Let's take a look at a suspicious apple like usb charger, i say apple like because this one cost a pound from a shop in sucky hall street in glasgow, and i get the feeling it's probably not going to live up to apple expectations. It does have a plastic film round as if it is a real apple type product and actually the best way to get this off. I show you is a pointy object to get it off. That's probably the best thing to whiz it off.

That's your unboxing experience. Actually, the full unboxing experience would have been this crinkly plastic bag. It came in which i believe apple doesn't supply those either. Let's get down close to this i'll.

Tell you what let's do some tests, so i'm going to plug it in gingerly. It's not exploded. Yet this is good, let's plug the little tester into it, which is upside down. That's not helpful right! Okay, let's test! Oh, it's not going in that way either! Yes, it is it's going to go into the wiggling.

It's definitely not apple uh and the plugin is putting out five volts, pretty solid five volts, let's put a little load on it and see where it goes to keep my hand away from it, because i don't even know what the electrical isolation is like in this. Okay, i think you should be able to see it. Okay, the voltage is still holding if anything, which is rising very slightly uh we're at 900 milliamps we're now at an amp and that's what it's rated at. Let's keep turning up and see if we could bang or if it suddenly cuts out it cuts out just over an amp and keeps cutting in and out again it's gone into little hiccup mode and it's back.

Okay, that's a promising start, let's open it up! This came from ebay, don't ask me where i got it from. I got it a while back. It's a little load. Tester.

I've got a few of them here, that's the first one that came to hand. I've got a spudger, i shall spudge it open. The spudger might look different because it is a new spudger, not because the other one failed, but because i've somehow lost it. So, let's see how this thing opens up.

It should be quite hard to open it's not too hard to open. Often these cheap ones will open themselves when plugged into a standard british ring main or ring final circuit as we're supposed to call these days interesting construction, little circuit board, let's see if we can get this out. Oh it comes out quite easily. It's got a chip on the back um electrical separation.

There's the output socket, there's the circuitry surrounding it. There's the main type socket tree. There's not much separation right! Tell you what i'm going to take a picture of this uh, we'll reverse engineer and then i'll. Do a little high voltage test in the transformer one moment please and we're back.

So if you want to try and trace this out yourself here is the top uh? Well the component side? Well, it's the through component side. It's got the little transform which will inevitably get hot. It's got the output smoothing capacitor. It's got the incoming supply really on generous 2.2 mega far capacitor, and it's got a little bootstrap, capacitor, 2.2 megafart and a presumably one nanofarad.

Does it say anything on it, uh 2.2 nanofarads uh 1kv? Is that really going to be a class y? I doubt it 2.2. Nano 1 kv, okay right, that is the suppress capacitor. So let's take a look at the back which i've flipped around, so everything correlates. So the incoming supply comes in, and these two connections here skirts around the usb output connector and goes to the bridge.

Rectifier comes out the pre-direct fire and gets smoothed by that really on generous 2.2 microfarad capacitor um. The negative rail comes round here. To the other side, the positive goes to that capacitor goes to this uh very high value 5.1 mega ohm resistor, which is used to initially trickle charge, the bootstrap circuit, that's the bit that was cut in and out, and it also goes to the primary winding here. I shall write primary primary secondary and i'll just call that fb boot - i'm writing over this bit.

Here. I should be writing over this bit here because see that little tiny little tiny bit there that i can just fill in here. That's your isolation between the main side and the low voltage side. It's not actually good enough.

Really. They should have routed it out there if they were going to go so thin as that, but it is what they've done other things it's using an fm3773be switchboard chip, which is breathable enough. I've got a data sheet for that showing the schematic. It's not going to be a case of just comparing it.

It's going to be case to see what bits they've missed off. There's a 2.4 ohm resistor shown in this orange bit. The reason for the orange is that the current flowing through the primary goes in and these two pins the chip and then it passes through switch through to that and then there's a 2.4 ohm resistor and then a sense input from that to actually detect the voltage Across that to tell when to cut the coil off, if it's exceeding a certain threshold uh, there is a feedback winding which has the negative connection going to it, and it also has the other end of the winding goes via this diode to charge up the bootstrap Capacitor, which is here which powers this chip so initially when you power it up this capacitor here charges up until the voltages reaches a threshold, and then it will start running and if everything is okay, if it's not got a shunting load like i put on it, There it will then power itself. If it doesn't, it will cut back off again this resistor here and the value of this capacitor determine the time of that startup delay.

Uh, there's that little presumably a high speed died. It didn't check if it shot k or not. I could check if it's shot kid. No, i guess that's going to be short k.

Let's put this through to diode test and if it's 0.6 volts, it's not shotkey. If it's point, two uh should save it yep. If it's point six volts a silicon diode. If it's point two volts, it's short k, let's see if i can even get a reading here like this, it's a silicon, diode just a standard one.

Well, the other one's silicon as well, but it's more traditional diode. What about this one? The output diode is short. I would expect that, because it's going to be dissipating a lot of power, this little diode is going to get quite hot okay proceed. The other thing that happens with the feedback winding is.

It goes along this track here to a resistive divider that goes to ground and then goes to feedback pin and that kind of monitors the voltage and probably regulates the voltage on the secondary side. On the secondary side, we have the schottky diode for rectification going over to this fairly large 470 microfarad 10 volt capacitor. Let's get a little resistor across it to put a slight load in it, which is normal enough, five, one one which is five one and one zero 510 ohms. And then there is the class y capacitor, which connects the negative side of this to the negative side.

And output, it's designed to provide a path for current to come back that has leaked between the adjacent windings of the transformer, but it also passes enough current to give you a tingling output, usually positive more than a tingle uh. I shall do a high voltage test than this afterwards, we'll test that i'll remove the transformable test up to 4 000 volts i'll, maybe leave it in situation, actually we'll see how well the little capacitor fares too. So, let's take a look at the schematic by the manufacturer. Here is the schematic.

Are you going to be able to see this? It's very spindly, it's not very thick. Maybe i should have drawn out the big black pen, but i didn't we're just going to have to deal with it. The incoming supply shows a fuse, a fusible resistor. Actually, so that's missing uh.

It then shows two capacitors 4.7 megafar they've actually used 2.2. It's a 400 volt capacitor, so it's used for death beams too. Obviously it also shows the inductor another capacitor they're missing. That's missing um! It's missing one of these resistors.

Instead of the 1.5 mega ohm, they used 5.1 mega ohm, so i'll scrub that out um and this snubber network, which is quite important to protect the transistor here and we'll make it last a lot longer. That's all missing as well. So, let's scrub that out and there's also a little resistor, which is quite odd as a crude snubbing device, it's missing across the schottky diode uh, which would normally provide protection against a back emf, a spike from that that could actually damage the diode. Oh and, for instance, third temple, yes, it's got one of those, not two of them.

There we go. It shows the uh one nano one. Nano 101 is not one. That's a hundred pico, okay, it's bigger than that.

So there's a little um. There's a cent resistor going to the zero volt rail there's the little smoothing capacitor the bootstrap class, there's a little diode and there's a little divider coming from the uh feedback, wang that powers that chip, that is it right. Okay, now, let's do the exciting stuff. So i'm going to pause momentarily while i solder some wires onto the back of this.

I'm going to bridge all the windings together on this side, um and then i'm going to bridge the two connections here on this side and i'm going to like basically whack it with up to 4 000 volts. One moment: please: okay, i've hooked it up. I think we're ready for a high voltage moment we're at 4kv setting it's set to burn, which means it'll just sustain the current when it starts arcing uh. Let's go for three milliamps: let's just go with the full whack and uh begin.

So let's turn the voltage up and see how far we get so we're. Currently at 600, volts 700 800 were up to the first thousand volts so far, so good 0.26 milliamps leakage that may be through the class y suppression capacitor and between the adjacent windings. Now we're going up to 2000 volts, something just sparked: it just failed at 2, 000 volts. Okay.

What actually failed was the transformer that failed. Well, that has actually flashed over completely. It's not sustaining. It has completely failed right, interesting, uh right.

The next thing to do is to take transformer apart. That's a failure! One moment, please, quick updating that the transformer may not be the guilty party. It looks like it's the safety capacitor that failed. What's the next thing that's going to fail, so i'm going to turn the voltage up here.

I'm going to see at what point so it's now arc. I don't know if you can see that i'm going to turn the lighting down here, i'm going to take the exposure off here. That's why you have proper electrical separation, because if you don't uh things just arc and spark across circuit boards, but it's doing very well. It's not going through transformer, it's just arcing across that totally unsatisfactory gap.

Okay, yeah! That's not too impressive, not to worry right. Let's take the transformer to bits. The unwinding of the transformer was torturous and didn't go to plan. I did record something video then gave up, but the main thing is the outer binding was the sense feedback winding wound straight on top of the secondary, which is taboo, because the the feedback winding is reference to the mains, then there's a layer of tape and then There's a primary winding inside so, basically speaking, it's just uh the only thing insulating the layers is just it's not even double insulated wire.

It looks like standard thickness wire that it was just copper, wired, copper, wire with just a layer of lacquer insulating between them. I'm surprised it didn't flash over in the test. It shows you how good that insulation is um, how it would fare over time with thermal cycling and expansion, contraction of the windings against each other and a hot little transformer um. I don't know if it would fare that well with them, and it could eventually just you know it might start off fine.

Then it could go short circuit, um and when it does go short circuit, it would make the output live and if you were had your iphone or whatever plugged and not it'd be terribly happening, a one amp charger. If you had a metro, cased phone, you would by default. Let me demonstrate metal, cased phone if your hand made contact with that, you wouldn't be able to let go of your phone you'd be gripping tightly onto that phone just by the muscle of contraction. So not ideal, so i do not rate these uh.

I suppose really. It was a 1000 volt safety capacitor that failed at 2000 volts. It did manage twice its rating, but i'm pretty sure they're supposed to go higher than that. I'm not sure the exact specifications for switch mode power supplies, but the fact that the copper was just basically wound on copper in the transformer without any layer of insulation, without the double insulation that thicker sleeved uh, copper winding that's commonly used in the secondaries.

Well, that just gives me absolutely no confidence in it whatsoever. So that's uh, interesting, but ultimately is a case of what do you expect for a pound? Well, this is it. This is what you get for a pound, something that is not necessarily going to be, that safe and well might damage your equipment and might damage you.