Wireless Power system does not... well... power anything

I am trying to make a wireless charger for a small handheld device. I've tried searching on the internet for something read-made, but everything I could find was much larger than I could fit.

The way I understand how wireless power works is that you basically have an air-core transformer with a 1:1 turn ratio - which in theory should pass the power through the air. I therefore made two little coils of 0.1mm magnet wire with ~250 turns each and with a resistance of 15 Ohms.

I used an H-Bridge driver to create a square AC voltage with peak to peak of 5V on the primary coil, but even my best attempt only created ~1V RMS on the secondary coil. The main problem is also, that if I try to power anything from the secondary, the voltage immediately drops to 0. My ampmeter registers 2µA flowing in the secondary, while it consumes up to 1A on the primary (which makes the primary very hot).

I tried with AC frequencies from 100 Hz to 60 kHz and I had the best results with around 4-10 kHz.

I thought I could just rectify the output of the sec and use those 5V to charge my battery, but this is apparently more complicated than I thought :)

Do I need some sort of resonating circuit? What am I doing wrong? Any help is much appreciated

Here is an image of my setup:

And here is my schematic (really basic circuit): OUT1 and OUT2 is where I put the probes of my oscilloscope for measuring output voltage. +5V goes to my bench supply. The Arduino is powered from USB.

• A dimensioned drawing might help or a good sharp, cropped photo of your construction with something in it to give scale. Jun 26, 2021 at 13:39
• Resonant power transfer would be more efficient, but the efficiency you're getting now is very low regardless of resonance, so I suspect something else is wrong. Jun 26, 2021 at 13:47
• @Transistor I've added a photo of my setup Jun 26, 2021 at 13:47
• More turns is better here, don't bother with less. I'm not sure how many turns is typical for wireless power transfer, but 250 sounds a little low. Jun 26, 2021 at 13:52
• @ChristianidisVasileios I've added a schematic Jun 26, 2021 at 14:07

What you are trying to achieve is a DIY Qi-Charger! Awesome!

Qi-charger usually work at an operating frequency between 87 to 205 kHz. Usually 100kHz is fine. To assure neat power transmission, it is essential, that both coils are more or less in sync, so you want the receiver coil to have a resonating frequency at about 100kHz (or whatever you choose).

Additionally, a Qi-charger provides a simplex communication from the receiver coil to the transmitter coil to adapt power flow, I think you can omit this, but be aware, that the voltage will rise, when the drawn current drops, and vice versa. If this happens in a Qi certified device, the receiver coil tells the transmitter coil to adapt power, which is done by slightly changing the frequency out of sync (or more in sync), which changes the efficiency ratio between the two coils.

$$$$f_0 = \frac{1}{2 \pi \cdot \sqrt{L \cdot C}} \approx 100kHz$$$$

So you can measure the inductor of your receiver coil and then calculate the necessary capacitor you have to put in series with the coil.

• The turns ratio does not have to be 1:1 but something that suits your application. Adding metal or ferite will help the energy transfer.
– Gil
Jun 30, 2021 at 20:00
• Thank you very much for your answer, I will try to build your design! :) But I still have some questions: When I increased the frequency to around 60 kHz, the current through my transmitter coil was only around 0.09A. How do Qi chargers transmit more power at even higher frequencies? Do I need thicker wire (I'm using 0.1mm currently)? Also, what would be the best way to measure the inductor, I don't have an LCR meter currently. Jul 1, 2021 at 14:20
• @Gil Would you recommend a higher number of turns on the transmitter or the receiver side? Also, where should the ferrite/metal be put? Should it be inside the core of each coil, or between the coils? Jul 1, 2021 at 14:22
• That depends on your design and the current and voltages you need. Remember watts in = watts out + loses.
– Gil
Jul 1, 2021 at 14:28
• You can try with thicker wire. Have a look at actual Qi charger coils (e.g. here ) so you can get an idea, what to design yourself. Also have a look at the Q-factor of those coils and at which frequency it reaches the maximum. A good explanation is on Wikipedia (link ). Jul 2, 2021 at 13:58

Reason for drop in the supply Voltage: From the image, a thin wire with turns is used as a coil. Higher frequencies due to the skin effect, the concentration of charge is more near the surface as compared to the core of the conductor. The ohmic resistance of the conductor is increased due to the concentration of current on the surface of the conductor. Hence Voltage drop increases.

The solution will be to use a thick, stranded coil

In order to increase the coupling between primary and secondary coils, a Ferrite plate can be placed in between the coils.

The coil will acts as a series RLC circuit, So operating at resonant frequency will increase the efficiency of the system

• This answer would be much better if you could calculate the skin depth at 100 Hz and show that the depth is much smaller than the wire diameter of 0.1 mm. Also, can you explain how the OP got better results at 4 kHz than 100 Hz if the skin effect is to blame? Otherwise I am skeptical of your answer. Jun 26, 2021 at 17:00
• @ElliotAlderson Obviously at 100 Hz, there will be no effect of Skin depth. But there are other factors included right! Induced emf = 4.44*fNphi, When the frequency is low, Induced emf is also proportionately low. With the increase in Frequency, the effect of Skin depth is realised. Jun 26, 2021 at 17:16
• So can you still say that the skin effect is "Reason for drop in the supply Voltage"? It doesn't sound like this is a valid explanation to me. Jun 26, 2021 at 17:34
• When you say a "thick, stranded coil", you're referring to Litz wire, not just ordinary stranded wire, right? Because normal stranded wire provides no advantage whatsoever here. Jun 26, 2021 at 17:36
• @ElliotAlderson - An approximation of Skin effect shall be considered as - R(AC) = R(DC)*Gauge factor * Sqrt(frequency). Hence Resistance of the wire increases with an increase in Frequency. Primary coil resistance is increased now. As we are loading in secondary coil, drop in voltage is observed Jun 27, 2021 at 4:54