Wireless Power Transfer Using Capacitive Plates

It seems like wireless power transfer is all the rage these days. There’s wireless charging mats, special battery packs, heck, even some phones have it built in! And they all use inductive coils to transfer the power — but what if there was another way? Coils of copper wire aren’t always that easy to fit inside of a product…

As an experiment, [Josh Levine] decided to try making a proof of concept for capacitive power transfer.

He first demonstrates inductive power transfer using two coils of copper wire to power up an LED. The charging coil is supplied with 15V peak-to-peak at 1MHz which is a fairly typical value for inductive charging. He then shows us two glass plates with some tinfoil taped to it. Two LEDs bridge the gap alternating polarity — since the power is oscillating, so we need a path for electrons to flow in both directions. There is no connection through the glass, but when it is set on the charging plate, the LEDs light up. The charging plate is supplied with 30V peak-to-peak at 5MHz.

It works using the concept of capacitive coupling, or electrostatic induction. The main reason it isn’t used as widely as inductive power transfer is because it requires higher voltages to transmit significant power — which can be dangerous! But one good thing is it doesn’t cause as much interference because the magnetic field is largely confined between the two plates.

Now this is only just part 1, [Josh] is planning on continuing to research this and see if he can create a practical system for use — we’ll keep you posted!

We wonder what the next big thing in wireless power transfer will be? Will the next Tesla vehicle charge wirelessly in your driveway? Or maybe charging pads for quadcopters?

33 thoughts on “Wireless Power Transfer Using Capacitive Plates

  1. I thought that capacitive coupling required the plates to be very close whereas inductive couple could tolerate some gap and yet still be efficient. Although less RF noise would imply more efficiency for capacitive coupling. Has anyone calculated the theoretical limits for both?
    Pyrulux would be a good candidate for capacitive coupling experiments.

  2. “it doesn’t cause as much interference because the magnetic field is largely confined between the two plates”…

    Electric field, not magnetic.

    THere is a magnetic field here, but it is not confined between the plates. It is likely small compared to magnetic induction, though, and its strength and geometry are critically dependent on the geometry of the circuit. It can be made quite small outside the circuit bounds.

    1. I would tend to call a magnetic field that is generated by Eddie Currents – an ‘Electro magnetic field’ to make it clear that it has dependence on a potential difference and also that it is likely to be very small in magnitude as you stated.

  3. ” The main reason it isn’t used as widely as inductive power transfer is because it requires higher voltages to transmit significant power”

    The main reasons it isn’t used is

    1) it needs perfect alignment to work properly
    2) dielectric losses in the insulator, which leads to low efficiency

    If you cross the two pads on the transmitter with your recieving pads, such as turning the phone sideways, then you’ve simply short-circuited the transmitter.

    1. Also, biological systems such as your fingers react to radio frequency electric fields by acting as conductors, so you get RF burns from the pads.

      Magnetic induction systems don’t have that property.

        1. It’s also similiar to the machine used to glue high density board together by blasting it with an RF field while under compression. The dielectric losses lead to heating, which cures the epoxy.

    2. I wonder if instead of two large pads, you could have multiple small pads (a grid?) and somehow detect the orientation of the device and send the power to the appropriate pads. Of course, due to the additional detection/switching it becomes more expensive.

      Regarding the alignment, keep in mind that the magnetic ones need to have the two coils just as aligned; at least with my wireless charger if I shift my phone a bit it stops charging.

      1. There is nothing to be gained with that added complexity at all. The portion of the plate that is getting less coupling does not “drag down” anything. Larger plates capture more of the electric fields will get higher power transfer.

        1. On the contrary. Such complexity would be necessary to prevent you from crossing two of the transmitting pads over with the recieving pad and causing a short circuit. Think of doing it not capacitively, but simply with touching metal plates – it’s the same principle.

          The recieving pads can be large, but the transmitting pad needs to be a grid of smaller pads that are switched depending on how the recieving pads are oriented. They need to be probed first to find out which transmitter pads are being shorted out by the reciever, so they can be assigned to the same polarity.

          1. Why does everyone say this? The receivent pads and transmitting pads will be separated by an insulator, hence “capacitive coupling”. If they were exposed and made direct contact with each other, then DC current could just be passed through the pads directly and there would be no need for capacitive “wireless” charging in the first place…

      2. The mangetic coils don’t need to be aligned perfectly. Misalignment simply weakens the EMF in the recieving coil – it reduces the voltage you get out. In resonant coupled systems the misalignment has almost no effect on the output because the “unused” field is returned back to the transmitter, which is why they can manage relatively high efficiency over several times the transmitter coil diameter.

        Your wireless charger probably has a detector that deliberately cuts the power when the phone is misaligned, because it has some limited power budget for the transmitter for EMI reasons.

    1. It’s not the capacitor, but the use. I don’t think I’ve heard anyone purposely use dismountable caps to transfer energy. Sure, anyone who sees it can instantly tell how it works and the math behind it is not groundbreaking, but the idea is at the very least not common.

    2. The person who wrote the article and titled it “power transfer using capacitive plates” and us dumb readers had no idea this was using capacitive effects until you pointed it out. Now that you have proven that you are superior, I look forward to your future groundbreaking contributions which don’t make use of any existing technology.

  4. this is one of those things you ask your self what were we thinking? they over complicate so many thing just by not understanding what they are doing…good work..air capacitance is old tech i recently learned about..btw how may watts (volts/amps) you putting threw the pads

  5. Well, I am going to ask the “stupid question” I see no one has asked.

    I understand how the coil one works. It emits a magnetic field and it is induced into the other coil. Energy transfer.

    But to me this looks a bit “fake”.
    It isn’t transferring energy between the “source” and the “receiver”.
    All it is doing is making the existing electrons flow between the two plates on the “receiver” making it look like energy transfer.

    When the plates reverse polarity on the “source”, the “receiver’s electrons swap sides too, but I still fail to see how it moves energy from the “source” to the “receiver”.

  6. there is no need for shielding as one suggested, cause the capacitive coupling is different from magnetic coupling. the electric lines terminated on the secondary plate and hence there is no EMI and RIF issues like some of you’ll are worried. Ppl have used capacitive coupling to charge EV’s wirelessly. I can send you a link to the paper if you can log into IEEE. i know this for a fact cause m working on this system. The technology is still in its infant stage and is actually a complete alternative to IPT for small distance applications. Thank You

Leave a Reply

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