ARRL Testing of Broadband Over Power Line Systems Ed Hare, W1RFI ARRL Laboratory Manager 225 Main St Newington,CT 06111

1. ARRL Testing of Broadband Over Power Line SystemsEd Hare, W1RFIARRL Laboratory Manager225 Main StNewington,CT 06111w1rfi@arrl.org860-594-0318

2. Background • ARRL is the US national association for Amateur Radio • Represents Amateur Service in regulatory and industrial venues • ARRL’s interest in BPL is related only to its EMC aspects • Other than EMC issues, BPL should be allowed to succeed or fail on its own merits • Amateurs are often early adopters of new technology

3. Scope of ARRL testing • ARRL staff have done testing in 12 cites where BPL is located • Other radio Amateurs have done testing in several other areas • ARRL’s testing capability has been evolving over time, so the nature of testing has also varied from system to system • ARRL testing done for EMC assessment, not for compliance purposes

4. Types of ARRL testing • Computational, mainly using NEC-4 • Interference assessment • Site survey • Measurement of noise floor • Measurement of ambient signal levels • Relative measurements of noise-floor degradation • Field-strength measurements

6. Measurement of noise floor • Ambient levels of man-made noise can range down to –20 dBuV/m at HF-station antennas • It is not possible to measure this level with spectrum analyzer and typical EMC antenna • Such measurements, even with active loops, are really measuring the noise floor of the test equipment • Communications receivers and real-world antennas are much more sensitive than EMC test equipment • To measure ambient levels, as a minimum, an EMC receiver and 8-foot monopole antenna, tuned to resonance with inductive loading, would be needed.

7. Receivers and spectrum analyzers see the world differently • Sensitivity • Antenna gain • Overload • Those who use test equipment to analyze the EMC aspects of BPL will see the results differently than radiocommunications users

8. 14 MHz along a length of Road as seen by receiver

9. Same Data: Simulated Spectrum Analyzer

10. Calculations • Primarily done with antenna modeling • No power-line model can represent actual real-world conditions • However, models can be used to determine principles behind how power lines radiate as antennas • ARRL used EZNEC Pro with NEC-4 method-of-moments electromagnetics calculation engine

11. Areas investigated • Gain of power line as antennas • Pattern of radiation pattern • Near-field E and H fields • Standing wave of emissions along power line • Relationship between E and H field • Distance extrapolation

13. Antenna gain and pattern • Gain of power-line antenna ranged over a range of approximately –20 dBi to +10 dBi • Varied with frequency, programmed losses, feed-point method, number of phases and spacing of the lines • Actual gain is not all that important, because the limiting factor is the permitted emissions level • Pattern shows that below 30 MHz, in the far field, most of the energy from a low horizontal antenna of any type is radiated upward

17. Fields Near Large Radiators – 14 MHz30 meter/3 meter ratio 16 dB

22. Field testing – typical test fixture

23. AntennaAH Systems loopor ¼-wave mobile whip 0-70 dB Step Attenuator ESH-2 EMC Receiver Field testing – typical test fixture A AntennaAH Systems loopor ¼-wave mobile whip B 0-70 dB Step Attenuator Icom PCR-1000 Receiver Laptop P.C.Windows 98Soundcard

24. Measurements of ambient noise level without BPL

27. Measurements of field strength in areas where BPL is deployed(vendor and city location not included)

31. 14.3 MHz: amplitude vs distance along a power line

32. 14.3 MHz: amplitude vs distance along a power line

33. 28.5 MHz: amplitude vs distance along a power line

34. Frequency sweep 18.9-22.9 MHz

35. Frequency sweep 28.0 – 29.7 MHz

38. Frequency sweep: 21.0-21.015 MHz – close-in

40. 28.5 MHz vs time

41. FCC Limits were apparently exceeded by at least 22 dB in this system: The test fixture and measurement software made the following measurements, estimated as quasi peak field strength in a 9 kHz measurement bandwidth. These data are not extrapolated to distance. 3.52 MHz: 69.2 dBuV/m 68.7 dBuV/m 69.1 dBuV/m 69.0 dBuV/m 70.9 dBuV/m Average: 69.4 dBuV/m

46. Other detailed reports • Raleigh, NC • Cedar Rapids, IA • Cottonwood, AZ • Cape Girardeau, MO • Manassas, VA • Emmaus, PA • ARRL CD

47. Bring the mountain to Mohammed A number of BPL manufacturers have taken out experimental licenses. One of the conditions of their license is that they file 6-month reports with the FCC, showing the measurements they make to determine compliance with the emissions limits. The following are from some of their reports, or represent an ARRL analysis of same.