This light is the sort of light you'd expect to see in an all-terrain vehicle, but in reality this one comes from a very unusual location. This was given to me by a local fisherman called jiffy. It's from a fishing boat, a commercial fishing boat that operated in the irish sea will still operate that i receive, but this light is no longer operating the irish sea. Let me show you why.

If i turn it on half of it lights, the other half was lighting earlier on, but it's cut out. So i've tested it and seems to have voltage regulation. It accommodates from about 12 to 24 volts. The power rating of this is effectively 2 watts per led, which gives the whole fixture a rating about 80 watts, with each half section being 40 watts.

It's interesting to note that it has a couple of different types: leds. It's got the fairly highly focused center point leds here and then it's got a broader spread of leds at the edge. So, let's see if we can open this up and see what's inside and how it fared in the irish sea, because the irish sea is a ferocious place, not just full of lots of salty water, but lots of salty water thrashing around violently. So, let's uh i've already looked out the correct screwdriver a bit: let's unscrew the end cap and see if any of that salty water got in to the light fitting jiffy said that they swapped from using these to the broader led floodlights, and they were just much Better because these are very directional, they actually project a beam which you'd expect for a sort of like an all-terrain vehicle, but not so ideal for people working a fishing boat, uh pulling nets up over the side.

So one of the things i'd expect for this is water, ingress, salty water and grass causing corrosion. But having said that, the inside looks unusually immaculate i'm taking these screws out. I don't even know if that's going to give me decent access to the inside i'd, expect it to be an extrusion with uh the components sliding out that is not easily coming out. I think there's another screw in there.

There is another screw all right. Okay, tell me what i'm gon na have to take this out, there's another screw in there right. One moment please, oh, this is not helpful. The end has come off revealing a rubber seal then, but also lots of potting compound.

Now this is kind of big to actually fit in the bench here. So now, i'm going to have to try and remove the circuit boards, look as though they're probably screwed onto this aluminum fin, but unfortunately the glass. If it's been sealed in that's going to make it extremely difficult to access to this, but i shall give it a go one moment. Please progress report i have tried removing some of the sealant at the end.

It's just not happening. This glass has been slid in after the circuit board's been screwed onto the back of the heatsink here, so i'm going to keep trying, but this glass is proving very difficult to remove. I've already had to tidy up the shards of broken glass from trying to push this out. It is just not sliding.

I shall continue just give me a moment to see what i can do with this. It may involve glass, breakage, okay. Well, that's the glass out! It was somewhat harder than expected somehow they managed to glue the glass in the full length. I had to dremel the side, the metal off and then break the adhesion under the glass to get it out, but it's out now we can continue with the rest of the disassembly and resume the teardown.

I do not know how they got that black strip in there was that black strip in there to start with and the o-ring the white o-ring at the bottom, and then they slid the glass in with maybe some sort of thin sort of thing to let the Glass slide and then extracted a tape or something i'm not really sure, but that it made it extremely difficult once the glass is in it is in, but now let's take the reflectors out. Each reflector is screwed in by a central screw into the aluminium housing itself. So if i just lift this out trying not to well it's too late to not get my fingers on it, the reflector just is that single screw and it sort of allows it to sit on top of the leds. The leds are standard.

One watt three watt type, luxian style leds, i say luxian they're not actually locks in they're, a ripoff of the luxon. What the chinese call led, beads and some initial probing uh indicates how the wiring is done in this. So these lenses are the wide-angle ones and they're different they've got the same sort of housing here, but they've also got a little plastic double lens clipped in that uh hovers in front of the led and gives a different wider angle. I guess it's just to provide a full light.

The circuit board is divided into two sections which are effectively connected in parallel for power for the drivers there's a driver at one end if we go down to the other end, there's another driver down here. So, let's see, if i can take this out the midpoint of the circuit board, it's kind of a universal design. They've got a couple of pads. Maybe it's just for power, but they've got some bus bars going along the full length on the circuit board.

It says yf008 led 120 watts, suggesting i think it's 40 leds in this suggesting perhaps that they were running at about uh. Was it about 80? What i'm trying to remember now? They were another two watts. Each um just give me a second one. Two three, four.

Five: six: seven, eight nine ten, eleven twelve thirteen, forty fifty sixteen, seventy eighteen, nineteen, twenty! So there's twenty leds in each section: forty leds it would. They were running about 80 watts, but it has the facility to run these leds at the full three watts, which would probably shorten their life greatly as it is uh. A couple of leds on this have failed from proving about. Have i shown you it lit.

I don't think i've shown you it lit. I should show you that the bench power supply is on here. We can also do some diagnostics when i do that there are signs of corrosion. Oddly, these pads are super clean, but where the leds are soldered on, there is corrosion, maybe that's just the flux.

The inside is extraordinarily clean and dry, which is surprising, given that this uh isn't exactly a new fixture. It's been out on a boat bobbing about an irish sea for quite some time, so i shall take just off shot. I shall take the last screw out. It's kind of big thing.

Oh there's a couple of screws. I should take that one out as well, and then i shall try and get the circuit board out and reverse engineer it, and we shall see what we can find on it, or at least just find what the driver is. It will just be a standard current regulated driver. I guess oh, that's it lifting up nicely.

It's lifting up. Is there any more hidden screws doing it use too much force that is glued in there, where the cables come in they've put glue over the end of the cable? Here, that's presumably to try and uh prevent water ingress through the cables. They've actually put resin hard. Resin there this might be tricky right too.

Well, i'm just going to connect power to it. So you can see what happens. So it's not the full power. So it's not too dazzling.

It's uh. It is just nine volts and if i turn the voltage down, you can immediately spot part of the circuitry here see how these two leds are. The only two that are lit. They just happen to be next to two resistors in parallel and they've done that to allow it to they've, got the leds as c series circuits of threes, but these one they're just two to pack up the number and they've used resistors to balance that there is An led at the end there can you see it? Is it gon na oblige by flickering uh, there's one that is flickering on and off here, which uh there? It goes a bit of flickering and that is in sears of these other two, the other panel down here lights when it's in the mood but uh it spends most of its time, not lit.

I'm not sure if corrosion has occurred there, because it's affecting all the leds right tell you what i'll get that circuit board out and we shall explore further. Reverse engineering is done and it's fixed. So, let's explore and see what actually went wrong with the section that was out. The circuitry has three parallel schottky diodes for reverse polarity protection on the negative rail.

It has two smoothing capacitors, plus a little transient voltage suppressor in here and they're all in parallel uh. It has the chip itself tb99, which is a buck regulator. Chip using this mosfet to actually switch current through this inductor and the leds, and these are little two little parallel flyback diodes. It's got two current sense resistors.

In parallel, it's got a resistor to set the internal frequency of the chip and it's got its own little capacitor for a power supply and after that the leds are in series circuits of three: let's bring in the doodle. So here's the power supply 12 to 24 volts here are the three schottky diodes in parallel that are designed to protect against reverse polarity they're, the two smoothing capacitors at 47, microfarad, 35, volt and ck 56. Not quite sure, i've come across that number before didn't. I drew a blank on that.

I've found data sheet websites which just randomly point at random components that were nothing to do with it, but that's what usually happens, but i've drawn our transient voltage suppressor in there. Here's the chip tb99 notable for a few things. It has that resistor that sets its oscillator. It has a little capacitor and it sets no matter what you supply it with theoretically up to 500 volts, not sure about that, but it will generate its own little 7.5 volt supply for internal use.

It uses that to switch this mosfet, which is a csd50no6, and that allows current when it's turned on current flows through the leds, the large parallel array of led sections through this inductor and in the process of creating the magnetic field inductor that limits the current when It reaches a certain threshold, these two parallel resistors here - 0.069 ohms, when you add them together when it reaches that it turns off the magnetic fieldness collapses when it does so. The current goes through the freewheel diodes here, which are two shortcuts in parallel, and it goes through the leds. So it's a very efficient circuit. They claim about 90 efficiency.

There are six sets of three series leds to make up the panel plus the two that are, the odd number have two resistors and 22 ohm resistors, so a total of eleven ohms. So what went wrong with it? Well, this capacitor here went wrong with it. I checked the voltage across the circuit board. Let me bring the circuit board back in and i'll zoom back out, so the unit generates its own internal supply that chip for the uh to power itself, and it has that little capacitor.

I can show you on this better. It has that little capacitor there um, that is, that supply of 7.5 volts. If it goes below six point, symbols the thing locks out. It just means it's uh.

Under voltage, situational lockout, the section that was working was showing the correct voltage. This one was showing about five volts, so it was locked out that could have been for various reasons. It could have been that the chip had failed or the mosfet gate had failed, allowing current flow to the zero volt rail, basically acting like a resistor which can happen. So i looked at it with a thermal imaging camera to see if anything was giving off a little bit of warmth and the other main suspect here was the mlcc multi-layer ceramic capacitor, uh very notorious these days for failure, because they're cramming a lot of capacitance into A small area everything's over miniaturized and it was showing a little bit of warmth.

It was actually roughly the same temperatures the chip, but it was, they shouldn't show that there should be pretty much ambient temperature unless they're stressed so experimentally. I removed it and it was pretty cruddy underneath and i replaced it with just an ordinary louis s. Arc pastor. It doesn't need to be louis sr, but that's just what i had handy.

So i put in a 10 microfarad capacitor and now the whole panel is working again and the voltage measured across that. Well, let's bring the meter in and show you the voltage measured across that uh. Let's put the meter there, i shall measure the voltage across the other one's capacitor. So let's go into the common negative circuit negative here and this capacitor 7.46 volts.

That's fine uh and this one is now probe probe probe. Oh short, it out, uh 7.48 volts. So that's absolutely fine, that's a good voltage, so it was one of those pesky little multi-layer ceramic capacitors, going resistive again. That was basically dragging the circuit down so uh there we go interesting light.

It looks like it's been very waterproof. There is that sign of corrosion inside, but only on the led solar joints, maybe a metal reaction, and it is fairly lightweight. It's not that bad. I wonder if that's, though, what had affected the mlcc, the capacitor in here, but that's it have i have.

I got everything here. I think i've covered everything that little oscillator set resistor, that's more or less it yep. So um. That's the uh fishing vessel light given to me by jiffy from one of the local fishing boats.

They do prefer their flood lights because they're much better. It's like that's what this bench is lit with. It just provides a better wash of illumination, because this thing was definitely designed as a spotlight, so it had been quite hot spots of light, but there we go interesting well worth taking apart.