Hi, it's repair video time and well, this is my own bit of uh kit in the lab here used for shooting videos. These are my um Aputure Amaran hr672c um studio lights that I normally have on either side. uh you know things that I'm shooting on the bench so there'll be one that's sort of like sort of just lighting up like the front panel of a product or something like that. Just to you know like help add some extra fill light and the more light I have.

Of course when I'm shooting the videos the greater depth of field I can get. Uh because I shoot at a fixed aperture, I'm currently shooting this at F 5.2 fixed aperture so that gives like a decent depth of field so that like the front of the shot is in focus as well as like the back of the board because I shouldn't an angle so you don't want it to be all that defocused, arty fatty rubbish. Anyway, so the more light you have the more better. So I usually just start power them up from the mains input like this.

but they can also take um just standard Sony batteries and I have these um large Nwf 970 970s here which is 48 watt hour. uh Jobbies so they're quite large and it can take two of these. Um, if I actually disconnect it and I turn it on, it won't do anything. It actually needs a second battery in here like this and it works fine.

There you go running at 50 and then that can change the color temperature. You can see that changing there. and then of course if I plug it in that will, um, start. Well, hopefully no, it's not going to charge.

Why these are already full? I think like normally like, I have them just here and if I have to move them around the lab like if I'm doing a mailbag or something, um, it's useful to have the batteries working now. Um, I noticed that one of my batteries failed once one of the lights actually stopped working. It was this one here. One of them failed to work on the battery and then I measured, uh, one of the batteries.

and sure enough, if I measure that, it measures zero. So I thought, oh, you know, look, these are okay brand. These are newer ones, but you know they are like off-brand They're not, uh, genuine. Don't pay for the uh, general ones just for an application like this.

I thought, okay, well, yeah, a dodgy Lithium-ion battery, right? No worries happens. Right and then. So I got another one and it worked for a bit and then I sort of realized like the same thing's happening again. and sure enough, I got another battery that's failed.

The other batteries inside that were on the other side of this thing, I'm not sure which side it was. they're actually good, right? So I ended up with, um, two dead batteries from this thing. So I think what I've got here is a killer light. um, and that is actually killing one of the batteries plugged into which? Uh, channel? I do don't know.

at this stage I haven't like experimented to find out which is which and it's Look, it's trying to charge when there's nothing in there. So clearly there's something wrong with the charging circuitry inside this thing. Actually, the first thing I notice, check this out right? They're both off but plugged in 8.3 volts on there, right? And that's where you'd expect. But this one over here getting 0.9 volts and 0.9 volts.

So what the? um, if I switch it on? I mean the the light works Actually fine. Um, so that goes and switch it on. That goes up to 6.7 Oh no point. What that was? 6.7 Wasn't it? Yeah, Is it? Is it going to drop? Is it going to suddenly drop? Yep, look.

6.7 and then it drops to 0.9 Aha, that one's got zippity-doo-dah Yeah, so this is different to this one. This goes up to 6.7 and then drops. This one starts out with nothing and then drops 8.3 So yeah, that's what it's supposed to be. You'd expect to see the float voltage there like that, so it's ready to detect the battery when it's plugged in and then you know it's the charging chipset detects it.

So um, yeah, this one something wrong. There's the inside array for those playing along at home. Uh, and it's got two color uh leads in there and then it just uh, switches them in a uh pattern depending on the color that you're actually in the color balance that you actually want. All right.

So that front panel lifted off, whole bunch of self-tapping screws around there and we're in there. We go. So there's our array. They've got the uh, just the holes in the board for the cooling there.

Um yeah, there's no fans in this. It's all just, uh, passively cooling. Um, there's a little antenna down there. These aren't Wi-fi but they like do have like a wireless.

uh, like you can synchronize them. You know, like like on big um, you know, studio shoots and things like that. It looks like this charging board up here. It's upside down so all the electrons are going to fall out.

That's annoying. So there's the main board. Not easy. Unfortunately to, uh, get out.

Anyway, we have some uh, dodgy looking electrolytics on there. So yeah, you got to, uh, uh, suspect those. Anyway, there's our um, did they call it modem version 1.4 Um, these are these these? This a 2015 design? Um, then we've got our dual gang uh pots. Then we've got the various Leds for the Uh channels and the modes.

And and the Uh. seven Seg display. Looks like we've got some power supply stuff up there. Got our main micro? don't know what that is.

Um, I have to get my Uv filter out. Um, I'll find out what that is in a minute. But uh, yeah. here is your um drivers all down here.

So yeah that I hate black solder mask Pcbs and this is not going to go under the tagano very well. Uh, looks like we've got some current, um, sense shunts there. Uh, they're 0.1 ohms? Yep. Then we've got three, uh, presumably like, um, constant current, string drivers or whatnot.

Um, got a Dc to Dc converter there, so they all work fine. so there's nothing, uh, wrong with that. So what we want is the charging circuitry. Aha, there you go.

That's interesting. Here's our two battery terminals. That one's got a diode across it, a little diode that one doesn't what I just did then. Uh, I didn't know the camera rolling, but I just plugged a battery into.

uh, this side here. and the battery that came out of there is now measuring two volts. So yeah, um, so it's this one. It's this one here.

Let's go in and have a squeeze, shall we? Let's zoom, zoom zoom. So here's our two battery connections here and check out this uh, diode age just bodged on uh, pretty how you doing onto the terminals of the set. only the second battery. and of course that's not a pad.

so it wasn't designed to be like that. It was an afterthought. Um, in there so I'm not sure if like these are in series or not. I guess we'd have to measure that.

Anyway, if you're wondering what all the chippies are around there, we can use our light trick. Look at that magic. Um, we've got some uh, mosfets. By the looks of it, you can tell from the uh traces.

Unfortunately, the black solder mask is really difficult to see the traces in here. I just hate black solder masks in particular. gloss. Um, like this one.

It's just absolutely awful. But yeah, there. obviously. Uh, 4307, 4303, and 4831 by looking at the pads on those.

if you get the light at the right angle, you can see that all the pins are tied together like that. So you know that's these are mosfettis. So this one over here also look. Three pins tied together.

Four pins tied together with a gate. So we've got three mosfets here. a couple of more diodes there. um, they aren't the same as this one up here.

Then we've got another one over here as well. So, um, there's no bulges in the caps. But uh, let's have a look see if we can find some data on those. These are all cyanopower.

uh, brand. I think I vaguely recall the brand, but they're just one of these. Like non-mainstream um, Asian manufacturers, but incredibly common in stuff like this. You just won't.

but you might not find these on digikey. Anyway, so the 48.31 that's an N-channel enhancement mosfet and the 4307 that's a P-channel enhancement mosfet and interestingly, the 4303 is another P-channel enhancement mode mosfet. So why you'd have two, um, almost like 30 minus 30 volts, 11 amps, right? 30 volts, 17 amps. Okay, so that one's a bit higher.

Nine milliohms max idea. Okay, so this one's grantee, but why you wouldn't use this one? In all of the every situation in here you can see around here, this one has Esd protection on the gate here. Yeah, here we go. h esd protection 8 kilovolts whereas this one does not have the Esd protection.

So maybe um, there's a reason there. Perhaps otherwise like this would only be like one cent cheaper than you know and this one here. so I you know you wouldn't penny pinch that much. You just would have done bomb consolidation to use this one part everywhere.

So there must be some reason why they needed. If you need like gate protection on here and you wanted to save parts, didn't want to put external gate protection for example, then you could have used this. but then you couldn't found you couldn't use it somewhere else because of that gate protection, I don't know. Anyway, we don't know unless we've got a schematic.

and then we've got another one here which is a 1430 instead of the 1431 and uh, another. again again 43.07 So we zoom into this part over here you can see down the bottom. There also got two 4307 mosfets down there as well. and we've got the same uh diode here as well that we had bodged across.

but aha, there you go. That looks like it could be our Lithium-ion charger. Looks like it's a consonants brand. Um, I've got no idea.

Cn3702 Lithium-ion Battery Charge Controller for two cell lithium-ion battery in smaller package. This looks like it only works with a single 8.4 volt um battery that we've got here, so I don't know how they're putting them both in series. Anyway, there you go. There's a typical topology there.

we're going to external mosfetty there external diodes. We've got the big inductor, which we'll find it's at 220 there and then this current shunt resistor will be equivalent to that Will be this Rcs up here for the output current shunt to measure the battery. and that's it. So yeah, all this data here shows that it's um, it's not going to handle two batteries.

um, in series. It's only going to do one. So those who desperately need to see the microcontroller sync Moss. So yeah, there you go.

This thing is just full of um, weird ass Asian sauce star chips that we just don't see here in the west. Um, I don't know. Hands up. If you leave it in the comments, if you've used a sink, Moss Technologies micro okay, you can probably get them, unlike if you use like an St32 or something.

Okay, I've got the multimeter on screen. What I'm going to do is, uh, measure this diode here and um, see what we get. There you go. Diode is good.

Nothing wrong with that. I mean, you know I'm not worried about leakage or anything like that. That's not going to be a problem. So yeah, that's that's good to go.

Okay, so while we're here, I'll just measure a few other deities. Oh, 0.15 Okay, Ss, that could be shot key. This one measures the same. You'll know.

Yeah, there you go. 0.15 volts. So they're all good and you'll notice how that I can just make that number appear on that diode, but just shadowing a bit of light there. There you go and it pops up.

It's all about the shadows. You don't need anything, that filter rubbish. I just did a video on that one's good as well and that one, and that one. Uh, the good thing about diodes is like they're really easy to measure and you can generally measure them in circuit.

so why not? It takes you like seconds, right? So now you've got to decide, um, what your next plan of attack is. I mean, we've done our visuals and everything. It doesn't look like there to be the caps, but I'll get the Esc Esr meter on the caps and just make sure. But um, really, I like it.

Wouldn't um, explain the symptoms that we're seeing? It looks like we've got a couple of mosfets up the top there as well. Are they? Yep, There you go. 4307s again. But that's just.

um. switching of the input. So our charges here. Our switching trenny is associated with that.

We've got another one up here. Um, so this is all of our like. This is our main charge controller here, but this these would be for the switching between the battery and the load. because this thing works fine.

If you put in the external voltage, it works just fine. It just bypasses, uh, the batteries. Just the fact that the batteries don't power this thing. Uh, even if you put good already charged batteries on there, it doesn't power it and it won't charge them.

So if you put good batteries on there and it won't charge, that indicates that it's more likely maybe something to do with the switching. Perhaps then it is the actual charge controller. That would be my guess. Hmm, I don't know.

Low mosfet. Here's the schematic, right? This is the charger chip. and so I'm going to probe this inductor, right? So I'm going to see if there's any switching action happening on the inductor here, but that wouldn't explain why. like it's draining the battery.

So that's why I think it is one of the fits gone. But just as a matter of course I do want to check that. Yeah, so it's switching. so the charge.

so the lights actually switched off at the moment. it's in like off mode. But the charger chip still works because it charges the batteries when you switch the light off, that that charger chip is switching. So it's it's doing its business.

No workers, different things. Seriously, When I switch this off and I have no batteries connected like in this one, right? the micro seems to think that charging is happening because it the bar graph happens. but I think it's for I thought it was four Leds. Um, so I still don't know about the first Led.

So if the sensor resistor is okay, the charger chip must be the problem. I don't think so because the charger chip is switching. it's doing its switching right. And if the charger chip was at fault, it just wouldn't charge right.

You would think, right? So I think there's something fishy going on with. Like, because they've got lots of mosfet switching in here, right? There's lots. and it would be an interesting schematic I think. And it's not easy to reverse engineer because the black solder mask.

No, it's some sort of drain. I can probably get a current clamp and put like a battery on here. I think that they're switching the cells in parallel. But the interesting thing is even on the good one, if you only plug in one battery, it does not work.

It needs two. The charging chip is not capable of charging two batteries in series. I think the Cpu is being filled full full pier. Yep.

But the micros got to be powered from the same supply that the charges that the residual charge is being powered from. So there's mosfets after the charger chip. That's what's happening. And I think they're controlled by the micro.

So yes, I think Colin, Yes, they are using the charge. Led from the charger ic is an input to the Mcu that is that is likely and then doing smart stuff to detect battery voltage and displaying bar graph. Something like that's happening. I think because the charger chip does not have the ability to switch external mosfets for battery packs if the charging chip was dead or the charging chip wasn't enabled or some other thing wrong with the charging chip, why does that kill the battery? Here's the charger chip.

It's doing its regular charging thing. This matches the application Note: Okay, we've got the switching inductor. We've got the switching diodes here. Okay.

And then after this, we've got one, two, three, four, five mosfets around the batteries. Okay, and the charger chip has no ability to control those mosfets. I think it's this channel which is being killed. So it is.

It is discharging those batteries. It's doing something that right. There's some I don't know. shorted dead mosfet or something like that is my guess.

Yeah, my. My next step was going to be um, do the mosfets The problem with these? They got really high contact resistance right? But they're really sexy because they're um, pogo pins and they're really fine. Yeah, Mosfets? No. No batteries? No power, right? So I checked and measure the mosfets, make sure they're not shorted, and then I'd plug in some charged batteries and um, and start trying to figure out what's heating up for 500k.

Not shorted. Hello! Ding-a-ling Ding-ding-ding-ding One-ohm one-ohm one-arm Anyway, we might have found the culprit. I think that's a three hang up. Put a bit of spit on it.

That's a three. It was a little. There's a little thing on there which made it look like an eight 4303? Yeah, look at that. Can anyone come up with a reason why that should be one Ohm in circuit.

Everyone in the chat seems to think that this is failed. I asked everyone to press F in the chat if they think the mosfets failed and it's failed. By engineering consensus, that mosfet is kaput. It's reliable and rugged.

Here it is apparently not. so it looks like I'm going to have to try and find a replacement. I'll get out my uh Bm786 0.2 Ohms I know I was going to say is my flux pin did. Got it? Ta-da 1-0 0.33 Ohms got him? Yeah.

Failed Mosfet. What took it out? I don't know. Do the legs blink now? Yeah, yeah yeah. Let's uh.

power it up. Oh I saw some no it doesn't flash. look at that and hopefully we haven't taken anything else out there. You go.

it's it's not. It's not doing it anymore. in theory, right? and in practice, right? Taking out that mosfet could actually like take out other things. Who knows? There's a non-zero chance that mosfet in there shorted was actually actually protecting other stuff.

and then when you take it out, it's dead. Um, unlikely. But I don't know. I think there's got to be a it's got to be a scenario where that's possible.

Have I got any spare mosfets? Um, let's have a look. They're all D-packs I've got through-hole um. mosfets. I've got to 220 mosfets.

That one turns on, but that one does not. Yeah, I think we've got a dead lead. What are the odds of having a dead lead as well as a shorted mosfet? The Lcsc do have them in stock. 50 cents so I can get an exact replacement.

I really like the idea of Bodgins something I think. so let's see if we can find a replacement mosfet inside a dumpster laptop. This is an Acer jobby uh, Aspire 9420 and it has Intel Centrino duo technology and it's Windows Vista. ready.

Are the dream combination of Centrino and Vista beautiful? Now, the reason why I'm doing this is because laptops are usually a decent source of like little uh, Smd power converters. And right off the bat, I can see two there and I haven't even flipped it over. This will be a double-sided mount jobby. We're on the bottom of the board.

There you go. These are clear. you can clearly see pin out two mosfets, so I have to have a look at those part numbers. um, anywhere else.

Well, turns out none of those on the back, uh were any good. We had two end channels and another one which I thought was a mosfet wasn't It was actually a Usb power switch type thing. So you flip it over and here's where you start getting the good stuff. Look at this.

there you go. Whole bunch of power mosfets along here and possibly in here as well. and over here. Another core, uh, generator as well.

Yeah, that must be the um. Nvidia graphics, uh, co-processor so that would be the core for that. So anyway, there's probably a few others around the board as well. like up here, maybe? Yeah, there's a couple at the top there.

So yeah, heaps of power. So 8 power mosfets on here. I'll check them out. and well, you guessed it.

Um, they're all N channels all along here. and all these ones over here. these are all in channel everything's end channel. had to look around, but I finally found a P-channel jobby right up here.

and it's a 30 volt 12 amp jobby. So yeah, that should do the business. And I also found an additional two P channel mosfets up here which are different. again, uh, slightly different.

but these aren't um as higher current rated as uh, the one that we have got down here. so I think I'll use this one even though it doesn't quite match. uh, the cyanopower one that we've taken out. but it's probably good enough for Australia, it's worth trying.

Okay, there it is. The 4407. Bingo. We just wick up some of this, cleaned it up, and don't worry about those pins shorted out they're supposed to be.

That's just the solder mask coming off, and see, the solder mask has come off between those pads in there. So yeah, there we go. Okay, so I'm just going to measure the resistance across that mosfet just shorter before it's no longer shorted. So yep, no workers.

So let's uh, power this sucker up. I've got no batteries in it, so we should. let's see if we get now. No.

there you go. We don't get the leads doing their cycle thing anymore. Let's see if we get our voltage on the terminals over here. Tada 8.3 and 8.3 There you go.

Win a winner chicken dinner. Or at least well, it's at least back to what looks like normal functionality. Whether or not that muscle is good enough, a path isn't passing the current and all that sort of stuff, but we're only talking like we're not talking anywhere near like 12 amps. um, or something like that.

So like. I think these mosfets are like grossly overrated. uh, for the task. and there's nothing wrong with that.

It's very common. Um, so yeah. putting in that lower rated, uh, current mosfet. I don't think it's going to do a huge business.

Anyway, the actual switching functionality seems to be there. Okay, I'm going to put two batteries into it that are currently 8.1 and 8.4 Okay, they're a bit offsety, but no wuckers. I'm going to put those in there and see what happens. So here we go and it won't start charging until we put the second one in and it's not charging well.

There we go. Okay, so a light still works okay, and those batteries should have been good enough to make that work. So no, there's something else given at the old finger test. Nothing's getting warm.

No wait. I I put in the uh, bigger batteries. I took out these, um, smaller, uh batteries that I had, so maybe it was the mismatched voltages between them. It wouldn't start it up and now it starts up.

and I this is not uh, plugged in. So now it's now. it's working from the batteries. so that's working.

That's working fine and dandy. Let's plug that in. Okay, so that's got a full battery. I still don't know what that Led there's doing.

Um, I don't know what failure that is. Um, that is weird because my other one will show four leads. this one only shows three. So yeah, that's strange.

Anyway, what I'm going to do is I'm just going to leave this running for a while, discharge these batteries, and then because it won't charge at the moment, because the batteries are too full. So I'm going to let them discharge and then I'm going to plug it back in see if it charges and just to check the temperature again. Finger test: No, as mosfets run as cool as an ice cold stubby. No worries.

Okay, it's been running for ages but still hasn't gotten down there. But these are big ass batteries. so let me plug this up. There you go.

Charging charging. Yep, I think we're good. So yeah, I still need you know more experiments. but I've had this like running for like an hour now or something on full brightness and nothing gets warm here.

Not nothing at all so I won't even bother breaking out the thermal camera. No worries at all. The charger chip? I mean, we just started so you know, but no, I don't think it. um, I don't know how long it takes to charge this.

I'm not sure the charge rate. I think it's like it's pretty slow. Um, so yeah. I don't think there's any worries whatsoever with that mosfet in there.

So even though it's lower, uh, rated, it seems to be doing its function just fine. And in these sort of applications, you'll find that almost any mosfet would have done the job provided has the maximum voltage rating. That's one of the most important things, and the maximum current. But because it's like a a 12 amp rated uh, mosfet? Like, we're never going to get 12 amps.

So it's You know it isn't really a problem. The higher Rds on at, uh, the certain gate voltage. Whatever it's being used in here. Um, yeah, okay, it's might be a little bit higher, might display a little bit more heat, but can't even feel it.

So no worries. Now, the really interesting part about this is the switching topology of these mosfets that we saw here. You know? remember, we only had the one charger chip and it's only capable of charging one battery which contains two cells. Now, I've only got the one battery in there at the moment, and it's charging.

so it actually supports charging of only one battery. But if I switch that off, then it does not work from that battery even though it has more than enough charge to actually, uh, power these leads. So what they're doing here is really interesting. There's only the one switching controller here, and but it has no ability as I said, to control the mosfets.

Um, in here. So these must be either controlled via the micro which is powered uh, when it's off, by the way. So it must be using the same supply as the charger chip over here. So it must be charging them one at a time, because in theory, you could put them in parallel, like, and charge.

uh, you know, split your charge between uh, two, uh batteries. but then if you put them in parallel um, to charge, then you wouldn't be able to put them in series at the same time to power the product. And because it can actually charge uh, the batteries when the products being used and the product has to be used with them in series. Therefore, it must be alternating uh, between.

it must charge one and then charge the other. Um, and that's all under control of the micro. Neat. And if I probe both of those packs there either side, you can see that it switched them in series there to provide operation.

And the only voltages I can measure around here like on the big caps for example, I'm getting like 11.3 and over here 11.2 So yeah, it looks like it's uh, dropping those a bit to run the array, but obviously a single battery with a nominal eight volts or whatever. Um, is just not enough. uh, to run this. So they need to do all this fancy switching.

It's most likely reading like a charge, complete line from the charger or something. and it's also measuring the voltages on the individual Um cells as well. So yeah, like they're probably like, um, like switching the grounds in there as well. It's not a common ground, in fact, I can check that.

So here we go. I've got the batteries removed now so we can measure those without any voltage present. and there you go: 0.1 ohms. And the other one over here is not that is open so you can see that's uh, 12 meg.

So yeah, that second battery or what they call the first battery, the one that, uh, that channel that failed, Um, that's got mosfet, uh, switching in there. This one will also have uh, some mosfets associated with likely these two, um, that can disconnect or and or reconnect. Um, either of those cells and charge them independently. So why didn't they go with the charger chip that actually supported um, the multiple batteries, Um, in series like this? Meh, I don't know.

Um, they could have also designed the, uh, change the design topology so that it actually had a boost converter to work from one battery and then, um, yeah, you could like have separate batteries and you could switch them and stuff like this. So they've gone with basically the most complicated solution possible putting cells in series and or parallel for this operation. So yeah, they've gone to town. love to see the schematic on it, but interesting, huh? So if you enjoyed that, please give it a big thumbs up.

And as always, you can discuss it down below and check out the follow-up video to this which is, uh, the one looking at this particular mosfet and how to find a just searching for the data sheet for that and finding a suitable replacement. catch you next time you.