Hi I Thought we'd take a quick look at some common acquisition modes on modern digital storage oscilloscopes because it can actually be quite confusing and not at all obvious what the different modes are between the high-resolution or but sometimes called boxcar averaging mode and the mathematical averaging mode on a scope. So pretty much any modern digital scope should have both of these modes. If we go into the acquire menu here and then we've got the acquisition mode usually it's called the acquire menu or something like that and we've got different modes here Normal mode which is what your oscilloscope is normally at hence the name peak detector which out we won't discuss here and but what we want to have a look at is the difference between averaging and high resolution mode. So what are the differences between this traditional mathematical average in mode or this high resolution or sometimes called Boxcar average in mode? Well, it can range anywhere from between very little to essentially nothing in a practical measurement application, to being a huge difference which can greatly impact you measurement accuracy or what your entire waveform even looks like.

So it's really important to understand the differences between these two modes, and not everyone does because it can be quite subtle. So what I'm going to do is take that an example of this square wave here, which is sinusoidal modulated and it's got some noise on top of the sinusoidal part there, so it's coupled in to the square wave here and if we zoom right in like this, you can see it's actually updating like that. and if we zoom right in And have a look and we change between averaging and high resolution mode, look, it doesn't look to be much difference, does there at all? Can you see any difference? It's very, very subtle, effective, bordering on almost zero for this particular application, but that's not always the case. These are two entirely different ways of averaging, so this is just one practical example where there's almost a very little difference between them.

But as I'll show later, we will actually be able to measure the difference between them. But if we have a look at something like this: 1 megahertz FM modulated sine wave which is actually modulated at here, we go modulated at 10 Hertz here. I Think you'll see quite a substantial difference here between the modes that's normal. That's Peat detect.

There's your averaging mode really cleans it up a lot. and there's your high resolution mode. There seems to be very little difference there between your high resolution mode and your normal mode. Not a huge amount, but measurable.

But watch this if we zoom out like this. Oh my goodness, what's happening? Look at this. Oh, it's horrible. So that's normal mode.

Peat detect High resolution. Not much difference between those, but averaging. Wow This is clearly not the correct mode to use for this particular signal, and this could easily confuse you. You can think that's a real signal that's really happening, because if you've got your mode, like, where does it tell you what mode you're actually in right? A lot of scopes do not tell you this.

so you're sitting there and well, you're going. Oh, what's going on here And I Look at this weird our signal I've got and you don't know that you're in average in mode. It can cause huge problems. Huge misinterpretation of your signal.

So what's going on here with this averaging mode? How does it work? Well, You're actually likely quite familiar with this mode. It's your traditional averaging mode and what it does, you can set the number of averages here so we can. Actually, we can change the number of areas right down to two, so it looks more like what we're used to. But if you really go up in your averages, look at what we get in here.

That is because of the way averaging mode works. So we're in normal run mode here. If I actually single-shot capture that look, it's just fine. How signals are there.

Everything's hunky-dory and we go single single like I'm just single-shot Pressing the single shot capture button and everything's hunky-dory But when we run it, why we get that? Why is it? So the way averaging mode works is that it takes these individual snapshots like this and then does a mathematical averaging on them. For in this particular case, we've got it set to 16 waveforms. So that's why with this FM modulated signal, look, it's chittering around like this. That's why that upsets that averaging mode.

The Algorithm: A thematical averaging algorithm, can't handle a signal that doesn't trigger properly like that. So we'll find that this modulation here is a mathematical artifact of our FM deviation frequency. Del. Here, you'll notice that it's at 10 kilohertz.

notice that it's one division there, and we're on 550 microseconds per division. So the FM frequency if we double that, will find that these artifacts will actually change here. and there you go. That's the exact same signal again.

I Haven't changed the time base or ever in anything but I've changed the FM deviation to 20 Kilohertz. It says mathematical artifact, but you don't get any of that if you use your high resolution averaging mode. It just looks fine and dandy. so you can see that.

Our first example showed that there's very little difference between these, but when we got another signal which doesn't trigger repeatedly every time, it's got some jitter on there. It's got FM modulation for example, so which is essentially art the same thing then our higher, there's massive difference. Averaging is not the correct mode to use for this type of signal, but let's turn our modulation off here and just have a look at a regular sinusoidal signal. and I've actually added some noise in there.

I've added 5% of noise and you can see it's a fairly noisy waveform. What does what do the different modes do here? Well, high resolution mode doesn't seem to do anything at all, but averaging Wham There we go. That's what averaging mode is useful for. It's useful for hateable signals that are repeatedly triggered and they have random noise on them.

It reduces that noise right down. whereas high resolution mode even though it's doing averaging, it's doing a different type of averaging called Boxcar Average in in single-shot mode and its high resolution doesn't give you any real value there, but averaging does. So averaging is the mode to use when you've got random noise that you want to take out of a periodic signal. And that's the key.

a periodic signal. And that is why if we actually single-shot capture this, you'll notice that our waveform is still noisy. Because the averaging mode the mathematical average in mode does not work. In single-shot mode, it relies on those averages.

This case, I've got it set to 8,192 so it'll average 8,192 complete waveform samples to give us that very nice looking thin line that we've got there. So let's have a look at how high resolution mode here works now and we're in normal mode at the moment. I've got a 1 kilohertz sine wave with random noise added on it. Okay, if we actually stop it and zoom in, you can see it's got all that random noise so it's not a particularly nice signal.

em we might want to clean this up with in single-shot mode. Okay, without I mean averaging will work absolutely brilliantly with that. But let's have a look at the difference that high resolution mode can make. Now take a look.

We're in normal mode at the moment and we go down to high resolution mode. It's automatically like it's start. It's kind of like doing some averaging. It's getting better.

It's not as good as average mode, but let's have a look at the difference if we zoom right out like this. So let's use normal mode here at 200 milliseconds per division. and let's single-shot capture that. Okay, and let's zoom in to that and you can see all that noises there, right? All that there.

and that's single-shot capture. All that noise. But if we go into high resolution mode here and we go out to the same 200 milliseconds per division and we single-shot capture that, you notice it actually looks cleaner at this particular point. And if we zoom in, look at that, it's cleaned it up.

In single shot capture mode, High resolution mode uses what's called a boxcar averaging technique to actually average the samples as it's going so it doesn't rely on multiple waveform captures. It's doing it in real time in the hardware to actually a ver egde the samples on the fly. It actually uses a higher sample rate, so if you're at a very low time base like 200 milliseconds per division like we were, it uses the full sample rate of the scope and actually does sort of like sample the sample average in as the a higher frequency than the actual current sample rate on the thing. So if we have a look at the current sample rate if we get out of here so you can see that we were out 1 Meg sample per second here because we're at a very low time base.

but if so that the high frequency sample Boxcar averaging of the scope doing it in the hard way actually made a difference at that low time base. But if we if I push the single shot button again, it'll now sample at a different time base. It's now 200 microseconds per division. If I do that, you'll find it'll be noisier.

Tada because it's doing the Boxcar averaging over a different time base at a different sample rate 625 Meg samples per second now. So if we go back here and then turn off our acquire mode, here we go. If we go normal back to there, you'll see and we could do a single shot cap shot. You'll notice that it's noisier right down in there.

it is actually noisier. So I Found these graphics from a Keysight article on the subject and I think it actually represents that quite well and explains what I'm trying to say here. In high resolution mode, it takes a single shot capture ie. in real time and it takes all those samples like a 10 or 12 here and it gives you one and averages those and gives you one sample period.

But if you have a look at the graphic for the average mode, then the average mode works differently. It makes multiple waveform in this case, and with an average value of four, takes me four waveform captures. takes those four data points at each data point in your waveform and then gives you one average value so you can see how if you don't have a properly triggered and properly repetitive waveform, if the waveforms jitter in and going all over the place and not triggering properly, then your averaging is not going to work and you're going to get anomalies and things like that. it's going to completely screw up just like we saw.

So that's a fairly stark demonstration the difference between averaging and high resolution mode. We saw an example where they can be near identical, and we saw examples where they can be grossly different and you have to use them in different scenarios, otherwise you can actually really come. Agata But high resolution mode doesn't have the same sort of really gross traps for young players that the averaging mode does that we actually saw before. Our waveform was completely different and will get in all sorts of sampling artifacts and everything that mathematical averaging artifacts and things like that.

high resolution mode is more forgiving, and that high resolution mode of course, does exactly what it implies. It gives you a greater number of bits, It gives you a higher resolution. Effectively, a higher resolution. ADC Most digital scopes like this one are only 8 bit converters.

We turn on high resolution mode, and depending on your sampling and all that sort of stuff, it can give you, you know, 9, 10, 12, up to maybe even 14 bits of resolution. It can really increase the resolution of your signal. So let's go back to our original signal where we didn't really see any major difference between the averaging and the high resolution and the normal modes. Look, there's an open.

you can maybe see a subtle difference in there if you're really looking and I hold your tongue at the right angle. but we can. actually as I said, we can actually measure the difference between those if we turn on our peak to peak voltage measurement and we turn on the statistics and we go into the statistics mode or down here and we reset the statistics. We're currently in the normal mode here and we can actually get the standard deviation value.

So remember that. Look, we're up to a thousand council. We're averaging a thousand, were taking the statistics over now. 2,000 waveforms are around about that seventeen millivolt standard deviation on our peak to peak value.

Let's change the mode and see what we get. If we actually go down here into high resolution mode, we've got to go in there and reset our statistics. Otherwise, it takes some time to come down. Bingo.

Look at that. The standard deviation is now like a seven. It's more than half, so it's gone down fairly drastically because there's some high-frequency single-shot Boxcar averaging happening there. So it is.

Even though it's not hugely visible on the display, you can actually measure the difference in there. Now, if we go into the average mode here with 16 averages, it's now like hived again to 3.6 millivolts standard deviation. So they're just like you can't actually measure the subtle differences there. But this is an example of a repeatable waveform that does not have random noise.

You'll notice that this noise is the same I Have not single shot captured that look so that they are real deviations in the signal there. This is not random noise. So that's why averaging mode isn't having a huge effect like it was when we were introducing random noise. And that's the advantage of your traditional averaging mode.

here. when you've got random noise and a stable trigger that you want to remove the noise on averaging mode works fantastic, but it can have some traps too. young players. If you leave it turned on and you don't know it's turned on and the signal looks all weird and does all sorts of weird and funky stuff, then check that you haven't got every G mode turned on.

but high resolution mode is the one to use when you've got single-shot signals or things that aren't repeatedly triggering. you know? I haven't got a stable trigger, they're modulated or they jitter in or something like that and you want to clean up your signal. And you can also get increased resolution as well. So there you go.

That's a look at the difference between high resolution mode, average mode, and normal mode on your oscilloscope. I Hope you learnt something valuable there and the differences can be subtle or huge depending on how you actually use. The thing could be trust for young players there. So I hope you found that really useful.

You learned something and if you want to discuss it, leave it in the comments or the Eevblog forum down below. Catch you next time! Hi! Now there's a myth regarding oscilloscopes that simply will not go away. and that is traditional analog oscilloscopes like this: Tektronix Triple 2/5 or any analog oscilloscope is in quote marks lower noise than a digital scope and well, that's not actually true And I want to explain it to you today. So let's start off by take a look at this.

Tektronix Triple 2:5.