Hi, welcome to another teardown! This one's been anticipated on the Forum ever since. Uh, somebody on the Forum actually found out about this puppy before it was released. It's the New Agilent Keysight. Look at this.

Look at this ridiculous looking logo. Isn't that the most boring, pathetic looking, bland black logo you've ever seen? Unbelievable. Anyway, this new up Three, Double Four Seven Zero, a seven-and-a-half digit multimeter. Agilent Keysight I have never made actually a Seven a half digit multimeter before.

Apparently they've done these six and a half digit ones with the new Three Double For Six One A series we've seen before. They've also done of course they're famous are eight and a half digit, you know, transfer standard meter. But this is their first entry to the Seven and a half digit market. And of course they're using the same model and form factor as the Three Double Four Six One is their true Volt series, so they're now actually got two models.

There's also the Three Double Four, Six, Five A and which is around about fourteen hundred bucks and it's a six and a half did you want? It's got. Its some improvements over the Three, Double Four, Six One A, which have all these bloody per model numbers. So confusing. Anyway, we've looked at the Three Double Four, Six One ie.

before. there's a new Three Double Four Six Five A which is a bit more expensive and there's a couple of people on the Forum actually who've traded in already. They're almost brand new. Three, Double Four, Six One A for the Six Five A SOHO because it offers some nice advantages for not much extra cost.

Anyway, this one is twenty nine hundred US dollars or there abouts. that's the street price that I found and it's for the seven and a half digit model. So and it's got some nice new features on it which I'm really interested in seeing, particularly the current ranges its own. It still only got the three amp jack on here, but it actually has a one micro amp full scale current range Wow So I can't wait to do a review on this and have a play around with that.

But anyway, that's not what we're doing today. We're going to take a look inside because a lot of people have been asking what is the voltage reference used inside this thing because it's a lot better than the Three Double Four, Six 1a and also the Six 5a which still uses the LM Three Nine nine and we suspect that this one doesn't use the Ln 399 because its specs probably not good enough or the specs are too good to be using that reference which is It's like they quote about us 16 ppm basic DC volts accuracy or something for this thing. Pretty kick-ass meter and it's reasonable value for a seven-and-a-half digit meter at that. You know, almost three grand price point.

Yeah, it's not, sort of, you know, hobbyist or or your basic lab pricing, but hey, you know when you're paying for seven half digit meter, you're paying for the huge performance in the thing. And as you can see, practically identical to the three Double for Six One ie. here. So nothing is really changed except this one does have auto Cow.

you can go shift auto Cal Like that that one the Six One A doesn't have that. But apart from that, everything looks identical and absolutely no difference on the back either. If I swap those over, you wouldn't have even known which ones which so you know we say you're on the eevblog, don't turn it on, take it apart and here we go. Of course we expect it to look near identical to the previous one.

Three, Double Four, Six One A. Tada, we're in like Flynn look at that. Isn't that really neat and tidy? I've mentioned this before with the three, Double Four, Six one a teardown, but that is just yeah. That is just gorgeous construction, well thought out and nice tidy wiring on their mains transformer there got a good old fashioned linear one there.

I Like that, although there is a switch mode down in there. Hmm, let's we'll get into that. But yeah, that's all the magic is going to happen. I'm yeah, I think I might see the magic under the shield.

here. there's some sort of plastic holder that's got to be the reference. That's got to be it. Absolutely So Yeah, let me try and take that off and we'll be even like Linkous Everything looks pretty much the same as what we've got last time, so I probably won't go through all the details.

It looks near-identical How nice. Huge HRC Fuses over here. Absolutely massive. They're fairly easily user replaceable.

There's only three screws to get this case cover off. There is a bit of a trap for young players these little clips here. you've got to sort of like pull this out a bit before the case will actually pull off, so it's a bit annoying and not obvious at first. Anyway, see about brand fuses in there, so have spared no expense.

We've got our Tellurium that exactly the same are custom-molded tellurium, copper contacts, low thermal EMF contacts front and rear of course, and the big physical clunking switching here going into this gang switch down here, which then what's not gang switch? it's a just a multi multi way switch down in there, which are switches between the front and rear panel terminals and everything looks very similar. If memory serves me correctly, of course, I'll have all the high-res photos available on Eevblog comm and you'll be able to compare it with the three Double four, Six, one I got ourselves a nice spark gap over there noise isolation slots Anyway, that is just gorgeous and I Also loved how my point is gone I Loved how they have these little hooks in here. The board just starts, slides into these hooks and then just slides forward and out of the case. It is very, very nice, but that's what everyone wants to see.

That's the money shot under there. Let's get to it. Oh Check this out. This is really interesting.

This is more than what I expected in this thing. I Expected like a just a linear tech LTC 1000 volt Aegean Sand We may still have that, but look at this like module here. this is like you can bet I Don't know what's under here, but you can bet your bottom dollar this is the reference. They don't go to this amount of trouble to make a second daughter board like this with some custom plastic carrier like this and a header on top.

And just for you know, some miscellaneous thing, this is definitely the voltage reference, which they have. no doubt The reason they've put on the second board is so that they can maybe yacht thermally aged. These things are separately and test and characterize each individual one as a separate process to building the multimeter itself, and this is really interesting. They've got an 8 pin header on there, so that's obviously how they plug into their our production test jig for these things.

So whether or not that's I it's likely to be analog. you know, just like you know power. So I can supply power to the theme from the from the production chest SG They probably got a huge board. they plug like a hundred of these things.

Maybe you know, upside down or something. Perhaps I don't know. It doesn't matter. Anyway, they plug them in.

It might have to be upside down because this head is on the top, so you might have to actually flip it upside down to put it inside the inside on onto the production art test board. Anyway, regardless of the physical arrangement, they probably got like a hundred of these. put them on a big tray, and then they maybe put them in a big thermal chamber and they can then read out and log, no doubt data from each individual one. So, but this opens the possible that maybe is their digital compensation on here? perhaps? Mmm, where they can actually software trim this thing, but that's probably unlikely.

I'm getting a bit too excited there because that would that would be a bit too much because you don't need to go to that sort of effort for a voltage reference on here. As I've always said with voltage reference is is all about the stability. The absolute accuracy of the voltage reference does not matter. This voltage reference board could be.

you know, one or two percent accurate absolute accurate. It doesn't matter whether it's you know ten point, one volts or ten volts because when you put it in to the multimeter, you can software calibrate that value out. It's all about the drift. In this case, it's a sixteen ppm accurate one.

or I think it's a point five ppm per degrees C temperature drift I Have to double check that. So let's take a brief look at the specs here because it's very worthwhile. Look at this. it's best spec here is going to be on is on the ten volt range here.

and like usually like on cheaper meters, they might use like a two hundred milli volt reference or or something like that. So the best DC volts accuracy is going to be on your millivolt range. but this is likely using a 10 volt reference in here. so it's no surprise for finding our best spec here.

Absolute spec is on our 10 volt range for DC volts. And of course for any multimeter that's the DC voltage range which is going to have the best spec. Current is going to be lower AC voltage is going to be worse. Ac voltage is going to be worse resistance current.

All that sort of stuff at DC Volts is where the business is that and this is how they get the banner spec of 16 ppm and you can see where they get their been a spec of 16 ppm here at point Troopa low 1 6 % absolute after one year for a plus minus 5 degree differential from the temperature that it was calibrated at. But look at the 24 hour spec here this 8 ppm who absolutely are Now we're talking that's the percentage of the reading plus the percentage of the range. you're going to have that 2 ppm percentage of the range. When you're down there counting you, you know, counting your least significant digits.

that's all going to add up and matter. But anyway, this one is like only like a transfer standard spec because it's within plus minus 1 degree C of when it was calibrated. So you usually can't like ship the thing within 24 hours and then use it somewhere else. So that's really like within the same lab standard, but even something more practical here.

Look a 90 day spec, it's still 13 ppm salut and then that goes up to 16 ppm after a year, 20 ppm after 2 years. So very, very nice. Now here's one of the interesting things. the Auto Cow feature which we saw on the front panel before.

What it does is it compensates well improves the temperature coefficient, the temperature drift of this thing. So without the Auto Cal feature ie. just like the three double Four, Six, one A, for example, then it's going to drift by 5 ppm per degree C of the reading plus 1 ppm of the full scale range as well. And that's basically either doesn't matter whether it's within this plus minus 5 degrees or outside it, it's just going to always drift by that amount.

But you turn Auto Cowl on. Bingo! It drops that from 5 ppm down to 1 ppm plus one down here of course. and it's a bit worse down here. probably as you'd expect on the hundred millivolt range there.

But yeah, it gives you down to 1 ppm. So you basically with the Auto Cow feature on, you're looking at sort of like a worst case of that 2 ppm per degree C temperature drift within the operational range of the instrument. But as with all datasheet values like this from a reputable manufacturer like Keysight who really know what they're doing and do take a care with this stuff, these specs are going to be conservative, so I'm pretty sure it's actually going to be a better temperature coefficient than what you get here. This is just what is guaranteed.

So yeah, it could certainly be better. One thing I really like though, is that they do give you a two-year spec for plus minus 5 degrees. A lot of companies won't give you a two-year spec on the thing. they'll even give you a year or some only give you like a 90-day spec.

In fact, I think HP Used to do that back in the day on some of them heaters that only give you a 90-day spec, but they're pretty confident on this sucker. So they've given you a two-year spec. Very nice 20 ppm class instrument. After 2 years switch on the auto.

Cal It compensates for the temperature coefficient in that thing, so it must have a good temperature sensor inside this thing. It must be pretty schmick. We might find that on the board there somewhere perhaps. I'm wondering, is that puppy there a temperature sensor 3-pin TI 92 package could very well be, although it is labeled Qq1 there.

So hmm. Q is for transistor. Ha. It's a bottom entry socket.

That's the thing I was talking about before how they would put these on a big production test panel and they might have to flip it upside down. No, this is a bottom entry socket. So if I just Oh Wiggles look at that. I'm wiggling the voltage reference.

If I just pull on that, it should just pull up. Yep, there we go because we've got our pin header on the board and that just fits neatly around that crystal. There they've designed that in Matata there is L Voltage reference is not much on there at all, so I suspect that's just like an LTC 1000 volt. adjourns.

Let's take that plastic clip off and see if we're right and I was right there you go. LT said 1000. No surprises whatsoever for finding that in there The day code on that the 27th week 2013. So that's a reasonably old beast.

Anyway, it is the LTZ 1000 A CH I'll link it in down below. It's one of the best voltage references you can get. And to get the specs in this thing, it couldn't have used anything else. Basically, it couldn't have used your traditional LM mark 3, 9, 9 which we're so used to seeing in you know, six and a half digit meters.

It's been the standard in six ARF digit meters for like ever. So yeah, no surprises. It's still in the new model of the What Is that? the three, double four, Six, Five A, as well as the six one A we've seen before. But anyway, we have ourselves a linear tech.

Probably that's just a Op-amp Is it? Yeah, A 1013 I Think We just got ourselves an Op-amp there and a couple of our miscellaneous parts that's just forming the voltage reference. Probably just based on the application note from the LTZ 1000 and that's all she wrote. There's nothing on the bottom here. We've got our holes for our now header to come through, so that's quite clever.

So as I said, there's a big big production test board. No doubt that individually, um, soaks each one of these. and of course, each one is individually serial numbered. There we go: Three, Double Four, Seven, Zero A Specifically designed for this.

Any case you're wondering, Well, it looks like they had a looks like they had some pads on there. Maybe for a bypass cap or something. They decided not to fit. anyway.

in case you're wondering, what these slots around here are This big one here and around like that that is to take the stress off the leads. That's for thermal expansion of the board of the PCB material looks like it's standard. You know, if our four class fiberglass board may not be, it may have a lower thermal expansion coefficient perhaps. but that's what they're trying to do there because any expansion of this bore, they're just trying to take out the stress from the leads.

And that's one technique of doing moons like that. Often you'd like put it on like a cutout right on the edge of the board for example. Either way, you know works. I Think the this one's probably not quite as efficient as the other way just having like it right on the edge of the board and then cutting a slot out like that.

But anyway, it's obviously going to do the business and that's what it's there for because any stress on those leads, any mechanical stress on those leads translates into a stress inside the die. I mean we're really down to the physics level here and it stress on the die and there or whatever and it can cause the thing to drift. So these things are very, very susceptible. And of course, the reason that they've designed this custom little case is to keep that nice and toasty inside.

They're not not toasty, but keep it at a constant temperature so that in your unit like this because it's got a fan inside it. Okay, and there's air blowing through because hey, we have to cool down the other parts and stuff like that. You don't want that airflow going over your voltage reference down in here. That's just going to ruin your day.

So you don't want that turbulent airflow blowing over that thing. So that's why they've done it. They've put it in its own little custom plastic and clip and that one clips on the bottom and it is all very, very nice and well designed. That's brilliant.

Look at that. that one goes over there and there's a screw to hold it all together. Ah, they know what they're doing. Agilent Keysight and there you go.

There's the money shot for you volt nuts. Lt's Ed 1000 for the win. It's a 2n 3904 NPN transistor. Oh, what a letdown.

But hey, actually with hindsight, it's not that surprising because they need a drive transistor to drive the heater inside this puppy because this is a a heated reference. There's a heating element in there, which I'll show you in a second. hence why they put it inside here. and so that's the that's the idea behind that.

So let's take a look at the application note and they're almost certainly running exactly the same application. Notice our LT are supplying in their datasheet. Let's take a look. So this is what we've got inside the LT said 1000.

We've got a couple of our transistors, the Buried Xena reference of course, that's that's doing all the main magic here, and a heater element shown like this. this. is just that substrate diodes they tell you there, so they're just, yeah, not actually a physical diode as such, but that's just, um, clamping as part of the resistive element in there. So that's basically all there is in there.

And this puppy is capable of look a temperature drifts of better than 0.03 ppm per degree C and Longstone term stability 1 microvolt per month, and it's about 0.1 5 ppm. Noise is also obtained, so that's you know. the noise figure is basically what edgelet might be. Keysight might be telling us here with the spec, because it's the reference he using is capable of better than the speck they've got here, But hey, that's not uncommon for high-end manufacturers like our keysight to over spec their gear.

It's sort of like a worst-case guaranteed figure they put in the datasheet, so the reference is certainly capable of better than what they've had here. In their reference of 0.1 ppm. it's capable of almost an order of magnitude better than that. And no surprise is that the application node here has exactly the same chip.

We've got the LTC 101:3 It's a joule precision reference here. There's our 2 and 3904 which is driving all of the current for our header element that takes a significant amount of our current I'm not sure what it is, but yeah, it's not trivial, hence why they've got the probably the Tio 92 a package there and also have isolated it outside here. because you don't want that device getting warm inside here? That would be a big no-no So you want that out there where the airflow, the main airflow you've designed into your product can actually take that? You know that that heat away can dissipate if you put it under there. Even that would have been a bit of a fail.

So yes, I Have no doubt that this circuit matches theirs pretty close to precisely, because hey, why would you do it? You know, why would you try and do it against what linear technology? one of the masters in the field has spent a lot of time and effort building the one of the world's best voltage references here and getting and characterizing a circuit. Well, you just use it and go. Yep, of course you'd do your own tests and performance checks on it. I'm shocked aside, have done that.

They're not just going to blindly build this up and go. Oh yeah, she'll be right. no worries, whacking in there and off to market. We go.

No, they would have seriously are characterized the design of this thing. and as I said, they've really thought about it. From a a thermal airflow point of view and thermal stress on the board and everything, they've really done it right. That's a brilliant example of how to design a nice precision reference, so it would be interesting to know exactly what burning they do on these, because I'm sure that they do so.

And of course, there's no trimmers here. As I said, they're not worried about trimming this to an absolute value. It doesn't matter because that's taken care of in the software calibration of the meter itself, so that capability is already built into the multimeter. There's no point trying to get an absolute reference here, so this could be a couple of percent off.

Doesn't matter, it's all about the drift. It'll be interesting no, and maybe a key so I can get back to us about, um, what sort of production testing? Maybe if they've even got some photos, they can share their production test facility for these things or something like that. I'm sure they work them in the thermal chamber and characterize each one. That's your bottom door.

Okay, so a web boy with any more details of the teardown in here because I've done it all before in the three Double Four, Six Zero, a video which I'll link in down below, and I've done like a little Dave CAD explanations of how there are multi slope, our converter works, and all that sort of stuff. So there's some real interesting info in the previous art teardown because it's an identical design basically, and you'll see in a second. I've taken some high-res photos of both of these boards as aligned as I possibly can and well, let's go to the videotape. Alright, here we go.

Let's compare the two boards and this is the original one. This is the Three Double Four, Six One A. You can see the Agilent logo up in the top right corner here and you can see that this here's the part number Three, Double Four, Six Zero A which is Six Zero, Six One and it's a Rev Four Board. Now look at this.

I can switch between them is fantastic. I Love it and I could play with this all day long. Look at this, you can see that there's hardly any differences at all. It's terrific.

But anyway, here's the new Keysight one. This is the Three Double Four Six Five so the same PCB is shared between the Six Five model and the Seven Zero, which we're tearing down here once again. Rev Oh Four Again, if you have a look at that so you'll notice that one of the first thing you notice is that well the relays have changed. change to yellow Whew! Excellent arm.

they've changed brands have from Omron but look, there's a missing relay on the old Six60 aboard. I'll call it and the relay is populated. And also there's another relay here and two So8 packages here as well. and you'll see that they just pop in there and so maybe they had the design of the you know, they were thinking about this right back when they design the Uh-60 and Six-one so it looks like they've had the capability in there.

But a few minor things have changed. Look the keysight one up here, where the cursor is. They've added a couple of parts up there, so it's squeezed them in there and they've changed the layout of this. You see how on the old one, they've got the large footprint for that oscillator there.

But then they decided our, well, no, we're going on this smaller package and they've changed it and they've sort of just shuffled a few parts around there around the analog to digital converter section around here, which is based on the latest part. The latest part is the exactly the same device. It's the LF x P2 - 5e, but you'll notice that look, there's a part that's a resistor. They've moved that, but I think something else is added in there now.

Anyway, there's minor differences in the tracking and things like that, but G's is not not much at all. difference. Check out this down here in the bottom left corner. Look, This R102 here is no.

it's mounted upside down. all the electrons are going to fall out. oops. But it's look.

It's everything's pretty identical. It's got all the same input protection, they still don't have that second mob fitted there, and this precision resistor are four one six is still here. These two ceramic packages here and here are still the same. All the analog to digital converter is all still the same.

They're just trying to squeeze out some extra resolution from it, of course, which they can because the meters always had that extra resolution in there. If you see my review last time, it actually does. A software calculation gives a single precision floating point output value which is like M equivalent to like eight-and-a-half digits or or something like that. It's more than the quoted six-and-a-half digits that was on the previous unit, so it was obviously good enough for this seven and a half digit meter.

And you'll notice here we go. Here's the change. Here's the voltage reference the LM three, nine, nine in there and you'll notice that they've basically just ripped out the LM 399. Even that look, even that capacitor in there and that diode in that twenty three package they're They've basically just ripped out the Ln 1309 and replaced it with the LT said 1000.

So there you go. Well, look, there's an extra resistor there. look R 310 maybe? I Don't know, Is that a current shunt for the one micro amp range? That's the other thing. I Haven't found how they're doing the one micro out range because that's two orders of magnitude better than the Six Zero model here, which only went down to a hundred micro amps full-scale range.

So this one can do one micro amp. How is it doing it? I Don't know. Everything looks identical on the bottom side of the board or the current shunts look the same there. So I'm not sure what can anyone spot it? Can anyone spot it? I don't know.

Anyway, we've got our high voltage resistor Hybrid over here. they've got a high voltage Network here and everything looks pretty identical. So oh, here we go. No, sorry, never no look here.

Well they've changed the type of current shunt resistor used here. Look are 231 here. You see how it's like that's very similar to one I use in my micro current and very similar as sort of designed. but look, they've changed that to a much larger package.

Now it's only a 1% value. but as I said, it doesn't need to be precise because it's software calibrated out. It would be incredibly low drift. and that's a that's a weird-ass 4-pin package.

You'll notice like surface mount. It's still a surface mount package, but it's a much bigger physically. much bigger so I can hear. It has a greater power dissipation in it, but it's still a 4 terminal current shunt resistor in there.

So maybe they've changed that and maybe you're amplifying. Maybe that's maybe those two there are an amplifier perhaps with you know that boosts it up I'm gonna have to have a look. Oh I boot. What does that mean? Current boot.

Hmm. interesting. Hi Yes, my hunch was correct these devices here that they've added two extra devices here. you'll notice I can I sue min on that? No, I can't But you'll notice that there's a guard ring around that trace there and whenever you see guard rings around there sorry about the small cursory or I probably should change to a really large cursor so that you can see it.

but when you see guard ring like that, you know it's a sensitive node. So we're talking. you know, low current stuff into there. So bingo this chip.

You probably can't read it there, but it is none other than a Analog Devices 8080, 638 Bingo. Exactly what you need for. Like, it's like my microcurrent capability. It's got an auto zero rail-to-rail Op-amp in it.

ie. it's got zero DC offset in it. There's zero offset voltage and that's exactly what you need for measuring down at that are sort of. You know that low end.

That's This is how I'm sure that they're getting their one my can't range and they've got two of those in there. Actually, if we go back here, we'll see that this one here, it's still fitted. That one is still one of those devices, but they've fitted another one in here. You - OH - there I think you 203 is just a TL O7.

While it's just a Joe Bloggs Op amp, nothing fancy going on there at all. But yeah, I I Actually looked at using this in my microcurrent at one stage. I think and this one's a pretty big Op-amp in very low input bias current 40 40 puffs. And what's the typical low offset voltage? nine micro volts maximum.

So it's actually not as good as my microcurrent, but that doesn't necessarily matter, it's going to be the biggest once again, they can calibrate that out so it's the offset drift here which is going to matter. So it's 0.04 micro volts per degree C maximum. So that's how they're probably getting using this device to get their 1 micro amp full scale range and they can still get presumably there's 7 1/2 digit precision on that. but if we, if you actually went back and checked out the spec, the drift spec would actually be higher for the current ranges.

it's just other little tiny changes here. And notice, just under the lattice chip here, you'll notice how they've just changed the just rotated that component designator our 808 there so somebody the PCB designer was just fussing around I did I think I'd rather have it in that orientation. That's a bit neater though. we're just you know, gilding the lily there a bit and like down here around this in 94 V-0 they've added a dash in there for some reason, who knows, and but like, apart from this, like this bugger, all difference aren't really really surprised.

And the other noticeable difference here is what looks like a crystal there, but it's not actually sorry. I've only got the top down photo here, but it's actually a Vishay Zed foil precision resistor. so it's a 10k precision resistor. This one was not fitted.

as you can see, it was not fitted on the original 6:06 one, but it is fitted so it's our 324 there. But it's yet a really precision 10k resistor. so why they've added that in? Perhaps it has something to do with the current shunts, but it seems like yeah, it's just out of the way tucked over there. so I'm not sure why they've put the extra one in there.

Anyway, it is a very mik resistor so very extremely low art tempo. It's one of EA's our best ones. It's the Vh 102 RZ T. You'll notice that look here.

they've changed manufacturer of that. What is it to the UM What Is it? A seven four? It's a HC four, Oh Five three analog. MUX You'll notice that the change manufacturers from Ti to somewhere else. having zoomed in on that one can't see detail.

But but obviously they therefore like they are not critical parts. they're just, you know, jelly bean parts. So really it doesn't matter what that manufacturer you're put in there pretty much although they would have characterized it, specified it and put it in their bill of materials as that is an authorized you know a replacement part. So you know a company the with the reputation of Agilent they wouldn't just get whatever they can get from the Shenzhen market this week.

It doesn't work like that in these kind of things, but you'll notice that the in with that time amplifier over there I still don't know it that I boot is so yeah. I don't know what that means. It's been removed here. I Think on the new unit, so that's rather interesting.

and they've added that extra resistor there as well. So they certainly have changed the current range configuration because they've got two additional current ranges as a 10 micro and the one micro amp. so you expect a few changes around there. But yeah, basically identical.

Very, very interesting. So they haven't added a huge amount in here for the price difference. What? They're basically chart. You know they've got a couple of extra Op amps in here.

Okay, fine and dandy, but you're basically paying for the LTZ 1000 reference and the characterization of that reference. as I said, they probably go to a lot of effort to, you know, thermal, individually, thermal chamber, test each one of those, etc. So yeah, that's and here's another interesting thing. Look, you'll notice just around that chip there that has a thermal pad on the bottom.

that's why it's that gold color it's got exposed a copy there. They're using that as a little heatsink. It's got some got some solder. it looks like they had excess solder paste or it's flow through the videos from the bottom or something.

so there may be maybe change. There are pasted. Well, they had to have changed their paste stencil because it is. There are a few differences in the layout, so instantly have to go to a new solder paste stencil.

but maybe they've applied some some extra paste on there and it's just flowed out anyway. Just an interesting little insight there. But yep, everything looks pretty much damn well identical. I Love it all.

The value is in the voltage reference and if we have a look at the bottom of the board I won't do the same hi Reza Photo thing here once again. teardown photos are on Eevblog Comm. If you want to have a look, this is A actually I've lost track. This is the old one.

the six one A and this is the new Six Six seven, eight. Basically identical except look I Can actually see this one actually has an extra resistor populated in there where this one doesn't But yeah, I don't Maybe a few little miscellaneous differences, but Bayes's big can't shut resistor up. There is exactly the same from Dale Yep, it's basically identical. So there you go.

I Hope you enjoyed that teardown of the new Keysight! 3 double for seven zero a So thanks for Keysight for getting this to me. It was a bit late I Was hoping to get this before it was released, but yeah, there was a few shipment delay problems. but thank you very much Agilent Very, very interesting and absolutely brilliant First-class quality bench multimeter. It's it's almost impossible to fault this thing.

Really fantastic. So I hope you enjoyed that. If you did, please give it a big thumbs up. And as always, if you want to discuss it, jump on over to the Eevblog forum link down below or leave youtube comments or blog comments.

Catch you next time. So the longer your capacitor takes to charge up, the greater your resolution. But therefore, you're trading off measurement resolution versus speed. But a modern high-end bench multimeter like this Agilent, or even the previous model or the previous model before that, or before that? Um, they don't use just basic run-of-the-mill Joule slope integration techniques.

They use what's called multi slope integrating ADC and it works exactly the same way. It's just that it has some extra switching in here. You.