Hi Welcome to part Four of the HP 35 670a Dynamic Signal Analyzer Repair video. and if you haven't seen the three previous videos, I'll link them in down below. Otherwise, hey yeah, not following the story may not make much sense, but anyway, this has been a real pain in the ass. There's been lots of red herrings on this one and but we found a few faults on the way and a brief recap is that uh, the power supply on this looks like it failed on the I think the negative Rail and 15 volt Rail and it took out at least one opamp plus three voltage Regulators that's what we found in the previous videos and we've uh, fixed the voltage Regulators So the voltage rails are just fine now in this thing, but we did find a faulty Op amp on the plusus 15 volt Rail And of course, if you find one faulty uh component on a rail which is a voltage rail which has gone over voltage in this case by the looks of it, cuz that's the usual failure mode, you're not going to failure mode on under voltage unless it there's Scr, latchup or something, some bizarre thing like that happening then well, if One Op amp can fail, well, this thing's got about 30 opamps or something 20 or something opamps on the same voltage rail, so that was my guess last time.

Is that possibly uh, a whole bunch of opamps on this uh card which we pretty much nailed well have narrowed down to at least the fault on this analog card in the thing and also the error messages and stuff are basically telling us the same thing, something wrong with this card in it and well, that was my guess that maybe some other opamps are fail but I couldn't get in there because it was all tucked away in this shazzy. But thankfully. um, some viewers have sent in some extender cards. Uh DIN 41612 Connector Extender cards Fantastic! So I'm now able to get the board extended out so I can actually probe it and do the thing.

And one thing we didn't have before either apart from the extended cards was TDA my flare thermal camera. So we're going to use this as one of our main tools of attack this time on this sucker because if these Op Amps are failed, generally they're going to I think we found it failed short inside but I couldn't measure the temperature usually you know, bit of spit on the back of the finger. go around and you can test the chips well only for the temperature, well only if you got access to them. So we now have the card and yeah, I could go around and test them with my finger to make sure uh, that none of the chips are overheating because that's a typical failure mode with uh, chips like this that fail.

they short or go low impedance, internally, draw excess current and they heat up more than they should. So um, that's what we're going to attack First, We may or may not find anything. We'll find out. Let's go.

So yes, some viewers very kindly sent in, uh, various solutions for extending these things out. and we've got a basic extender card from Uh Roth electronic here, and this one is actually quite neat in that you can have a dip switch actually disable every single one of the lines cuz this is a DIN 41612 connector. It's got a whole crapload of data lines, and they're each wired individually across and go through a dip switch, so that's fantastic, although we won't really need that to isolate any signals. Although, if we got really, really desperate deep down into the system architecture of this thing, yeah, something like that may come in handy.

Someone else sent in, uh, these Sony extender cards and these ones I could have you know, cut down the middle, but unfortunately. um, these ones have big Power tracers and shorted out multiple pins. So these ones are actually designed not like a universal type one. They're designed for some particular Sony product.

So I have no idea what that is. Ex51 obviously designed for, uh, troubleshooting and repairing some particular Sony product that uses it. The other solution is of course, a uh cable like this to extend out. There we go.

an IDC ribbon cable to extend out like that. Just got a gender changer on here that's just to protect the pins, but male to female so we can use that one. Although I really do like these cards so this is the one we're going to use today now. of course.

A real trap for young players that you got to watch out for. I Can put this card in here. It extends out a couple in is past here. No problems.

I can plug my board in Prop It Up Off the Bench and Bob's your uncle right? Well what not? Look at this right. We've got through hole components pins right about there look and all these exposed pins you don't want those short in onto your metal shazzy down in here. when this board inevitably bends and touches down. oh what just fell out there? A screw screw just fell out.

What on Earth is that goodness? I Don't know, it's falling apart. it come out of yeah, it came out of here. It was just one of those hold in. yep it looks like one of those studs there.

In fact, that one right there just fell out. no problem. Anyway, you don't want them shorting out to the metal shazzy there. So what I've got is I've just put a stiff cardboard envelope on there, just taped it under so we can safely now just plug this sucker in and then plug our board in here and extend away.

But of course I've got to prop up and that is absolutely perfect. Look perfect height like that. We can extend it. now.

we've got all our test points, our ground points, our probing points, everything We can access the chips. We can check the temperatures on the chips. We can do everything. The access fantastic.

So now we should have no problems. um, at least seeing what's going on here and finding an overview. So as I said before first Port of attack on this I've basically I was able to get like the probe and meter and scope stuff in there before and measure things. I We know that the voltage Regulators are now fine.

We this is a chip which uh, we repaired before there it is down in its socket. That's one of the Op amps that failed on the plusus 15 Volt or 18 Vol rail or something like that. but there's all these other opamps. Look at them.

they're everywhere. I think there's like 20 total just on this board or something or 12 15 I don't know. there's a lot so if one of them can fail I'm hoping that uh, you know the rest of them haven't failed. or maybe that's a good thing.

At least you know finding a fault is good if hey, worst comes to worst. I can actually replace all the Op ANS but gez that' be a hell of a job. so there's only one way to find out that power this thing up and before you even probe it, uh, just go around and look at the temperature of these things. And as I said I can wet the back of my finger, go around test all the Op amps.

but hey, the thermal camera is much better and much more fun. And as before, I'm ping it from my DC Source cuz there was something wrong with the AC Source here. So I've got 15.4 volts which is the maximum voltage of that supply and let me switch this puppy on and it's drawing 5 amps. So there we go.

I'll just let that warm up a bit because uh, if the chips are foldy, you need to bring them up to temperature. don't measure it straight away. otherwise you may not find the C for it. So I'll give that 5 minutes.

We'll come back. The screen yep is booting up just like it did last time. Now just while that's booting up, uh, some of you may be thinking, well, what are the effects of putting these extender cards in here cuz they are very long. You know it's like a foot long or something like that.

30 40 cm long and well, yes, they can have an effect If you've got a system, which is, you know, running at relatively high speeds, you know this can make a huge difference. Although, any well-designed product like this will pretty much have local voltage regulation on here, so it's not like they've been powered from somewhere else. So all the Voltag on this board should be just fine. so the board should just operate fine and this is all.

5V TTL Logic In this particular case, not operating very quick. You know it might be Opera I think 10 mahz maximum something like that. That's really not much to worry about going over an extended board like this, but if you had you know a system at a higher frequency, well, you could be in deep trouble. But generally, if you're using old school uh D 41612 connectors like this, these aren't highspeed connectors you're not going to be operating at.

you know, hundreds of mehz across that bus typically. So yeah, in this case, it's pretty safe. I'm fairly confident that this board will operate and function correctly. uh, with that board in place.

but even if it didn't, even if we were getting data issues and like that at the what we're going to test for first is just to make sure that nothing is heating up on this board and we might test a few clocks and voltage rails and things like that. Um, but as I said, I've done that before and they look to be just fine. but we'll do it again with the extender board. but I'm curious to know if any of the other Op amps on that plus minus rail have failed.

So first thing, get the flare meter out and see if there's any temperature difference. so if they are, if there is a fold in them that has caused low impedance, or a short across the rail, or you know, some sort of latch up fold or something like that, we should be able to see, uh, thermal differences in these chips. Let's see what we can find. So yes, I've got my new Flur E8 and it really is a fantastic tool.

It's booting up at the moment I I Can actually capture video with this I can hook it up to the USB port on the top I believe and and actually get you know, 9 frames per second streaming video out of it. but I W out of that trouble today, I'll just point the camera at the screen. so excuse the crudity of this thing. it's still booting up and uh oh no, there we go.

I got to turn the shutter off and this thing is really, really nice. Oh there we go. We're starting to see some stuff so let's oh yeah, look at that few chips. Oh hello.

Okay, let me set up the camera and the angles properly and let's have a look Now of course one of the issues with this camera is that it's not designed for closeup work. and of course you can actually hack on uh, an external lens on this thing and you can actually get a tool which is on thingy verse that allows you to uh, tweak the lens in there so it can focus at a closer distance. but it's basically uh, 30 cm. That's basically it.

So if we go into the menu here, we can actually settings. Yep, there it is we can I've already got the alignment disc set for the lowest setting of 0.3 M So there we go. We should be able to focus on uh items if we're 30 cm away and the built-in camera with the MSX technology that o that does all that neat overlaying of the real image and the thermal image. there's my hand.

Um, then it will align at if you're at least uh, 30 cm or3 M away from this thing. But jeez, look at all that stuff over there. That part of the board is heating up like a like a torch. look at that.

Wow, that doesn't that doesn't look good. and a couple of other hot spots as well which is rather interesting. So yeah, let me set the camera and give you a good view of this. Now what we're looking at here is the temperature on the screen Here is telling us that is the maximum temperature across every pixel on that screen.

so it ranges from room temperature 23.8 to almost 70 degrees there. And then we can see the individual components. You can see the advantage of the MSX technology. here.

you can see the outline of those chips sorry about the reflection of the Uh lights on the roof. that's a pain in the ass look at that. Oh but anyway, what we can see. We can clearly see a couple of the chips there.

Light like that one there. It's Oh I have to wait for it to calibrate. Look, look at that and then we can see the temperature in the top left there. 40 6 for that little puppy there.

45 There's our Three Voltage Regulators up there 70. That one's red hot. Yeah, 50 Something like that. but that poor little what looks like a A Dip Eight package is already at 40 and a couple of the others are lighting up whereas the ones next to them look are not lighting up at all.

And they look to be the ones that are lighting up look to be the opamps. The 8 Pin opamps that are powered on the rail. The Plusus 15 Volt rail that we know has failed. so that's just on this left hand side of the board, let alone all the way over there on the right hand side.

Oh, this looks awful. So that one there if I take my finger away Tada There it is. That one is an Op amp. and yes, it's an Ad 845.

It should not be getting no Op amp should be getting that hot. 45 Are you kidding me? But do what I'm forgetting here is that these things are Plusus 15 Volt Rails: Hey, that's a 30 volt Supply that's a lot. and well, hey, if this has got a quiescent current of you know, even a milliamp, then you know it's a reasonable amount of power dissipation in this device. So let me check the data sheet for that.

So let's do the proper engineering check on this by looking at the data sheet. Aha, this thing has a quiescent current of around about 10 milliamps that's actually quite large. And when you operate this thing at 30 volts, well. V * I What's the power dissipation in that chip? Just for the quiescent current? Well, you're looking AT3 Watts.

and that's actually quite significant for a Dip 8 package. And if you look further down in the data sheets, down in the all the weird stuff hidden away down in there, you'll find the uh thermal resistance of this dip package. It's different between the dip package and the ceramic package, but this one. the plastic dip package is.

uh, 100 C per watt. So 100 C per watt times our. 3 Watts We'd expect this thing to be 30 C above ambient. So what am I measuring? Yeah, there it is.

Look around about 50 C for that chip there. and well, Bingo that is roughly 30 that 30 above our ambient temperature. So you know you might look around your board and go, aha, an opamp. Look at that.

it's you know it stands out like a dog's hind leg. Look at that Awful. That must be the culprit. No, it's not.

Do your engineering check the data sheet every time, so there's actually nothing wrong with that. It's operating exactly to spec, but wasn't this side of the board lit up like a Christmas tree? Yes! I need to go back and check the other side, but hey, this one was really looked red hot before and a quick check. Let's go over here again. and yep, a lot of the chips are.

Look at that. 60. That one's up to 67 66 67. That one's That's pretty disturbing.

There's a metal can. got to make sure I've got the alignment uh, properly on this. No, I don't I'm not 30 cm away so that's a trap. Let me get it more accurate.

There we go. Yeah, basically up to 68 OD de. but it's pretty much the same. I mean that one chip is a little bit hotter than the others, but the others are still.

probably. you know they? I don't know. it's yeah. there we go.

63 65 I Don't know. it's hard to get this, especially when I'm doing it on camera, but uh yeah, they all look. All those opamps look to be dissipating the same amount of power so they're all foldy. or they have actually just got a high quiescent current operating on that 30 volt rail.

Time to check them too. Well, they're all exactly the same ad uh, 845 so well yeah, I'd expect them all to get quite warm, but they're a good 7 to 10 warmer than the other one on the other side of the board. So either they're uh, you know they're that's just a quiescent. uh Power Of course that's not the active power of it driving anything.

So although it does look suspicious on the thermal camera there, I I Think this is probably right. we're probably getting cuz there's multiple ones here, we're probably getting some, you know, heat spread between them. so the whole area is raising up by a bit more temperature. I mean this thing's been running for, you know, 15 20 minutes or something now.

And yeah, I you know the whole board's probably warming up and just raising the whole uh temperature around the board and they're probably driving uh something as well. So although that did look bad on camera, No. I I Don't suspect those at all anymore now. curiously though, this LM 6321 here, which, which is a uh Power buffer amplifier basically driving the 50 ohm Source I believe it is, then um, yeah, you'd expect that one to get pretty hot.

check the data sheet for that. Sure enough, 20 milliamps quiescent current. so geez, I'd expect that one to be significantly warmer than the others and well, there you go. 46 No, it's very similar.

So yeah, 50. so I would have expected that to be much higher, but it ain't So I've really had a good look around this board and well, nothing stands out at all. The opamps are as per their data sheet values for the quiescent current. the uh Ron uh, chip down in there I expect that to get a little bit warm and uh yeah, it is a bit toasty.

I'm still using my finger even though I've got a Flur camera in front of me. Crazy. And the As6 gets as warm as you'd expect. 55 and that sort of stuff.

But apart from that, nothing really scorching, hot or out of the ordinary. so. well. yep, great tool but didn't find anything in this case.

bummer. And I've had a probe around with the scope on all the various test points cuz there are lots of nice uh test points in here even though I don't know what a lot of them do, but I'm getting. you know the activity I Expect a clock line. You get clock line.

You get clock data there. All the system clocks are working. all the voltage uh Regulators are still working the supplies for all the Op amps and everything. As I said, just fine.

There's no noise on them. no issues whatsoever. So yeah, I don't like that. you know gets back to we looking at the Uh Source before the source wasn't working and trying to track that down.

Well, I'm no closer to that cuz I don't actually have a schematic for this analog board. Uh, somebody did send me the schematic for the main processor board, but unfortunately that doesn't help us with this board that I'm pretty sure is the culate cuz our source is not working and our source is generated from this board. So yeah. I'm uh, going to have to call it quits I'm afraid for uh, today.

once again. sorry. um yeah. no more progress on this.

Um, well, there's progress I Mean not finding an issue is progress. You know we've looked at it thermally with the camera and you know nothing stands out. Although I wouldn't rule out that the Op ANS could be faulty, but they're at least not overheating so that could still be an issue. and well, if it is, then you know we might be screwed.

cuz tracking those down. If it's not a thermal issue, tracking them down is basically you know you got to have the circuits. You got to trace it all the way through. You got to put test signals into this thing, know how it samples, and know how it does everything else, and you know.

Yeah, that's not easy. Uh, and anyway. I hope you found that mildly interesting, if uneventful, just like the previous videos. Oh well, if you want to discuss it, jump on over to the Evev blog.

Forum Link is down below and if you like these sorts of videos, please give them a big thumbs up. Although yeah, the EV blog repair curse strikes. I Never find anything that's either repair. either it's trivial or it's just.

you know, unrepairable or so difficult or whatever to track down. And this one I think is going to be a real dog but I'm not done yet. If I can get the schematic for this uh, analog board, then we will certainly be on our way again.