Power transfer from power lines using induction

[ptricks]

I live in a rural area and internet access is poor at best. One of the possible solutions is wireless internet or wisp.  The issues stopping this is lack of good LOS (line of sight) due to dense trees and putting up towers everywhere is very expensive and labor intensive, but the area does have electricity lines everywhere and it is almost 100% provided by power poles.   The thing about power poles is they are LOS to each other and that LOS is maintained free from trees due to having to keep the lines clear for power.  The idea I am trying to research is if it is possible to harness 15-25 watts from a power line by means of induction without making physical connection in any way to the power line itself.  A transceiver can then be attached to the pole , powered by the overhead lines and relay signal to the next pole that has LOS.

I realize there are all sort of regulatory things involved,  networking  issues,  but I want to focus on whether it is possible to do the power side of things in the way I described.
Voltages in the area on the lines are 10KV+ and some are 3 phase but majority are single phase.  Homes are fed from those lines with every home having its own pole mounted transformer which is center tapped to provide the home with 120VAC from each secondary to the center tap or 240VAC secondary to secondary.   The access point would be mounted so that all it takes for an installer is to place it on a pole a few feet below the overhead line and align it with the next radio powered by induction from the overhead line.

I have already discussed the possibility with local engineer at the utility and he likes the idea but wasn't able to help on whether it was doable or not, so reaching out to the forum to see if anyone has some ideas, theories, links, math involved to test or comments.
thanks in advance !

[Siwastaja]

Interesting idea.

The first thing you should concentrate on is to optimize your plan on power usage. 15-25W for relaying some data (<100Mbps I guess?) for a few dozen meters sounds at least an order of magnitude (if not two) high to me. Remember your links are directional.

Even on a bog-standard WiFi link we are talking about 1W, and that would have a longer communication range than what you need, and it would be omni-directional (spherical), and WiFi is notorious for poor power efficiency. LTE is able to communicate with cell tower kilometers away at well below 1W. You need power-optimized communication to need less energy harvesting.

I guess you would replicate hundreds if not thousands of these repeaters; it would mean that they are not one-offs, and you could concentrate in optimizing them. While 15-25W is AFAIK impossible to safely harness from the wires, if you can drop the link consumption to, say, 1W or less, then a small solar cell with a li-ion cell becomes a real alternative, even if you failed harvesting energy from the power line itself.

The delay caused by the number of repeaters would become an issue (latency is really important in modern Internet usage), unless you can come up with some really clever low-level protocol stack to do the job. I guess you should still minimize the number of repeaters as well, most of the time the power poles are in a nice straight line so if you have a LOS between ten poles, you also have a LOS from pole 1 to pole 10, directly. Then you'd use higher density of repeaters only when the power line changes directions.

In any case, I guess this will be a demanding project, requiring a lot of RF and network protocol expertise.

Finally, if you can afford to install devices to each or most power poles, don't you think you couldn't just as well install some optical fiber there, hanging well below the power lines, but still out of reach? Wouldn't that do the trick much more easily?

[ptricks]

I did set the power a bit high ,  the low end could be a couple watts, I'm  mulling the idea around thinking about the worst case.
I thought of solar but the issue becomes the battery and run time issue with something that would essentially depend on one radio working for the next in the line to continue working. One radio loses power and you would lose all the remaining, of course you can plan a network for things like that too.

One of the reasons I was discouraged about fiber is that the cost is  high when you have to put it on every pole down a road, some of these roads might have 20 poles between a house and for those houses 1 radio at the beginning and then another at the house is much cheaper .   Looking around the area I see areas where a single radio could cover a mile or more because the poles are in line on straight roads but would need a relay around curves to reach the next stretch of road and houses.

There is always the option of installing physical taps on the line, but that is another layer of cost and labor and would require the power utility to do the install, not impossible to do at all since they do add taps to the lines with smaller step down transformers whenever a customer request an outdoor utility owned light install.


Currently people here spend  $65 USD to get as low as 3Mbps/256Kbps  through dsl, often a home isn't close enough to get access,  and the company has no plans now or in the future to upgrade, no cable, so people can choose cell, which also doesn't cover here well, or satellite but have a cap on data. The telecom company even boasted in shareholder meetings how they plan to dump the aging copper infrastructure to focus on business or more profitable areas if they can find a buyer. If even 20Mbps were offered at the same cost people would jump at the option.

One benefit I have is that the local utility is member owned or co-op so they are more open to ideas from customers than huge corporations.

[T3sl4co1l]

Anyone know offhand if a Goubau line works with typical single-phase medium voltage distribution lines?

I would guess the insulators and links screw up any propagating fields.  Might be okay for a few poles, but not a whole run?

The significance of a Goubau line is, it's a propagating mode for a single wire (usually a dielectric-clad one), and it's done at frequencies where high-speed internet could be practicable.

Otherwise, you're stuck with twin-line or round-wire-over-ground modes, with modest bandwidths (~MHz) and terrible losses both due to radiation (piss off all the nearby hams?) and interconnects (especially transformers?).

As for parasite power from distribution lines: not much.  I don't know that medium-voltage potential transformers come very small (i.e., ~100VA) such that power could be drawn directly.  Current transformers may be able to do that, but delivered power depends on burden voltage times line current, so it's a pain to get dependable power.

Even if technically possible, the power company will detect the parasite drain, and won't have any good way of metering it.  (The transformer approach would need a disconnect and an energy meter at each drop.)

Far easier is probably just putting up solar panels.

"Water, water, all around..."

Tim

[MagicSmoker]

Well, Tim beat me to it as I was going to say the exact same thing: use the utility poles for a convenient mounting point and rely on solar power instead of trying to steal energy via inductive (or capacitive) coupling.

[Neomys Sapiens]

As for parasite power from distribution lines: not much.  I don't know that medium-voltage potential transformers come very small (i.e., ~100VA) such that power could be drawn directly.  Current transformers may be able to do that, but delivered power depends on burden voltage times line current, so it's a pain to get dependable power.

HV measurement transformers seem to be what you are thinking about, but it seems not to be what he wanted.

Even if technically possible, the power company will detect the parasite drain, and won't have any good way of metering it.  (The transformer approach would need a disconnect and an energy meter at each drop.)

Tim

Yes. All inductive power transfer mechanisms with increased separation can be seen as VERY inefficient transformers. Also, your primary winding (ONE) leaves a lot to be desired. So you will not even get a few watts out of it while keeping the safe distance to the 10kV. If you tried. you would have to go closer and closer with your coil becoming more and more massive. And if cou are close enough to get such a significant magnetic flux that your secondary can derive the needed output, you would still draw just more power than you use from the line because of the infernal efficiency. Also it depends on the current flow in the line.

[T3sl4co1l]

The examples I've seen (fault current monitor devices, very little power consumption) use a rather large, clamp-on CT (about the size of a standard issue donut, with about 1/4th the cross-sectional area).  To get tens of watts, you'd need a quite large CT indeed.

Insulation I guess isn't such a big deal -- it would be potted, or hovering at line level perhaps.

Tim

[james_s]

I know from experience that if you stand outside under a high tension line holding a fluorescent tube it will glow so there is *some* power transferred, however I doubt it's very much. I'm sure one could optimize and make significant improvements but I have to agree that solar is likely much more practical. You're still harvesting "free" energy.

[T3sl4co1l]

Electric field is actually really awful for drawing power.  I ran some numbers on it once.

It would be interesting with a very high Q inductor -- in effect making a resonant Tesla-style receiver.  You'd still need quite a large pickup plate, and needless to say such an inductor will be very large indeed.

Tim

[Neomys Sapiens]

Electric field is actually really awful for drawing power.  I ran some numbers on it once.

It would be interesting with a very high Q inductor -- in effect making a resonant Tesla-style receiver.  You'd still need quite a large pickup plate, and needless to say such an inductor will be very large indeed.

Tim

Yes, while writing the earlier comment, I also thought when I totally needed to do this, I would go capacitive. But I did not want to suggest that to someone without extensive HV background. Your idea would improve on that, but it is still capacitive coupling, and maybe you have heard some people go ape about capacitively coupled supplies from a normal line.

[IanB]

When I drive around I frequently see isolated monitoring stations next to highways mounted on a pole with a solar panel and a box of electronics. I am almost certain these are free standing, self-powered devices with a radio link to some base station somewhere. So I think the device you are looking for may already have been invented and you just have to adapt/re-purpose it somehow?

[ejeffrey]

I think those stations use GSM radios and probably only power up for a few minutes a day.  So the power budget for an always on wifi repeater would be higher.

[Marco]

Capacitive can light fluorescent tubes with comparitively small electrodes ... the available power between a small "antenna" and the grounding conductor on the pole might be enough.

[ejeffrey]

By my estimate, the available power for a "small" magnetic pickup coil near a power line (but not encircling it) is:

\$P_{max} = \frac{V \mu I_0^2 f}{8 \pi d^2}\$

Where V is the volume of the core, \$\mu\$ is the permeability of the core, Io is the current in the transmission line and d is the distance to the power line.

I'm not sure what the current is in your transmission line or what you consider a safe distance, but putting in my guesses, it doesn't look good for getting 10+ watts.

[Siwastaja]

If you start aiming for <100mW average consumption, things suddenly start getting feasible. I don't know offhand if leaching 100mW from the power line is feasible, but it would be in the right ballpark. But even if it didn't work out, a modern 20W solar panel would be fairly small & cheap, and should be able to support 100mW average, even during long-term bad weather, excluding the completely dark winter in the arctic area. For storing 100mW average over 4 days, for example, the required battery is just 10Wh, or a single $3 18650 cell.

I see no reason to spend watts of power, except to simplify design by using general-purpose networking/computing parts. But I think you need to go beyond that, and optimize for the power. Your links are fixed and directional, so little actual power is needed for radio communication, even assuming less-than-ideal modulation schemes. It all boils down to being efficient and using the right RF tools for the job; custom if necessary. I know this isn't the easy road.

[maxwell3e10]

For capacitive coupling the key is a long length of the pick-up wire. Suppose that you run a wire from one pole to the next, parallel to the HV wire. If the distance between poles is 100 meters and the pick-up wire is 3 meters away from the 10 kV line, then the mutual capacitance is about 400 pF. For optimal power extraction one needs a HV transformer that steps down the voltage from 5 kV to the usable level. The power that can be extracted if the pick-up wire is maintained at 5 kV is 4 Watts. So, perhaps it is not so far off.

[Siwastaja]

Although, if you run a capacitive pickup wire full-length from pole to pole just to extract power, you could have just as well ran a cable (fiber preferably) for the communication directly...

[maxwell3e10]

I think the idea is that for straight lines one could get away with having a relay station every 10th pole or something.

[Someone]

Lets see if this link works:
https://researchbank.rmit.edu.au/eserv/rmit:162090/Rodriguez.pdf
Rodriguez, J., 2017. Electric field energy harvesting from medium voltage power lines.

Fairly recent and comprehensive review of the options.

[L_Euler]

For the cost to design and implement an inductive capture system, do you think solar may be just as economical.  The trees are already cleared.

[ptricks]

I want to thank everyone for the comments, lots of input has really helped me flesh out the concept.
I think I need to determine what the exact minimum power requirements are for the radio to work.
I'm not giving up completely on power from the lines but it does seem that if that aspect is going to work the power requirements are going to have to be pretty low.
The local utility is a big proponent of solar , they have a 65MW solar array built just a few years ago and another combined 80MW from other solar sites in the area so I don't seem being opposed to the idea of small panels on each radio.