A while ago I bought a box of LED bulbs from Teu 24 of them, six of each color red, yellow, green and blue and the listing did not say what voltage they were, but I thought well, isn't it reasonable enough that whichever country you order them to, they're going to send the correct voltage. Well no, they sent a 120 volt lamps to a country that has 240 volts. and because they're ordering and well, they do have a return system. Never tested it I Was going to try it but kind of stalled and delayed because I wasn't confident work from here owing to the weird status of this place that I am in.

Anyway, last night I opened one of them and it turns out that it was a good thing I kept them because they're very interesting inside. Let's open one! Let's cut straight to the chase. I Have already hacked one of these to change its voltage rating and the circuitry inside is amazingly simple. It's really interesting, so if I use, use the spudger to pop out the little end cap here, thus releasing that wire from the end.

And unfortunately, because these are crimped around, this is not a problem cuz I can put it back together again with this replacement Crim I'm going to have to actually hack into it with a pair of side Cutters and basically just Tin Tin opener It Off by unraveling all the metor. Oh, it does actually have tax it interesting. So I'm going to do that right now and show you the very, very minimalist construction. The LED filament itself is very Interesting.

but the power supply for it is even more interesting because it really is relying on a lot of the work being done by the filament itself. So almost there lied Clive struggling to open it because he wasn't being able to gain proper purchase with the side. Cutters This is where people will say couldn't you use a Uh Dremel or a pipe cut cut to open the lamp and well, no, it doesn't really work as well that is. The Dremel makes a huge mess and destroys everything and the pipe cutter doesn't really work in thin metal like this.

It is off. I should keep this little stud although I Don't know if we can really use it again, but I shall keep it anyway and I shall sweep the rest into the recycling container. And here is what's inside the lamp. now.

There are two wirs coming off the bottom with what looks suspiciously like resistors. Let me Zoom down this largely because they are resistors. This one I think it's this one is a 15K resistor and this one is an odd resistor that just doesn't have any specific value on it and really strangely see the filament supports here. they're the leads of the resistors.

That is very, very odd. So let's cut this one and cut this one because we want to strip the resistors out of this and replace them with other ones. Interestingly, they've been locked in by basically just fusing them in from the side. I Think they've used heated pins in here to actually mold the plastic in.

so we're going to have to try and get these out by other means that might not necessarily turn out to be so easy. Uh, this is where in the background I should actually turn the soldier on cuz I'm going to need that in a moment. So here is Mystery Resistor number one and it has no markings on it. It has a very low resistor.

It is a resistor. You can see where the laser has cut the spir around it, but it is being used just as a generic Link in this instance, which is quite strange. they could have used two resistors to reduce the dissipation and increase the power of the light. but they didn't As it is, the resistors they've used seem to be S half watsh Am I going to get this out? Oh, that's s gone horribly wrong.

Oh good. So let's prize this out there. It is brown, green, orange, 15 and 30 15K right? I Can't use that in the UK because with Uh, our higher voltage 240 volts is a lot more dropped across the Uh resistors. That results in much higher dissipation, exponentially higher dissipation.

But let me show you what's inside one of the filaments because I took a closeup and it's quite neat. I may actually have to zoom. No, I won't zoom out I'll just show you as it is. So here's the filament.

Here's a zoom up on one section of it and it shows that each end there are two diodes and there's two bus bar rails positive and negative and then hidden from view under the gel are individual LEDs Now I Thought these were going to be flip chip. the ones that physically the LEDs have no bonds onto them, they just physically so on to the PCB. But these do have little Bond wires and I can show you a close-up of that by showing you our doodle. Here is my doodle.

So the end of the ceramic strip and the LEDs are just on one side has a Metalized tab here with the two diodes on it one up one way, one up the other way. there is a little X I Don't know if that means positive. not sure about that. Um, but there's a crimp here and the crimp literally just sits over the uh ceramic end and then folds round it.

There's one of those at both ends and the four diodes effectively form a bridge, rectifier feeding the positive and negative rail. Let's assume maybe that this cross means positive. Maybe that's positive. Maybe that's a negative? Not really sure, but then there are discrete LEDs placed in the middle of that and then there are wire bonds linking them all together and then total there were 19 LEDs giving 55 volts.

And if you do the mathematics, 120 volts minus 55 is 65 volts to drop across the external resistors IAL v r 655 Vols to drop divided by the 15,000 ohms gives 4.3 milliamps, which gives roughly .5 watt dissipation from the whole light. However, half of that is across the resistor just over half of it. so the LEDs are only emitting about2 watts of light Seems inefficient, but the up upside of it is it's super, super simple and reliable. There's not really much to go wrong except the LEDs is worth mentioning.

I previously took apart lights that look like this mangled light, but they were. uh, they were 240 volt lights and they had a similar rangement but they had used uh LEDs a matrix of LEDs as a brid direct far and that was probably why they're not very reliable because that that's not a good thing to do to LEDs. But anyway, I have done the mathematical calculations and come up with the conclusion that I need 47k resistors for my application I've only got quar Watt they kind of used half Wat I could have maybe used higher, but I'm also quite wary of having something dissipating a lot of power in these little plastic cups. It's also worth mentioning there is a middle one.

not sure what that's for. Oh, actually, I could put the resistor into that one. That would work, but I'm not going to I'm going to stick it in here. Uh, and stuff it through that existing hole.

if I can stuff it through the existing hole, so much go wrong and this one, uh, can fold across and this can be the one that can next to the bottom of the filament. so I shall crop that now I've folded it and the other one assuming I can even get it up there. H The other resistor I shall stuff it through that hole. Hopefully it will go through what's left of that hole having pulled the bigger lead out.

Is this going to work? or is this going to mess up completely? It's kind of going through, it's going through I'm kind of holding in place cuz it's a friction fit. This is good enough. a friction fit. It's been lead as it goes through.

Okay, right. let's get some soda and Tin these leads. Hopefully that lead is going to be long enough. I might actually just weasle that down a bit.

So I've got a bit of extra length to go into the base. so I'm going to apply some Soder onto here and here and I'm going to Tin this bit here and I'll probably tin roughly up about I'll tin a whole load of it there I'll tin a lot of that lead that's good, can can crop it to length and now I'm going to solder this lead. Oh, that's quite hot. Yes, onto here.

watch everything just start moving around and being awkward when I Try this, it will. It's guaranteed. let's sploosh that uh Soldier about a bit. it's not going dead Central But that's all right.

and now this one's just about the right height actually. I shall just Reflow That messy. but you know what it will do? That's fine. See how sloppy I am? It will do indeed.

No. I'm going to push this one back in and note that H The one I took a picture of was uh, blue so had clear phosphor. These are all blue chips. These are based on just CU.

It's the easiest way to manufacture it by changing the color of the gel. the phosphor gel. They change the color of the light. so I'm going to put that back in there and this lead here simply folds up the side and it's going to get trapped on by the cover.

This one here though is going to go out the back so let's see if I can do that. Now it is sticking out the back. so I'm going to fold this one over and I'm going to get the little pin that you get with these and I'm going to stuff it in to hold it in place. Then I'm going to get my magic crimping tool from China where else this when you uh, have made a separate video about this.

When you turn the handle in this, you'll see the metal pins push in. So basically Al speaking, you, push the cap on. Is that nicely seated? I Think it's nicely seated. It doesn't feel that nicely seated.

doesn't feel fully seated. No, it's not fully Seated on that's annoying I think it's being stopped by the existing crimp holes. Let me try and push on a bit further without destroying everything at every opportunity. This video could go horribly wrong.

That's semi- promising. It's not very promising. Am I going to have to pause while I try and get this on I may have to fou those existing dimples down. Hold on I should just use Brit Force here on my bench.

Oh Brit force is not working. Uh, maybe I'll just do it as it is then in a halfast manner. If I push that, I'm already pushed this down to the point it's folding in. That's not very good is it? hold on H I'm just going to pause momentarily and resume.

So because I don't work in a Chinese Factory I didn't know the super secret technique. Once you've put this in, you actually screw it on. It physically screws on because there is a slight thread in that. and now I poke this little lead down and put the little end crimp on the little.

uh, the little pin that goes in. Where is my little pin? There's my little pin. Oh I've just put on the original one one moment. please.

Got not very organized Am I No I'm not very organized. Let me have another go at this. But at least you get to see the whole process as I screw it up repeatedly. I'm going to have to straighten this up now.

It's also worth mentioning the resistors they used because they were also the LED carrier were pretty heavy. uh set leads that may also be they s of weld them in there cuz with very little effort it's quite easy to drag this lead and then it makes the filament wobble at a funny angle. That's all right. if it ends up at a funny angle, it ends up at a funny angle.

Let's try it again, shall we? So there's a lead up the side. uh, which uh is going to in here first then that one's going to get trapped up the side. This one's going to go into the end cap I shall screw it on. that's better Third Time Lucky you reckon? I hope so right? the little pin smoosh the pin in and then this is where I drop it into this.

but because it takes a spare amount of force much more Force than I can do with my hand here. I'm going to have to put it off the edge of the bench. just give me a second here. So I'm bracing on the bench, turning it one indent and you can see it's pushed the little crimps in there so this is ready to plug into a holder.

This is going to be a good idea. I've also knocked LED filament at a fun angle, but a brief tap will probably get it back into position. Here's what: I made earlier: Uh, here is the new one with its 47k resistors which should actually be a bit brighter. Well, that is notably brighter, right? T I'll get the meter up and for this I think we'll use the anti because although it's got a dimmer display, it does theoretically have higher resolution for lower current load.

So I'm going to plug this in here. Uh, and I'll actually take the exposure off. Oh no, You know what? It's not bad is it you can see the display, it's not shimmering too much. I plug this LED light in and it's showing 2 milliamps which is what I wanted .5 Wats um, and9 uh 97 factor which is actually really good now that 05 Watts um, there's a a reasonable amount being dropped across the filment itself, but the rest is being spread across those two resistors.

I would guess uh, there's roughly about2 wats being dropped across each of those resistors which are .25 W resistors. so that should be all right. So the answer is and I'll Zoom back out here. Uh, the Temu 120 volt lamps can actually be hacked and it's really interesting.

the circuitry use I like the fact that they basically just use a couple of resistors, one of which in the case, 120 volt one is a dummy resistor and the other is the 15K. But by using 247, you can actually get a decent brightness out of this. I can knock that filament back into shape. The other one, which is this one, almost put that one in.

That would have been great. It would burn out the resistor instantly. This one was my first experiment with just uh, a couple of 100 key resistors. It's much much dimmer.

It's only doing 1 milliamp and a02 Watts. But it's still surprisingly bright at night. But there we have it. The Tuu I mean intensely speaking.

it's is going to be very bright. It's not going to last very long. Uh, 13 milliamps? That's way too much. 3.3 Watts That's way too much for a half watt resistor.

Uh, But there we go. the Teu lamps. Very, very simple. Very interesting.

With all the work being done in the filament with the bridge wck far and the LEDs and then just a couple of resistors to balance the current, that's about as simple as you could get. Really quite interesting lamps.