6

I remember not long ago seeing a lot of chatter in the ham community about large-scale "Broadband over Powerline" (BPL) systems causing harmful interference across several bands. Does similar interference happen with SOHO "Ethernet over Power" (usually called "HomePlug") networks?

HomePlug networking is a convenient way to extend home computer networks, when WiFi is impractical or undesired, without having to run cables across the house. However, I would not want to recommend this to anyone if it is a known RFI offender.

If these devices are known to cause RFI, is there anything the end-users can do to reduce or eliminate their impact without having to actually replace them?

KJ4JTN
  • 161
  • 1
  • 4
  • I suspect they are liable to cause interference, especially on HF, and the interference might be somewhat mitigated by judicious use of common-mode chokes throughout the electrical system. – Phil Frost - W8II Jul 30 '14 at 21:29
  • @PhilFrost Thanks for the comment. I'm hoping someone can speak from experience, though. These are Part 15 devices after all, which should in theory be engineered to minimize interference, but we know how well that can go sometimes. – KJ4JTN Jul 30 '14 at 21:30
  • yes, I think that's an implicit requirement of all answers on stackexchange, which it why it was a comment and not an answer. Anyhow, the part 15 rules are pretty lax, and are designed for the vastly more common case where the part 15 equipment is not co-located with radio equipment. Consider the inverse square law, and the burden that would be imposed on equipment manufacturers if there was an assumption that there was an HF receiver 20 feet away. EMC certification is hard enough to pass as it is. – Phil Frost - W8II Jul 30 '14 at 21:37

2 Answers2

3

There has been an ongoing debate about this in the UK, where the system is called PLT for "Power Line Telecommunication".

See articles on The Register (slightly sarcastic tech news site) for more on the subject. Two such examples are here and here. Of course the general feeling the manufacturers are trying to engender is one of the Luddite radio hams being stuck in their ways and trying to stand in the way of progress, and my feeling is that they are succeeding in this goal.

It has been proven to exceed the level of radio emissions allowed by the relevant EU regulations, but equipment is still being manufactured, sold, installed and used.

Yes, it interferes with HF. No, the regulatory bodies don't seem to care.

Scott Earle
  • 4,599
  • 1
  • 17
  • 30
  • Thanks for the information and links. Is there anything, short of actually changing the wires in the walls or just not using HomePlug/PLT, that end-users can practically do to prevent or reduce the interference these devices may cause? – KJ4JTN Jul 31 '14 at 14:52
  • Lots of chokes on the mains wiring. I think if you used enough ferrite, and filtered out the rubbish before it leaves the house, you should be OK – Scott Earle Jul 31 '14 at 14:53
  • How "practical" is that? Your phrasing "I think if you used enough ferrite..." seems to imply "not very". – KJ4JTN Jul 31 '14 at 14:54
  • Yeah, I am not particularly optimistic ... – Scott Earle Jul 31 '14 at 14:55
  • I would imagine that a lot of ferrite cores would interfere with the operation of your ethernet adapters as well. I've seen some anecdotes that power strip surge suppressors would kill X10 networks. – W5VO Aug 01 '14 at 17:54
  • @W5VO I'd probably expect that from a power conditioner, but not a simple surge suppressor. Interesting to know, though. – KJ4JTN Aug 01 '14 at 19:22
  • 1
    @W5VO I wouldn't think so. Ferrite beads, over both conductors, dissipate only common-mode currents. Differential-mode currents have no net magnetic field, and so pass through the ferrite unaffected. This is exactly what you want for reducing interference: an electrical supply wire is approximately a twin-lead transmission line, which does not radiate only if there are no common-mode currents. – Phil Frost - W8II Aug 05 '14 at 11:39
2

Short answer: Yes.

Longer answer: Homeplug devices use HF frequencies (0-30 MHz) to transport data. Active networks cause a lot of noise, idle networks will cause a recognizable ticking sound. "Good" homeplug devices have specific notches for air traffic control, shortwave broadcast and amateur frequencies. I have tested this myself with a Devolo interface and visualized the results with Gqrx at To my shame as a radio amateur I must admit I still use a PLC (ethernet over powerline) connection in my home network (the devices are now gone). A clear notch for the 20 meter band, but serious amounts of noise outside that band.

Koos van den Hout
  • 141
  • 4