Hi, yes, it's New Year's Day and yes, I'm at work. Ah, go figure. Anyway, um I thought I'd do a uh quick follow up and have some more play around with this tectronics. TDS 3054 Oscilloscope If you haven't seen the previous video, I'll link it in down below.

didn't really have any time to uh, troubleshoot it uh, properly. Last time I was just a tear down to see if there was anything obvious there at all and have a poke around under the microscope. There was nothing. uh, obvious.

Of course we've got a faulty channel 3 here and we're not sure if it's in the hybrid or whether or not it's in the ADC or whether or not it's in part of the sample memory or something like that. but uh I Asked for comments on the Forum and sure enough I got them and it turns out somebody had uh, well, Vincent actually. uh, free electron on the Forum Um, had the info that the hybrid amplifier uh chip in here actually has a high frequency and a low frequency uh differential driver. Our put so splits the signal into high frequency and low frequency component and there's two line uh differential pairs which go over to the ADC there.

So um, and that might kind of explain it cuz it looked like the low frequency content worked but the high frequency content didn't so possibly there could be something there. So I'll have another look at the that under the microscope to see what's going on there. A few people ask, well what's the next step of course Well the next step are I'll try and find some power supply rails that are common between these things because we've got different channels. Um, then it's good cuz we've got something to compare it with.

so maybe I'll try and find some power rails in there, measure those, make sure they're okay. Um, but as I said, you know, look that it's the ADC there is digitizing that low frequency component so you know I'd be a bit surprised if there was any uh issue with the uh power supply there. Maybe, but hey, you never know and so check that out. Uh, we'll also probe The Next Step would be to probe the differential output of the Uh channel here and cuz as I said, we've got another Channel we can compare, feed the same signal into a working Channel and the foldy channel here and we can check to see what the differential output there is on both that high frequency and the low frequency um pair on there so we can actually, uh, do some uh, simple measurements, quick measurements, and uh, check that out and that'll tell us if there's anything wrong with this hybrid at all.

cuz it' be really nice to completely eliminate this hybrid module and you know, is it in here or is it up in the more conventional circuitry up in here? That would be real nice to know. So let's get to it. So let's check this hybrid module again and see where these diff pairs come out. As I mentioned previously, obviously these two pins here coming out of the hybrid are one of the differential pairs.

We've got some capacitors there, we've got some resistors by the looks of it. uh, laser trim resistors there, and these two pins are your differential pair output. And if you can see down in there sorry, it's not going to. Yeah, it hasn't focused on that, but you can see the differential pair winding its way out there like that and they're nice thick traces, so you know that's a controlled there's a ground plane under there.

You know that's a controlled impedance pair running to the ADC chip over here. and if you have a look at these two over here right next to it, Well, they're also running as a different differential pair down in there, but thinner. Well, that makes sense. That means that this one over here with the thicker tracers is probably the high frequency one would be my guess and this pair over here is the uh, low frequency pair if um, Vincent has got that right and they do split the signal out and go over to the ADC like that.

So there you go. Um, so one of the first things I I'm probably going to do because we can get the probes down on there I might just do get the meter out multimeter and just do some resistance measurements and compare between the two modules. Actually, this is channel three here that we're looking at and you can see the same we're looking at Channel 2 before. So these are the high frequency pins there and then you've got your two low frequency pins coming out there and going over to the ADC over there.

So that's definitely a differential pair. So I'm going to assume I can't see sort of any other ones sort of nearby. So I'm going to assume you know you would think that they would be like right next to each other on the hybrid chip chip. so that's definitely the best guess.

So in circuit resistance measurements are actually can be quite handy if you've got like an AB uh comparison here. So let like what I want to check is the capacitors are making contact down to the hybrid down in there and the way I can do that is I can contact the top of the capacitor up there. This is a good Channel There you go 21 .8 Ohms on one cap this is on presumably that that high frequency pair 21.8 the same. So let's go over here, make sure they're the same.

Yep and drat. I Was hoping it'd be that easy. you know I'd be able to find something. but no, that's um, fine on that pair there Now I can actually test the other pair.

Let's just uh, put the pins across there. 785 Meg Okay, swap them. not measuring anything. low impedance there.

787 Okay, but anyway, the actual values don't matter because you can just probe around. Ah, there you go. There you go. That's that's slightly different.

Let's measure a third. Channel Over here. Ah, there you go. Hey hello, let's go over to Channel 4 over here.

Did I get the right pins on Channel 3. Hang on. Hang on. Oh no.

There we go. Okay, so Channel 4 is no. All right. No, that's a bit of a fury.

Okay, no, we're chasing a red hair in there right now. I'm feeding in a 1 KZ uh 1vt Peak to Peak sine wave into both channels and you can see Channel three there. That high frequency content is just awful. The blue one there is channel 2, so that's working just fine.

And as I said, the amplitude is just fine and the low frequency waveform performance of Channel 3 there is just fine. It's just got all this high frequency crap on it, so you know that's that's really interesting in what Vincent said with those separate pairs with the high frequency and the low frequency content. So anyway, let's have a probe around now. Of course, when you're working on stuff like this this, you have to be very careful.

I've got the main section. uh, complet. you know, separate over there like that, which is really good. Uh, the low voltage.

uh CA in comes over quite nicely to the main board. and so I can actually, uh, probe around this while I'm still feeding in the signals to the front panel here, so access is actually really quite good. But anyway, just want to keep away from that main stuff over there. And also, there's a high frequency, uh, backlight? uh, high voltage backlight inverter here as well and it's got a little Ion on there.

Want to stay around with that. Now to probe this. Uh, what we can do, we can just use a single probe and do it single ended. Um, that's just going to work fine for this.

uh, differential signal probe each side independently. but we'll use uh, two probes and we'll uh, do a poor man's uh, differential probe. So we we'll subtract one signal from the other, but in this case, it's it's You know, it's probably not going to make a huge difference I'm going to connect the grounds through to the common, uh, ground on here. otherwise we're going to pick up, uh, too much noise.

I'll probably, uh, show you the effects of that in a minute. and um, yeah, we've got uh, both channels one and channel two set up. The math operation is Channel A minus Channel B So let's have a probe of this thing and see what we get right? I'm going to have a look at Channel 2 which is the good Channel and I'm looking at the high frequency output and bingo, that's the differential pair there and you can see clearly see the two Uh signals. the uh, blue one and the yellow one there and the purple one is the math.

This um Ryol scope isn't the fastest on this uh math operation. It's nothing like a real good analog scope with its uh, subtraction mode that's for sure. um to SLS a wet week really on that math function. But anyway, we can see our 1 khz and you can see that the signal sorry I can't push the stop button and capture that.

Um, I'm using both hands here. but um yeah, we're getting our 1 khz signal and we can see the differential signal there, which isn't really helping us a huge amount. Now as I said, let's disconnect these ground wires here. In fact, we'll take them off like that and let's probe that again.

and oh, here we go and you'll see out. Look, look at that noisy ass that is an awful waveform. so you really need to connect your grounds in there to uh uh, get your um noise down on that And there we go. I've run some uh average in on that and you can see it's bang on 1 khz getting nice beautiful waveforms out of the high frequency.

uh part of Channel 2 there so that's our good Channel All right now let's take a look at Channel two here. I've got averaging on so that's what. Uh, it cleans up the waveform but uh, doesn't allow us really to do proper realtime probing. But there's the low frequency output and that is just fine.

Let me, uh, switch that average in. uh, off there. That's a bit annoying. Okay, so this is Channel Two, what we'll call the low frequency output.

Okay, fine, and this is Channel Two what we'll call The High Frequency output. And that's just fine as well. We're getting the same signal on both both of those differential pairs. Now let's go over to the faulty: Channel This is the low frequency pin.

That's just fine. There's our 1 KZ signal. not a problem at all. and there's our high frequency pin.

Not a problem at all. So Bingo we've just narrowed it down to something being wrong outside of this hybrid module. It is not that hybrid module there at all. at least not in this instance for this test signal.

And really, it should be the same case for every uh, you know, every test signal. Really, there's nothing wrong with that hybrid module at all. We can completely eliminate that. I was going to suggest.

You know, if that was an issue, the next step in there might be to, you know, get the freezer can in there or something on that uh chip and freeze it down. See if you know there's a thermal. um, you know a like a a joint issue or something like that. A bond like a, you know what is that like? a little BGA device or something.

um, solded down to that uh, ceramic hybrid module or something like that could be a bad joint under there. something like that which would ordinarily show up with, um, some thermal cycling so you might get the freezer spray on there or something like that. So, but anyway, we don't have to do that because it is not an issue in there. we're getting exactly the same.

This is the advantage of having a working Channel Side by Side you can just prob both and we know it's not that hybrid module so it's something in here. So next thing is the touch test yo that's pretty hot. That's pretty hot too and so all of those adcs are getting pretty darn hot. In fact, I might get the thermal camera out for this.

I've got my new Flur infrared camera which I haven't showed yet and woo check this out. This is the new Flur E Series This is the E8 This is the topof the range E8 so thank you very much Flur for sending this in. It's going to be a permanent fixture here in the lab. They've donated this to the lab here so that we can use it for specifically stuff like this and um, yeah, I will be doing a full review of it in the uh, well, it is a New Year now isn't it? But yeah, go figure now.

I can actually capture video from this Flur camera via the USB port. but uh, please forgive me I Couldn't be bothered dragging The Notebook over and uh, you know, hooking it up and and doing uh, that sort of stuff so you'll have have to be content with just viewing the screen here unfortunately. But um, yeah, it still gives us a very good result here. And you can see the span where we're going from 22 sort of.

you know, on the bench. Uh, this is the temperature scale. It's automatically adjusted. it's up to 96 so you can see that the hottest objects which are obviously that main chip the die inside that main chip.

There we go. 101. Well, you know, 90. It's basically 100 there on that main As chip.

Here comes my fing. There it is that main As6 chip down there and you can see that the the four Adcs one there, one there, one there and there and you can see the hybrid um chips up there, those hybrid As6 and which is you know, a differential amplifier and stuff. And then there's that mysterious um, sort of, you know, fourth, uh, fifth, uh, bridging chip in there. That one's also pretty darn hot running at 90.

Anyway, the Adcs there are running at about 80, but the thing I want to check is that they're all running at about the same temperature and you can pretty much see that there in the color gradient. I mean I'm not going to dick around with getting the exact you know pointer on that. And the interesting thing about this watch this is if if I go closer, you'll notice that the hotspot offsets won't will be off on the chip because this is the parallx error due to the inbuilt camera, you know, So please forgive me. I Don't have a uh tripod for this thing either Now I've got my pointer in here and you can see that it's This is a really awesome camera because it has this um MSX technology which overlays a real image of the board like that over the thermal image and it's really the resolution is really fantastic.

I can read serial numbers of chips and everything, but anyway, you can see as it gets close here there's parallx are you can see that The Closer I Get I'm moving the camera towards the thing, then see that hot spot there that my point is pointing at should be in the center of that chip. but it's not. So that's the parallx error due to the camera and the thermal sensor because Tada the camera is up here and the thermal sensor is down in here so there's a you know you're going to get a parallx error and there is a software option in here to actually um set that when you're up close and you can see that we're already already at our closest alignment distance there of 0.3 So yeah. anyway.

um, this is. this is a really neat camera and uh I will have to do a full review of this. but you can change the image mode. The user interface is a bit annoying.

See, here's the difference between the MSX mode which overlays the camera look. you can read all the serial numbers on there. It's just beautiful and as opposed to just normal thermal image mode like that. which is, you know, pretty boring even though this thing has three 20x uh, 200 resolution on the damn thing.

So the resolution on the thermal is awesome. or you can just do picture in picture like that. and or you can take uh, digital photos and things like that. Anyway, um yeah, this is not a review.

I'm just playing around. but as you can see there, the uh. as we get further away, that hot spot will line up properly with the center of the chip. There we go.

it's getting much closer so you can see that they're lined up in there. But anyway, um, enough mucking around with my new toy here. Um, that tells me that you know that ADC chip there isn't on Channel 3 isn't um, you know, faulty or anything like that. It's not heating up so no real thermal issues there.

That's interesting. input termination overheat message. Maybe because uh yeah, the fan is not on and we're getting no Cooling in this thing. Whoops.

All right. So what I've started to do here is just have a prob around at some of the voltages I won't uh, show that up close You' seen in the previous video but there's a couple of like So 23s under there and I've been probing around at those so 23s between the channels and I found something rather interesting. there's a three pin hang on. let's I put my Probe on there on the shield for ground and I probe all three.

Ah jeez, that's hot damn it. Probe all three pinss of this so 23 and there's - 2576 - 2576 and 2576. So we've got 2.57 6 on all three pins of that sock 23 down in there. Whatever that device does, it's got the same voltage on all the pins and I've checked the other So 23s on the other channel here and they've got various voltages on them.

one is 2.5 minus 2 2576 but other ones have got you know, 3.3 and other voltages on there. so that is rather interesting. So what I'll do is I'll um, turn the power off and I'll measure the resistance of that and uh, see if things are shorted. Um, just be careful when you're probing around here.

by the way, cuz I was using as I said before, this is ground this metal shield around here so if you're accidentally probing and accidentally touch things, you can short stuff out. but the power's off at the moment cuz I'm measuring resistance and and uh, these two pins on that so 23. Look at that completely shorted. huh? I don't measure that on that same presumably the same to so 23.

I'm going to assume that all four channels are identical. There's like a different layout in terms of where they're placed on each channel on the board, but there seems to be you know the same sort of components on each channel. So the the two pins on that so 23 the the third one over here is not really anything. There's like you know, a k between there or something like that.

But if I go over to one of the channels over here for example this one and measure that so 23 that. so 23 of the pins there you go. it's like 55k and I can do that on these other channels over here and look there it is. 54k and I think the well, that's for completeness.

Let's measure this channel over channel one over here. a 2.5k Yeah, I don't know. Anyway, still right, it's not a dead short. so whatever.

that little beasty there is a bloody thing I think is shorted out and that could certainly explain what's happening here. I'll try the thermal camera again and try and get a closeup of that uh little so 23 No. I can't really see anything obvious. It' be like up under there.

it's like there, but on the uh, thermal imaging camera it would be. You know it would show up as a as a hot spot if there was significant current flowing through there. but that sucker is is short. I I Really suspect it? I mean you know I could be wrong, right? but you know, just it just doesn't doesn't feel right.

Aha, there's the little culprit that I suspect and under the Mantis microscope because it's got better Optics and you know and the 3d effect. it looks like there is a split in the top of the case on that little tiny split. I'm not sure how this is going to show up in HD on my camcorder camera here. um it's not showing up I'm viewing this through my macro lens on the camcorder I Was hoping that it would uh do it but no I'm at full Zoom here and that's the best I can get.

Let's see if I can film this under the Mantis and there we go. You can see the top of that so this is really hard to get in Focus but look look at that just on the seven there there is a looks like a big split right in the center of that so 23 that for all the world looks like a blowout hole that I think the magic smoke has escaped from that thing I Have no idea what it is. it's just got 27 uh marked on it. but those two pins on the Uh left there of course are shed and the other devices which are also marked at 27 in the circuit.

Those ones aren't a dead short. So magic smoke every time. the active ingredient in every component, if it escapes it, don't work anymore. I Completely forgot that I had these real sexy tiny Pomona uh test leads someone sent into the mail bag.

these are perfect for Really T White quarters probing on boards like this I'm going to get back in there with these. Oh I Just love them. They're so look at needle point sharp. Oh beautiful.

So I'm back probing around on here. And yeah, the other so 23s on the other channels they're measuring uh plus 3.3 Vol uh, presumably 3.3 volt Rail and then uh, plus 2.5 volts. But the blowing so 23 in here is measuring - 2.5 Vols on its pin. So yeah, that is just that is rather curious.

See there's 3.3 on one of the pins of uh, one of the SL 23s one of the good ones and there's you know, 2.44 and 2.45 S whereas our foldy one with the magic smoke that's escaped presumably - 2.57 and- 2.5 and Well - 2.5 on all the the pins. So yep, something ain't right. So it's marked uh 27 on the top of it and these bloody cryptic so 23 things I Don't do you know I'm not in the repair trade if you're doing this all the time, you might sort of, you know, have an inkl in what uh, that could be? Is it a little uh, thre pin? So 23 uh voltage regulator I don't know, could be cuz the others have a 3.3 volt input and then um 2 and half output. so it could certainly be a 2.5 Vol regulator.

That doesn't explain why there's not 3.3 Vols on one pin of that thing though. So I don't know. Is it like a a dual diode or something like that? but it is labeled. it looks like CP but I think it might be CR cuz the others are CR It's labeled CR 140 on the Um silk screen there now of course.

Uh, CR is a Um standard designator. It means diode so possibly it's some sort of di. you know, could be like a dual diode or a single diode or something like that like a Bav 99 or um, something equivalent like that. No I checked and it's definitely CP 140.

It doesn't look like the r has been uh, chopped off like the you know, the little leg of the r has been chopped off by one of the Uh viers there. But considering that an identical one on another channel is labeled CR which indicates diode, it has the same marking on the top 27. Well, yeah, my best guess is that's a diet of some sort. Do silly me.

I Just realized that the two leftand pins there on that so 23 are actually uh, shorted on the PCB so that's not inside the package. but anyway, um, that is definitely I reckon there's a blo hole in the top of that sucker. and yeah, I really suspecting something's wrong there and a bit of Googling and I tracked it down to possibly a Fairchild mm uh BD D 1204 cuz it shows 24 there, but 27 is used for the 124 which is a R pin Arrangement Uh well, the 124 pin Arrangement which is the common cathode small signal diode just as I suspected Bingo that's got to be it now. I should probably also double check the signals going to the ADC down here to make sure those differential signals are getting from that hybrid down to there as well.

I mean this diet I I reckon that diode is gone I mean I reckon you know it's blowing the ass out of that diode. but you know, usually that they might only be in there for protection or something like that, so you know why it's blowing. Well, let's ignore that for the minute. but the fact is, you know that may not affect, or you know, affect its current operation.

There could be something else, um, causing the issue. But anyway, I probably should double check, pair it up again, feeding the signals and just double check that the signals are going there. but I'll have to use. um, you know, something like these.

Uh Pros because the regular Um scope probes I just can't get in there in that tiny pin pitch. especially right next to that Shield there and probe those pins. It's really quite annoying. Anyway, that diode there.

is it? A diode? There we go. But let's check it the other direction. and yeah, it's exactly the same. So eh, it's in circuit.

What do you do? Actually, instead of probing those really, you know, cuz there's nothing going to be wrong with the PCB traces from there to you know, the 1 cm it takes to get from the hybrid module to the ADC So you know if there was anything wrong, it would be the solder joints on the chips. So I just reflowed those. so I just used I just dabbed my flux pen on there and just you know, got down there and just reflowed the joints down in there and well, let's power it back up, see if it made a difference I doubt it. but hey, it's worth a shot and of course no.

I'm never that lucky. Bummer. and well, I went the whole hog and I uh, lifted that one pin on that uh, foldy diode there and N makes no difference whatsoever. So ah goodness, let's have a look at that diode there that I've I've lifted the pin up on so we can actually access the pin.

There we go. It's still acting as a diode. There we go. 0.55 sounds a bit right.

haven't checked the data sheet, but that does sound right other direction. Oh hello, hello. I'm making correct contact there. Yes, I am there you go? Oh yeah, that thing is.

yep, that's there's something wrong with that diode. Yeah, that doesn't work well at all. What I'm going to do now is just apply power to the bare board here. so I can access the backside cuz so we shouldn't need any of the uh, you know, the screen and keyboard and stuff.

uh, hooked up for this sort of thing. So I can access possibly some uh, you know, power stuff on the bottom. I mean you know we' got lots of bypass caps here I'd like to test, uh, the voltage rails. Here's the third.

Channel you know, 1, 2, 3, 4, uh, there's some Ferite uh, beads in there for those channels I've already tested those. those are uh, fine, they're not open or anything like that. I'm not sure if like you know, there's one of those per Channel or something like that. but anyway, all the bypass caps there are on the bottom I can maybe check some uh, Supply rails for that.

uh third Channel there. So because you know like there's nothing on the top and there's no, uh, test points, you know it'd be nice if there was. You know, here's a 2.5 volt rail. Here's the 3.3 volt rail.

Please test it. but No, the good thing about these extra channels again, as you can see is that uh, I'll put my Probe on the ground over here and I can probe these caps and you can see we got minus 2.5 volts there on that cap sorry, you can't see it and then plus 2.5 volts on that channel. So it looks like the ADC is operating plus - 2.5 at least on those caps. Um, so there's some Rail and look, this is the FY Channel This is channel 3 minus 2.5 plus 2.5 They're all probably running from the same regulator I Don't know if they have local regulation or not they could do.

that's what. one of those little so 23s - 2 uh, plus 2.5 - 2.5 So and all the other channels there you go, they're doing exactly the same thing and the layout is duplicated across the channels. The PCB layout designer. of course, it's not going to make their life hard, so they're going to duplicate the layout on all of them.

so you know you can easily compare channels. So this is our folding channel 3 and those two caps. there are just hunky dory in the 3 .3 Vol rail on the main processor over here, powering all the uh, you know, the heavyduty digital stuff. That's all just fine as well.

I can't find any other voltages on these ADC channels. It looks like it's just operating from Plusus 2.5 Vols So yeah, all the the voltage rails are fine I Don't know. I'm starting to give up again and one thing left to try is the old freezer spray. I Don't have any freezer spray at the moment.

Going to use the air duster? just turn it upside down. Instant freezer spray. Brilliant. So I'm going to spray that third.

Channel ADC and uh, anything? I can't watch and spray at the same time? No, let's try that. Center chip. no memory. No no, no big Asic chip.

nothing. Not a sausage. no hybrid. just in case.

No hybrid's fine. let's really go to town on that. Oh there we go look. We got an offset.

Hello! Oh o o look at that. Look at that. Wow. Drifted back I Mean you'd expect that, but it's not like it suddenly vanished or anything.

Let's see if I can isolate that right just doing the little chips surrounding it. And well. let's see if I can get an offset on Channel 2. the blue one.

Oh yeah, look look at that. look at that. It's coming back. Oh that's beautiful.

That is beautiful. That's fantastic. Look at Channel 2. It really it.

What? it hasn't recovered, hasn't recovered. it's giv Channel 2 another spray. Look at that. Woohoo! And so yeah, it's not surprising that of course Channel 3, but we didn't see Channel 2 do an offset.

Channel 3 is giving us an offset so that's interesting. Um, there we go anyway. that's I You know I would expect that it's not like it's changing the high frequency uh performance on there at all. So yeah, I don't know.

that's fun, but yeah, we haven't found anything so sorry. I think I'm going to call it uh, quits for today on this one again. I've already spent an hour or something on it and well, you know I've got some other things to do and well I think we've got some progress I'm pretty sure I found a faulty diode in there, but well, it's not affecting the channel at all I remove it from the circuit. The channel still does the business I've reflowed the Uh joints on that channel 3 there I've checked the voltage rails.

We've verified the signals coming out of the Uh hybrid there, so we pretty confident it's not the hybrid module at fault and well, I don't know. Is it one of the memories or something? but look, there's It's not like there's a memory per channel, so they're clearly like, um, sharing those between the Uh channels. And of course, the Adcs are nine bits each, so that's an oddball uh value, so you know you probably got to share the memories anyway. But I like I don't know once again.

I'll you know? open the comments, see if anyone's got any better ideas. But anyway, I hope you found that interesting. Um, that little troubleshooting procedure to at least I think find a faulty diode. That's at least progress.

Something's magic smokes escaped from that, so that's not bad. But unfortunately Murphy has conspired to ensure that we don't find the fault yet again. The EV V blog repair curse strikes again. This one will have to go on the work in Progress list.

Catch you next time.