Hi. When we last left our intrepid adventurer, he was, uh, caught up with this Ibm compatible, the world's first Ibm compatible compact portable machine from the early 80s and managed to get the motherboard up and working and determined that the power supply was a shot. We weren't getting any negative rails whatsoever, which actually stopped the main processor from booting up because there is a logic output from this, uh, comparator circuit around here that obviously, uh, checks the rails on all the voltages. There's plus 12, there's five, there's minus 12 and minus five at least I think and they're all obviously sum together and gives a signal.

Good output. and it was given a signal bad output and that was going to the Cpu holding in reset, stopping the machine from starting him up. But once we powered the motherboard up with an external Uh power supply and bodged in that power good signal. the motherboard came to life and were able to get video out of it.

it still worked. So let's get back to this: A power supply here. This pesky little power supply reason we didn't look at it last time is because, well, look at it. It.

it. look. it's like you can't even get in there to read or measure half of the parts. It's all higgledy piggledy.

It's all over the shop. It's just a really horrible layout to work on. and of course, Murphy, uh, ensures that any trace that you try to, uh, you know, trace on here. It's going to go through to the top layer where you can't damn well see anything.

So yeah, real pain in the butt. Anyway, we're going to solve our minus 5 volt and minus 12 volt issue on here, so let's take a closer look. By the way, we're going to be using the new Eevblog Bm786 multimeter available exclusively on the Eev blog available on the store as of today. I just got stocking.

so uh, yeah. Anyway, check that out. Um, so what we've got here is, well, obviously a switch in power supply because it tells you it's made in the United States of America Smps Switch Mode power supply. This is obviously the primary side here.

We've got 240 volt mains or one of the 110 Yankee rubbish. Uh, coming in here, and uh, then we've got oh yeah. a couple of little bridge rectifier. Two bridge rectifiers.

Anyway, we've got bridge rectifier down there. Then it just rectifies the input mains. We've got our high voltage caps here, and then, um, this contraption which looks like a giant inductor. It's not.

This is actually your switch mode. Uh, transformer. You can probably see deep down in there there's some wires running off there. so this is like it's really ugly.

They're just like hand wired over to points down in there, there, there, and all over the shop. Um. Anyway, so yeah, essentially our primary side here. And uh, we've got a made in Britain Made in Great Britain.

Thank you very much. A 2n65 45 for those playing along at home Anyway, um, obviously primary side and obviously the rest of this is secondary side. With this is our isolation transformer and the taps for this thing are right down in there. Some like secondary Uh taps.

So these are all of our secondary Uh regulation capacitors. Now, we're basically just looking at the Uh topology. Basically because we've already gone over this ball. We've done the smell test, we've done the visual test to look for you know, a blown stuff or dry joints.

you know things like that. and we did actually detect a few dodgy joints on the Lm338k. But that's actually working because this is actually this is an adjustable voltage regulator. Here's the data sheet.

So this is going to be for either the plus 5 volt or the plus 12 volt rail. So obviously right off the bat, and because, look at the size of the heatsink on this thing. um, it's well, it's basically the largest one on the board here. So that's telling us that.

Well, we've got a linear voltage regulator. So even though this is switch mode, uh, power supply, we're looking at possibly, um, not any active switch mode regulation on the secondary side. Over here, we've got a linear regulator. So what that points to is just a simple Uh bridge rectifier.

Or you know, a diode halfway from the secondary side and then just going into the caps. And then we've just got linear regulators. Now there's another device down in here. You know you can't get in here to read any of this stuff.

It's ridiculous. Anyway, now these two here on small heatsinks. These are diodes. Not sure if they're primary or secondary, but they've got Cr on them and I can see that's an actual diode in there.

Trust me about that. Here's our mains coming in over here. We've got our bridge rectifiers going to our main filter caps here and then obviously. yeah, this is.

this is all the connections for the Uh primary side of the transformer there. and then we're going to have some opto coupler feedback somewhere. There they are optocouplers. Um, they look weird, but uh, yep.

Old school package opto couplers there. So I've got some feedback, but I don't necessarily think it's doing much in terms of secondary regulation. As I said, we've got that linear regulator there. So anyway, uh.

primary side and then secondary side here. We've got some extra stuff in here. So yep, this diode over here is on the secondary side and the other diode on the heatsink here, also on the secondary side. So what we're looking for probably is like a classic culprit in these is our blow-in rectifier diodes.

Um, yeah, I have to link in the uh, it was the Hp Oscilloscope. uh, repair where we had a really remarkable and, uh, really elusive uh, thermal fault inside. Oh no. spoiler alert? No, no, no follow along.

I'll link it in anyway. Interesting. Now my first suspect would be a uh, diode in there somewhere. So um, we'll just check.

uh, these two biggies and then, well, I can see a couple of diodes really deep down in there. I don't know how I'm going to get down and measure a couple of those. We have to flip it over and measure from the bottom. Man, cross our fingers.

We don't get the wrong pin. Um, they're measuring diodes in circuit. Not always the best thing. The one over here is here.

So 0.07 of volts? 0.072 Aha. Well, that could be low in circuit impedance? 0.03 No, okay, that may be uh, low in circuit impedance. We can go over and measure our Ohms, but I suspect so yeah. 67 one way.

Yep. so they're not short it, but that doesn't indicate that they're a dead ski. So um, what we're looking for of course is not a, uh, shorted die because we're measuring like nothing. I believe there was like nothing on those rails so it wasn't popping any caps.

It wasn't doing anything. so the diet was shorted. We might be in a spot of bother. Um, you know, could blow something up or you know it's going to cause issues.

So I would suspect that probably like more than likely like a diode's gone open somewhere. So a bit of light to get down in there. And what do we got? Aha, Seven, nine, one, two. There you go.

Negative: That's our negative 12 volt regulator. that's not much of a heat sink is it for your negative minus 12? So yeah, but probably not. Like for the minus 12 volts, it'd be like for Rs 232 or something. You wouldn't really need it for much, hence the tiny heatsink.

But there's another device on the other side in there, which I can. Ah, I can't. Yeah, so some people are wondering why we can't measure those diodes. Aha, are they shorter? Are they open? or whatever.

Well look at these big ass resistors down in here. There probably are bleed resistors directly across these caps here. These are very common. They discharge the caps very quickly if you turn the supply off so it's a regular half wave rectifier.

and then you've got a resistor in there. You've got the coupler ohms in the, uh, secondary side turn in here. Plus, you know, however you know they might be. You know, a couple of hundred ohms down there.

So effectively. Um, either way, you're measuring the diode. You've effectively got like, maybe you know, 100 or a couple hundred ohms in parallel with your diode. And yeah, that doesn't really let you measure the diode in circuit.

You'd probably have to take these out to measure them. but that would be a red herring because I know they're not associated with this and I can't really film down there. But I was able to, uh, get in with a magnifying loop. I get in there tongue at the right angle with the light at the right angle and that's a 79.

So bingo. There's our minus five and Minus 12 volt regulators on the one heatsink, so hopefully you can see the cramped area I've got to work with anyway. I can see two diodes down there. one is horizontal and the other is tucked under that white gunk between the white silicon between the capacitors there and that one's a vertical joby that looks bigger, so I'd say the bigger one is probably for the like.

the minus 12. Okay, so what? We've got three pins there. Three pins there and the center pin on both of those is going to be the input. Well, any Uh 79 series, uh, voltage rig, and the one on this side, then one.

Sorry, this is the 12 volt. This is the five volts. so this one ground over here should be ground. and certainly it is.

And then of course the trace goes around there to match this one that's ground. Okay, so I can't remember which pin is our negative 5 volts, but we can find that there it is. There's our negative 5 volts and our negative 12 volts will be there. You go.

There's our negative 12. 12 because the 5 volts is almost certainly fed from the 12 volts, so that would be my guess. Any, but we can check that. So this is our 5 volt input here.

is that coming from our 12 volt output there? And yes it is and no workers. So that explains why both rails are dead because the if the when we don't get our minus 12 volts out, then we're not going to get our minus 5 volts out. so both of those are cactus. So either we're not getting voltage to the 12 volt airport or the 12 or the 7912 is dead.

It could be either of those things. So our Dave Cad reverse engineering drawing is going to look something like this: We've got our transformer, we've got our secondary uh tap. I think it's probably only going to be a uh, like half wave rectifier diode in there. And yes, you'll notice that I drew my capacitors backwards because I'm in an absolute twit.

Anyway, yeah, positive on the on the ground side there, because it's negative. So we're getting negative. Whatever voltage out of here, you know, a couple of volts above? uh, 12 at least. Um, and then you've got to take into account ripple and all the rest of it.

So anyway, it's probably significantly higher. so I've got a one, two, and then that just powers the 7905 and then we'll have some uh output uh, capacitance as well, also drawn in backwards. but I got that one right. Got that one right? And what you'd do as well is you'd go in there and you'd have a squeeze around the magnifying loop to which I use my macro lens for my camera Works great times.

10 macro lens and just look for any cracked joints or anything like that because that could be the problem may not actually be a component thing and with this power? good here. With this Lmr 339 Quad comparator, what's going on here is a rough sketch. Um, we've got all the power supplies coming in. There might be more than this will be.

Plus five plus twelve minus twelve minus five. Should have drawn an extra one there. but they all basically I go into uh, comparators here. Uh, there'll be a voltage reference on these pins.

So it's basically you know, are they uh, like below a certain spec say five volts? You might put you know four point seven five. is it above that? your typical five percent? Then it gives. Okay, I'm good. and how they're doing this diet arrangement here.

You could do it like as an and array or an or array. It depends how you actually configure it and how you configure the positive negative inputs here and how you actually, uh, compare them. But anyway, basically the idea is that yeah, this will give a output happy if all of the inputs here are above their thresholds. That's it.

Okay, so let's follow the input pin of that regulator and the there's one diode here, but it's got a little signal trace going out of it. Not sure if you should be able to see that. Um, so I'm not sure. It's that and the other big one here.

So let's try this. So it should be a note and nobody dope. And nobody dope. Oh, that one's there.

We go. We've got a charge discharge thing and so therefore, yes, it must be. And of course it's the last one we check. So yeah, so that large diode, that's the vertical one in there that looks like a big, uh, one watt? Uh Jobby, let's go to diode mode.

That's all right. and the other way around is open. Okay, I'm just going to, uh, power this sucker up here. I managed to when I disconnected the fan.

I'll manage to, uh, get it out of the chassis. Just no touchy. All of this section here on the primary side. Here we go.

It shouldn't need a load, it should be okay. But yeah, there's our five Volt rail 5.1 No worries. and that drains pretty quick. with no load.

they must have a output drain resistor on it and we'll just verify. Plus 12. Yep, no worries. and that one drains.

Yep, quick too. Okay, I've got a clip in there go into, don't our output of our transformer tap there so let's have a look so we should. Um, well, it's it'll. give us ac won't it? Nothing.

Wow, really nothing out of the transformer and Vfd mode on there. 0.3 out of the tap of the transformer. So I yeah, I've got circuit ground and that doesn't make sense at all. and I can't actually, um, show you it's too deep down in there.

But I can actually see that the uh, point of the diet I'm probing actually goes over to, um, the pin on the transformer. So oh, what the ah, you're not going to believe it, please. And the answers in the comments down below, I found it. I found the problem.

Hang on. I'll get that. Come out your bastard. Show yourself.

there's the ugly turd. There it is. Can you see that orange wire down in there? That is the transformer tap for the negative, Um, the 12 volts and negative 5 volt rail? it was. It's like it's come off So like, I don't know.

Is this just like because this is just flapping around in the breeze? I think this has just vibrated loose over the years. Um, you know, the fan. It's got a you know, 240 volt fan on there. I'm not sure how much this thing thing's going to vibrate, but that's what it was.

So tracing down all that, well, there could be something wrong with the circuitry. I don't know. But um, yeah. the bloody wire.

Unbelievable. Wow. Could have really chased a red herring down a rabbit hole there. Um, just for the sake of a wire, Because like you know, like you see, it looked like for all the world that it was connected down to that point.

But it damn well wasn't unbelievable. No wonder we'll get naf all out of it. There it is there, can't quite see it. Is that a single strand? Yeah, that's a single strand jobby.

I think it's just broken off from the board. Unbelievable. There's not much length left on that. Gotta attempt to get in there and strip that.

This is a real dog. The hell kind of insulation is on there. Unbelievable. That was a real dog.

Get that out. Let me tell you. So yeah. I got to.

uh, get in there with the needle nose pliers and feed that back down. Yeah, it's not easy to get these damn connectors out either. Um, so I might take the whole chassis over or bring the soldering iron over to here. So here's the tap here.

these two pins. So this one goes to ground obviously and this one goes over to our diode over here like this. There'll be a filter caps in there and then or somewhere and then that'll go directly into over to there. Um, and that's the input to our negative 12 volt regulator.

So I've cleaned out that hole so I should be able to feed in the wire from the top. Um, and hold my tongue at the right angle and hopefully it'll get through and then hold it with your finger on the other side and then solder from this side because you don't want to be soldering from the top. This is just ridiculous. Seriously got no idea.

it's right back under the transformer. Sorry, I can't shut Oy, did I get it? I think I might have got it. Hang on. Stay in there, you bastard.

I think it's going to stay in there because it's a single core so it's stiff as oh yeah. yeah, Good enough for Australia. Yep, there we go. You can see something there, so let's solder that back.

It's got enough length on there. she'll be right. Flow that through and we're good to go, but always give it a tug afterwards just to make sure. Yep, Yep.

Tug test complete. I think we're good to go. All right, let's power this thing up. Uh, so this probably hasn't been powered up for quite some time.

Look, I don't know when it, uh, broke, or how under what circumstances. but obviously our negative 12 and negative five have, well, I think have never powered up since I've had it. So uh yeah, it could blow something who knows. Release the magic, Uh, smoke.

But anyway, let's give it a bowl. Minus five. Yes, winner winner chicken dinner. So if our minus five works, that means our minus twelve works as well.

Hello Minus twelve. Let's do that again. Minus five. Minus da twelve.

There we go. I think I think the I just moved my ground. uh probe. I don't think it was making loosey goosey in the contacts.

but there you go. Fixed! Winner winner chicken dinner. So I have absolutely no doubt that we'll be getting our uh power good signing out of here because but murphy. um I might say that well something in the power good circuitry's failed as well.

but now look, it's going to be doing going to be doing its business. There you go. That would have been my last guess. that was a wiretap on the secondary of the switching transformer.

Um, you don't normally get this because usually they go in via a connector of course. um, but this one. No, it's hard wired into the board. But not only is it hard, biting in the board doesn't use any of that stranded rubbish uses single core.

and of course, uh, the thing with single core is that if it gets, uh, vibration and flex and everything else, um, then it can. you know it can break off fairly easily. That's why. like your high quality multimeter pros, for example, if we cut these apart, these might be like, you know, a couple of hundred strands of wire.

Like the real good silicone ones, with the like, the real super flexible ones. But basically, you know regular, uh, stranded wire might be. you know, seven strands of point, one millimeter or something like that. Uh, for example.

and of course you know it gives you a redundancy. it makes them reasonably flexible. But no, using the solid core wires on there causes to come a gutser. Wow, I'm actually glad it was that and just not.

You know, which is the most common fault? Might be a like an open, um, a diode or something like that, or even the regulator. Really. Although you know usually diodes would fail before the regulators would. But yeah, at least we've got something interesting out of that.

How long has this been gone? 20 minutes at least and we can actually measure our power. Good signal, which is pin two here and there you go. It's five volts. Yep, I will get in zero before, won't we? So the power Good.

So I have no doubt that this thing will now power up our motherboard over there. No workers, but I've got to assemble it all in the case for it to do this. so that's really annoying. So that's not something I'm going to do for today's video.

So yeah, I gotta have to like reassemble the whole thing and then you know, see if the Crt works and all that. but you've already seen that the uh motherboard works. We got video out of that and we knew, uh, we traced it down, that it actually came from uh, the failed uh power good on the power supply and that failed because our negative five and negative 12 were gone. But even if we had one of those rails, go anyway.

That was simple. But even simple ones can be interesting. so hope you learned something and found that useful. If you did, please give it a big thumbs up.

As always, discuss comment and down below catch you next time. But wait, hang on. I just remembered that somebody, uh, viewer, thank you very much. uh, sent me in the link to the schematic which I had.

I got this a while back. I forgot I had it so that would have helped. Anyway, let's go in and have a look. I believe it's the same one.

It's the compact Portable. This is the portable plus, but that's pretty much only. Uh, it has like a hard drive and some upgraded rom or something like that. So anyway, how are W.

Sams and Co. computer facts and um, we'll see why This is a Howard W. Samson Co. Uh is important in a minute printing in the United States? I'm no, I can't see that, but printing in the United States of America anyway? Um, so yeah, for those aficionados here, here's all the digitally stuff.

But we do have the power supply in here taken while pressing spacebar. There you go look. Nice, nice annotations on the schematic. Absolutely brilliant.

And anyway, let's go down to the power supply. Um, that's the part that we want and we do have it. Thank you very much. There you go.

Um, beautiful, all on one, mostly on one sheet. Although I don't like how it sort of like goes off here. It's like, you know, all these like jump off like down here it's like, uh, it's yeah. Anyway, um, and we do have an extra one over here, which just has.

Then these letters come over here and then these are the ones that actually go to the connectors over here. But anyway, as you can see it, mains in here. We've got some chocolaty chokes, we've got a full wave bridge rectifier, Then we've got our big mains filter caps. They've got some big bleed resistors across there, and basically, yeah, this is all just primary side switching here.

and then it looks like it does have a feedback coil here because that's feeding back without opto isolation into there. Yeah, and yeah, and they've got little, you know, if you really want to go in there and analyze it. They got, uh, scope shots, but of course we didn't need the schematic. We didn't need the scope.

it was fairly easy, but you know if we had to go in there and you know this diet here was. but you know something in here or this cap was bust or something, you know like you might have some issues, you might have to go in there and scope out waveforms and stuff. but luckily we just had a broken wire. so but yeah, this is handy.

And of course on the secondary r side. And yep, as I suspected, here's our 12 Volt Lmr 338k. I suspected that was the highest power one. Uh, and it was.

certainly. And yep, we've just got a single, uh, halfway rectifier there. Um, there's no filter caps for that one, so they must be on the other page. Yes, they are.

There you go. That's the 12 volt source. Um, and as you can see, yes, it's got a 680 Ohm resistor across here. This one up here has got two 33k 2 watt resistors.

There you go. So that's why the voltage was bleeding down and that's why I'll show you in a sec. In fact, let's go into this. So if we zoom into here.

Okay, so we've got our 12 volt uh, regulator here. No worries. But this is the uh diode that we were. These are the two diodes that we were measuring here.

And of course, look this one here. Look here. It is 68 Ohms. This is your loop right here.

So if you're trying to probe your diode here and here, regardless of which way you put your probes, you're essentially got 68 Ohms here in series. With whatever this coil is, it's going to be, you know, tens of ohms tops, right? So you got well under 100 ohms in parallel with the diode you're trying to measure. This is why you can come a gutser trying to read diodes in circuit like this. You might think, oh, it's measuring, you know, 50 ohms both ways or something and it must be shorter.

You know it must have like A or a highest impedance short out or something like that. Well, no, you've got to think about the rest of the circuit here and how there might be a bleed resistors like this in parallel. So yep, that's what they've got here. Oh sorry, I was getting carried away.

I think this is the 5 volt. C. Let's let's go to the next page. C C C Yes, 5 volts here? That's it.

Um, what? So there's no regulation for the five volts. Really, It looks like they've just got a big zener. If we look at that one one up, I think we'll find that's a 5.1 volt center. Yeah, couldn't get that readily.

But yeah, oh, it's 5.6 volts. Uh, 5 watt general purpose voltage reference regulator died? Well, that's a bit how you're doing, isn't it? I mean, they go to the effort to use a proper linear reg up here for the 12 volt roll. but for the 5 volt rail, they've just got a lousy Zenna in there. Oh, that's terrible.

Muriel. Anyway, here's our Minus 12 and Minus five. And yep, the minus five is connected just through to the minus 12 out there. And yep, uh, we've also got a 330 ohm 2 watt in parallel here.

So that's why you have a hard time measuring these diodes in circuit. So now if we go, check out our uh, power good circuit. So yep, there you go. There's our diode and gate there.

and yep, Lm339 voltage regulator and this will be. as I said, it'll be coming. So you've got plus 12, minus five, minus 12 uh et cetera. And then yeah, you've got some resistory dividers and then a voltage reference will become.

yep, it even says reference over here. Voltage Reference: Yep, there it is. 2.5 volt voltage reference Tl431 Classic for those playing along at home. And then we've got some Opto isolators and stuff like that.

So there you go. that is. That is the schematic. So that's exactly what uh, we deduced it to be.

Uh, during the troubleshooting that, you know it was just obvious this is what it was. that it wasn't actually that it had linear regulation on the secondary side and it wasn't doing any secondary side switch mode regulation. Now this is the interesting thing. I posted this on Twitter: a photo Fact: Standard notation schematic with Circuit trace Copyright: Howard W.

Sams and Co. 1987. Well, apparently this dates from like the 60s or something. We'll briefly look at this because it's kind of interesting.

So I posted this on Twitter and the tweeps of course came through. Are these: Scott: Uh, Def Pom: The Sam's Photo Fact: The service Manuals Published by Sam: They did a lot of manuals for old Cb and amateur radios in the 80s, but they did manuals for other types of equipment too. And Peter, thank you very much Barang you. I'm butchering that pronunciation.

um, unlocking the component Pcb mystery so I still don't quite understand it. but we've got an info thing, but it seems to be basically, uh, putting annotation and stuff on uh, schematics and things like that. And here's a I guess their uh, brochure Now photo fact helps you lick printed circuit board troubles in seconds. Exclusive Use: Sam's Circuit Trace features eliminate costly hunting for test points.

No more maze to trace, no need to flip-flop board. which means you know, turn up and up bottom to top, which is what we were doing trying to trace out where the traces are going. Here's how Circe Trace works for you. All test points are clearly shown on the schematic and is plainly coded.

The test points are similarly coded on the printer circuit board, so you instantly know are they, um, Well, I did like, is it just like they take photos and so I don't actually know what the product actually is? You know, write for your free photo Fact: I think like September 1958. That is great, right? So this is like the late 50s, Um, this circuitry stuff and like it was still still using it in the 80s apparently. Um, so yeah, it was still a thing. Please leave it in the comments down below if you've used uh, circuit trace or not.

So yeah, I still don't know. exactly like what they're actually selling here. Is it some sort of documentation camera then that links? How does it like? I don't? I have no idea. Anyway, I hope you found that interesting.

And once again, if you did, give it a big thumbs up. catch you next time you.