A little dig through the archives of the Channel shows that the last time I took an Ikea coupler power supply to bits was approximately six years ago and since that time, this very Supply I got I was so impressed I bought another one after destroying the other one completely. This has been plugged in 24 7 since then. it is my main charging device for all the devices because it was very good quality, but now the coupler has been superseded apparently by the Smile Gal. Now here's another thing: Coppola means connect in Swedish but Mahagal means small hail I'm not sure why they chose that name.

it's a nice enough name. It's kind of odd and the specifications are almost identical. Between these, they look identical and the cases? well, I guess it's the design is accredited to David Wow. I Guess that's just the case.

I'm not sure about the electronics. Oh, the reason it says John I'd like to thank John for actually sending me a couple of these power supplies because I don't really have access to Ikea stuff over here. they don't ship to Isle of Man bastards anyway. Uh, the specifications are almost the same except that the new SMA High gal if that's even the correct pronunciation for it.

Instead of just being at 200 to 240 volt unit, this one goes from 100 to 240 volts. so it's totally Universal For travel and stuff like that, it is the same power rating in the sense this output is a maximum of 3.4 amps total at 5 volts and uh, 2.4 amps per output. Um, so let's compare them now: I Get the feeling that this is going to be destructive. The last one if I recall was kind of glued together I think this one by a how do we go to I think it is ultrasonically welded together.

so I think this is going to require the Dremel So I shall use the Dremel and I'll be back in one moment. one moment please and resume. and I'll Zoom down this a little bit so that took quite some time to reverse engineer. As you'll see when you see the other side of the circuit board, we have the incoming Supply live in neutral going to a 2 Amp Fuse We've got an in-rush limiting thermistor here.

We've got a metal oxide wristered clamp Spike sorry about all the black, the white Speckles over this it I really had to dremel that big time to get it open. May have actually touched the capacitors a little bit in the process and scuffed them. but there's two capacitors and an inductor. We've got the mosfet that's switching this Transformer on its little heatsink.

We've got the class wire capacitor. We've got a big link over here a little tiny capacitor, which is the main smoothing capacitor on the secondary, and then we've got three parts. If we take a look at the other side of the circuit board, it gets more complicated, very complicated. I shall zoom in just a tiny bit more.

but I'm going to abbreviate this. Won't spend too much here because it is. Well, that's way too much because it is a very densely packed circuit board. Here's the incoming Supply that Matlocks have Riska fuse NDC It's got the bridge rectifier the two capacitors and it's got that little inductor for interference suppression with a 4.7 K resistor cross it I've shown the color-coded the sections.

we've got the chip and it's interesting that it's quite a complex chip and it's got a unique feature involving this little transistor. something I have never seen before and it saves a bit of power, but it seems a bit Overkill to be honest. Um, we've got the mosfet over there or the drive circuitry. We've got the bootstrap circuitry and processing and feedback circuitry quite a lot.

This is what took so long to reverse engineer really. Uh, but let's say that that and that it's just like it's been a long. it's been a long shift uh, but very densely packed. uh, little resistors? thanks for the numbers on them, but the tracks good weaving in between them.

We've also got the snubber network here, which uh Clips the slight transient when this Transformer under here turns off uh to protect the mosfet. Um, and then we jump over this big separation Gap no slot, but that this is good. It's a good wide gen separation Gap and we've got the class Y suppression capacitor jumping across and we've also got a little spark gap there to allow elk just to crack across quite a large gap. Mixed thoughts about stuff like that sometimes I wish they'd just use some high value resistors perhaps to augment that, but they don't.

I've got my reasons for that. Some of the things that get plugged into these like Tesla coils and ionizers and stuff that generates high voltage like plasma Globes and that can put a lot of stress between the primary and the secondary side of the power supply. On the secondary side is another big huge change versus The Cobbler This is a mosfet and this little chip here is what's called an asynchronous rectifier control chip. I Shall show you that in a moment.

and basically speaking, it replaces what was a Shopkin diode before, and what it does is when it detects the point that this: Transformers applying power to this side, this little chip turns the mosfet on it effectively acts like a switch, a very high speed switch and avoids the drop you'd normally get across diodes and means this power supply should theoretically run quite cool. Not sure what it's going to do in the primary side, but this side will run cool. Um, also held by the other little thing that I'll show you in a moment about the power saving other side I'll show you in the schematic. Lots of little chips here.

These three chips are all overload protection chips. They are programmed to allow about 2.4 amps or so per output. If you exceed that excessively, they'll basically cut that output off. Then we've got two other chips down here that connect the data pins.

This one just connects to one of the three ports and this one connects to two of the ports. And the point of these chips is they actually whatever device it plugs in, whatever it's looking for, somehow these would be chips. Decipher that and say things like yes, I'm a real Apple product you you may take charge. It's worth mentioning that these units are just five volt output.

They're not going to step the voltage up, it's not going to do quick charge, but this is good because you can over you could if you try to charge your phones too quickly on other devices, it can basically shorten the life of the Lithium selling them, right. Schematic: Let's start with this chip and then we'll move on to this side. I've broken it down in debate size pieces to make it more manageable. Is this going to fit in? I Shall Yeah, it's going to fit in.

Hold on. I'll just tweak that down just a little bit. It's fitting in. So what we have here.

This is the chip that we're using. I've used the original manufacturer's schematic, but I have augmented it with the extra components and some of the values that they've put in. So here's the supply command goes through that fuse. There is the NDC thermistor which is actually over on this side and that just basically limits the inrush current.

It starts off the higher resistance and then as it heats up, it drops in resistance. Then it's got the metal oxide. Barista The Vdr A MAV MAV Vdr many names the clamps spikes. We've got the bridge react far.

We've got the two 400 volt death Beam 15 microfire capacitors giving a total about 30 megawatts of super mega Death beam. Um, and we have the inductor between them with this little resistor across it. I've drawn the resistors in the same style as they drew the resistors, which is the old-fashioned zigzaga style. We don't really do that in Europe Drop the zigzags a while ago for uh I Kind of like the zigzags to be honest.

I Had a difficult time changing from the zigzags to the the little rectangles. They also change all the logic symbols to rectangles with symbols in them. It's just it just seemed like idiots were involved. But you know that happens a lot.

So when you power the circuit up to start off with current flows through these resistors. It's a classic bootstrap circuit. And this mosfet. I Think this might be a mosfet that defaults to on, but you turn it off.

Not sure, but uh, it charges up these capacitors here and once the capacitor reaches a high enough voltage, it kicks the chip into action and it starts driving this mosfet and pulsing the primary winding of the Transformer. At that point, if everything is going to plan, if there's no short circuit on the other side, you get feedback through this winding which doesn't just go to the sense circuitry for the voltage on the other side because that winding will reflect the secondary winding depending on how it's being loaded. But it also provides power to this capacitor and once a trip detects everything is up and running. It turns this transistor off and the reason it turns a transistor off is because normally those resistors would be trickling current, but you know it's not a lot of power.

Let me just get the Kink Power Glitter and try and work out how much it's saving by doing that. So typically in the UK our voltage will be the peak voltage of our 204 volts. play. I'll be roughly about say 300 and let's make it 350 volts, 350 volts and it's got two 402 200k resistors in series for that.

So 400k divided by 400 okay gives a current of about 0.8 milliamps. 0.875 times that 350 volts gives a power dissipation about 0.3 Watts. So by cutting these resistors out, it is reducing the dissipation from the unit by 0.3 Watts I Suppose that's useful enough. It certainly will appeal to the Eco people who want the power of these things to be zero when they're on standby.

To be honest, they're normally pretty low anyway. Um, but once I started up and that transistor has turned off, everything's running. Uh, we've got the output driving the mosfet, turning it on and off to basically build a magnetic field in this. Then it turns off and the field collapse isn't the secondary.

It takes the energy from the coil and also the feedback. Something worth noticing here there is the feedback. uh, these sounds which they've used two resistors in parallel and a series resistor to finely calibrated to give five volts in the secondary because that does reflect what's happening on this side, but they also have unusually I didn't see this on the data sheet they've used. Two resistors in series defines tuner value.

Another resistor down here going into the configuration pin and it's a multi-function pin and one of the functions is it can respond to over voltage. Possibly if something really major goes wrong. the sense circuit doesn't work and the voltage goes too high. This is a second layer of protection.

It'll shut the chip down. The other thing that can shut the chip down is a thermistor. They've used a 100K thermistor and they've balanced it with a series resistor. I Think that was 5.6 K Hold on.

Where is it? Uh, 5.6 k562 Uh, let's just write that in 5K 6 5.6 K and that thermistor there? If this circuit board I'll show you where it is in the board, it's right here. If the general mass of the socket board and everything inside keep in mind this does the other side of that. Transformer If it gets too hot, it will shut the chip down. It will protect against overheating.

This is good. Other things where they've not before we move on. The mosfet has two cent resistors in parallel down here 1.511.3 Ohm providing feedback. It's got a 10K pull down resistor in the mosfet which isn't shown in the data sheet.

Normally they have a diode to make sure the mosfet turns off really quickly as opposed to turning on. So it's got a slight soft start of the 60 ohm, 8 ohm and then a sharp turn off. Then it's got what looks like a capacitor. Um, between.

Yeah, this is hard. It's connected between the drain of the mosfet and the gate. I Don't know why they've done that. Part of me wanted it is a high value resistor upside down or something like that or or is it an arch capacitor? Uh, not sure that is.

but I measured it and it seemed to come in at 54 nanofarads if that wasn't just a skewed reading, so it might be a capacitor. Anything else on this. So basically speaking, they've gone above and beyond with the primary side. This is good.

Next, we want to take a look at the secondary side and this is where it gets really interesting. I should have locked the exposure off from that, but apparently I didn't lock the exposure off from that has that been flickering up and down? That would be annoying if it had. Here is the secondary. There is the capacitor coming from the negative of the primary to the negative the secondary just to provide a path back for past of the coupled current that can just reduce radio interference, particularly with long leads connect.

Output: Here is the mosfet and a control chip. This control chip took ages to identify. This is possibly and I say possibly because it's the only one that had a printout that roughly fitted. It may be a Dialogue Semiconductor Iw673 Digital Green Mode Synchronous rectifier controller.

Most of the other chips had pin one driving the Gate of the mosfet. This one pin three drives to get the mosfet. Very hard to find, but this one does tally up with all the pins. Iw673 Let's just write in.

Iw673: Uh, it looks as though they've left their options open. There's a zero Ohm resistor used as a link with nothing passing under it that looks as though they've got the option to add their resistor in there. They've got a resistor to the gate and this is a system that technically speak when it Powers it up I think current I'm not sure how these if they purely pair themselves From there, there's no real yes, there is a current path via this. Um, but the main thing is that when it detects that uh, this is going negative down here because it's the positive is straight through when it touch the screen negative, it turns the mosfet on and bridges it to the Uh zero volt Rail And that basically just acts like a diode.

Every time it gets that pulse of power from the Uh switch from power supply, this turns on solidly, which means there should be virtually no heat dissipation that then goes to charge a 1000 megafied 6.3 volt capacitor, which is actually a fairly tiny looking capacitor when you see it. I Wonder how long that'll last 1000 volt 6.3 microfarad at 1000 Microfarad? 6.3 volt? That one that would be completely different if it was there. One thousand Volt 6.3 Magnifier wouldn't do much, but after that, then it goes to all the other circuitry and it's basically this multiplied by three. We have the five volt bus coming in and the zero volt reel and the first thing it comes to is this Uh current monitoring chip and it's got a decoupling capacitor.

Fairly high value on this side I Think high value I Measured about one megafide across these. Um, but it has a resistor 51k that sets a threshold that will allow current up to that threshold and then it will potentially cut off if there's too much current drawing. It's basically to protect against short circuits and then it's got a 10K resistor just basically as enable so that puts the supply out the 5 volts out to the USB port and for the data to allow for uh, different things looking for different charge rates. It's got a dedicated chip which was also quite hard to find.

Let me see if I can find that chip. Um, this is the Td9521 Td9521 was the power chip and then after that um, we've got if this is it Pg3w2se which basically has a the ground connection and the five volt connection. and then it's got the two data lines data minus data positive and uh, that goes to the data connectors and just signals to the device You're plugged into a charger no matter what it is. There is another version of this with six pins and it's got the positive.

It's got the zero volt reel and then it's got a data minus one data plus one data minus two data plus two. So it can actually do two ports and that's how it's doing the third one that the two other ports. So there's one of these chips for one port and then there's another one of the other one that is doing the two separate ports. But that does mean because it pairs itself in this chip.

If that Port gets overloaded, it'll actually knock out the data signal for the port next as well. but that is fundamentally it. so that just leaves the Transformer to take a look at because the Transformer is the bit that mainly separates you. You've got this class Y capacitor.

It's usually a nice thick fact as it is, but this is the one with its thick windings. Come out here that uh is a bit that separates you from Mains voltage. should something go wrong? So uh, let's take a look at that one moment, please. I have removed the Transformer and it's housing into that plastic outer shell with uh, the FedEx core and on the primary side with a little thin wire here, just laid and taped onto the ferrite core, presumably as part of the screening.

although I wouldn't really consider ferrite as terribly conductive. Let's Zoom down on this and see if we can peel some layers off. So if I peel this outer tape off, it may reveal part of the primary winding. since this is someone who sometimes how these things work or a feedback winding.

So that's one end of the low voltage winding is laid over that, which seems to have good insulation on it. What do we have? What do we have to get the tip off here? Or just, well, basically rip it off. So there is one of the feedback whinings I Think let's cut that wire. Is that going on to the outside? It's highly probable that it has so I filled that up.

Pull this tape off and we'll unwind it. The tape is sticking everywhere, so this looks like possibly the feedback wind. We'll see if it. rolls back onto those terminals and if that is the case, then I'd Next I'd expect to see either the secondary winding with good insulation.

so this is going back to it is going back to the feedback winding position. So that was the feedback. When let's get that wire off and see if I can find the end of the tape here. This is sometimes tricky because uh, when they wrap the layers of wire over the top, it pushes the tape together so tightly that you can't sometimes unravel it.

but this looks promising. This looks okay. So this looks like a primary part of the primary winding, which is also expected. Let's see if I can get the outer Prem winding off.

Sometimes the frame rate is wound as two layers, a layer of the primary, then the secondary, then another layer of the primary just for good coupling, and then the second. Uh, then the feedback and the outside of that. So now we're down to what I think is going to be the important bit. noting that the cable is sleeved quite heavily the secondary, which is good.

that gives it a good electrical separation. This is one of these bits of tape that is sandwiched down so tightly. or is that the bits underneath? Where is the end? Not necessarily seeing it. This does look like it make our screen the wrong direction.

so tightly taped? Yeah, that's it. It's coming off, it's coming off, We'll get it. Oh, we'll get a bit of it. Oh, that is so tight.

Let's do like this: Yellow Tape Although it is just standard transformative, it's kind of frustrating at times for the exact reason we're experiencing here. It's is, do they call that double insulated or triple insulated, But it's completely kept, well away and it's only a few turns. Oh, look at that. That's it.

Super well insulated, three cores for a flexibility and for laying them flat. Uh, and well sleeved is that super thick insulation. So the transformer in short is well up to spec. That is good.

So this thing looks pretty good. There is one thing that I didn't mention in the first part of the video. Uh, there is a snubber network. They've kind of doubled up in components here as well.

What that does is that when the mosfet turns off, you tend to get a collapsing field in this before the other side catches up. The secondary can clamp that. so to clap that tiny Spike you get there's a diode here, some current limiting resistors, and then a capacitor that just absorbs that Spike and then a discharge resistor across that to keep that capacitor from just charging up continually. and it just basically it resets it, ready to absorb the next.

Spike The other thing that's worth mentioning that I Kind of missed couple of things Actually, Most remiss of me. Uh, this little snubber. Network which is, let me just grab the the picture. if I can find it is here.

it's across the mosfet and what it does is that uh, when the other side turns back on, you sometimes get a voltage Spike and just to protect the mosfet from that voltage Spike This acts as a little low energy shunt. it just absorbs that Spike again just to protect from excessive voltage. The other thing is worth mentioning. It's a trivial thing, but also worth mentioning the data chip.

Its power supply is basically decoupled from the five volt rail by one key resistor and a capacitor just to provide that stability. That's what they're for, just in case you're wondering what they are. uh, it just decoupling it just gives its nice stable. Supply So in summary, the new power supply the magical.

If that's even the correct way to pronounce it, it's made up to the usual standards. It's made to a very high standard. It's a very, very safe power supply. Um, and I think I'll probably switch to using this one as my replacement for the other one.

not that it needs replaced yet, but it's just good to refresh your power supplies every so often. And uh, it should be good for hopefully another six years. but I'll find out. The only way to find out is to test it.

but um, everything seems to tally out. It seems to be made to the standard you would expect from Ikea and uh, that totally beats the crap you get from eBay and Amazon that is a either cheap no-name brand. Oh, this thing isn't expensive, but I mean this thing is dirt cheap. This thing is cheaper than Dangerous Ones sold on eBay and Amazon So uh, it's one of these things.

Ikea have a reputation to uphold. The standard's very good. It's actually made to eco-european standards. Um, and it's super safe to plug your devices into.

So what more could you ask for? For something that is so important as a USB power supply, it's a Gooden.