Hi Part Three in the HP 35670 a DSA repair Previous two videos will be linked down below. If you haven't seen them last time, we uh, finally found that there were three failed uh, negative voltage Regulators in this thing and it still didn't work. We're getting that ADC Gatoray uh error message ADC ASC on this thing and uh, no, it's most likely as I said before, not the ADC as chip itself. it's I think it's like more likely to be something else surrounding the circuitry in there.

Um, so yeah, we're going to, uh, have a read of the service menual and uh, poke around at the block diagram, see if there's some other stuff we can measure some reference voltages or other things and uh, see what we can come up with and track it down. Hopefully now. I've had a look at the service manual for this thing and that ADC Gatoray error message is actually measured in there and it basically Bally has to do with the uh Source output on this thing. So this output Source here and what I've done is I've actually measured that.

It tells you some uh, you know, basically Turn The Source on but I did this in the previous video and I was getting nothing out of it. Um, and that's basically what it's telling you to do and it says well, if that Source fails there's a uh test point just there just to eliminate that. it's the actually the coax uh, not at fault, but it's giving me like a you know, a fixed like 5 Vols output. There's a test point there, and if that test point is still folding, not giving your uh Source output signal, then why you're going to get your ADC Gat aray error message presumably Anyway, And then what does it say? Um, how it just says, well, this board, which is the A6 board is faulty.

It doesn't tell you anything else. it's just like, well, sorry, faulty board, go and replace the board. You know, because we can't get the schematics for this thing and uh, well, nobody's been able to dig up the schematics anyway. So the service manual just tells you you, you know, eh, that's it.

That's the FY board. So that's all the service manual gets you down to is a board level uh type thing at this sort of stage. So there's our source output down there. and uh, I don't know, You know, There's nothing obvious around there.

so that ADC uh, Gatoray test when you actually run that. obviously it's turning on the source signal and this thing has lots of uh routing build in so that the source signal can actually switch through to the ADC and it can measure all sorts of stuff. So as I said, it's most like, not the ADC, Asic or the ADC failing. it's most likely the source.

Well, we know the source is failing. So really, um, that's where we need to have a look. And here's the block diagram for the A5 board. I think I may have mistakenly said A6 before.

It's actually the A5 is the analog board. the A6 is the digital one above it. So this is the board that it's basically telling us is foldy uh with the ADC uh, Gatoray presumably that's it. It doesn't sell it, it just says ADC Controller but presumably that's the big Actel Fpga, uh, Plcc chip on there and uh, something in here is, uh, giving us a problem Maybe I Mean here's here's all our power supplies which we fixed last time.

You remember we had that, uh, negative, uh 5 and 15 issues but they've been fixed, but it could have taken out something else and what? I See here this looks interesting Here we go: 6 Uh, plus - 6.2 Vol Voltage Reference Used as the reference for the ADC circuits and two test points. So let's get in there. measure that. That's always the first thing.

If you can get in there and measure any sort of voltages, reference voltages, power supply voltages. uh, you know, DC quent voltages or something like that, you should definitely get in there and measure them as a first pass. So there you go, let's have a look at those and I think I can see those on the board here. There they are.

Let's give those a bill and there's our first one thereus 6.5 It's supposed to be It's marked as 6.2 but we've seen this uh before in the previous video where it wasn't as per uh set I Mean it's suspiciously close to spot on 6.5 So uh, you know the exact value may not actually matter the fact that it's um, you know, set to something alth they. You know the firm could have changed so they could have offset it in software or whatever. but 6.5 I'm not going to worry about that at this stage. So let's measure.

so the negative reference is okay. positive one. hello, 3.7 and bouncing around. Well, the whole idea of a voltage reference for an ADC is that it's supposed to be stable is the number one requirement.

Not necessarily accurate, but stable And that's jumping around like a Jack In The Box So something has failed. Let's check it out. Now this is where a schematic would come in pretty darn handy. I mean I have no idea of the layout of this? I Mean here's the 6.2 which we think is uh, fine.

Well, let's just assume it's fine. Anyway, it's at least stable, so I'm happy with that. But that plus 6.2 is the problem there. Now there's a national semiconductor chip here, but it looks like it's got one of those pain in the ass HP part numbers on it.

There's an LF 356 Op Amp over here and that looks like maybe it could be involved in. uh, the. it certainly could be involved in part of the offset part of the reference circuitry. there.

There's a 4053 analog switch? there's that Ron uh chip I was telling you about last time. No idea what that suck is doing at all. but uh oh yeah, we got to at least track down that part number. I Mean there's nothing obviously blown around here.

There's a couple of dodes here I could maybe check, but yeah, I mean just there's nothing obvious. So what do you do? and a couple of seconds on? Google Bought up a cross reference for that HP Part number 1826 0962 It's actually an LFR 412 dual. uh, low noise jfet opamp. so I don't know.

has that been taken out by our power supply? fure Plus - 15 volts certainly could have been. Actually, there's something funny going on here I Just realized that uh I was measuring the wrong uh when I had the wires hanging out measuring the wrong one. If I actually probe the negative well according to the uh silk screen -6.2 Vol one, then I'm getting that positive value that we have before. and if I hand probe the positive 6.2v reference, we're getting that minus 6.5 So that's complete opposite polarity to what's showing on the silk screen.

Unbelievable. And it says TP 413 is negative. Yep, so the silk screen matches up to what's on the block diagram here. So jeez.

I don't know what's going on there at all. You can see how important these voltage reference rails are there if you have a look at that internal layer, see that dark Trace going down there and flowing through there, through there, probably down into this Dck over here, along there, all up into here, all up into all of these Op ANS for all the channels and they'd all be going all over this shop all over here. So these, uh, plus minus references, of course nothing is going to work at all. Um, unless we get these references going.

But but how do we go about fixing it? Well, there's no obvious signs of Fa. There's no like voltage like a uh, you know, voltage reference chip anywhere here that I can uh see. Anyway, not around this part. so um, I'm not sure what's uh, going on there, but anyway, um, you know the LF 412 we've got here has basic, um, you know, high voltage opamps got like absolute maximum rating plusus 18 volts, you know.

So you got to think, well, if this one failed, then pretty much every other opamp with a similar rating could have failed as well. and they're all over the shop. Look at them. just you know.

Dozens and dozens of opamps spread all over not only this board, but the other board. So if the power supply Fa did take out this op amp, well, what else is it taking out? but I Guess there's only one way to find out. Uh, really? I can't do anything else around here without a schematic. We're just Flying Blind Really, it's probably worth sucking that out and uh, and putting a a socket in there and putting in another chip.

I Don't think I have an LF 412 lying around but another opab? uh, it'll at least, uh, test to see if that's an issue. Anyway, there is one thing we'll just do. as a matter. of course, we'll just, uh, take this Plcc chip out.

and I've got one of these Plcc extractors here. If you don't have them, you can get a screwdriver in there and a little flatl of screwdriver in each side and gently lever it up. Alternate between the two sides like that, but hey, we've got the right tool for the job here. We'll push in there, grab this sucker and it's out, and what you want to do is just get in there and uh, just have a look around.

make sure there's no uh, corrosion or growth or anything like that oxidization on the on the pins. they all look, uh, pretty good. You put, you know, if you were really fussy, you could put some uh contact cleaner on that uh sucker just for Just for kicks. but n that looks pretty good to me.

and physically the chip near pin one has a little notch in it and it won't let you install that the wrong way. So here's the uh. here's a little cut out in the socket and it allows us now to fit that in and you just push down like that. and Bob's your uncle.

Make sure it's seated really well. well. I Replaced the Op amp I didn't have an LF 412, but I did have something even better and 8712 which is according to the data sheet, an enhanced replacement for the LF 412. But I still get unfortunately on the screen the ADC Gatoray uh error message.

So let's check the test points. that's says the negative on the silk screen, but that's positive 6.5 and negative 6.5 on the test point. So it looks like we fixed the problem so it looks like that opamp did actually fail. Which is rather disturbing because that means if One Op Amp on the plus- 15 volt rail can fail.

Yeah, what about the other 30 or something that are on there? Oh, and the source output ain't working either. That's supposed to be 10 khz at 1V RMS and we're just getting no, what is it? Uh, well actually um, 15 volts There You Go full scale output. So we fixed our reference voltages, but we haven't fixed anything else. I Mean our source still isn't working and we still got of course the over range on Channel One and Channnel 2 hasn't fixed that.

But as I said, I think those are dedicated uh, circuitry. uh for those work independent of that reference voltage or the analog. well at least the analog to digital uh converter Anyway, so ah jeez, it's not looking good. So without a schematic, what do you do? you be chasing your tail until the cows come home? I Mean it's not like we're getting like a classic uh, short inside one of the opamps or something like that and it's drawing excess current and then we'd be able to narrow it down and we'd be able to find any C corporates and take them out and stuff like that.

No, it's drawing what seems to be typical power consumption, so we can't go around and find which ones are failed. and just on this board, there's probably you know, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12. There's like 15 I don't know, 15 or 16 or more opamps just in here. And then we've got lots of things like uh, adg switches in here multi way and then there's 4,000 series switches and there's all sorts of stuff and probably all of it is running on that.

plus Uh- 15 Vol Rail And we definitely know that uh, We've we've found at least one faulty IC on that 15 volt rail. I Mean there could be I I don't know, right? there could only be like one more or something like or one other thing you know. I could be one step away from fixing this thing. uh, unfortunately, that's where it uh, falls out from underneath your feet because well, Murphy will get you every time.

You just don't know where to go. Now it' be. It'd be a bit easier if I had a schematic. If I had a schematic, you know I'd start by looking around another no and failure mode which is the source output is uh, not working cuz we're getting nothing out of our tp8.

There's a um LP 63 uh sorry, um LM 6321 uh driver buffer there to drive the output, a couple of um, classic Any5 34s in there and all that we've got. Looks like we've got a Analog Devices uh Dack up there and you know a whole bunch of other stuff. so maybe you'd you know start be oh sorry, you didn't see that up there. uh The Analog Devices Dack around there you know you start be looking around all this Source part of it.

but you know without a schematic I don't know I just don't want to be going. uh, you know, sucking and seeing if things work. That's ah, you can't beat Murphy that way. But I do see a test point there Dack output so that's worth a probe at least.

And there we go. That's our Dack output and I can turn the source off and on there we go Source off and uh, channel one there and uh, the green Channel up the top. that's 15 volts. That's actually the output on the front panel B and C so it looks like the dack's working I've got just random noise selected there.

We can choose a periodic chirp and you can see the data changing. Uh, pink noise. There we go and fixed sign. There we go.

we can see it. It's uh, it's changing there, so looks like our DAC output is certainly doing something at the very least. But we're getting nothing out on our BNC connector here. And as I said before, when you run the ADC gate array, uh, test.

it actually turns the source out. Uh, Source on. So let's actually have a look at that ADC gate. uh gate array.

Test. There we go and it's generating Source signal and then says, well, you know ADC Gator Ray failed. Now what I'm doing is I'm actually probing a point further on past the deck. So we've looked at the deck.

we were getting the data out. This is actually labeled low pass filter. Or there's two of these. There's low pass first, low pass filter, and second.

So looking at the first one. So we're at least getting data out of our first low pass filter. but I've got my source set to a sine wave at 1 Hertz And of course, you know. um, as we saw before, that sine wave just looks like it's got a whole bunch of like.

you know, you can see that the period is there. but apart from that, it's just I don't even know what that is. but the frequency's right cuz it's bang on 1 Hertz From there to there 200 milliseconds per division. No worries.

without a schematic, I'm just pushing brown stuff up a hill with a pointy stick pissing up a flag pole. It's just, you know what am I going to do? Start sucking out Op amps and think, oh no, this is this is getting ridiculous I don't know. going to go away, have a head scratch and hopefully someone will find a schematic somewhere which may help. but otherwise this one is not looking good at all folks.

Um, it could be. well, not. Beyond economical repair, but Beyond Time Repair really I mean to suck out everything? You know? who knows what's actually at fault, but it you know it looks to be doing the business like I could solve this uh, source output issue right? I could find out what's going on with this Source circuitry. Okay, we get our source H working, but who's to say, that's the only problem.

Odds are, um, based on what we've seen that this Op amp here, uh, definitely failed then. well, it's it's not looking good. I Expect, uh, some more things to be taken out rather than just the source. So that's it for today.

Sorry guys, still couldn't fix it. But if you can find the schematic, please let me know. And no, it's not in the service manual. I've had quite a few people emailed me the service manual.

Yes, I've got it. Thank you. It hasn't got this commics in it unfortunately. Catch you next time.