All right time to bring out the big guns. The Zen tricks have attained the end of this one. I'm not sure. Anyway, the Zen tracks XFR 300 - for 300 volt 4 amp that's 1200 watts for those playing along at home I set it to 55 volts because well, 55 sounds like a good number and we've got some quiescent current and it's working.

There you go, we have the magic smoke. The magic smoke is escaped. Holy crap Quick smoke alarm. Well that ended very badly.

The Magic Mike escaped I don't know what the hell went wrong and I was just feeding 55 volts into it and it whoop. Still smoking and he's still what's mine I don't know if you can see it and still smoke coming out. Damn yeah she's still wafting out there. You go, you can see it.

Love the smell of a bit power supply. in the morning it's still smoking, it's still coming out of the vent. This is hilarious. I was just sitting here, didn't even have a load.

actually hooked up on the thing and wah wah wah. I've actually got this sitting right down near the inlet of my arm carbon air purifier system here and I'm still having a hard time getting the smell out of the lab I'd been gone for I would just went away for half an hour. came back and it's still not great. Hi Well I was gonna do a video in fact I shot a lot of intro material for this Ardi tech what is it DPS 50 20 power supply module that you've uh seen in previous videos which are linking at the end of this video and also down below and the designer of these modules Glenn from Rd Tecna who sell C's on Aliexpress he's the designer and manufacturer of these and he saw my video I liked it and kindly sent me this very nice case for it which I mentioned in the previous video it's 20 to 24 dollars for the case including all of the fan and the switch and the binding posts and everything else and the wires and the whole kit and kaboodle this board up here which are fan, controller board and everything else and kindly this our Dps 50:20 power supply which is a 50 volt output 20 amp 1000 watt module in this tiny little thing but you seen in previous video that these are very efficient.

like over 95% efficient but still even at 1 kilowatt claimed output power then well that's like 50 watts in that tiny little heatsink and everything else. So I was going to do a video I built this I got time-lapse footage of building everything else. got to do a little review of this thing and I built it up and hooked it up to my Zen tricks power supply here and wah wah wah The magic smoke escaped as you saw all the drama at the start. so let's obviously I'm yeah it's not gonna work anymore and that's probably it for the review unless I can repair it.

so let's take a look at what went wrong. What I think happened I didn't have the camera running at the time unfortunately when it went bang but I didn't have a load on the thing I was feeding it with fifty five volts on the input. It has an input voltage range up to sixty volts so I wasn't even at the maximum and I just pressed the voltage button here and I started adjusting the voltage up and I just went all the way with LBJ right up and then all the sudden it just went poof and went. and then things started catching on fire.

Smoke billowed out of the thing. There was literally something on fire in here, so let's take a look at what went wrong now, just in case. uh, somebody are some people might think, well I don't have any rubber feet on this thing. metal on top of metal here.

Maybe there's some sort of earthing issue and that's what I shorted out. No, it's not That because the output of the Zen tricks our power supply is floating didn't have it's not our mains referenced at all at, so it's definitely not something to do with that. it's something into it. Must be something internal in the power supply because I like these Zandt Rex's are bulletproof.

These are the Ducks guts in the industry and I die and it still works I don't think there's anything wrong there at all. it's still still hunky-dory because you kind of kill these things. So yeah, and I briefly saw this voltage when I turn this up briefly saw and I'm bang and I look down here and this had actually dropped down to zero. So I think it had hit it's our 4 amp I can't I can't remember what I had the current limit set to at the time, but it goes up to our 4 amps Obviously just I've dumped all the power because this has a 1200 watt capability so this is a really high power supply.

It can deliver a lot of up see into there as something goes wrong. Alright so let's have a look here and see if we can see what's gone wrong. First thing you do, of course. weak.

Well use all your senses aren't We smelt it. We saw it, it was smoking and now we're gonna check. visually. Like you know, none of the caps have exploded like there's nothing around.

like the MOSFETs nothing seems to be blowing except Tada. look at that under the board obviously. Um, there's flames that is clearly flame type. You know to produce so much smoke it might like and it kept on producing the smoke.

It wasn't just puff of smoke, then gone like a component caught on fire under this thing. So let's take that off and see what's on the bottom. Just get all the connections off there. we'll flip it out and oh Gonski Wow Look at that.

What was there Other any components on the bottom? I'm gonna have to check the previous video where I think I briefly showed the bottom of this thing. My first guess would be there the two pads. that's the negative pad. that's the positive pad.

I had no load connected to this thing and all. It just went poof when it went up near towards full-scale 50 volts. So maybe there was a cap under there. that was that.

Would there be like a ceramic cap that caught fire? We've seen that in the Ness video that we saw a long time ago which might have to link in all there's the that's a trace under there. Yeah, let me have a look at the old footage, see if I can get a picture of what was actually there must have been something otherwise like what's causing the short. and sure enough yep, here's a screen capture from the previous video. Luckily, I did actually capture the components on the in that video and sure enough there are two capacitors under.

there. Looks like only one was fitted though. What does it see? 34 there? It's a quite large a ceramic cap and that has clearly caught on fire here and it's just like it's completely God like it is vaporized but it actually caught on fire just like that Ness one that we saw before and the problem with these are ceramic and problem with ceramic caps is when they fail, they fail. usually fail short.

So if you want the utmost in reliability from a design, you might put say two of them in series. Yes, you have the capacitance, but then if one of them breaks down for whatever met reason it's a manufacturing problems and over voltage, stretches, issue, whatever then it's covered by the other one in series with it. So it's not a problem. one of them short, one of them's just an open cap so you just lose half your capacitance is not a problem, but given that we only had the one capacitor across here either, it was a wrong rated part ie.

wasn't rated for 50 volts, you know, was it like a 25 volt happened? It went poof because this thing, it only failed when I turned that I don't know exactly what voltage you got to, but I think it was getting pretty close. It was going up to like 30 40 volts or something and maybe a like when it topped out. something's gone wrong. So either they've got the wrong stress component in there or or it's just a faulty cap.

It happens. You know they can. got it from the One Hung Low company at their stand at the Shenzhen market that morning and there's going to be like a manufacturing bell curve or these things, you're just going to get a certain number of defects. which as I said I fail short and that's what's happened here.

It just caught on fire and because I think the default I haven't hadn't I just powered it on factory-fresh module probably by default is set to like what was it, the 20 amp current limit or whatever and they just delivered all that power to the output because this thing's capable of delivering a kilowatt. Remember that. So if you got something shorted on the airport 20 years flown through the cab, it's gonna catch on fire. and that's what's happened here.

So the only question there is why it failed I'm so I could clean this up, have an attempt to clean it up, and power this puppy back on. Are there any like or any protection fuses anywhere around? you know? I don't know, nothing else. Looks are smoked so everything looks pretty intact. So I think that's all it did and it did.

The power supply did its job. It delivered that maximum like 1 kilowatt output that thousand watt output into that load. It was just dumping as much power as the capacitor was saying give me, give me, give me more power I want to catch on fire and the power supply said yeah, no worries. Here you go.

Therefore, all the power you want. What? What? What? wha? Magic smoke escaped. Well there you go. There's the aftermath of that Wow that poor cap that used to live in there is now dead.

It looks like. um, the cap was soldered directly across the two pins and that's got big thermal syncs with the pins for this thing and all the copper around there. so they would have had to dump a lot of heat into that cap. Poor old cap to actually get that on the output.

So that's a bad design decision because you want those caps to be, you know, refloat. Well, in this case, it's the only component on the bottom is it? So I Probably hand soldered, but you don't want to put it directly on those large mass components like that. It's almost as if it's an afterthought. like they just didn't design it.

and oops, you know we need this extra bigger cap on there. That's probably why the other one wasn't fitted. would be my guess. But yeah, so it could have done some damage to the cap which then cause premature failure even if it was aspect correctly.

Wow, that really is quite amazing. You can see the charring of the fiberglass in there. Ah, Wow I was just trying to I sold of that. And look, what happened to the pin.

The pin just like sheared off in half. It was weakened by the heat of the fire. Wow So what value capacitor was actually here? Well, I'm Glenn will have to tell us that I'm sure he'll respond to tell us what the issue is here. but if say for example, you go to Digi-key and you search for ceramic capacitors.

multi-layer ceramic capacitors above 50 volts because you don't want to use a 50 volt, this is zero to 50 volt output, so you wouldn't choose a 50 volt rated part for this particular case. So would it have to be 63 volts, 80 volts, hundred volts or you know one of those are preferred values Voltage values over 50 volts. If you have a look at Digi-key it's basically got to be something under 10 microfarads, probably 1 micro farad or something like that because you because you the higher you get in capacitance in multi-layer ceramic capacitors at that particular high voltage which is quite an unusually high voltage for a ceramic capacitor. They're either like especially manufactured, there are special physical size or they're like in like a a little lead frame stacked array which is our rather interesting and I'll show you that here so it's probably like in the case size we saw like you know a 1206 type size package or you know something a bit larger than that, then well you know it's probably a micro farad all right.

So I've cleaned it up I haven't added a cap on the bottom because I don't have any readily available high voltage caps. I could put 2 in series but me, whatever we burn through a sense line coming back you can see that the two sense lines they're coming back. So since in the voltage directly on the output, I've just added a two hundred volt 22 mic cap. Much larger value then what was there before, but it doesn't matter that a lot.

just get us up and running. I Mean probably work fine without a cap, but we'll just add something there. So let's pair it on this time our external supply just to limit the potential damage that can happen here. So let's give it a ball.

All right. So I got a 40 volt input 1/2 ampere current limit. So let's switch that on and it's booty. It's booting.

It still works. Look at that. It's alive. Beautiful.

Let's actually see if it outputs a voltage. Oh sorry. the output has to be on. but no.

look at that. it reset. No, no, that's one sick puppy. It's set to 10 volts at 20 ampere current limit.

Let's go I set. So let's turn the current limit right down there and let's turn the output on. So we've got 10 volts at. you know, 1 amp on the output.

No. no load at all and well. seven nine, seven point, two volts not. She's gone ski and it's reading a nap At eight watts, there's not nothing on the output.

not. hang on I think I can smell something again. Hmm St. good.

Yeah. I mean like you know it's drawing One point you like? One point knowing what's quiescent with nothing on. and if we switch that out, put that out, put on again. Well yep, now it was drawing like 10.

plus what's up there? No, it's one one sick puppy. Yeah, it's actually drawing 17 watts. Yes, is one sick puppy. I don't think I'm going to try in trouble.

Shoot that without a schematic I think she's well, we could you know I could try and trace it out, reverse-engineer it, but I'm not going to do that right now. It's like 10:30 at night and I need to video and get it up and go home so missus Eevblog doesn't get too upset that I work too much. Hmm yeah, for those you want to see what's under the heatsink got ourselves a sill pad their trademark um, oh, there we go. Hey, is that there you go? Um, let me get those under the microscope and see if I get a part number.

We've got ourselves 4 N channel mosfet s-- these are our Alpha Omega I love that company name. Terrific. These are D 2810 That one's upside down so the electrons are fall now. So that was what our problem was.

clearly. um, but there are. They look like they're certainly not from the same same batch. are they there? They're quite different, but yet we've got four of those.

They're 80 volt our 40 amp and channel MOSFETs So fairly fairly grunty little beasts. and also under here that's obviously not a switching converter. You can switch in control I cuz I it's got a designator Q which is a transistor and it's got your three pins tied together here, all four pins tied together over there and that should gate down in there your classic MOSFET configuration once again. Alpha Omega AOR 42 64 and this is a 60 volt ah, 12 amp job in series with mind you, a poly fuse.

look at that probably put the kettle on. So what that one's actually doing I don't really know because it's not the switching controller main switching controllers the XL last 705 I believe we saw that last time on the modules. We've just got some input transistors over here and so that's coupled with the To5 9 for the classic, our PWM controller in there, and then we've just got the quarter of amp up there and all together, it's actually quite a complex beast. I'd Love to see the full schematic of it reverse engineering.

This is quite possible, but it's quite a task so I'll leave that to someone else to do. I'm not sure if Oakland wants to share the schematic with us, but yeah, like there seems to be no other damage though. but what? the physical damage but obviously electrically something else is. Gonski So there you have it.

that was supposed to be a review video on the DPS are 50 20 and the nice little lower case it comes in and everything else. but in the end it was a complete balls up. And it's just like bad luck that I got a unit that possibly had a damage cap I Don't know once. I As I said before, I don't know if it's like under SPECT or something like that, whether or not it was solder and damage, whether or not it was just unlucky.

that is, that ceramic cap failed. You know, infant part component infant mortality is a thing and they often fail short and catch a lot like that. Not particularly common, but it does actually happen sure enough, but it happened to me. Sometimes you get lucky I'm off to buy a Lotto ticket.

Catch you next time.