Hi, it's tear down Tuesday Again, got something a little bit different. It's one of these Brwn electric toothbrushes I use them I Love them. They're fantastic. Can't do without my electric toothbrush and you've seen them.

It sits on one of these. Chargers Wireless uh Power transfer to charge the internal battery so I thought this one, uh, crapped itself I think it at least needs a new battery. so I thought we crack it open and check it out. Not only what's inside here, but what's inside the charger as well.

Let's take a look. could be interesting. Now here's the charger part of it. This one's specifically uh, 220 to 240 Vol So it's not the universal type.

this is model number 4729 made in Germany Hello to all my German viewers Beauty I know you love it and uh It 2 Wats 50 60 Hertz only and uh, it doesn't look like this sucker is going to be easy to open. It looks like it's um, maybe, uh, you know, heat sealed around the edges or something like that. So uh, before I uh, really crack this thing open and see what's inside I thought we'd um, just, uh, run a simple uh test on it to see what we're getting out of the coil here because the coil sticks up in this part like this. So when the toothbrush mates in there like that, it just it just sits in the bottom and that gets proper coupling from coil to coil.

because um, you need proper coupling for these Wireless power transfer systems to be anywhere near efficient. they're still going to be probably horribly, uh, inefficient. but um, I Expect this thing not to be operating at 50. HZ Of course it'll be operating at a much higher frequency because there's no way you're going to get a loose coupling like that working at 50 htz.

In these sort of sizes, it's just not doable. so it's probably working at a couple hundred ktz or something like that, perhaps. So maybe this is even potted inside I Don't know if we'll even be able to see the electronics in here. so um, I don't know I might have to get the Dremel and take that sucker apart.

but uh. anyway. I thought we'd uh, hook on and see what we're getting out of this coil. Now how do you do that? Well, um, unless you have a matching coil to sit on there, there's a very simple thing you can do with your Crow probe.

Crow stands for Cathode Ray Oscilloscope. It's an Australian term. sorry old habits I Keep using it. the oscilloscope, probe or scope probe.

You can actually form a little transformal with it. A single Loop like that and that can be used for RF Uh. pickup for high frequency? Um, pickup. It's It's pretty horrible and uh, very inefficient.

But it does actually work. It acts as a single turn, and it does actually work at high frequency. So let's uh, just put that over here. I'm sure it's going to be absolutely, uh, horrible.

In fact, actually I'm not even going to bother cuz I know it's not going to work at the couple hundred khz we've got here. So what can we do? Simple, we can use our ground lead and wrap that a couple of times around like that and that will give us better coupling. And if we go up here and take a look at the oscilloscope, let's see what we get. Aha, we have something.

Let's have a look. What have we got here? we've got. let's Center that expand it out and Bingo! Yep, oh no. I was way off on the couple hundred khz.

I was off by an order of magnitude. It's only 20 khz. There you go. and it's uh, you know, around 700 MTS Peak to Peak the absolute the uh voltage of course.

Um, it doesn't really matter cuz it's just what's being picked up by our coil. But there you go. that's roughly sinusoidal. There's a little bit of something happening on the top there I'm not sure what, but uh, there you go.

that's 22 odd. KZ All right. so you'll notice that with three turns of my scope. uh, ground lead on there I'm getting about 250 molts uh RMS on there.

and if I remove one Loop there, let's let's have a look. There you go. It drops down and if you remove another one, it drops down again. But actually surprisingly, we are still picking up 20 khz there with just the single.

The single Loop like that is just enough. very Loosely Coupled like that is just enough to pick up that signal. All right. So what we'll do is, we'll just clean that up with an average in a bit.

So we go into our choir menu. We'll just turn on, uh, some average in there. Let's do more than two averages. Let's say say take it up to eight averages.

There we go. We've cleaned up our waveform a little bit. Okay, what I'm going to do just to get consistent results? I'm going to Sticky Tape a just a single Loop down on there like that. So it's fairly consistent as you can see and uh, it won't change around if we wiggle our probe and play around with it like that.

It's pretty steady like that and as you can see, it's beautifully sinusoidal on the bottom. Like that. Almost perfect. And then there's that distortion on the top there.

Let's see what happens if we add a load by putting on the extra coil on there. Look at that, you can see the rising slope at the bottom. There start to distort around about there. Look at that as we load it down.

the amplitude really isn't changing, it's staying the same. And then there's the peak. There's that little Peak up the top which is changing a bit as we load that down now. I'm not sure this actually doesn't uh charge anymore.

so the lights don't come on. so I'm not sure what's actually gone wrong in here. Maybe it's not actually the full load, but uh, just the act of putting that coil on there does change that. Waveform does distort it a little bit more so you can see how it's certainly possible.

Um, using this Uh, Poor Man's um, inductive? Uh, pickup here to actually? um, you know, do some troubleshooting on these type of Wireless uh, power transfer circuits. You can, at least, uh, play around and have some fun. Anyway, it's by no means absolute accurate, but it does allow you to couple into a system like this. just using your scope ground lead.

It's a really neat trick. Now, as for taking this sucker apart, it's uh, pretty trivial, and it even provides instructions on the back here. It shows you to put it on the stand like this and then give it a little bit of a Twist and Pop Goes the Weasel there. it is awesome! So let's hey, look at that is our battery going to fall out? We've got a contact down in there, so let's take a see if we can.

Uh, get that out and take a closer look. Now you should be able to see inside there. It's actually got three wires coming off the coil down in there and you can see the coil just down in there. you can see the multiple wraps in the coil.

You might be able to see it. Yeah, a little bit better through there. There we go, but we're got two what look like little Ferite beads just sort of stuck in there not doing anything. um I'm not sure what their purpose is and there they are.

I've taken them out and they just sat in there. There's just two of them sitting in there like that. I'm going to stick this back on the scope up there and uh, see if there's any difference in the waveform. and I did stick it back on there and uh, trust me, there was absolutely no difference whatsoever.

I won't even bother showing it. So um, I'm not. I'm unsure how to actually get the rest of this stuff in here out. You can see the battery in there with the worlded uh contact terminal.

that would be a uh, nickel metal hydride battery and it's sort of. it's pushed. all of this stuff if I try and pull it out, it sort of pulls all this stuff with it. and that bar the uh, the uh vibration bar there doesn't um, slide through or anything like that.

so I don't know. um requires some more percussive maintenance I think when in doubt, just yank harder. here it is. it just uh, popped it pulled out.

look at that. um I think I've uh yeah I've busted uh, something there from these two things. Here's the uh, here's the battery. There we go.

What is it? a 607 M hey, it's Soo excellent. They put a top quality SEO in there nickel metal hydride. There you go. so uh, technically that's replaceable except I think I've uh, busted the whole show.

so I'm not sure. um, it's probably some sort of, you know, oneoff, uh, press fit type thing I don't know, but I can't find any way to get the rest of the case open so uh, to get a look at the motor and stuff which would be up the top end? um I'll have to crack that open but we have some. Electronics Beauty and sorry I was mistaken about the number of wires coming out of there. It is actually uh, four, three didn't make sense of course.

So there's uh, two pairs coming out there. One of them is common in there on that terminal, so there's only three terminals on the board. All right. it's time 10 macro lens time and this is going to be, uh, really good.

We should be able to trace out these circuits. look like looks like there's part readable part numbers on both of those chips will get into and uh, not much, uh doing on the bottom there. um, some test points of course and uh, a couple of tracks. so we should be able to trace this thing.

actually reverse engineer it pretty easily. It's a 668 331 7415 R8 Well, doesn't ring a bell. might have to Google it? Well, this one, uh, wasn't trivial to uh, find and I think think I've found it possibly um, it. it all seems to uh, match.

but you Google uh, you know 668 331 is really the only thing that you get and it turns out that's an Em look. It's in the exact package you want. uh, t-p I'm got on on one of these. um Ali You know one of these Chinese uh, broker, uh websites and it's a Tsop 14 package.

Exactly what we got and that's presumably that's obviously the manufacturer. So you Google Em 6683 31 and you end up popping over to uh, various sites. But you get on to a company called Em Micro Electronics Um, and they're part of the Swatch group go figure. And they claim that they're uh leaders in ultra low power and low voltage uh Solutions including micro controllers.

and as it turns out, if you go into their microcontroller section, they've got um, em, uh, let's have a look here in MCU Waiting waiting. they've got Em 6682 but they don't have a 6683 so go figure. I don't know whether or not it's an older type I still I Googled that and I still couldn't find it, but who knows. maybe it's a custom uh part specifically for Brawn Who knows.

But uh, you go take a look at it in the EM 6682 and it's a rather interesting thing. I would have had uh, no idea that these things are even existed if I didn't open this toothbrush. That's what tear Downs can do. It can lead you onto Parts your search from and it can lead you to manufacturers you didn't know about I didn't know about these little ultra low Power 8 Pin Micr controls I Had no idea this company even existed.

but uh, here you go. They've got some, uh, looks like some uh, novel ultra low power eight pin microcontrol. They work from uh low 0.9 volts um to 5.5 So that that's pretty much ideal for single cell, um, single cell operation like single cell alkaline? Uh, cuz really, their um. energy density the energy usable energy in an alkaline battery, for example, pretty much exhausts itself at around about that 0.9 volts figure.

Some people take it as 0.8 volts. That's a more common thing, but basically from a single alkaline cell, you can use up, um, all of the energy in say, a AA alkaline battery battery for example, with this microcontroller. So that's really quite neat. Another interesting thing has got a 4bit ADC or 12 levels of the supply voltage.

So obviously there's no built-in reference by the sounds of it, just uses your supply uh, voltage to do that. but it's only got a 4bit ADC which might be all you need and these things are I don't know, they might be really, uh, Dirt Cheap or something like that, but it's got a mask uh ROM built in. So these are not flash devices so you can expect these to be really, uh, cheap because they don't have to have all the extra um expense of making a die with uh, you know, flash memory reprogrammable flash? They're a mask ROM device. Um, and they've probably got development kits to go along with it.

If you're you know that's why you've probably never heard of these, because, well, you know, hobbyists and uh, and uh, you know, professionals just working on your standard. you know, stuff off the shelf from Dig Key Have no idea about these sort of things and they're certainly probably not. uh E Easy to start developing with, and the core is an Em 6600 that sounds like their own uh core. It's probably based on some derivative of something somewhere along the line, who knows.

But anyway, it's a complete single chip solution. Uh, 4 micro amps in active mode. the Um built-in oscillator. It's designed for ultra low power stuff.

It only goes from 32 khz to 800 khz. The main oscillator. It's got a watchdog, uh, timer. It's got a sleep controller uh, 10bit Universal counter timer, some interrupts in there, and a 4bit ADC So it's really, um, you know, a pretty specific, uh, low power sort of.

you know, consumer microcontroller in these sort of consumer appliances. Cuz you can bet your bottom dollar that Brawn paid absolute lowest possible. They shaved every cent lowest possible cost off the price of this thing and they would have ordered millions and millions of these parts to be using these. uh, toothbrushes.

So um, no wonder they picked, uh, possibly, um, something like this. This company may have offered them the cheapest price with their Mass ROM and their 800 khz and no flash and you don't need anything fancy. There you go. Typical applications household appliances and if you go down further and take a look at the CPU it's only a 4-bit CPU There you go, they still make them 4bit processor.

Here's a classic example of one using your two brush. It doesn't need anything more cuz this is an intelligent in quote marks. uh, toothbrush. It you know, has timers in there and uh and uh, stuff like that.

So it uh, you know after a set time it'll beep at you and stuff like that it'll pulse the motor and uh, do uh, you know various intelligent things to tell you that, um, uh, you know you've been brushing for, you know, 3 minutes, So it'll it'll vibrate the motor and actually warn you that, okay, you've been brushing for 3 minutes, don't you know overbrush or something like that? So it needs some intelligent control in there. that's why I'm pretty darn sure. even though our number on the chip is the 6683 I think we've actually got the controller here. It matches up in terms of the um uh package and the functionality.

It's You know exactly what I would expect I Would have expected a super cheap, low-end consumer microcontroller in this and you don't get much more lowend than a built-in mask ROM and a 4bit core. There you go and it's two cycles per instruction. 74 instructions. Um, and there you go.

It's worth taking a look at this thing and if we uh, go down to the bottom if we take a look at this, it's uh, 2008 Copyright 2008 There we go. Rev: D and uh package marking There you go em 6682 No, it doesn't mention anything to do with 6683. So there you go. These are s soic but I'm sure it can come in.

no. there we go. 4 Imp Tssop and uh yeah, I mean we do have um, those sort of numbers. We've got R8 on our Um package as well, so who knows.

But uh yeah, I think we've definitely got the right device here. and our other device here is a TSM 7401 with a marking 7K 4 Once again, not familiar. going to go have to look it up. The micro controller isn't going to be able to drive the motor directly, so uh, let's go look that up.

and curiously, when you're looking at these things, you'll notice sometimes you really have to get the right angle of light on these chips to read them. as soon as you get that right angle. you know, bang. It just really stands out.

But uh, sometimes if you don't get the right angle, you just cannot read these things at all and they appear just as a, you know, a completely black, uh, surface. You can't read anything at all. It's all about the light and the angle. Well, there's nothing complicated happening there at all.

It's just an ordinary N Channel mosfet in an So8 package from Taiwan Semiconductor. It's only a 20 volt job at Uh 4 1/2 amps. Not much doing there at all. You could have used, uh, any, one, of, uh, countless number of Um N Channel mosfets on the market, So clearly it's just a, you know, as you'd expect, a standard DC motor and they're just hooking that directly across the battery and turn it off and on with the mosfet.

That's it. Now let's have a look at what else we've got on the board. We've got a attacked switch obviously, which goes through to the button on the front of the thing to switch it on and off, and uh, got a few passives, a couple of uh dodes here in a mou uh package, and uh, which you don't see too much these days, and a couple of So 23s there which are OB or like might be uh, transistors or maybe even uh, dodes or uh, or something like that. so maybe even a regulator perhaps.

So um, and we've got a couple of LEDs here. Two of them that, uh, match different colors that, uh, match up to the uh light pipe or the Um or the cover on the front. The clear cover where they shine directly through there. And that's it.

There's not much happening at all. We'll have to trace the circuit out. Uh, shouldn't be too hard. Now let's have a quick look at the input circuitry here and you can see it's got two coils here, with one common.

Why they've got two coils? I Don't know. your guess is as good as mine. If you're into Wireless power transfer technology and stuff like that, Maybe it increases. You know, the the coupling coefficient between the coils or something like that.

These things are fairly, uh, critical when you design these power coupling, uh, systems like this. You know just the physical, uh, aspects of how they're wound, how they're mounted, the physical spacing of the coupling, the rotation, all that sort of stuff can really matter. So it would have been engineered precisely to give a particular output voltage here on the rail. So basically, we've got a rectifier on, uh, each part there.

This is obviously a charging lead here, which comes from the microcontroller. and uh, we've got another diode here and we've got a um, a filter cap. So we're going to generate a DC voltage here on our rails. Okay, I've got my probe hooked up to the power rail there.

and as you can see, we're getting 4.53 volts. There's a bit of uh bit of noise on there, but it's uh, not a big deal. You'll find that will be the Uh 20 khz switching frequency and we'll take a look at that in a minute. and if I wiggle that around a little bit just back and forth.

there's really no change there. But if I start to lift it up, you'll notice the voltage. You can see the voltage dropping. Got the DVM feature here and uh, you'll notice that voltage drops as I lift that up I don't have to lift that up much.

only a like a millimeter. You can visually start to see that voltage drop and it's practically I'm not going to say it's linear, but it's almost a linear relationship between the Uh height and the voltage there. There you go, not not quite. You'll see you might notice that the LED actually flashed there when we got to a point.

It's probably not going to do it. Probably a bit of hysteresis there. There we go. If you saw that there we go, the LED.

It's just fractionally switching on when it transitions through that voltage there and if we probe one of the coils there, check it out. That's the waveform we get and it's about 13 just over 13 volts Peak to Peak there. and if we freeze that you can see you can see this ringing on the rising Edge there and Associated ringing on the lower edge with a flat top and a flatter Bottom Now as opposed to what we were getting when it was unloaded. and a bit more interesting is if we probe the other channel, look at what we get.

There we go. They're inverse waveforms. Look at that. I'll zoom in and give you a good look at that.

There you go. Once again, they're about just on 13 volts Peak to Peak and we're talking. there's that 21.8 khz there. that's our switching frequency, but you can see they've certainly flattened on the top.

Um, this? uh. this channel here is more rounded on the bottom than this one is. but they're ENT ually inverse waveforms. And if we have a look at the Uh Ripple on our Uh 4 1/2 volt rail, here, you can see the Uh 22 khz there.

There it is 22 khz. but of course you can see that second Rise there. which Uh is due to the fact that we're at full W rectifying this thing from the second coil. and if I actually cut one of those coils off there as you can see, it's not.

you know as there it is, it's we don't have that second little rise in there. Well, maybe when it's uh, charging. if it was charging the battery properly, perhaps it might matter a bit more, but I think maybe it's got something more to do with the just you know better coupling coefficients between the Uh coil. Perhaps when they're uh, do you know when they've got larger charging currents? maybe? but I don't know.

It' be interesting to see if um, if we actually got full charge current on this thing. and as it turns out, the pinout for our microcontroller by the way on the board doesn't actually match that uh, data sheet we saw. So um, it's but I still think it is one of those, uh, one of those devices though. Maybe a custom derivative for Brawn or something like that.

perhaps. All right now, let's hook up a new battery to this thing and uh, see what we get charging current wise? let's give it a go. 100 There you go. Just over 100 milliamps and you can see the charging.

LED Flash there. I'm going to rise. Lift this up. Whoa.

Yeah, you only lift it up a millimeter and you can really see that charge current instantly drop away to whoa to nothing like it's halfway up the stem, there, the supporting stem and it's it's dropped away to nothing. It just doesn't kick in in at all. Now let's have a look at what happens to our voltage rail up here. our 4.5 volt rail.

when we hook up our battery. Let's give it a go. Whoa. It drops to just over 3 volts and it drops dramatically down in frequency.

We'll freeze that and uh, what are we getting there? We're getting uh, 800 odd Hertz There you go. The switching frequency drops dramatically when we're drawing um 100 milliamps charging current from this thing. And that gives us another thing to check. What happens to our the actual output waveform from the Transformer.

I've only got one side here of the Transformer When we turn it into charging mode Let's have a look. Oh, look at that. it's dropped. Looks like there's some.

hey, there's something. There's some pulse thing happening there. Let's try and capture that. There we go.

Look at that. There's these pulses in there that are, uh, look at. Check that out. Look at that pulses at what interval we're talking.

You know a millisecond interval or there abouts just over. Well, that did correspond to our 800 Hertz that we were seen before and certainly there. It is. No surprise that's the 800 htz will will get in on the Uh 5V.

Well, it dropped down to 3.3 volt. the voltage rail. um, after the rectification and filtering. So it's using some sort of it's deliberately pulsing something I'm not you, that would probably it' be on the charger um, side.

Actually, draw I would assume from the charger side drawing these pulses like that. Remarkable. And indeed, if we take a second Channel and uh, measure the battery charging voltage and we freeze that we can see that the battery charging voltage. Look at that, it's got Ripple on it, then it doesn't for that period where it jumps back to its original um original position.

So obviously it's doing some sort of pulse charging of the battery in here at 800 odd Herz So what we're actually seeing here is a pulse uh period charging period between here where we've got the Ripple on uh, the uh Well, the charge Ripple on the battery and the amplitude of our Uh waveform from our Transformer of course drops down due to the Um extra current being drawn from the power coupling scheme there. and then you have a period where it just uh goes um, it. It switches off the charging during this period, which is why you get a flat line on the battery because you're not measuring the charging Ripple anymore, you've just got the flat battery voltage on there and our waveform returns to normal as we saw, which is the same waveform we saw before with the Uh no load with the no Uh battery load on there. And if we have a look at the basic reverse engineered Uh circuit here, we've got the Dual coils over here full wave rectified uh by these two diodes and then goes through another series diode here which generates our positive voltage rail uh which you saw was 4.2 volts or drops to like a 33 Vols uh during Uh charging.

But when it's in operation and there's no Uh Power coupling through the coil here, of course the voltage rail for the IC uh the microcontroller is from the 1 2v rechargeable nickel metal hydride battery minus uh, the diode drop here. So that's why we need a really ultra low voltage uh microcontroller that can operate from anywhere as we saw in the data sheet from 0.9 volts all the way up to Uh 5 plus volts. So you need that entire span to have such uh, simplistic circuitry like this and powered from the Single Cell Now uh, what we've got here is uh, of course the Uh motor is connected directly across the battery with the end Channel mosfet goes directly to the IC control and it can switch that off or on um under intelligent control and it is actually um, this particular toothbrush is the like the topof the range professional model and it um does uh things like actually up pulses the motor to after a certain time to let you know you know. oh, you've been brushing your teeth long enough so you know so it can do smart stuff like that because it's under Uh IC control like that and it's very simple.

just a standard N channel. Uh, switch there and we've got a simple RC filter here which goes off which allows the uh that 4bit analog to digital Uh converter inside the Um microcontroller or might even be a a you know, a better resolution analog to digital converter allows it to measure the battery voltage during charging. Now as for charging itself, it's very simplistic. We've got a switch up here I don't know whether not it's a Um, whether or not it's a Moser or it's a bipolar device.

it's little so 23 package and it basically um directly. There doesn't seem to be any current limiting uh in there in series with it. It basically it looks like it connects directly from the Uh fourway rectified Um voltage on the power coupling coil here, straight through to the battery like that. So really, they're they're relying on the Um.

the maximum current available extracted from the power coupling coil is the maximum charging current for the battery and as you saw, they actually pulse that uh charging at around about 800 Hertz or there abouts. and uh, we've got some filtering here of course for the Uh Rail and uh, some reverse Doo protection there and the charging LED uh over here, which you saw that they flash it once every 2 seconds or whatever while it's Charing and that's basically all there is to one of these rechargeable toothbrushes. And as for the base unit, yeah, unfortunately I was right I took off this uh back thing here and you can see the potting compound in there. Really freaking annoying.

So I'm afraid. uh, we're not going to be able to see inside that one cuz I quite frankly couldn't be bothered. Actually, it's not that I couldn't be bothered I've actually, uh, run out of time. time for tear down Tuesday I've got to head home and edit this video to make sure it's up on Tuesday But uh, there you go.

that is. Um, inside one of these electric toothbrushes. They're rather interesting I hope you liked that. Uh, tear down.

And if you want to discuss any of this, if you're into, um, all this, uh, Power uh, Wireless power transfer technology and uh, stuff like that, there's a lot of, uh, lot of art involved in this sort of stuff, and if you want to discuss it, jump on over to the Eev blog forum and remember, as always, if you like tear Down Tuesday Please give the video a big thumbs up and we'll catch you next time.