Hi In a previous video which I'll link it at the end and down below if you haven't seen it. we looked at the advantages of using the vernier control on your oscilloscope. This is when your signal might be too big to display on your screen like this and therefore you're not able to. It's outside the ADC range of your converter front-end and it can't do measurements on that.

And if you turn it down like this, you're wasting a lot of your analog to digital converter range on there. and you may not get as good a resolution or accuracy on your automated measurements and things like that, so you would. What you do is you hit the vernier control like this: a fine adjustment mode. It says variable down there on the scope and you can bring it so the waveforms just within full scale range like that and you take advantage maximum advantage of your analog to digital converter range.

All those eight bits that you've got available if you've got a fancy pantsy scope, ten bits or whatever, so that's just a handy little tip to get that. But a lot of people ask the valid question: do these modern scopes when you actually put it in vernier mode like this? So is this just like a software trick? Does it actually do? The vernier adjustment doesn't actually change the gain in the hardware on the front end. And hey, let's answer that question right now by doing a teardown and some probing to find out if a modern scope like this siglent SDS 1104 XC actually does do real Hardware gain control just like the old school analog scopes. So we showed in the previous video that there you can.

The results you actually get from doing this do seem to be real. so it does seem to be, you know, doing something useful. But is it actually doing that in the hardware or is it some sort of you know software trick And then it's It's giving you some extra, you know, warm and fuzzy software resolution where it really shouldn't be there? Well, it's only one way to find out. Its teardown, a scope and probe up.

Its clacker, right? So I have actually taken apart this poor victims 11:04 XE and I have probed right up. It's clicker and we can actually see and probe the waveforms of the vertical gain amplifier front-end But let's take a quick look at the teardown of the front end of this thing and see what Actually if there is a digitally variable adjustable gain type circuit or chip in there. Alright, so let's take a look inside the front end here. this is from my tear down the photos I have to link given the teardown video.

if you haven't seen that, this is the bottom side of the analogue front end. It's shielded as you can see, there's just a bunch of and diodes on there and a bunch of passives. Not much else so it ain't there. but if we have a look at the top side here and it comes in on the left, that's the B and C there and it's got the requisite relays and some trimmer caps, we can tweak it.

But if we have a look at some of the chippies in here, this one up the top is a Sim for HC V 9 v serial to parallel interface so that they can. they don't have to have all the lines coming from the main controller over to the front end. Anyway, what we're interested in is this one here. Look at this: 83 70 that's an Analog Devices 83 a 70.

Let's have a look here it is and a low frequency. the Seven 50 Megahertz digitally controlled VGA That's not video graphics that after that's a variable gain amplifier and it's got programmable low and high gain. listen to DB resolution so you can adjust from Minus 11 DB to + 17 DB gain two different ranges 6 DB 234 TB So I can go to Minus 11 to Plus 34 DB gain and it's a differential input differential output and as you can see it's got an adjustable preamp here and adjustable art transconductance which is the gain here and it claims to be have like precision gain range. and it's got a serial 8-bit digital interface which we're going to tap into to see and that's basically all it is is.

it's a digitally controlled or digital gain controlled amplifier perfect for differential ADC drivers and oscilloscopes and stuff like that. Surprise, they don't actually have a saw scope there as one of the functional of things for it anyway. a low-cost digitally controlled variable gain amplifier that provides precision gain control, low noise, excellent distortion performance, white bear with for modern receiver designs, etc, etc. And here's the magic word.

A Vanya seven bit transconductance stage provides 28 DB gain range at better than 2 DB resolution and 22 DB of gain better than 1 DB resolution. So you can. The software can potentially adjust the gain of the front end that goes into the ADC in one DB resolution steps. So as you adjust that Vernier that fine control.

In theory, if the software supports it, it can adjust this, but we won't know until we actually measure it whether or not it's actually sending the codes using the Vernier. It's obviously going to be using these particular different gain stages for the different fixed millivolt ranges. you know, one volt per division, 100 millivolts per division, 10 millivolts per division, etc. It's going to be switching all those ranges, but does it do it on the Vernier? It's more than capable with this chip.

Theory of operation is fabricated with the 25 gig Silicon Bipolar process for those playing along at home and we're seeing the block our tech architecture. This transconductance stage is digitally programmable gain and this is quite complex actually how it has different performance characteristics depending on the particular range that you're in less than one day. by resolution. listen to DB resolution.

It's got two different gain stages and then a programmable transconductance amplifier inside. That and the gains actually are load dependent to for those playing along at home. But here's the digital interface. It's a simple 8-bit digital interface.

It just accepts a single word here. by looks of it, that's it. There's a latch signal that goes low When that does, it has a clock and just a data stream on the input starting with most significant bit to least significant bit. so it looks like you only feed in the eight bits and that's it.

You can configure all the gain stages. There's a typical example application and here is the gain code that we're interested in: I Expected to find like a table of like okay this is one DB for each bit or whatever, but it's actually a formula here which you've got to put in. The total gain is the game code with the Vernier plus a pre game like the little most significant bit and it actually you know it comes out at X amount X point X X X amount are volts per volt. So obviously you've got to calibrate these ranges in software and that's what Siglent would do at the factory.

or you could do in some sort of that user calibration or something like that. But once you've done the fixed ranges then the other Rangers would be knowing steps so you wouldn't have to recalibrate that. But yeah, that's what it does. You just put in the value of your register.

the first most significant bit is whether or not you're in the high gain range or the logo range, but after that it's a seven bit or a hundred and twenty-eight step a gain stage. So yeah, it's more than capable of doing this. So let's hook up the scope and see if it actually does adjust the gain of this thing in the Vernier stage right? So now comes probing this thing. Unfortunately, it's a bit of a pain in the butt because it's an 0.65 millimeter T Sob package and that's you know, too small to get your traditional like easy hooks in there on the individual pins like you can on a regular Esso package so that's not going to cut the mustard.

and if you just got the one probe of course you can get in there and touch adjust if you hold the tongue at the right angle. but then if you slip you got to short the pins out and you can potentially use like a like a stand like this. one of these flexi stands. you can put your probes in here like this and you can muck around and you can precision locate them on there.

but then they spring back. but they you know you can do that kind of thing. But then the problem is we've got to operate the control. So even the most minut thing in this, let alone getting three of these probes on there.

That's just going to be a complete no-show so we've got no option but to actually solder some wires onto there and then you put some strain relief on here because usually those joints aren't going to be that strong, especially your lights. Really difficult to get in there and the sight. You know the joints might be a bit how you're doing, but as long as they're touching so we just take the stress off those, then we just has some fly wires coming out. we can touch with our regular probes either scope probes or logic analyzer probes.

Beauty Got three signals: channel ones data channel twos o'clock Channel three is the latch signal and I've got it set up for normal triggering here. You don't want it in Auto mode because then you'll just continually get your signals and you don't want to always be single-shot capturing this so you want normal modes so that you can run it. and then every time you get a signal like this, bingo It triggers like that and we can see it over here. So every time we change our volts per division setting, we will get a yo pro.

You can see it change a little bit over there, but every time we change it we will get a new trigger over there and ordinarily it's not actually triggering I'm actually triggering from the latch signal. So when it goes negative, that's the only time that the data is actually valid to that particular chip. So you can see here that there's multiple packets of data and clock here, but only one of them is going to be accepted by that chip when that latch goes negative. So that's the one we want in there.

You can see it stays low for a long period of time, but it's only sending one little packet of data to that particular chip. The reason is there's all this other stuff which we ignore over here. In fact, let's go right out on the time-base Seven different packets there every time we change the volts per division setting. So it must be updating all of the different channels even though we're only changing channel one.

So that may not be the smartest thing to do in software. but the chips that aren't latch will just ignore this. So obviously there's multiple chips on that same bus and that's you. Do it.

That's why you got the latch line. So I'd say it's up. Yeah, it's probably sending out data to all the chips and that just saves lines, of course. So, and you don't have to have a discrete set of outlines going to each chip.

So there it is. there's our data, there's our clock going into that chip. and if I then change the volts per division. you can see our data is definitely changing.

So that's it. Sorry you can't see it on the screen. Take my word for it. 10 volts per division, 5 volts, 2 volts, 1 volt, 500 millivolts, 200 millivolts, 150 millivolts, 20 millivolts, 10 millivolts, 5 millivolts - and 1 millivolt I Know we'd get a 500 mic.

You can see that there's no difference and now actually switch in between 1 millivolt and 500 micro volts per division. You can see that there's absolutely no difference in the data whatsoever, so we're not getting any extra gain there. All we're doing is software magnifying that in software. So it's not a true 500 micro volt scales and what we want.

What we want to know is ok. The data changes I Don't care what the data is. Ok, it's in there. It's choosing the particular register settings.

As we change the millivolts per division, it's choosing one of those fixed gain settings that it's actually calibrated for, which will match the range of the analog to digital converter as best it can. But what we want to do is if we press the Vernier here, what happens? So I'm going to put it fine. Adjustment Mode: Oh yeah, 494. You see, it's updating every time.

Oh, for 8 - look at that. it's jumping. It's changing. Its changing.

So it's definitely doing something. It really is changing that data arrow actually every step. I Don't think it's actually repeated. A step has It so it looks like it's increasing.

probably like a 1 DB gain for each time we adjust that Vernier. That's interesting. so to know what the exact range is, you'd have to decode the data here. There are some combinations of our code where it doesn't change at all like I'm flipping between those two there and there's data is exactly the same so it's obviously gone well I don't need to change the range there.

you go for that one there as well. now. I Initially thought this was it actually counting up in binary and that was strange because the formula is the gain bit which is the first one here which is always zero by the looks of it. when we're in the Vernier, you'll notice how that those some tight just changed it.

By one Vernier position, you notice how it actually had this extra pulse here. this is actually nothing but that's not going to count because it only registers the data on the positive edge. Here is so the most significant bit which is the first one is always zero. So therefore it's always in low gain mode.

so the rest of the so you can ignore that if that just start pops up. It might just be a quirk of the software algorithm that they're using that's very common. There's nothing wrong with that as long as the data is valid when that clock signal comes along. So for just that, vernier.

again, sorry, it's jumping around because there does seem to be a bit of variable timing in there. once again. I'm not sure why that's doing the real-time operating system. The scope has to do stuff I guess Ok, so let's start at this one.

Perhaps we've got one, one, one one there. So we've got four ones and then we've only got the one one at the end, so it skips through the one zero. Then we've got one zero. and so if you look at those four bits there, it went from 1 1, 1, 1, 2, 0, 0 0 1.

Now at 0 0 1 0. and if we keep going 0 0 1 1 so it is. See, it is cut. Yeah, it is kind of counting up.

Yep, and that's what you'd expect because it needs to increment a little bit again each time. I'm actually surprised that it a sort of like fine, compensates the range almost every turn of that Vernier. That's really quite remarkable. You know, gain control than that.

And of course I you rely on the gain. If they can't individua, there's no way. at the factory they've individually calibrated the gain on each one of these Vernier steps so that only do it on the particular course sequence like this that so this is calibrate each one of those ranges and then you're relying on the fine software steps, the accuracy of the steps with inside the the variable gain amplifier itself to give you your calibration on your fine vernier control and there's nothing wrong with that. That's exactly what I expected.

So there you go: I Hope I Answered that question: Is doing modern digital oscilloscopes actually should have digital vernier control in there? Do they actually adjust the gain? And in the case of the Sieglin ah, the answer is yes it does because it's got a pretty funky variable gain amplifier in In, they're designed for just such a vernier control, but not every digital scope is going to have this like the Wray goal. For example: I can't find in the front end of reverse I haven't done a whole video which you can see at the end. It's really cool video of how to how to do reverse engineering essentially and I reverse engineer the front end of the RAI Goldie's 1054 Zed and it doesn't have such a variable gain amplifier, but it does have some digital gain control with inside the analog to digital converter chip itself. But yeah, that's not as good as this.

so effectively I Don't think the RAI go away does it. but the Cygwin does about other scopes. so have a look at the TM Leave it in the comments down below. Well, I've done various Ted outs of various oscilloscopes over the years.

You can go look over my high-res photos on my Eevblog Flickr account. You'll be able to see the front ends for yourself for various scopes and see if there's a variable gain amplifier in there. See there you go. Umm, it's not just a software thing, it's actually doing real Hardware gain.

So there are advantages to making your waveform using your very control and making your waveform as large as possible when you do measurements because it's got more bits of the analog to digital converter to work with and the being inside the variable gain amplifier is gonna take care of your right calibration for you. Remember, scopes like this are only a percent. You know, two percent accurate stuff like that on their vertical, so not not a huge amount of absolute accuracy. but hey, Oh Once your ranges are calibrated, you can actually do comparative measurements and things like they can get the resolution and if you use more of your ADC range then you're gonna get more bits to play with and it's going to give you a more accurate in quote marks, measurement anyway.

I Hope you like that. If you did, please give it a big thumbs up. And as always discussed down below, catch you next time.