL is at the time of writing challenging in the US Federal Court, an FCC ruling relating to interference protection to mobile stations and the interpolation / extrapolation factors specified in Part 15 and relating to 15.209 compliance measurement.

Whilst it is interesting to see the 15.209 limits shown on the charts because much of the BPL equipment is manufactured to exploit the quite high limits of 15.209, it is risky to opine what may or may not happen in another jurisdiction.

In the Australian context, the more relevant questions relate to the meaning of "substantial interference" as used in the Radiocommunications Act 1992 (Cth), and the effectiveness of S.197 of the act.

Assumptions

BPL emissions under traffic are essentially wideband emissions with reasonably uniform power density with frequency.

S-meter equivalences are offered here on the basis that S9 is equivalent to 50µV in 50Ω, S units are 6dB each, values are rounded to whole S-units below S9 and to 5dB above S9. It should be noted that S-meters in receivers are notoriously unreliable for the following reasons:

  • calibration depends on receiver attenuator, preamp, IF bandwidth, noise blanker and noise reduction settings, and they are often not stated in user manuals;
  • some receivers use different calibration equivalence (eg some Yaesu equipment uses S9=100μV);
  • often quite inaccurate below about S6 and above S9+30dB;
  • LCD or LED displays often have very low resolution;
  • they usually depend on AGC voltage and are closer to a quasi peak response than true power; and
  • receiver effective noise bandwidth is an important factor in calibration and not usually stated in specifications.

There is no reliable information in the ACMA report to characterise the ambient noise level at the study sites. Expected ambient noise level in this article is then taken as the median residential man-made noise as characterised in ITU-R P.372-8. Not that they are median levels which means that half the time, the levels are lower than used, and indeed, some observation of ambient noise in Australia suggest that it may be nearer the rural category.

Conclusion

The ACMA have measured and reported emissions associated with a long running BPL trial in Hobart, and the report provides a deal of valuable information in understanding the problem. The ACMA are to be congratulated for publishing the information.

The ACMA report details measurements of BPL emissions that are well above the expected ambient noise level.

Although the BPL infrastructure's notching facility has been used to limit emissions in at least one amateur band, the emissions in that band still increased the ambient noise by 11.4dB over that which would be expected under ITU-R P.372-8.

V1.1 20/02/09 19:23:37 -0700

 


© Copyright: Owen Duffy 1995, 2021. All rights reserved. Disclaimer.