Power Over WiFi Might Not Be A Unicorn After All

There have been a few reports of power over WiFi (PoWiFi) on the intertubes lately. If this is a real thing it’s definitely going to blow all of the IoT fanboys skirts up (sorry to the rest of you *buzzword* fanboys, the IoT kids flash-mobbed the scene and they mean business).

All of the recent information we found points to an article by [Popular Science] titled “Best of What’s New 2015”. The brief write up includes a short summary lacking technical info, and fair play to [PopSci] as it’s a “Best Of” list for which they hadn’t advertised as an in-depth investigation.

However, we tend to live by the “If you’re gonna get wet, you might as well swim.” mentality, so we decided to get a little more information on the subject. After a bit of digging around we came across the actual article on [Cornell University]’s e-print archive where you can download the PDF that was published.

USB energy harvesting dongle.
USB energy harvesting dongle.

The paper goes into detailed explanation of the power harvesting theory including a schematic of the receiving end hardware. They had to create a constant transmission for the harvester to get over its minimum required voltage of operation. This was done with one of the wireless router’s unused channels to fill the voids of packet-less silence between normal WiFi communication.

As you can imagine PoWiFi is currently limited to powering/charging very low power devices that are used intermittently. The research team was able to charge a Jawbone headset at a rate of 2.3mA for 2.5 hours which resulted in the battery going from 0-41%. The punchline here is the distance, the device being charged was only 5-7cm from the PoWiFi router which is getting close to inductive charging range. The researchers stated in the paper that they were looking into integrating the harvesting circuitry and antenna into the headset while working towards a larger charging distance.

At the time of writing this article it seems that PoWiFi is best suited for devices such as: low powered sensors and motion activated cameras that have increased energy storage capacity, which the team mentioned as one of the continued research possibilities.

We’ve covered numerous wireless power projects before, some legit and some we still get a kick out of. Where do you think this one falls on that spectrum? Let us know in the comments below.

Thanks to [ScottVR] for the tip.

54 thoughts on “Power Over WiFi Might Not Be A Unicorn After All

  1. The tyranny of the inverse square law. 5-7cm in poor, even for this. Each time you double the range you have to square the power. In order to send this amount of power a couple of meters will require stupid amount of power to be emitted, enough to drown out wifi for several miles and potentially damage human tissue at close range.

    The only other option is to go directional. This could work, but is a completely different kettle of fish (personally I’d just go with a laser pen and solar cell).

    1. This isn’t about *delivering* power specifically to a device. This is about *harvesting* power that’s already available. Yes, the power received sucks, which is why you need to do significant engineering to get it to be usable. The idea of ‘using it to recharge batteries’ is stupid, I agree, but the idea of powering sensors is a lot more interesting.

        1. luckily normal wifi is readily available and as far as i can see there is little to prevent a feature like this in household and commercial routers, allowing people to siphon power even if they don’t have access to the network traffic itself.

          i think the idea is to use the otherwise “wasted” wifi power already out there in general life.
          no idea if it will ever work or be practical but with proper engineering it would be similar to some already used radio modulation.

          1. Ok don’t think you realise how little power there is in a WiFi signal. Emission limits are 100mW. At 5m there is only 300pW/cm^2 available. Even if you could harness 100% of this you wont have much to work with. In practice, the efficiencies will be far lower, a fey % maybe.

            In short, you will get a lot more power from a cheap solar cell and a 40W lightbulb.

          2. The line of thinking in your response (tell me if I am incorrect please, making assumptions here), is that we do not need to make the one transmitter stronger, we just need to make use of all of the myriad transmitters out there already.

            There are so many people broadcasting wifi, that with all of those signals combined, we have a net amount of power.

            And this is a reasonable stance. But it still must answer to math. The inverse square law says if we move 10 times farther away (from 5cm to 50cm, half a meter. A reasonable distance), then we need 100 times the power. So as long as you have 100 routers within a half meter of you at any given moment… power achieved.

            So… even with a distributed network of wifi signals being harvested… not practical. Even if we include radio and cell tower signals into the equation, and even if we assume those have the same power level as the wifi router, we fall short by a landslide.

          3. mate, this is already the case with hertzian noise from all the waves surrounding yourself, and it would be like if you took a coil and a capacitor, connected it to a schottky diode and hoped to charge your phone with it…

          4. to all of the above, i readily admitted i had no clue about the practicality nor feasibility of the project in my post, nor did i deny that the inverse square law holds true, i even put wasted in quotation marks to imply the spurious nature of that concept in relation to wifi, since any signal intercepted will mean a small but existent loss of signal.

            i was commenting to try to illustrate the concepts presented as i understood them and as presented, if the intention is to drive ultra low power devices then the very small amount of power “harvested” from already present fields might just do the trick, charging a phone is a fairly high power application to begin with.

            it isn’t exactly a case of the base principles being physically impossible here, only a matter of if, when or where it would be worth it compared to other options, something i already admitted i have no clue about.
            what i can say is that it probably depends entirely on what you want to do with it, you are probably never gonna run your tv or interior light off of it.

          5. Use solar cells. Even those on calculators ten or more years ago will yield more power than this method, except in a very dark environment. Besides: Harvesting power from electromagnetic radiation is forbidden by law in a number of countries, to avoid people hijacking broadcast station signals. This has happened before.

            “Wireless charging” is nice for putting a phone on a charging bed for the night, without the need to insert a plug. But that’s about it.

  2. I still feel like somebody with the right skillset needs too go over all of tesla’s notes ever uncovered, and decide if theres something too his more elaborate wireless power plans (the ones that dont revolve around a teslacoil powering things in throwing distance, but the ones that turn the entire globe into a power grid)

    1. The problem I always had with the Tesla power distribution, is nobody ever says what constitutes a receiver of that power. You can’t broadcast power if every piece of metal in the world is going to start receiving that power.

    2. The magic of Tesla rears its head yet again

      There is no magic

      That’s why he couldn’t raise money for his harebrained ideas.

      Tesla was a showman, exploiting patent-court victories and the newly emerging media (newreels) of his day. That pic of him reading a book with sparks flying everywhere is classical Nikky Tesla bullshitting – it was a time lapse pic.

      Anything worthwhile from Tesla was pretty much stolen, or had minor tweaks, from Michael Faraday and others. Even the Tesla turbine came about as a total accident.

      Tesla was a PR genius, but a sub-par engineer who believed more in magic than science.

  3. Using WiFi is too weak and it seems to me that anything capturing significant energy will create a WiFi “shadow” behind itself.

    This makes me wonder about having a low power household transmitter station specifically for this though. Perhaps a 5 watt transmitter at some frequency that would deliver enough energy for remote sensors. If this was done to some kind of standard, you could buy them off the shelf (like WiFi routers) but their purpose would be to power other stuff in the home or office.

    1. You would need ‘lots’ of transmitter stations. Remember the inverse square law mentioned above?
      One transmitter a few cm from each sensor… you might as well just plug in the sensors…
      If you really want to do something like this, I guess it’s time to look into long-range passive RFID again ;)

  4. I suppose the idea is that a greater volume of signals to harvest is better than trying to harvest from fewer, more powerful signals? Or, maybe not fewer, but just harder-to-receive, more powerful signals. I don’t know a lot about the methods and theory behind it all, but my gut feeling is that harvesting from “high-power” signals would yield the best results in the short term.

    1. Think about kids splashing in a pool.
      If you have a lot of kids splashing, you end up with indistinguishable ripples at the edge of the water.
      Same thing will happen with multiple signals unless they can actually transmit at the same frequency in phase with each other… of course, you run into funny issues where these waves propagate differently based on the building and your position… so you don’t really get anything useful out of it except for a little noise.

  5. Respectfully, a quick calculation shows that this is just impossible. A typical 802.11 router is limited to 100mW of broadcast power, but because of the inverse square law the power density decreases rapidly as you move even a small distance away from the router. The maximum received power (-10dBm) corresponds to 100uW of received power, and the minimum that Wifi tends to work at is -100dBm, or 0.1pW, a room or two away. Assuming a 100% conversion efficiency (it’s probably closer to 1%), and a 5V output from the USB port on the back-end of their charger, that gives anywhere between 20uA (best case) to a few picoamps of current. At the high end, assuming real efficiencies, that’s about enough to keep a real-time clock from losing time, but not much more. Even with the antenna from the wifi router directly coupled to the “harvester”, somehow harvesting at 100% efficiency @ 100mW, that’s only 20ma@5V, which would take forever to charge most things.

    1. Okay, replying to myself here — I just skimmed the paper, and it looks like they’ve made their own transmitter that operates when normal wifi is not transmitting, they use their antenna to charge a supercap, and when it achieves a threshold voltage of ~2.4V, they run an ultra-low power microcontroller for a few milliseconds to wake, read a sensor value, and transmit that somewhere else. Their plots show that the frequency that this can be done is only a few times an hour up to about 20 feet away from the transmitter. So, not as impossible as it first seems, just powering devices engineered to consume ultra-low amounts of power and operate for milliseconds per hour, so just very limited in scope.

      1. With such short ranges and almost anemic sampling rates, this seems to better suited to sensors in tanks or some such. Such as a sensor floating around in a sealed container of… I have no idea… toxic waste? With one less seal (the wire) to deal with, the tank could be made to handle higher pressures or what have you.

        Yeah… it’s a stretch and adding random noisy crap over WiFi competing with an increasingly crowded spectrum strikes me as nonsensical.

  6. There’s a device out there that already does this, and it was invented over 100 years ago. In fact, it’s so simple that you can build one powerful enough to produce audible signals in a piezo element using only a coil of wire and a germanium diode.

    It’s called a radio receiver.

    Using radio to recharge batteries is just plain asinine. It doesn’t make sense on any level. Whoever came up with this obviously forgot about basic radio theory, the inverse square law, and every other piece of relevant data on the subject.

    1. Arnt there problems with the schematic provided at that page? The screw terminal on an auto ignition col are connected to the battery voltage The coil’s metal case and the center tower are the coil’s high voltage terminal. the schematic doesn’t show a path for the high voltage through the coil when the capacitor reaches a charge that will cause the plug gap to fire. I beliieve that capacitor should be across the spark plug. The pug’s ground and the coils cases connected to Earth ground, along with the batteries negative terminal. The spark plug’s top terminal connect to the coil’s center tower terminal. In that manner when the capacitor discharges over the plug gap the circuit will operate in reverse as it does when used to in normal engine operation. In normal operation one of the screw terminals is connected to the battery the other to the ungrounded side of the breaker points. That’s my take anyway, discounting the author may have been passing on an old man’s yarn.

    1. You can. Coaxial feed line is in effect a capacitor. Static electricity associated with thunderstorms can charge the feed line to a point the charge will jump a gap of a half inch. That is if the other end isn’t connected to a DC grounded antenna. Put up some dipole antennas run coax the where you can see the end build adjustable spark gaps. Just don’t get too careless.

  7. Didn’t we do this exact thing before, only with an article explaining why it’ll never work? “Inverse – square – law” seemed to be the gist of it.

    This is just advertising some shitty product. They haven’t reinvented physics, it’s all existing stuff, put into a box, that won’t work. At least wait til it’s on the market before doing a puff piece. Until then assume it’s all lies, something to fleece investors.

  8. A diificult and confusing read. I was lost as wast WIFI power; what wasted WIFi power My router is set to use on chanel only, and being a hamm I chose on in amateur rado allocation of the band. What a waste of limited spectrum if routers could set to broadcast garbage data on all channels My desk top where I do the majority og my work id hardwired to the router, so isn’t causing any RF transmissions to take place. The WiFi on my phone is generally off so it and router aren’t checking each others status. I rarely use the other devices in the house that rely on wifi for internet status, but the article suggests my router is always broadcasting, and I don’t understand why that would be other than a periodic beacon so other devices can find it

  9. Areas where this might be reasonable (obeying but decaying with inverse-square-law), and a battery is less able to be used
    – Sampling inside sealed chambers
    – Sampling in environments sensitive to chemistry
    – Sampling inside organisms (beyond inductive range)

    But in reality, with a maximum budget of 2.3ma, batteries plus other energy harvesting methods (eg piezoelectric, thermal) combined with low power and low frequency sampling methods are going to be more competitive.

  10. I really can’t understand this obsession with wireless power. The connector on my iPhone 5 illustrates a better solution. Simply design connectors to be non-polarized, easy-to-connect, and not likely to yank the device across the room if accidentally caught. That on the iPhone meets two of those three.

    Besides, the amount of power we’re talking about here is so small, I doubt the power company meter would even register it.

  11. “They had to create a constant transmission for the harvester to get over its minimum required voltage of operation. This was done with one of the wireless router’s unused channels to fill the voids of packet-less silence between normal WiFi communication.”

    Yeah I’m sure the neighbors would LOVE this.

    1. Yup. Maybe microwatt power and little solar panels will be the way of the future. Every home and office has light, people need light to function even more than they need wifi. Since wifi and the like are designed to be low-power as much as possible, trying to harvest it, going as far as blocking up the other 11 channels with “power” transmission, is barking up the wrong tree.

      Or as somebody else mentioned, just stick a same-sized battery on it. For the lifetime of the battery, it’ll do a better job. The only radio-harvesting that ever worked was enormous wire antennas running crystal sets, and people only did that until amplification was invented.

  12. The maximum allowed electromagnetic power density to remain safe for people, for Wifi is 20nW/cm2, that can give you an idea of the maximum power available since you know the surface of your antenna.

  13. I remember readin that paper a while ago and then I though: “Well, that’s a neat thing, but it’s a terrible waste of energy”. I’m still of that oponion, when it comes to all types of wireless charging. Use a solar panel instead. You will get way more energy in a lit room.
    Also, the other guys (can’t remember the names but it started with an “x”?) who claimed they had solved wireless power by using directed RF-beams and tracking where the device (in their case, a smartphone) was. I think they were supposed to launch a crowdfunding campaign about now. Still very sceptical.

  14. light beam li-fi converter. see that heat converted to power for the battery that self heals along with all that exxess tdp/heat to the battery? mmm. flips killswitch to end chanrging in software and on the laptop like on my smartphone

Leave a Reply to gregkennedyCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.