Hi. In a previous video, we took a look at this uh, fluke Pm 3370b that I scored on ebay for 99 bucks and it was almost a winner. Winner chicken dinner. Um, except that after about five minutes, it would just, um, that power down and the screen to just go bloop and actually collapse.

So obviously something's happening with the uh, like the high voltage, eht aspect of it. or at least you know that's one of the things that's involved here. So let's have a troubleshoot of this thing and see if we can find the culprit, shall we? So obviously it's going to be power supply related, so we're having a look at the power supply board here, which is, you know, a fairly complicated little beast. We've got our mains input here uh, there's no reefer, uh, caps.

On the input, we've got a full wave uh, Ptc, uh, we've got a full wave bridge rectifier. We've got the the high mains, high voltage caps, common mode uh, choke and then primary side uh, switch in here and our main switching transformer and then a big secondary side and look at all those Nippon Chemicon caps are beautiful. Now the first thing you might suspect, of course, is, uh, the caps. You know, electrolytic caps famous, uh for failing, especially on a 20 year old scope, but unfortunately that doesn't really fit the symptoms.

like it works. Absolutely perfect for five minutes and then switches off and then restarts immediately when you, uh, power it back on and then five minutes. So you know, like when you have something that fails after five minutes. like that's sort of like reasonably consistent.

You're looking at more of like, you know, the thermal side of things I've I've done a video before repairing an oscilloscope where spoiler alert Um, like one of the diodes actually had an internal thermal failure. After a set amount of time, it was actually dying internal to the diode like the junction in there, like the bond wires or whatever inside was like, you know, failing thermally when it heated up to analyze this. Um, yeah, you might go around with your freezer spray and stuff like that and start playing around with stuff. But of course the first thing you want to do is a visual inspection.

and I won't bore you with the details, but I cannot find anything visual on here in terms of you know, leaking caps or anything like that, any components that are heated up, diodes that have gone brown, resistors that have overheated or anything like that. There's no blow holes in any of the parts, so I cannot find anything visual on this thing at all. And that includes the bottom side surface mount uh parts as well. And you can see like the solder, uh, thieving on those larger pads.

This, uh means that this thing has been uh, reflow soldered. um which is you know, fairly common for these uh, prime, secondary, side, um, surface mount parts. And I've also gone around and looked at uh, the solder joints for example, like a classic thermal problem might be a dry joint and once the component heats up, then that actually causes you know the component expands slightly and uh, the joint actually breaks and I've gone around and you know, like the through-hole ones will be. Uh, the classic, Uh, for this, You know, large components, large like, uh, you know, power components and stuff like that.

But I've gone around the board, I can't see any dry joints on this thing at all, so I'm going to rule that out for this stage anyway. So step two would be measuring the diodes and I've gone and done that so I'll boil spare the boring details. But yeah, I could not find any failed uh diodes at all. But as I said, this is more like a thermal thing after five minutes kind of, uh, problem.

That's what it sort of indicates. So just measuring the diodes on the border like this, I didn't expect to find anything and I didn't. The only component that that seemed to be open was this one here and I suspect what it might be, but we might have to verify that with the Uh schematics. I think that's a high voltage diode.

Some commenters in the previous video mentioned on the Eht section and this is the tripler here. Obviously this is our Eht driver transformer here which will generate the anode voltage and then that goes into the tripler here and that will drive however many kilovolts are required for the Um see and that goes up to. that's the nasty, don't touchy, um in the Crt there it is so but it's all discharged now. Everything's hunky-dory but um, some people.

A couple of people in the comments said um, yeah, these Phillips scopes They are actually famous for having uh, the tripler like arc over and fail and then print some protection circuitry kicks in or something like that. But anyway, let's jump over the schematic, see what we can see, just get a bit more uh, background info before we start. uh, prodding and poking this thing. Unfortunately, there are no test points so once this thing gets plugged in um, you know it's not easy to like access stuff in there so it would have been nice if they had like various you know, uh, test points along the top that you could, uh, just probe.

So yeah, that could be annoying if we get desperate attached probes that the wires that then come out and then attach to probes later. Let's hope we don't have to resort to that though. And some commenters also talked about upgrading the memory on this thing. And here's this.

uh, sample memory. There's eight of these four on the other side and these are 2k high speed s rams and you'll notice that there's an extra two pins on each one of those, an extra footprint and you might even notice up here. Tada Jumper Links 32k 2k. These are 2k chips and these are 8k chips over here to give you like 32k of sample memory.

In theory, I think it's probably possible to simply move these jumpers over here and then populate these with the larger uh part. And I have checked the manual and it does actually have the part number for the 8k parts instead of the 32k parts. So in theory that's possible. but uh, you know, I just maybe.

But anyway, that's not for this video. We have the service manual. Here are dates from 2000. It's very comprehensive, only 478 pages so hats off.

Unfortunately, it doesn't give a page number. it gives a section number. First section that's of interest to us is like the main system block diagram here. So like we've got the front panel board here.

We've got the main processor unit. This looks like the motherboard, the base motherboard or whatever. We've got a Crt, of course. Um, and then we've got this.

Looks like that's our power supply board over there. So that's one of the things we're interested in. Shouldn't really have anything to do with the drive section. I think the eht when you saw that thing actually collapse that's an eht high voltage eht is extra high tension.

That's a classic thing of where the eht is dropping down and then the capacitors goes and all your image just collapses like it's not like the Xy drive or anything like that. So our power supply board is pretty basic stuff. We've got a line filter here, we've got our Uh bridge rectifier. there.

There's a 10 volt voltage reference, there's our flyback transformer, and and then the rectifier, and then the five volt regulator. It's got to have more than that. Um, it can't just be five volt, um. output.

There's got to be tons of Uh rails, especially for the analog and digital and all that. Anyway, Um, there is a protection circuit because some people mention the protection circuit like shutting it down so it does actually have a power fail output, but I suspect that's just going off to the processor so that it tells the processor or powers failing quick. Save your settings to your non-volatile ram and stuff like that because there's a temperature, um, overload going down into that. I'm not really concerned with that at this stage.

I think there's uh, something else. So the output from that comes into here, and that goes into our Eht converter. Aha, And that gives our negative 2.2 kilovolts there. that would be our anode uh voltage, and then 6.3 volts.

that's classic voltage for the heater, the filament heater, And then there's our tripler. Um, the high voltage multiplier as they call it. and then that goes off. That goes off.

Follow the money. follow the money as Deep Throat says. Um, And then here's our graticule. Yep, so that gives us 14 kilovolts.

Okay, so that's what we're dealing with. Yeah, there's the 2.2 kilovolts. Uh, there. That's the anode voltage.

I've got a five kilovolt, uh, probe. so we could actually measure that. Uh, we're not going to be able to measure the 14 kilovolts output of the tripler. which, uh, some people might have suspected.

It is okay. We've got our mains input here. Here's our bridge rectifier. Uh, input Varista.

So yeah. like there's no bad, uh, reefer caps or anything like that, so it's nothing to do with that and that doesn't fit the symptoms. But anyway, so this is our primary drive side here. Here's our transformer here, so it's all switching.

One thing I want to do when I put this back together is the first thing I would check for is whether or not the processor is still operational like all of the main digital power supplies are still up and whether or not it's just an eht failure. So I can probably get that by just once. I put it back together. just operating the controls and see if it like still responds to the rotary encoders and stuff like that because you can hear the click, click click.

And if all that still works, I don't even have to measure voltages to know that the digital section is still working. It should be able to get some, um, sort of, you know, tactile, um feedback from that. Anyway, I'm not seeing anything in here that would shut it down that there's the feedback of course, coming via the opto coupler here. But uh, no.

here we go. Here's our eht converter here. There's a signal coming in here. Okay, so that that actually could shut it down.

I guess I won't go through the whole topology there, But yeah, does that come from the processor? I don't know. I might have to rtfm a bit further. but anyway, we've got 58 volts high voltage there, so you know we can measure that point after it fails. Of course, the whole idea is that you want to hook up meters.

This is where I've showed in previous oscilloscope troubleshooting videos. I think it was where that dire one went. I had like four or five multimeters set up. Um, just to monitor all the rails because I knew one of the rails was dropping after.

you know, five or ten minutes. This is a similar sort of thing. So why you need more than one multimeter? This is one of the things you might have to measure like. you know, several things at once.

You don't want to get in there and start probing around because there's dangerous voltages and everything. So you want to like attach all your probes first and then have them all coming out to various meters and then you can measure them all at once and just see where one, uh, drops out. So 58 volts Ht. You know if that's not happening, then uh, then your high voltage stuff over here is not going to happen.

We've got a switching transistor down here. Um, you know, so any of that fails. You know, so it could be. You know, I wouldn't rule out like a transistor failing thermally.

I wouldn't rule out like diodes failing thermally. So here's our tube. Eht, over here. 14 kilovolts here.

So this is our tripler. so that takes. Yep, sure enough. five.

That's why they call it a tripler. Five to fourteen. It's near enough. Um, so we get our five kilovolts here.

So this is our main. This is our eht, well, eht, converter as they call it. So we've got a single diode here and that, um, creates minus 2200 volts which then is smoothed out by these to give us. well, minus 2200 smoothed out.

Not switching. I'd more likely suggest like thermal failures in diodes or in here or something because there's diodes inside these high voltage multipliers here. But somebody mentioned that these actually contain reefer caps in them and they can arc over and that sort of stuff. I wouldn't rule that out.

Here's all the rails. So I've got Plus five. Minus five? Yeah, Minus five. Uh, plus Minus 58.

Minus 12. Minus eighteen? Yep. Plus eighteen. Plus twelve? Yep, Yep, Yep, Yep.

So there's lots of lots of voltage rails there, but I wouldn't be suspecting those. I reckon you know there's there's something in this. um, eht. high voltage section.

Yes, I think this is that component that was open, but I'm I'm gonna look up that. Uh, Byb412? Almost. Bingo. Um, I.

I searched for that and I got this by 8400.. Whether or not, that's a schematic error, because this is exactly what you. This is the exact diode you would have anyway. this is.

these are interesting beasts. So these are temporary high temperature high voltage parts and they're specifically for color televisions and monitors. High voltage applications for multipliers slot way on diode split transformers, so it uses a high temperature alloy construction. This package is hermetically sealed and fatigue free as coefficients of expansion of all used parts are matched.

Interesting. The package is designed to be used in an insulating medium such as resin, oil, or Sf6 gas. so they're designed to be like our potted. So the By 841.2 Yeah, we're talking 14 kilovolts, although it's only used in the two kilovolt section.

So here's the interesting bit and probably why I couldn't measure this. Because the By-8412 has a forward voltage. This is a forward voltage you're used to 0.6 volts on your diodes. Uh-uh these are special snowflakes that's a forward voltage of a maximum 52 volts.

It doesn't even tell you what the typical is, but that's at 100 milliamps. So look a forward voltage going up to like 45 volts. Unbelievable. So our multimeter is only going to push like a couple of milliamps through their top.

so we're like we're down here so we don't even know what we're going to measure. I will retry that with the 121 Gw. Next logical step is to reassemble it and then check as I said to see if the digital section is still up and running when the if it's just contained to the eht section and then that will rule out if it's a primary side mains thing. Uh, that's it.

could be primary side mains, but I remember hearing a high frequency hum when it's switched off. So all these things are important. Like all these symptoms that you get, they direct where you're going to shift your focus in the troubleshooting. So in this particular case, I know it's not like it, you know, failed Esr and failed a literally cap in the primary side for example.

That's just that's just not the symptom something heating up or building up charge that then arcs over or something like that and something in the high voltage side of things. So I'd be very surprised if it had say multiple faults. But it is possible that there is say a problem with a one of the digital suppliers and then that causes the eht that. remember we saw that that eht driver line coming in or or something like that so it could be so the digital could could be failing and then shutting down the Eht and stuff like that But yeah, that's why we need to check.

Um, if this thing's still operational and doing stuff. So let's go back. reassemble it so there's that little sucker down there. It doesn't look like any deity that, uh, you've ever seen.

It almost looks like shriveled up or something. like a little sausage or something. So if we measure that, let's get in there. I think I've got that around the right way.

but if I haven't let's swap it and you'll notice that that is open. And if we get ohms on that, there's something there, 30 meg or something and open the other direction. That's interesting. So, but if we get a 121 Gw that has a 15 volt uh, compliance voltage 10.7 volts they are.

And of course, in the other direction, we'll get Zippity Doodah. There you go. So um, if you didn't have a higher voltage Diode tester, then you wouldn't be able to test something like that. What a sneaky little buggy.

You might think that's open and you could chase a red herring down a rabbit hole thinking that that diode is no good. Of course, you could physically or lift one leg and then hook it up to your power supply and put some current through it. and uh, test it that way of course. And those things are annoying to test, but using a regular meter you would think that's busted any day of the week.

So you've got to be careful. I've got my Uh 5 5 kilovolt high voltage probe here, which is 100 to 1.. So I set up my scope with a 100 to 1 division ratio on 500 volts, Uh, per division and we know it's negative 2200 volt, right? So let's power it up. It's on.

Hey, negative 2200 volts. We're good. All I gotta do now is, uh, wait. So I'll wait for it to fail and then, um, if this doesn't go back.

If this doesn't collapse the 2200 volts, then we know it's just the tripler arcing over almost certainly. So I can hear the clicks as that processor encodes the knob. So when this thing fails, assuming it's going to fail, Murphy's going to give me a break. Then if we can hear still hear that, it means that almost certainly all the digital rails they're all still working.

You wouldn't believe it. I just pressed stop like I was going to go for another clip and it just happened. No, the digital rotor encoder that's dropped down to zero. Everything's gone.

so the whole thing's gone. So damn it, we're gonna have to probe. um, further. Okay, I've got some more probes hooked up now.

All right. So what I've got here is the yellow one, our 2.2 kilovolts. Then we've got the green one channel 2 that's at 20 volts per division. That's our 58 volt rail there, and the blue one that's our 12 volt rail there, So I'm measuring the three different rails on here, the 58 volt, the Uh 2.2 kilovolts which actually gets back uh, fed back here by the way, I just noticed that and the 12 volt here.

I'd like to probe this power Ht input, but unfortunately, that's a surface mount. Uh, there's both surface mount devices on the back side of the board and the back bottom side of the board is right down in there, right down the bottom, so you can't probe that and the board has to be in there unless you've got like service extender cards. Um, there's nothing you can do about it if you got desperate enough. of course you have to take the board out, then you, which is a pain in the ass and you've got to solder wires on and then you've got to have them coming out and then you've got to probe everything.

and well, we're not there yet. Okay, so I've set this up to trigger when the 2200 volt rail when it actually collapses. Uh, like we saw before. So I've got my Uh trigger source set to either slope there just so that you don't do a brain fart because it's got a it's negative voltage so it's transitioning high way.

We got it. there we go. I did actually turn it back to Uh, 10 milliseconds per division. I think I should have done it slower than that, but uh, that was our 12 volt rail.

That's uh, dropping. So our 12 volt rails switching off. Wow, that's interesting. Ah, the trends.

All that high pitched squeal is annoying. So I'm going to switch that off that we triggered on our 2.2 kilovolt. So as that transition up, you can see it hit the trigger point there. But you can see the 12 volt that was already decaying that was starting to decay before that.

But interestingly, the 58 volt rail has been zero all that time. So it looks like that 58 volt Ht rail there is dying. Uh, before we actually get, um, our 2.2 million kilovolts shutting off. And of course, if this dies, then there's no switching.

There's no voltage for, uh, the primary side switching. And then, of course, if this, this would fail first. and then this would fail after. Oh, we're getting somewhere.

I think I could now set my trigger point on there. If I really wanted to capture that 58 volt rail failing, why not? I'll actually wind that back to like 100 milliseconds uh, per division and I'll just leave that set up. I got it. Uh, don't worry about, like any sort of like little risers like that, Probing's not perfect on this.

I'm using Mains Earth grounding for this sort of thing. I don't actually have the ground leads from the any of these probes actually hooked up to these boards. It's going via the Mains Earth here and going back to the power strip under the floor here and going back. So you know that's just for ease.

Like we're not looking at signal integrity here. Our 58 volt rail has plummeted once that starts happening. Yeah, of course. then our 2.2 kilovolt rail starts then decaying or negative 2.2 kilovolts starts to cane, and our 12 volt is decaying as well.

So yep, that sudden drop on your 58 volts? That'll definitely do it because there's no more volts to switch. So yeah, no more high tension. So looking at the schematic here, this is the point. We're measuring plus 58 volts Vht, which is interesting because I can't yet find that anywhere else in the circuit.

It just says plus 58 volts. It doesn't have the Ht on the end of it, so I don't know if that's after some sort of switching element somewhere else in the schematic. I haven't found yet. The signal was coming from like Power Ht here.

If this was like shutting off this switching, then it wouldn't I would expect it wouldn't shut off the 58 volts. Uh, rail. The 58 volt rail would be there. and it'd simply switch off all the the switching transistor here and it just wouldn't switch.

but our power rails actually failing. So either there's a switch in there or somewhere else in the schematic. or if that's coming directly from the switching transformer, then it could actually be the primary side switcher failing because we saw that the 12 volt rail was also failing as well. And this 12 volt rail is in here.

That's why I was measuring the 12. the primary side of the mains, which then generates this 58 volts and 12 volt Vht on the secondary side. So if both of these are failing, maybe we have to go back and look at the actual primary side main switcher. Okay, the answer is always more probes.

So I've got two, uh, different high voltage differential probes now doing the primary side. Okay, so there we go. We've got some primary side switching. The red one there is our our mains input.

Uh, Dc. That's at 100 volts are per division. 330 odd volts something like that, and we'll wait for it to drop out again. It's just to rearrange those a little bit.

I got it. And there you go. That's interesting. Yeah, the primary side is still switching, but at a much lower rate like two kilohertz.

That's why I can hear that two kilohertz buzz. Yeah, you know it's roughly two kilohertz. Odd, whereas I didn't can't remember what frequency it was before, but it was much higher. Primary side, uh, failed.

But you can see that domains Dc is still there, so it's not any of the primary rectification diodes failing or anything like that. So I'll switch that off now. And um, there we go. So that means all this goodness here.

We were looking at the switching across this diode here because it was a convenient uh, location and yeah, it just it dropped in frequency and um, that would affect all of the outputs which are all of the outputs here. which are, you know, all of your voltage rails. so everything's dropping. So um, that includes your eht so uh, looks like um, people who, uh, suspected that it was like the uh, tripler failing and stuff like that.

It doesn't seem to be. It's like something in the primary side of the main switching that's going. Hmm, well this is interesting. It's been running for 31 minutes.

never had it go nearly this long. What I did is, I actually resolded um, Ohio, re-inspected it all, couldn't find anything, so I thought I'd re-solder all like the power uh components anyway and I put it back together and it failed after about four and a half minutes. But and then I thought, hmm, I'll time this um to see like you know if it's like if it's fairly repeatable or not And sure enough, the second time I do it, it's still been going 32 minutes so I don't know what the heck's going on. Anyway, it's 10 p.m I'm going home.

Well, I've come back the next day. 3 hours and 44 minutes later. Um, I powered it up this morning and it's still going so it's not very bright, but yeah, it's still going. has not cocked it since the very first time that I put it back together.

So really, you can only deem this fixed. Um, at this point because there's nothing, there's no failure that like to speak of that you can troubleshoot any further. So I think the next step is uh, to basically put the lid back on, seal it all up so it's a bit warmer inside and maybe you know it, Things might heat up in there or something like that, so I don't know, that's it's got to be the next step. So if I put the lid on, everything works and I leave it powered up for days.

Uh, you know, a day or something and it doesn't fail? Then well, like what's left. I can. kind of. Maybe I can freeze some stuff again to try and induce failure, but it fails after turning on.

which indicates that there's a heat thing. But of course doing thermal stressing and stuff like that might be able to induce some failures. but there's like there's nothing I can do. um, at this point.

So yeah, I'll put it back together. So maybe there's a component that's like was dodgy, but it's now healed itself. Maybe after some thermal stressing because what I've done is I I did freeze. um, the active uh diodes and some other active uh components.

I measure them just to make sure you know like they didn't go open. I've measured the Esr of every cap on there in circuit Esr It's a bit of trouble to actually remove every one, but I've done those um in circuit and they all, um, seem fine. Everything looks, uh, visually perfect. I've resoldered a lot of the, well, all the major Um power components like the uh, transformers and the devices connected to the heatsink.

like some of the uh, you know, pass switching transistors and stuff like that. So I've done. I I've done those. but as I said first time I powered it up, it failed.

but then after that boom it's like it. It seems rock solid. So anyway, I there's nothing more I can do. So this unfortunately is going to have to conclude part one and this is just like the nature of, uh, troubleshooting stuff like this.

Um, if there's no fault, then unless you can induce it to come back, then you know you can only speculate. Of course I can come up with, you know, half a dozen different theories about what could cause something like this. And please, if you've got your favorite theory, leave it in the comments down below, a link in the schematic and service manual down below so you can have a look yourself. but it does seem to be something at least isolated to the primary side of, uh, the mains, uh, part of it.

So yeah, I don't know. leave your favorite theory in the comments down below. But anyway, it's about the journey, so I hope you found that interesting. So at this stage it's fixed until a part two.

Anyway, I hope you found something useful in that video. Catch you next time. Nope. The bloody thing won't fail.

I've had it going for an hour and almost two hours now. Um, with the case on and it does actually get quite, uh, warm on top, but not nothing. So uh. you.