Hi welcome to Tear down Tuesday well it's actually more of a an attempted repair video cuz I've got this uh, nice tectronics TDS 3054 4 Channel 500 MHz oscilloscope here which I got in the auction score and I've done a video on that which will be linked down below if you haven't seen it and this particular one actually has a fault on Channel 3 here and we'll get into a bit of the detail before we crack it open. But anyway I thought I would um open this thing thing up and you know at least have a look around I don't like my chances of being able to repair this thing cuz these tectronic Scopes are famous for having custom uh, as6 and custom hybrid modules and stuff like that. So one if it's one of those that's uh blowing then well you know we're pretty much uh, screwed. but certainly it's worth a look.

and hey, we get a free tear down out of it because I haven't seen inside one of these TDS 305 4S before. So let's go now. the the first thing to think of when you find an oscilloscope like this with a problem on one of the channels is is the BNC dodgy you know, is it Loose has somebody you know put force on it broken off? You buy these things on auction, you know they could be manhandled and uh, you know, generally just beaten around. especially in in a um one which has come from the military which this one uh has potentially being uh uh, brutalized.

but there's nothing wrong with that. those Bnc's at all. So especially Channel 3 here. Nothing wrong at all.

So not a physical thing to do with the Uh BNC connector. And this isn't like the Uh Tectronics TDS 220s which are famous for their broken Uh Bnc's These ones are really incredibly rigidly bolted I think into the unit in there I think I can C see a couple of screws either side of that? so they're probably bolted into the um, uh steel frame of this thing. uh, probably the metal frame. So um, yeah, they're pretty d good.

So it's not a physical fold of the BNC So the next thing to consider is, well, have they overloaded uh, Channel 3 and well, we won't know that until we, uh, crack the thing open. But let's have a look at the waveform we get. Now here it is I'm feeding in the 1 khz calibration signal and it's weird. It is the correct voltage Here it is uh, 2 volts per division.

it's supposed to be 5 Vols so 245. so it's the correct value. But look at all this garbage above below all that is jumping all over the place like that. and if I press uh, stop on that and we zoom in.

I Mean you know, have a look at all this garbage. It's almost as if you know there's these huge bit errors in there. I Don't like that this thing doesn't have a um you can't press the horizontal position button to go back to the start. uh to go back to the center? Anyway, my first you know reaction to this is that it's some sort of uh fault in like the ADC and there's like a floating bit or something like that which is causing it to go to jump all over the shop like that.

but it's It Really is quite strange and if I dial back the time base look, it actually stops. okay. I can increase the time base like that. Not a problem at all.

Okay, but if I go down beyond that, it just vanishes. I mean there's nothing wrong with the trigger point I'm triggering off that channel and I can do all those setups and it still it still doesn't work. So there's it. It really is quite an unusual fault.

I Don't think I've ever seen anything like this before. so I don't know, you know, off the top of my head I'd say yeah. Look the you know, the front ends basically seems to be working so the odds of probably an overload on this blowing it. um, probably unlikely cuz we're getting the waveform.

I mean I can feed a sine wave into this and I get a sine wave out, but it has that same sort of you know, huge like it's almost as if the ninth bit cuz this is a 9bit ADC in this thing which is really quite nice and pretty unique about this scope. It's not just your usual eight bits, but it's almost as if like the ninth bit or something is just doing all sorts of random stuff or something like that. Anyway, cuz that that's just really weird. Look at that cuz it's just going way over like that.

but I don't know, it's you know it. It may not be the output of the ADC it could be something else I don't know, your guess is as good as mine I Haven't put much thought into it, but yeah, I don't think it's actually a blowing input because generally with a blowing input, you'd you know, blow the ass out of the fit or something like that and you'd really get no signal through. But we are getting a signal through on this thing so it's more of the sort of you know the process inside of it like the ADC or you know, after the ADC or the ADC hybrid I presume it's a hybrid module in here with these, uh, tectronics. So because it's 5 gig samples per second, it's uh, you know, woo, that's you know it's pretty darn good.

5 gig samples per second. So tectronics would have their own custom um ADC hybrid in there for doing this. But yeah, I don't know. Um, it really is a weird fault and why when you drop down in time base like that, it just doesn't do anything and it changes.

By the way, if I go into here and set fast trigger points, look at that. it just doesn't like it. It there it is. It disappears at a further point.

but there's look. there's a one line there. you may not be able to see that. but look now we're at 10 Vols per division so it's scaled that back.

It's weird. So the difference between 10K points and 500 K points is really quite unus. Oh now it's not doing anything at all. There we go.

So 10K points. it's that affects the horizontal where it disappears like for example, it's working there. but I put it on there and it's disappeared. If I turn the time base up, it comes back So yeah.

I That is really quite strange there. something to do with the acquisition engine or acquisition Asic perhaps. And if I turn uh dot mode on there instead of vectors, you can see the the dots up there sort of gathering around that position there. If I turn the waveform intensity right up cuz this is a DPO display of course.

um you can. just it's just it's very strange. That's with waveform intensity 100% Very, very strange. This one all right.

Now according to the Uh service manual of this thing, um, it's a bit tricky to uh, get this off. You have to start with the handle here. You don't undo the screw in there. You take this uh cap off and there's a little pin in there.

which apparently you have to put the handle upright like this and then pull this pin out about 5 mm. Actually, that is a rather interesting handle mechanism. It's uh, they've gone to quite a bit of effort to uh, design that. not sure why they had to go to that much effort, but okay, more Force required.

I Guess this pin ain't budging. and I've got serrated, uh pliers here. Oh no. There we go.

Pull it out. 5 mm. There we go. Okay now.

I don't do the other side. I Sort of put this back to the horizontal position and then apparently you can pop out. pop out the whole hub assembly. that's been the hub assembly inside there.

H There we go. It finally popped out I didn't pull the pin out far enough apparently. Yeah, go figure. Look at that.

A Unbelievable. Here we go. the other one should pop out. Tada There we go.

Oh yeah, piece of cake once, you know. and then the handle lifts off. But jeez, that's really convoluted. What were you thinking? Tectronics now? of course.

One of the most satisfying things of any repair, like this: voiding the calibration seal. You bet there we go, it's gone. Ski now. of course there are ways to try, you know, attempt to get this off and uh, and put it back on.

But hey, this isn't a warranty thing so we don't give a toss. All right now, we need a Torx driver to undo this screw here and we'll take out the comm's module. There we go. Taada, we're out all right now.

this back Chazzy should pop off. That's uh, that's the plan. There are no other screws I believe. So here we go.

Tada and yes, wait for it, folks wearing like Flynn who? I don't? Yeah, there we go. Can't let that rest on the floppy there. You got to flip it over, just flapping in the breeze. Look at that and there we go.

That looks pretty clean. I'm uh, happy with that. I don't mind the uh construction at all. This isn't too rigid on top here though, though.

but I uh I don't mind the arrangement of the uh power supply on there. It looks like we can probably uncp that top. Chazzy It's basically uh, single board really down in there. We got the main board there.

the um, there's the she for the front cans. You can see the dividers there, so we have to take that off and eventually get into the uh front end there. but yeah, I'm not keen on that floppy. just look, that's just flop around.

Maybe it. hang on. does that go into the case? Is there anything? Yeah, there's some I Think there's some, uh, sort of, you know, um. matching parts of the case in there to hold that floppy drive.

But the interesting thing is is that you can probably replace this. Thin Line floppy Drive Looks really easy. Uh, replace it with a modern Um flash unit cuz you can get these, you can get Slimline versions of these that have a USB port on here and it's got the same interface assuming it's got the Um same ribbon cable interface and it's not some sort of uh pin interface off hand. I don't know 100% on that.

but I know you can buy these things for you know they're about I don't know. 40 bucks on eBay or something. and they've got a USB stick in there and they operate and interface exactly like a floppy drive, so that would you know if you got one of these things, it would be a worthwhile addition. You know, cuz floppies are bloody hopelessly obsolete now.

Uh, much better if you can just plug a USB key into the thing and uh, save your files that way and you shouldn't have to change the firmware. This is interesting. Just notice this. Check out this uh regulator down there.

They've got this heat sink. Oh yeah, it's a PCB mount, but it's like a um, you know you don't often see just flimsy metal like that acting as a heat sink. Usually it's a proper, you know, nice little black anodized one or something like that. So yeah, that's a bit how you're doing.

Liking the look of this uh, power supply section down in here looks really beefy. nice big fat SMD Dodes down in there, huge big inductor bobbin. look at that massive I Like it. That's really quite neat and uh, what brand caps are they got in there? Not sure.

Nipon Chemicon Tada that one is there you go. Nipon K Chemicon. Absolutely First class as you'd expect in a uh, you know this is a real expensive Tectronics scope or it was, uh, back in a day. well even today.

It's uh, well, the modern version of this and this is like the older model I think it's the 305 4B now series but uh, as you'd expect. um they haven't cut any corners and there we go. we have our backlight driver down in there. That's one of the complaints I have with this scope, especially if you buy it old I'm not.

uh you know I don't have a brand new one to uh uh, compare it with but yeah, really. the backlight I think uh Fades with age and it's pretty horrible and dim and maybe it was like that, uh brand new I I don't know. my memory is are not that great on using these things. uh back in the day but yeah there separate driver and well maybe you could replace it with a nice super bright lead backlight or something like that.

Maybe that' be a nice upgrade and this is interesting. Check this out. They got a little two- pin header down on the main board there with these two wires. What they've done is, they've put them inside heat shrink surr just uh, you know, over the Main's wire in there, active and uh, neutral wires and what that's doing is just uh, there's no electrical contact in there, they just putting it next to there in some length inside that heing tube.

So um, the AC mains capacitively couples over and that's how they it gets in and that's detected by the processor and it can measure your Main's frequency and adjust for that automatically. So that's often a uh, cheaper and safer solution than actually having the circuitry on the power supply somewhere in here actually. electrically, uh, you know, tapping off the signal and then have it coming through the cable over here. But you do have that extra manufacturing step of having to heat shrink all that stuff in there.

but hey, you can't complain about the isolation, that's for sure. I'm not a fan of how this uh, extender board here this is the battery connector down in there sort of is not supported here. It may be inside the uh case of course, but when you click on that power button on the front, you know, um, there is a bit of stress on that board. but so unless that's really anchored well inside the other half of the case.

yeah, I don't know. that's not the greatest and took a couple of screws off the top here. Uh, disconnected the uh ribbon cable for the ancient parallel port and uh, looks like I can probably just get rid of a couple of other connectors here. Here we go: I can disconnect that from there and bring that up through there.

and then I should be able to disconnect the power supply connector over here. and Taada, we're in. There is a ground connection left H stubbornly attached to this shazzy. Now here's all the main circuitry and all I'm seeing here essentially is a bunch of uh, probably custom part numbers.

I mean what's this? Huge devices? National Semiconductor obviously Adg 360c I can find a couple of Uh references to that, but no data sheet on on first glance and then they've got Adg 365b, Adg 3618, probably an ADC per Channel and that's probably some sort of Max which is getting it into some sort of. that's probably the uh Pro the acquisition engine up there. as a custom device, we've got some memory surrounding that, of course, more memory over here. and then the main process of driving the screen and everything else is a um Motorola SL free scale um, Xpc 860 or you know, MPC 860 uh, network communications processor as they call it and well, there's not too much else on there, so you know you got to think that ADC per Channel because I'm pretty sure that's how this uh Tech operates.

I haven't double checked that. So four channels, four Adcs, um, something bridging those together. Main Acquisition engine: Yep, I think that's probably it. Now some good news on this: front end shielding can here.

while it it, while the walls are solded in there, it looks like it's a two-part and I can just probably Bend these and I should be able to lift off that top cover. That's the plan anyway. and so all the uh dividers will still be in there. but yeah, I might have to leave.

Uh, this thing this thing up. yeah, it may not come up easily. Oh, let me do this. There we go.

It is going to come up there gently and not easy. but yeah, it popped out and I should eventually be able to get all that off. Holy hybrids Batman Look at that. I Knew that'd be hybrids in there of uh, some description custom As6 and hybrids.

but jeez, look at that. The whole damn thing is a hybrid module soldered onto the board with these uh, with these connectors around the with the pins around the outside of there. Wow. and yeah, it does look like a C uh Tech Asic in there of course.

uh, it's got matex us I'm not sure Max Tech sorry I'm not sure who they are, but uh yeah, Tech part number on there. They've obviously, uh, rolled their own front end there and it is actually really quite impressive. There might be some RF Voodoo there on the Pcbs as well. Look here.

This is rather fascinating are these distributed element filters which I've looked at in uh previous? RF um, well, videos in um RF units? cuz cuz remember, this is a 500 mahz front end 5 gig sample per second. so have they been doing something special there? But look at that. I mean there's definitely a hybrid module. There's like a hybrid resistor or capacitor over there.

I'm not sure what and what this Arrangement is in here. This multi-layered thing with these you know that looks all the world like uh, you know, some form of distributed element uh, filter system right there and then what's underneath? There seems to be some maybe some printed hybrid components under that. and then they got that happening down there as well. and then they've just got regular, uh, you know, a regular cap down in there, but that is that is really quite interesting.

Look at that. Then they've got these l long carbon. You know, these long like carbon traces on the top there. So this is a really interesting hybrid construction, so that's very, very fascinating.

If anyone's got any details on this uh, Tectronics hybrid, then please leave it in the comments cuz I'm sure a lot of people would be interested in knowing exact details of what's going on here. and you can see the B Andc connector that's the third Channel there popping up into the hybrid. And as I said, well, that's clearly not, uh, broken. And of course it's as I said, uh, solidly connected.

um down into the shazzy down in there. so there's nothing wrong with that connection. otherwise we wouldn't be getting any signal um on this thing whatsoever. So I I don't know if there's anything under the hybrid I Can't know that unless I actually take the hybrid out, but it's unlikely.

It looks like it is flush mounted to the PCB Now I can actually get my torch down in there. Sorry, this doesn't show up on camera, but there is, uh, basically nothing that I can see on the PCB under there. I can see pretty clearly down in there. So on the main: PCB there's nothing under that hybrid module and there doesn't seem to be any um, at least, uh, you know, non-printed uh components on the bottom of that hybrid.

So yeah, that's uh, it looks like all the magic does happen on that hybrid. So hm, what's left? So if that hybrid looks like it's doing absolutely everything on the front end and we're pretty sure our issue is on the front end somewhere, of course, I will check the uh third Channel chip over here. It doesn't look to be anything wrong there. I Mean this is all perfectly reflowed.

You wouldn't expect any issues with, uh, any of you know there's a couple of little miscellaneous devices down in there. little so 23 stuff like that you wouldn't expect anything to happen on the main board, but the thing with these hybrid modules is that um, these are on a ceramic base and these have these are very temperature stable so they you know have a low uh coefficient of expansion versus Heat uh, much larger than PCB. So you've got this mounted onto the PCB which has a different thermal expansion coefficient so you can actually end up with possibly, uh, cracked solder joints on the hybrid. although the big springy uh leads on that would actually take out all of that you would think.

but really I mean that's you know at first guess that's really all I'm left with that. There's some sort of, you know, uh, dodgy joint either on the hybrid module itself or going down to the main. PCB and unfortunately, it's not easy to inspect the S of that cuz these damn metal walls are in the way and I can't take those out without, you know, ripping the whole board out and then desoldering it and it's just horrible. getting out these shields is just awful.

So I don't know I'm just going to have to go around with the microscope and inspect. This is actually Channel Three here I'm looking at so I have to go around and inspect every one of these joints around the outside and see if I can find anything funny I can't obviously see anything you know blowing. As such on that hybrid module, it looks in looks in pretty good Nick as does the main chip up there, but as I said, you know that must be working. it must be doing something otherwise we wouldn't get our main waveform so you know there could just be one dodgy joint somewhere.

oh what could it be? So I'd like to think that it's sort of more on this half of the side towards the ADC up there than it is on. You know, the front end hybrid down here cuz as I said, you know it seems to work and we're getting the correct amplitude and all that sort of stuff. So I think I'd probably be wasting in my time looking around all the bottom half of the hybrid module up here. I think it's you know, probably from here, up to probably up to the main ADC and maybe into this uh, whatever, this um chip is here, it's not too hard to work out where your differential output is on this uh, front end hybrid amp down here.

It's obviously these two caps here coming out and it's a dead giveaway because those two traces are fatter. I.E Controlled impedance and they're run as a differential pair from the hybrid there. Over to the ADC here and check this out. I'm viewing this through my mantis, so please forgive the uh, crudity of the image here.

You can see that Mark that L-shaped mark down in there that looks for all the world like a laser trim. Mark I reckon they've trimmed that and there's a couple of components that that have. You can see that one there has it and this component down here has it. They all have.

well, a lot of them have these little marks on them and uh, yep. I reckon they're a laser trim and check out that one there. Look at that, it's got multiple marks in that long Trace there. Fascinating.

And there we go. There's our solder joints down in there and I'm having a hard time seeing any issues with that at all. It's a, you know, it's a lot brighter and clearer through my mantis scope than what. I'm picking up on the camera here and really, I've looked around all of those joints.

This is on the third module and I can't see anything? There's the hybrid serial number. Woohoo! So yes, I can't see a single thing wrong with any of those joints on the hybrid module there. I'm I Good thing is, you got other hybrid modules to uh, compare it with. just nothing wrong at all.

Nothing looks like it's been blown or stressed or anything like that. Both the soldar joys on the Uh top of the module on the Uh, both on the top of the ceramic substrate there and down on the PCB look okay. I can get a much better result through the Mantis than a bigger, brighter, um, you know, more three-dimensional display than what's you're seeing on camera here and real and really I can't see anything so that's the main Asic there and you know it's not a of course that's not the ADC um that's just the you know, that's just the main uh, front end Asic and that would have a uh differential driver output which then would go over into the ADC and once again, the ADC here. I did notice a couple of maybe little, uh, crusty, uh, you know, a bit of uh, you know, fluff and crap around the PIN of of that presumably ADC up there for Channel 3 and I've sort of brushed that away.

but yeah, I man, I'm not seeing anything wrong with this. damn it. I Was hoping that you know there'd be something really obvious, but there's not. Well, the solder and IR was right there.

So I went around and reheated uh, all of the joints on that Channel 3 hybrid module and well n same result. So just a double check here. What I'm going to do is take out the entire ho board. There we go.

There was just a couple of screws here. I Like the fact that this, uh, holds wh there we go. I Like the fact that uh, you know this comes out as one separate module. you don't have individual little piss ant screws on the front.

That's really nice I Like that and uh, that just comes out as one module. and well, we can have a look on the bottom and well, look, there's really not much on the bottom. There's some ground plane and well, not much else to go wrong. So well I'm left wondering what the hell I can do? Um, you know there's just the hybrid module.

That's it. and I can't even get access to the side unless I really take out all this. um, this shield on the bottom I'd have to suck that out from the multi-layer boorder. It's just not nice at all.

And really I don't see what I have to gain for that and there's nothing obvious on that. uh, third module there at all. and well, I've reflowed the pins and jeez, what's left of you know there's some Gunk around the third Channel chip there I clean that off I don't know. left holding an empty bag.

not sure if you can see that, but the fifth pin from the right there, and the fifth and the sixth and the seventh pin say there seems to be some sort of little. you know, they seem to be bent or have possibly some sort of short between them. Now this is on the presumably ADC chip, but it's not on Channel 3. it's on Channel 4 so could be a completely red hering.

But I'm definitely going to measure that and just clean it up a bit and just make sure that's not shorted well. There you go. the Eev blog Repair Curse strikes again. Found nothing obvious at all in this stupid thing.

So I don't know if you got any good ideas. Uh, please leave it in the comments because well I don't know. There's not much there apart from the hybrid and the ADC Really? I don't know unless I missed it. Maybe somebody will have spotted something watching this in HD or something like that.

So if you have or if you have any better ideas, please share it. Ah well, catch you next time. Um.