These wireless LEDs seem to be aimed at modellers to allow them to add lights to models without any complex wiring. They work well, but are not bright. Fine for low level indoor lighting though, and the power use is less than one watt.
I've seen other videos about them, but they don't go into much technical detail, so here's the video that does.
It's intriguing that the system is based around a standard chip designed for inductive charging. It looks as though it could have other transformer drive applications too.
Here's a link to the aliexpress listing I bought the coil from. It wasn't shown as being shipped with LEDs, but did come with some in my order. The current price is £5.78 plus shipping.
https://www.aliexpress.com/item/4001225212166.html
The MOSFET package on the PCB remained cool throughout my experiments, even when I put over 100 LEDs inside the loop and the current doubled. It was typically about 15C above ambient room temperature.
If you enjoy these videos you can help support the channel with a dollar for coffee, cookies and random gadgets for disassembly at:-
http://www.bigclive.com/coffee.htm
This also keeps the channel independent of YouTube's advertising algorithms allowing it to be a bit more dangerous and naughty.
#ElectronicsCreators

By popular request, i'm going to be taking a look at this coil system commonly available on line, and it's aimed at modelers and you've got these little led receiver boards and when they get put into the coil like that, they glow and you can put a whole Load of these in, in fact, let me show you this is the best way to demonstrate i'm going to turn the light off. Take the exposure off and then show you what happens when i pass this little coil over the leds. It looks a bit like uh. Almost like a magnifying glass going over the leds.

Now it's not a bad representational intensity at the moment. Actually, it's pretty good representation intensity. It's not super mega bright, but it's it's bright enough um, but you can see the effect you're getting there that wherever the coil passes over them, they're picking some of the energy up and it's actually making the leds light up. It's quite a neat effect right watch.

Your eyes, i'm going to bring the light back, so this device isn't custom made specifically for this application. The circuitry here is specifically designed both the control circuitry and the power driver. It's designed to drive calls like this for inductive charging, so if you've got say, for instance, you've got a flashlight or a torch that you want to be able to just sit in a base and it charges. This is the sort of thing you use and it's only powered by five volts and the current is about 80 milliamps, and it doesn't seem to change that much even when you pile in quite a lot of leds, it just seems to remain fairly constant.

So if you do the maths um, it comes in about half a watt, so this thing is not super bright, but again it is just aimed at modelers. It's not going to be it's not going to light your room as such, but it is visually attractive in a dark room. So i've been doing lots of experiments. I'm going to push these out the way i'm going to bring in your first schematic.

So, let's uh put this out the way before i short something i will be completely reverse engineering. Everything you'll be glad to know. As usual, i'm going to turn the power off to that, so the first schematic is actually the receivers, so these are based around a standard little inductor and the inductor is wound with a two. So it's terminated in two little uh solar pads on top on.

That is a circuit board, which it turns out that you don't really 100 need it, but it probably assists them for manufacturing, but on one side the circuit board are these two pads that then melt onto the solder of the inductor on the other side coupled through, Is plated through holes are two pads that led and a couple of pads with a 330 pico fired capacitor, i'm not sure the exact purpose of the 330p fire capacitor i've seen this before i mean scotty of strange parts featured these ages ago. One of the first things he did in this video was: he dropped one and never found again. Oh he's right. These things are tiny.

If you drop them in your carpet, they will roll and they roll very easily. They will roll and you will not find them. Um, but when i saw them in scottish video, i thought why do they even have other components? I thought why doesn't the led get subjected to a reverse voltage, that's too high, and why? What is that low, capacitor and people were saying it's a tuned circuit? I'm not so sure about that the inductor value 2.2 millihenry um is quite high. Actually it's very, very fine windings.
I can show you what these things look like up close. That would be a good idea. Would it not here's what they look like up? Close there's inductors two, two, two uh two two and two zeros 2 200 uh micro henry, which is 2.2 millihenry, so you've got these very fine. Windings and they're coming up onto the circuit board.

You've got the led there and you've got a little tiny capacitor there um they have two sizes: they've got the four millimeter ones that came with it and the five millimeter. Now, when i ordered this from aliexpress the coil, what they didn't say, it was going to come with these leds, so i thought i was going. I thought i had to buy the coil and the leds. The coil wasn't that expensive.

It was about six pounds. I think that uh, the leds i ended up just going to town. It was one of those things that you can order a few or you can order a lot. There's, not a massive price difference, but uh as it is, ended up spending.

Far too much. We got a hundred of the leds for no good reason at all. It just seemed like a good idea at the time. I also wanted to put as many into here and have them all lit up as possible.

I shall do that. So what i'm going to do now, i'm going to take some scope measurements, i'm going to show you the results of some other experiments um, but first of all, i'm going to put as many of these into this coil as i can and then see. Well, how well they light up one moment please, and the answer is uh. Yes, you can shove them all in, but rather alarming.

The current has uh more than doubled in the process of that so uh yeah, i think i'll, be uh. Checking the temperature of this. So i'll be back in a moment, i'll tell you: if the there was a major temperature change in the actual control chip, you may notice that there's well that warm white one there it's very different to others. That's because it's a slight custom modification, as is this orange one, so i'll be back in a moment, i'm going to do the reverse engineering now and get the oscilloscope traces.

So we can explore this further one moment please and resume: let's get down close with this circuit board and we shall explore. Then i shall show you the schematic for this. The supply comes on here in this case it's 5 volts, and the first thing it hits is a 22 micro farad capacitor. I know this because i desolded it and measured it.

It then feeds this mosfet package, which seems to be dedicated. It's an xkt 333 nice memorable number, but it is just purely a mosfet, but it may be optimized the high frequency. This thing can operate at the main control chip. Here is a xkt001 again further memorable chip and it possibly has a frequency set by two resistors and that drives this mosfet.
The mosfet then drives the output coil and that's fundamentally it there's this capacitor, which is 100 nano third, which i thought was quite high value and that's the one that some people think it's a sort of resonant circuit, i'm not sure uh, i'm not really sure about That at all, if we take a look at the manufacturer's data sheet, it shows i shall zoom down further on to this. It shows exactly that setup, 22, meg farad. It shows the two resistors slightly different values here: the chip driving the mosfet and then the output, which is uh. It says five micro, henry inductor.

I did try and measure this uh inductor, but my my meter said no and uh the c1 here uh, they say 100 nano, that's what i measured 100 nano farad. So it's following the textbook. It's also worth mentioning that the data sheet i'll have to zoom back out a bit here is in chinese. That's not terribly helpful, but it basically says that the chip is designed for the 3.3 to 18 volts and also one kilohertz up to 3.5 megahertz um and they've got other information on there, saying that if you don't want everything to burn up when it malfunctions, you Might want to use a ptc thermistor, that's very helpful of them.

Let's bring in the schematic. Here is the original thing. I've already shown you the little circuit board with the inductor in the back and the led and capacitor i've done some experiments are they needed? Are they not? I shall show those experiments in a moment. This is the circuit as shown as used so there's the us.

I've shown it as a usb supply because that would be the most convenient supply 22 megafard capacitor across the supply rails. Just for that uh to provide a local buffer, this is the chip that does all the pulsing that's just dedicated. Its purpose is a charge chip, that's designed for inductive charging, and that shows a standard value of 200k. But then this value has been tweaked.

It's not what they showed in the drawing, but that will be used to match the coil here. In this case it's marked 71c, which equates to 553.6k. That then drives the mosfet there's the four turn coil point: five millimeter diameter uh wire. I did not measure the damage of the coil.

I shall measure the damage of the car. I shall use a nice ruler here. I shall measure it in millimeters and possibly inches. I'm not really sure see uh.

The coil is roughly seven centimeters round uh, which is approximately just under three inches worth mentioning when they've wound this coil, it's bonded together. They've obviously put it between a sandwich of a former where they've actually laid the wire in with adhesive, and it's actually bound together. So it's a very flat coil and then we wrap captain tape here when it comes out to this. I don't know how, if it can be wound loose or if it's just they've done this, for keeping it super flat or if it actually helps the function.
The receiver well there's the coil, there's the diode, the led and there's a capacitor, 2.2 millihenry and 330 picofarad things worth mentioning about the circuit board. It's more or less single-sided. Most the tracks are on this side, but there is one thick track on the other side. For the positive rail, because this uh is to ensure that this coil can be pulsed at super high current, so this uh rail here these two positive pads are linked by a big thick track in the back and these the coil output and the negative just have Their own little extra square pad in the back just for strength, but in this case this whole area here is the incoming negative going straight to the mosfet, and this whole area here is the output to the coil.

So it's all optimized for of high current right. What no i'm oscilloscope traces! This is what we want to see. First, oscilloscope trace proper pad here to purple pad here. It's the uh drive from this, the frequency and everything to the mosfet.

I scoped with the keysight, and it came in a slightly wavering at 223. Kilohertz, roughly, i was giving it slightly more than 5 volts. Fundamentally, it's just almost like a close to a square wave, but it's just basically just driving that uh, the um mosfet with just a square wave slight waiver here. But that's what you'd expect just for that was probably also being affected by fluctuations in the power supply as it switches the mosfet on it does monstrous things: um, the coil, here's the coil being driven this one here is the zero reference for this.

So it's actually pulling it down to about uh the five volt level and then, when it releases when it turns off it, just basically you get that inductive. Kick it just a little sort of almost like a sine wave. Look how symmetrical it is before it pulls. It down again releases it.

It is almost like tuned to that circuit. Maybe that is what the whole capacitor thing is. Maybe it is effectively a tuned circuit, i'm not an expert in tuned circuits. I try and do all digital if possible.

Here is what i picked up with one of the coils, with the circuit board removed and two wires putting it, and i scoped it again. Here is the reference to one of the leads here so on one half of the waveform it's pulling down to about. Well, this is two volts of division. It's pulling down to about five volts and then, when it's releasing it's going up in the opposite.

Half to about two four six, eight say: nine volts, so uh. Technically speaking, the there is a risk that the leds are being reverse biased by more than 5 volts. Hmm, oh, i should actually let you see details about this 219 kilohertz in this one. I'd lowered the voltage at that point, um anything else worth mentioning peak to peak.
It says: 13.8 uh. This one was 222.52 kilohertz. So it's just, i suppose, ultimate, it's just a it's, not super mega accurate, it's just possibly those resistors in this case 223.36 kilohertz. Okay.

So i did some experiments i modified some of these. I shall tell you what i did actually i'll turn the light off and show you, but on two of them i just put inverse parallel these or you can't see a thing they're miles away. Right say what there we go. I d sold the circuit boards completely in these two and replaced it with inverse parallel surface mount leds, orange ones and warm white ones.

This one is one of the original ones. The capacitor, this one is one of the original ones, but with the capacitor removed just to see what effect it has, i shall bring in the coil. I shall turn off the light actually and i shall um bring in the coil and energize these. So the first thing i notice is it's going to be bright enough.

It is the first thing i notice is that the first two to light up as i bring the coil up close them are the one with the capacitor and uh that may it just seems to affect its range, but look this one is with very little the One with inverse parallel, capacitors leds here that lights up quickly these ones - i would expect light up sooner, because they're lower voltage, they're orange leds in reverse parallel, and this one is the one without the capacitor and it's the last one to light up. But you know: there's not a huge difference, actually there's a modest difference in the the coil distance to it before it lights up, but the inverse parallel leds are the simplest approach here. I wonder if there's a downside to that, the one that's lighting up first, is that one one with the capacitor, the one that's sliding up. Second, is the one with inverse parallel orange.

The one that's laid out. Third, is the inverse parallel white and the one. That's lighting up last is the one with the capacitor removed. The one that's brightest here is the one with the inverse parallel warm white leds.

So maybe there's an advantage to doing that. This one won't look so bright because it's orange leds they're not very bright generally, but this one is absolutely blowing these two little white ones out the water for intensity, and that is just uh. There's no circuit board at all. It literally is just the leds soldered directly onto the pads in the back of that inductor and bring the light back watch your eyes.

So that concludes my exploration. It's an interesting little thing. It's basically been repurposed from that standard charging chip. It's a very minimalist circuit.

It's just those two components: i wonder what other applications they have um, basically driving a square wave high current output uh. It could be used for sounders, i suppose, or driving other transformer systems. The leds they're quite neat, um, quite a bit of work in making them they're, not cheap, because they're very trendy at the moment, if you, although they supply these in separate bags, if you want to know what colors they are, you can just basically hold it down. Next to here and move it about, and you can see all the leds in the bags changing color.
Oh something did i mention this. The data sheet also mentions 80 milliamps plus or minus 20 milliamps. I think that's what you tune this to by using your resistors, and once you get your coil done, i think they mentioned five micro henry in one of the data sheets: five micro, henrys, uh, five micro, henry indeed uh, but i think it's down to experimentation. They do suggest you know changing the number of turns to fine-tune things also, possibly change those resistors to match the resonant frequency, but they do in chinese mention a current of 80 milliamps plus or minus 20 milliamps and in its resting state without any leds in it.

It's 80 82 milliamps, so that might just be the defaults of value. It's very low, current uh when i had all 100 in it peaked at about 195 milliamps, which should be about at five volts about um one watt. Just under a watt. The passive standby is just half a watt and that's even with quite a few leds in it.

Oh here's, the other one i did uh. I just soldered some leds directly across it and uh the two blue leds light as well as you move up and down. One led goes out first before the others. Just because that's possibly that's the symmetry, isn't it that's the symmetry of the waveform uh, how it's a peakier and one half, but there we go interesting stuff.

I shall provide links in the description. I shall provide a link to this coil. Hopefully, it won't i'll, provide a link and i'll tell you the price just so that if they start price gouging, you can actually uh look at for other sellers, but there's loads of people selling these and again this one. I can't guarantee it will come with uh.

Some of the it came with the 10 of the four millimeter leds uh, but it didn't actually show that in the listing it didn't mention it, but it did come with them, but i don't know if that happened every time. So there we go interesting stuff. Certainly, food for thought:.

18 thoughts on “Exploring wireless leds with schematic and waveforms”
  1. Avataaar/Circle Created with python_avatars Pp Ss says:

    Brings to mind a couple questions… First, how large a coil could one get away with, possibly overdrive it? And, it would be interesting if you shaped the coils, i.e., formed tuo a square, triangle, star, etc, could you effectively manipulate a large layout of leds tuo confirm to the shape? In conclusion, would led reverse parallel circuitry result in better performance overall? Cool stuff Clive…

  2. Avataaar/Circle Created with python_avatars Fred Bloggs says:

    These things make you want to believe that a compelling use for them exists…

    …except you can’t quite work out what it is until someone else does it…

    …and then you loudly exclaim “of course that was obvious, I should have thought of that”.

  3. Avataaar/Circle Created with python_avatars Peter A1466 says:

    Did you try double reverse leds with a capacitor, or only without the capacitor?
    What do people use these for, besides making Youtube videos?

  4. Avataaar/Circle Created with python_avatars Tony Sheerness says:

    You will find the dropped led when you wake up in the middle of the night to go to the loo, when you are half asleep ans stomp around in bare feet. The pain will remind you not to drop them.

  5. Avataaar/Circle Created with python_avatars Nicholas Boyarko says:

    I've had my eye on these for a while but haven't purchased yet. Thanks for the breakdown, now I don't have to do it!
    It's one of those things that's neat, but I cant think of a practical application. The current being so low, the potential applications are even more limited.
    Could be cool for something wearable or a simple indicator light based on proximity.
    Now, if you could push a good 20W through this, there's tons of other applications.

  6. Avataaar/Circle Created with python_avatars SuperDd40 says:

    My comment is of topic, sorry. But ,Freedom need your help. The "Canadian Winter" is storming the planet, join in and together we'll make it snow in hell. Canada is leading a world wide movement, please support us, support freedom.

  7. Avataaar/Circle Created with python_avatars Heinz-57: A Lil Bit of Everything says:

    In case anyone wanted to know what the data sheet says translated, here it is, at least what was shown on the screen:
    XKT001
    I. Overview
    The XX1001 chip is a new wireless charging solution launched by Coretech Semiconductor. The working voltage is 3.3V to 18V. Under special requirements, a 4.2V lithium battery can be used to directly provide power for the transmitting part. The chip is packaged in SOT23-6. further compression. A lot of optimizations have also been made on peripheral devices to further compress the finished product size, and further optimize the production process and cost. The operating frequency range of the chip design is 1KHz-3. SMKHz, so that the chip has more frequency options in the circuit design, and its high frequency The frequency output can use PCB printed coil instead of winding coil, and realize high power output, which can greatly simplify the production process. The chip has reserved high-sensitivity control pins (pin 6). In engineering design, a control signal can be applied to it to meet special requirements such as low power consumption, or a control signal lower than the operating frequency can be applied to it to control the working state of the working end of the receiving part, making the back-end function design more efficient. Diversity and freedom.

    2. Features
    *Small size, packaged as SOT23-6
    *High working frequency
    *High integration, less peripheral devices
    *High output power
    *Wide range of applications
    * Freely design control functions
    *Under special design, the working state of the receiving part can be controlled
    *Working voltage: DC 3.3° 18V

  8. Avataaar/Circle Created with python_avatars hmarc417 says:

    Awesome…….> Did I miss the live stream for today, i got up early to be ready on your other channel but I guess i am 12Hr. off ?????

  9. Avataaar/Circle Created with python_avatars twocvbloke says:

    Well, they'd possibly make for an interesting blinky party if places around a mobile phone if they could pick up the signal from one… 😛

  10. Avataaar/Circle Created with python_avatars Hola! Freaky Jason says:

    I've made the equivalent of these using some NE-2 neon indicator lights. Just get a big enough inductor. Guaranteed proof of concept is the primary winding from an old 9V wall-wart on top of one of those Qi charger pads. Just connect the NE-2 bulb across the coil and you're in business.

  11. Avataaar/Circle Created with python_avatars Konstantin Grigoriadis says:

    The Leds with the LC Circuit form a Tuned Circuit and draw energy from the Electromagnetic Field of the Transmitter, like a Grid dip meter does.

  12. Avataaar/Circle Created with python_avatars Texas1FlyBoy says:

    I'm blaming Sir Clive for my electronics addiction. 😉 I have so much more enjoyment in life playing with electronics (and radio). Of course, I'm probably spending way too much money on all these things. Oh well… You only life once. Might as well enjoy it while you have it!

  13. Avataaar/Circle Created with python_avatars Peter Stimpel says:

    For the interested ones: atomic14, a scottish youtuber, was playing around with those, as well. He made an attempt to self build the LEDs in one of his videos.

  14. Avataaar/Circle Created with python_avatars DC Allan says:

    Wonder what an RGB led would do? With the diffrent forward voltage and the wave form make them stick on the first colour? great video 2x👍

  15. Avataaar/Circle Created with python_avatars MikeMike says:

    One way to think of how this works is to consider the big coil to be the primary winding of a transformer and the little coils to be secondary windings.

  16. Avataaar/Circle Created with python_avatars MikeMike says:

    Adafruit also sells a set with a larger, 20cm coil that runs on 24V and 10 wireless LEDs, which unfortunately is currently and most often out of stock.

  17. Avataaar/Circle Created with python_avatars Johan S says:

    It may be very well a resonance circuit; 2.2 mH + 330 pF gives about 187 kHz, which is near the 220 kHz you measured. It would also explain why the LED with the capacitor removed was dimmer at first.
    5 μH + 100 nF is 225 kHz. It all matches up. It will also mean that a larger coil, or one with more windings may perform worst.

  18. Avataaar/Circle Created with python_avatars BRUXXUS says:

    Wow, I've never seen those little lights before. Being assembled directly onto the SMD inductors is super cool! Might be a neat way to do light up clothing.

Leave a Reply to Johan S Cancel reply

Your email address will not be published. Required fields are marked *

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