And the lighting is back and by that I mean hopefully it won't just glitch out every so often. Also, the two hot spots are at the back from the flood lights. That's what happens. I Also wrote spring snow in this side I Was definitely trying to see how long these lasted.

Uh, to be honest, maybe not used these. So the construction of these is typically an aluminum. uh, Extrusion and then the end caps. and if you wiggle the end caps, they usually slide off revealing the potting compound.

Is it same going to happen? Other end, a little bit of pressure sometimes needed. Sometimes They come out easily. Sometimes they don't let me stick a screwdriver under that and use more pressure. Oh, that one is well in.

That's annoying, but that's okay. it will come out. now. what I'm going to do here once I've got this out is: I'm going to try and hopefully just chop those wires off Sniffy Snippy.

Hopefully I can grab the PLS here and just Flex this out a little bit and maybe encourage it to let go of what's inside and it'd be nice to slide it all out in one because you can reuse these little housings with the end caps for potting. Electronics which is quite useful. not when you do it far too much like I've just done though. Maybe that is going to help though.

I'll just do it like that. Screw it. Uh, is that going to make it come out? Am I going to cut myself horribly? now? No. I' I've really messed this up.

That's by using Over Zealous for, so that's fine. That's the way I shall go with this. Then the chances of me actually needing to pot something are fairly slim. Well, obviously they were until I said that.

So let's get this open. Oh, that's sticking. sticking hard so that would not have slid out easily. So this was the correct way to go violence.

Still not coming out that easily though. Uh, what if I stick a screwdriver up the side here? Or indeed just grab that and prye it off. It has bonded on. well, it really has bonded on well here.

We are right. Well, that's not potting anything anytime soon. In the recycling container, it goes right. What do we have here? We have the plastic film, which I'd normally expect to see over the circuit board but could also be over the components.

Let us, uh, try digging in here and it's one of these things. If this starts taking too long, I shall pause cuz sometimes the potting compound is super soft and this is fairly soft. It's a silicone type of material and sometimes it's not soft or it gets really WRA around the Co. So we're looking for this capacitor.

Anyway, that's one of the things I want to see that's the output capacitor, which is a prime suspect here here since they are often stressed quite significantly. Um, I wouldn't necessarily expect to see it do though they don't always do when they fail and it could also be other components. It could be the what they call the bootstrap past are just starting to drop a bit and uh, cause intermittent re-triggering It could even have been a bad connection cuz sometimes heat, uh, cycling can cause connections to fail. This capacitor here is probably the 240 volt input capacitor which is going to charge up and there's a little bootstrap capacitor there.

Nothing really obvious. This is the one that would hold a charge, but having said that because it was working when it failed, uh, it probably doesn't have a charge. But tell you what, let's just Bridge those connections anyway. shall we? nothing? If I got a good connection, here's the plastic film that is normally covering ing the circuit board, but it really has kind of crept right in everywhere.

This compound, which is what you want. It's supposed to make it water type for outdoor lights, better in a way than the one that I've just basically swapped in from a waterproofing perspective. I could zoom down this just a little bit, get closer I could have done that right at the beginning. Couldn't I yes, but I'm not very organized.

Get other bits off, see if we can find the diode. Is it a surface mount diode? Is it a through hole diode in the secondary side? I'm Reckoning that might be the diode here. maybe even more than one. Di here is oh, they've glued the stuff to the bottom of the circuit board and it's also, well, they' put a dot of that material in before kind of sticking it, just randomly playing myself.

Now, aren't I really? I Don't think this is going to find anything divisive decisive. Should I say because uh, the thing hadn't failed completely. It was just starting to show the earlier signs of glitching. Every so often, that glitching looked like the bootstrap circuit restarting.

Basically, when you turn one of these lights on, there's a slight delay before it lights as it charges a capacitor. and uh, that capacitor Then and Powers the power supply. Just long enough for the power supply to actually start powering itself. Um, what's the separation going to be like? Don't really know what the separation is going to be like in the Transformer Let's try.

let's try the spudger as a way to carefully slice this stuff off. Sometimes it's better to pick it off though because if you slice it sometimes uh, that just leaves a thin layer on that can hide details and become come hard to go off. Hopefully there is electrical separation from the input to the out. It would be nice if there was at the moment.

I'm not seeing a clear path of electrical separation. Oh there it is. Over the we slot through the circuit board as well. So they've made some kind of effort on electrical separation and I'm guessing that these down here are the diodes.

There two diodes. are they both. Associated There's a class Y I would guess. Very very small class y uh, interference suppression component.

but then there's that strange component actually that's in the primary side that may be possibly part of the bootstrap circuitry. Who knows. or it might be part of the filtering circuitry to protect the uh, the switching transistor in here. he said, point at the Transformer not the switching transistor I Tell you what.

Rather than Bore You by taking too long to do this, I shall pause a moment, turn and I'll see how much of this stuff I can get off one moment, please. Depotting is complete. What a mess that has made. it has been reverse engineered, all fairly logical.

Some suspect capacitor readings. Let's take a look at the circuitry so I'll zoom in a little bit. Sorry, there is some potting compound left. It's just pretty hard to remove it.

all. the incoming Supply goes to a component which just has one red band there. I Don't know if there were others at some point, but one red band might be it. It looks as though it is a 2 ohm fusible resistor.

Perhaps a dedicated resistor for that? Then we've got a bridge rectifier formed from four discrete diodes and just off shot here as a capacitor folded over the top of those. we have a PL 3537 Ch I Could only find a very vague schematic for that. however. Um, there is a partner trip to at the Pl 3535 with a lot more data and it's got the same schematic so it's from the same family.

There is a little wire link here I Don't know what that's for. It's very hard to read what that says MPT Something Is it? Perhaps a thermal fuse link or just a fusible link? I Don't think it would just be a fusible link because if uh, something failed in this side, it would potentially blow the diodes and this is after the diodes between them and the main reservoir. Capter: On the control chip side, it has two windings in the Transformer It's got the Uh primary winding and the feedback winding. and it's got the secondary winding that little capacitor there I Thought that was a Uh class Y.

Well, it is a class Y type capacitor for interference pression. it's not H They've just used it because it's a handy value that is is commonly used as a class Wi suppress capacitor and it's used in conjunction with this Di and a coup of resistance. On the other side, it's the snubber network across the primary. Um I'll just Mark that one as feedback and that is secretary, not necessarily their actual pin positions.

The feedback, although it's fairly well placed for this, is going through a diode, but it's also going through a resistive divider. I'll show you that the other side and uh, that charges this capacitor here. which uh Powers the chip, but it does also have another means to charge. it's a bootstrap capacitor.

it is suspect. There is also a little tiny I think it's 22 nanard capacitor for the chip for uh, internal references. uh, on the output, we've got a diode going to the capacitor and then there's a slight a resistor to provide a basic slow discharge just in case the LEDs go open circuit just so it can actually still, um, remain active but not actually pumping current through the LEDs On the other side, we have a chain of resistance here. uh, 750k 3 in series trickle charging the bootstrap capacitor which is down here that can also get charged by that diode.

There is a set of Uh resistors as a divider for from the feedback winding to actually uh and actually I think that is right. that is uh, correct. Yeah, cuz that's the positive and that's the switching through the okay chip, but these provide a divided voltage to an input pin in the chip. Just to actually indicate when things are going horribly wrong.

there is a current sensing Um array of resistors two 1.5 ohm resistors in parallel for limiting the amount of energy put into the coil on each cycle, and uh, there are the two resistors for discharging the snubber network 200k. And then there is that slight loow 22k. It's a fairly high value compared to normal just because it is quite high voltage output about 36 volts. Uh, what else is there? Not really much.

Let's take a look at their schematic which I filled the values in on. So if I bring this in here is that way too bright? It's not too bright. It's fine. So the main comes in here AC In there's that mysterious red banded just two component that looks like a fusible resistor and measures just over two ohms, which is coincidental, but are probably not coincidental.

There's a discrete Bridge W Farm There's that mysterious link which might be a fuse, a thermal fuse designed to actually trip out of the whole thing overheats, but they've not put that in. There is the main smoothing capacitor 22 megard, 400 volt. It measured 22 megard with a series resistance of 98 ohms. Oh, I should zoom out a little bit for that because uh I've got some numbers off shot here.

That's okay though, because this capacitor uh is a low frequency capacitor. It doesn't have to be low ESR So that 998 Ohms is Absol Absolutely fine. It's not uh, not a suspect in this case and it's measuring fairly sensibly for that. All it has to deal with is 100 to 120 Herz There's a chain of resistors that charge up that little capacitor 4.7 Mard 50 volt.

It measured 4.5 microfi at 1.7 ohm, which seems almost a be bit highish for a Louis Sr capacit which I think that should be H but it may. They may not have used a Louis Sr capacitor just to save money and to be honest I feel that should be a L Sr capacitor. There's a little 22 nanard capacitor used for internal voltage references. There is two 1.5 ohm resistors used for current sense and the current flows through the primary, winding through the switching device in here, and then through those resistors and when it detects it rises up to a level that uh, it reaches a voltage threshold.

It will then turn the mosfet in here off I Think it's a 600 volt mosfet it's got in there H When it turns off, there's a snubber network. A little diode diverts from the primary, the collapsing field. It just catches a tiniest little start before the Uh secondary ciruit can kick in and just shunts it into this tiny capacitor. the 2.2 Nano Blue Y class Y capacitor and it's got the two 200k resistors for a total of 100K across it just to keep that discharged so it's ready to take the next.

Spike It's covered across to the 220 megar 50 volt capacit that was measuring 24 microfarad uh 0.11 Ohms. It's not a bad resistance for our Lo Sr cap I don't think? not sure Uh, but I did measure a smaller capacitor Louis Sr with an equivalent Uh rating and it came out roughly the same, but having said the other one was much smaller so that may actually be quite a high resistance for that. Certainly the Panasonic equivalent said it should be something like Uh 03 or something like that. so it's possible I think that this capacitor here is the primary suspect for what's been going wrong here.

Uh, then there's that 22k discharge resistor and then the actual LED itself. and uh, the 22k resistor is if the voltage. if the uh LEDs go open circuit, it still poses a slight load because um, that just keeps things from climbing too high. It keeps things a bit stable, caps the voltage a little bit.

I guess um, not totally need needed because uh, the if that was, um, not sure they always put them in I've never quite worked out what that is there for I think it is to provide that slight load to stop the voltage creeping. High On That capacitor when it's open circuit, but I don't know if it poses also a slight load in the secondary to keep the primary running because there already is a slight load from uh, the feedback circuit. although it's not going to be huge because it is just feedback. um, but that is it.

um I'll guess capacitor degradation was the main issue here. Why the light failed? Um, if it was a one-off power supply that you couldn't get replacements for. what I'd do is I would uh carefully Depot it as I have fairly carefully. one of these leads is given often the track here, but that's all right.

it could be so direct onto the lead to repair it I'd change that capacitor and the bootstrap capacitor and while I was at I might just change the power spine capacitor because that should theoretically give it a new lease of life. But there we go. The light is fixed anyway. We'll see if it dips at any other time.

If it does, that would be extremely disappointing, but in the meantime that is it. It is fixed and we're ready to be in making videos again without those flashy interruptions.