Hi I've got a real interesting video for you today, you know I've done a lot of videos on oscilloscope probes. Uh, passive, uh active, uh, high voltage, uh, differential probes. I've got a video on how they work and reverse engineerings and tear downs and stuff like that. But one that I haven't done a video on is one of these new optical fiber probes and thank you very much mixing uh for sending this in even though I think the Uh labels upside down.

Anyway, check out this. It is their latest Um optical fiber probe and this is the fourth and most advanced type of oscilloscope probe and also the most expensive. Uh is type of oscilloscope probe you can get. Now the first type of probe is your regular passive probe.

You know, one to one, ten to one, uh, switchable, or times 10 fixed, uh, passive. Pro They're your general purpose probes, typically up to, you know, 500 megahertz, something like that, although there are some very special ones that might be able to scrape it up to one gig. But basically yeah, Um. and The Next Step Up is your Um active fit probes.

Ah, basically these are very high frequency probes, very low input capacitance, and they've got an active Uh fat amplifier on the input here and you can't These are only used for very low voltages and uh, basically single-ended or are differential but very high speed stuff. So if you want to measure like you know, high speed buses with great signal Fidelity Uh yeah. Active Fit probes. They can go up to, you know, gigahertz, tens of gigahertz.

Even so all your high frequency ones are your active fat probes and then your next type of course is your high voltage differential probe. and I have done uh, several videos on these um like actually reverse engineering these, explaining how they work and testing them and doing not common mode uh, voltage tests and stuff like that. So you want to use the high voltage differential probe when you basically want to measure high voltage stuff or things that are raised above uh, signal ground so that like a main switch mode up power supply, high side driving transistors in. you know, power supplies and stuff like that.

You can't do that with your active probe or your regular passive. Pro because you'll blow up your oscilloscope. I've done a video on that too. So um, yeah.

high voltage differential probes An essential bit of Uh kit, but unfortunately these are not high bandwidth. This is one of the highest bandwidth ones on the market. you know, 70 megahertz you can get, you know, 100. But as valuable and as great as a high voltage and essential as a high voltage differential probe is, you should have one if you're doing any sort of mains work or any sort of you know, higher voltage, uh, power supply type stuff.

Unfortunately, they're common mode rejection ratio can be a problem when you're talking about more modern uh switch mode power supply designs and really high voltage and and like high energy physics research and doing all sorts of advanced stuff. This is where you need one of these newfangled fiber optic probes and thank you very much mixing for sending this in. Um. these These are state of the art tick and it basically Uh contains the equivalent of a like a high frequency active Uh fret fat probe.

This particular model from mixing can go up to one gigahertz so quite high frequency. The one we've got here is only Uh 200 Meg but they do make models that go up that high. so it basically is an active fit probe front end like this. but instead of just being connected via a coax and having a common Uh Mains Earth this cable here is fiber optic.

There is no metallic conductors in here at all and it gets its power for the head over the fiber optic. So it's powered from this Uh fiber optic transmitter here and also receiver and this plugs into your oscilloscope and it can actually send power over the fiber optic to power the active fit a front end like this, but it then sends the signal back also via the fiber optic in the analog form they don't uh, sample. It's all done in the analog domain so they treat transfer power in this direction and they can get signal back in this direction and you get complete galvanic isolation. In this case, up to 60 kilovolts isolation because you've got no metallic conductors between your oscilloscope and your probe.

It's it's almost like magic. So you get the advantages of the high bandwidth fat probes with the advantages of the high common mode rejection ratio of uh, your high voltage differential probe It's the Best of Both Worlds but these are unfortunately incredibly uh, expensive. But these are essential bit of Kit as I'll show you in today's video. We're going to do some experiments and it's going to be amazing.

So there's a couple of other companies that manufacture these optical fiber. Pros Please correct me if I'm wrong but I think out of the tech. ISO View might have been uh, the first Uh one, the first famous one anyway. but mixing have now come out with this Sig Offater probe and it goes anywhere from 100 megahertz up to one gig model.

So we've got the Uh 200 Megahertz model here which is more than good enough for the experience that we want to do. Now this is Mix Sig's uh, comparison between uh, the Lacroix DL ISO The Tektronics Iso View and Mixing Off It So always take these manufacturers uh, comparisons with a grain of salt. But anyway, this is what they've got. Um, it starts.

Yes, then they're pretty pricey at 23.99 Uh, that's Yankee bucks. Um, but the Tektronics Iso View for the 200 Megahertz one starts at 10 800. So the one we've got here is uh 3700 US dollars as opposed to the 10 800 Tectronics one. And the Lacroix DSO is even higher.

but that's a higher bandwidth. uh, minimum there. But the amazing thing about these and the difference between your regular high voltage are differential probes and these optical fiber probes is that these have a massive, massive common mode voltage range. The 200 megahertz version that we've got here 106 DB common Mode rejection ratio I've done a video actually demonstrating this I'll link it in I won't go through.

You know the nuts and bolts of common mode uh rejection ratio here, but that is absolutely enormous and it will do that for the full 200 megahertz bandwidth. and compare that to Mixig's uh high voltage uh, differential uh probe which is almost identical specs to my EV blog Hvp uh 70 uh probe and they're great probes. They're great for high voltage use, so you know they've got decent common mode rejection ratio. but you'll notice at at one megahertz, it's minus 50 DB and you go up to 10 megahertz, it's minus 40 DB and it just gets worse and worse and worse.

So if you've got a one megahertz uh switching power supply for example, and you're trying to measure the high side of that as we'll do today in today's Uh experiment, stay around for it because it's fantastic. This 100 DB makes a massive difference like game changing difference. So we're going into today's video. we're going to compare this mixing optical fiber probe to a you know, pretty much one of the best high voltage differential probes on the market and watch this blow this out of the water for the specific use uh case that we're going to look at today so we'll have a very brief look at this.

The main advantage of this one over the Uh Tech and the Lacroix ones is it's a universal B and C interface. It can plug into any scope uh, 50 ohm output so 200 megahertz. but they've got models going up to one gig. So and it's got a built-in calibration mode which only takes a couple of seconds whereas the other models are even though I haven't used them.

Mixing claim that they can take up to minutes to do calibration because these things drift with temperature and stuff like that. So you've got to be careful to actually uh, calibrate these things before you, uh, you know, take your your uh critical measurements just so that you're you know, taking out DC offsets and other stuff. So yeah, they've got built-in signal generators in the probe so when you press the calibration button, it will actually run a uh, like there's a test generator in here which should generate a signal and then it can test it and it can calibrate for any uh offsets but you can manually um adjust the offset here. It's got a built-in fan here.

looks like it sucks in the end. Uh, from over here gets a little bit, uh, warm, but it's a little bit whiny. Um, it's not hugely loud, but it's not noticeable if you're in a quiet lab. But anyway, um, yeah.

so this has a uh Power uh optical fiber transmitter in it to generate to send the power over the optical fiber here. so you can't like Bend these really sharply. That's why breakable fiber cable warranty void if you bend that sucker. So yeah, don't go Bend in your really expensive fiber optic probe because you can pay up to 20 grand for like the one gig bandwidth version of this.

Oh, these are such sexy bits of Kit Wait until you see the demo and in the Box I've actually got two probes although I think it might only come with one. I've got a ten to one uh probe here. that's is an SMA interface so it's flexible so that's a ten to one probe. and I've got a 500 to 1 as well for high voltage use because as I said, these are active fit inputs.

So just like your active probe, don't go putting high voltage into this because you'll just blow the magic smoke out of your potentially twenty thousand dollar Pro But that'll ruin your day. but with the high attenuation input, they can go up to several kilovolts, no worries. And we could get a nice little desk stand off so that it isolates it from your bench. and I'll demonstrate that later.

Then we get some Mcxr connectors. uh here because this is an Mcxr connector interface. it's SMA here, but the actual probe tip itself is uh MCX. So yeah, um, the demo ball we're going to use today has an Mmcx connector and well, that causes a bit of problem.

But anyway, we'll get around it. And because the probe is not an Uh active probe interface so it can go onto any scope, it does need the five uh USB C here to power it. and it does come with a plug pack and cable to do that. Now as I mentioned, this is not a high voltage differential probe.

It does not use a differential probe architecture. I'll link in the video that I've done actually reverse engineering mix. eggs are Dp1007 differential probe I've also reverse engineered uh, the Sapphire Hvp 70 as well. So it is not a differential architecture.

It does not have the differential input this fat input that goes into a differential amplifier configuration. It is as I said, just like a single ended active uh, high frequency probe. It's very different to the high voltage differential probe here. so let's have a look with a practical example of where you can use an optical fiber probe like this and get the advantage of the huge, massive common mode.

almost practically infinite common mode rejection. Well, not infinite, you know, 100 plus DB common mode rejection ratio of these fiber optic probes and we'll see if we can actually get some advantages like this. Uh, compared to say a differential probe. Here's the yellow one.

There is what a differential probe is going to look like when we're going to probe the circuit that we're going to try out in a minute. And that's not the real response because it's got an extremely poor common mode rejection ratio of a regular high voltage differential probe. So and look, we should be able to get a nice beautiful response with a fiber optic probe. Let's see if it's possible.

Now one of the many useful applications for this is in real high Efficiency Modern Power Electronics which get uh like 99 plus percent efficiency you used to. Yeah, your regular DC to DC converters get in like you know 85 90 will be a good one. maybe into the 92 93 range but the more modern ones especially in higher really higher power. Really a dense uh Brick Designs can get over 99 efficient and the way they do this is using modern and mosfets called Gallium Nitride semiconductors organs as we'll refer to them here because it's it's like G-a-n again uh Power Transistors Here's an Infinian application note: Um, so we're going to actually look at what one of their cool Gan Uh Gallium uh Nitride power transistors here and one of the advantages if we go look at the topology, we won't go in into detail, but it's basically a there's no p in traditional PN Junction in again, uh, transistor.

So it's basically a planar device which means it basically flows on the surface. It doesn't the current flows on the surface, it doesn't flow into the PN Junctions And hence you can get lower uh conduction resistance of these things which make them more efficient at higher voltages. um, as well. So they work very differently to a regular uh mosfet, but we won't go into the details.

So here's another substrate. uh, diagram of how Argan device works like this. and basically, um, you're going to get like really high efficiencies I mean check this out, right? Here's a power factor correction: Uh, board 2.5 kilowatts in this tiny little form factor like this. But look at this.

we're getting. you know, 99 uh, percent plus efficiency on stuff like this. So you know this is how you can get really dense modern things and you get huge EV charges and you can get, you know, a really efficient power bricks and stuff like that because they're using these newfangled uh gallium nitride. and there's other uh types of modern uh Power transistors as well.

And they're called a high electron Mobility uh transistor as well. So a hemt. So if you hear the word hemped again, hemped something like that, it's just basically yeah, High electron Mobility it means more better. It means lower on on Resistance and lower on resistance in a switching converter.

Whatever it is. whether it's power factor correction, it's boost, it's back. It's it's whatever you're doing. um CPR converters or whatever, then you're going to get much higher efficiency due to the lower effective on resistance.

Yeah, I Know this is marketing blurred for Infinium but they're one of the leaders in these Gan devices. and they've got this cool demo board. But look, you can basically use them in the power factor correction circuit, the resonant uh converter here, and the synchronous rectifier on the output. So not only can use them as transistors, you can use them as very effective diodes as well.

So anyway, we are going to be using one of these 600 volt cool Gan Um, that's just their trademark thing, but Gallium nitride power transistor in a push-pull half Bridge configuration or a totem pole? uh, switch in configuration. We're not going to like do a full uh converter, but basically yeah, we're going to use one of these bad boys. It does have a Kelvin Connection in here, by the way, for the source down here. So anyway, we're going to use one of these application boards.

Uh, which, you can get really cool. Um, they only cost about 60 bucks or something like that if you want to play around with with these things too. It's got two of these again. uh, transistors.

So I've got this half Bridge R configuration like this. so we've got one on the top and one on the bottom. So this is a basically a totem pole output. and here's the drive circuit for it.

Don't worry, it looks a bit complicated I Won't go into all the details, but how you can achieve like 99 plus efficiency with these things is by carefully driving and this is what we want to monitor because so this can go up to 600 volts. But today we're due to the equipment limitations are only going to go up to 300 volts and it's basically impossible with a regular R probe, let alone a regular differential probe to actually measure the Gate Drive of a high side power transistor like this. Doesn't matter whether it's gallium nitride or anything else, it's just that gallium nitrizer. a modern application.

Uh, for it, because what you need to measure of course is the gate Source voltage like this on a power transistor and that's why we've got this uh connector here to actually do this. But the problem is, you can't do that when this Source here is switching like hundreds of volts up to like 600 volts or even higher. Um, you just can't do that. The common mode is just you get that common mode if you're switching at Megahertz, for example, which high efficiency converters are, you've got this huge common mode signal actually bouncing up and down hundreds of volts.

and your regular high voltage differential probes. even the best of the best ones can't handle that sort of thing and hopefully we'll get a demo of that today. Here's the complete schematic for the demo board that we're going to use. We've got our half Bridge again over here.

and then we've got these two, uh, driver chips. Have a look at the schematic in a minute, but these are actually isolated. so there's electrical isolation inside the chips here. Okay, and they're powered uh from Vdd and VSS here.

and they're powered from this, uh, isolated boost converter over here. So we've got like a five volts into the board here and then its powers the drive side of the mosfet and then into our half Ridge Here we're going to feed in there and here we're going to feed in up to 300 volts and we can vary that to see the effect of the common mode rejection ratio and just to ensure that you don't get what's called Shoot through which is both of these transistors turned on at the same time because you don't want to short out your like 300 or 600 volt power supply that's capable of like kilowatts or something like you're like, no, it's going to release the magic smoke. So um, though, what? uh, this does is this uh basically we've got an adjustable. There's a little trim pot uh here and here which you can adjust the off time of the transistor.

So this, basically um, we've got a 50 ohm input here which you can drive from the signal generator and that will just, uh, invert the signal and then add some delay here to ensure that these transistors. it's impossible to have both of these transistors on at the same time, so you can just adjust the dead time of the Gate Drive And then we've got a 500k low with some capacitors. We may not use that today. I'm not actually going to use a load so it's not going to be like an inductive load or it's not going to be a boost configuration.

We're just going to switch these transistors off and on at a fast rate like one megahertz and we should be able to see the effect of being able to probe because this is what we want to probe today. We want to be able to probe this gate drive because when you're developing these circuits, if you can't measure, the Gate Drive is like critical. I won't go into all the details of driving Gallium nitride again transistors. but I'll include some application notes down below if you really want to know about that sort of stuff.

but it's real tricky if you're you know, shooting for 99 efficiency. You need to get your gate drive right and you need to be able to probe it. and you can't do that with even the best high voltage differential probe. so we'll try that.

You can do the low side down here because this is basically um, ground right? And there we go. It goes through the Kelvin connection there, right? so you can measure that Gate Drive But this high side? When this code here of this source is switching by, you know, hundreds and hundreds of volts at one megahertz. It's gonna ruin your day. So here's these uh, specific Gan drivers and you can see here, there's actually isolation inside the chip here.

So uh, yeah, basically you're feeding your Pwm signal here. but it, basically, um, it's got to go over this electrical, a galvanically isolated interface. and then you've got another, uh, totem pole uh driver in here to actually, uh, drive your transistor here. and you need one of these driver tip chips per transistor and the output here not only drives the gate, but it also drives the source as well.

And this is why you use these Gan devices in really high efficiency converters, because you can really drive them in very specific ways targeted specifically for your application. So you've got to provide that isolated Supply here for actually driving the Gan transistor. Okay, I'll go through the setup here. We've got our demo board here and it's raised up with the pan of ice here and we've also got our our optical fiber Probe on its little Understand: there's a reason it comes with the stand because if you've got your board laying on your bench even though this is a non-conductive bench and you've got your probe down on the bench as well, you can get increased capacitive coupling and it's actually working at the moment.

and uh, have a look at the switching waveform here and watch what happens if I change it down towards the bottom like that, you'll notice there's much more ringing here, right? So you've got to keep this up off the ground like that. That's why they Supply the stand. And to befit our state-of-the-art optical fiber probe, we've got the no lesser, state-of-the-art Ron Schwartz mx04 uh scope here. so I'll be operating that uh remotely.

I've got a 5 volt uh, power supply here and that just up house the 5 volts in uh for the board, let me turn that voltage down. otherwise I should be using my takeaway protection container here. I'm highly recommended. Made in Australia Thank you very much.

We've got the optical fiber probe connected into Channel one here and I've run a calibration on that. We can just run that again. Boom Go! and it's all temperature stabilized and everything next on Channel 2 here. I've got my Uh high voltage R Pro Probably could get away with a regular uh times 10 passive uh Pro But just to be on the safe side, I'll use my 100 to 1 our probe here so that's on Channel two and that's across the Uh H Bridge output and uh, the low side so effectively uh, the ground here and I've got a 220k resistor load in here just going between the H Bridge output and ground down here I'm not going to do any fancy uh load today I'm just going to put something on it just so we have something there and we can of course uh Mains Earth common that uh, low side because um, everything else is, uh, floating so no problems whatsoever there.

We're not going to blow up our scope, done a video on that, and then either side of here. I've got my Uh 300 volt high voltage input which we can adjust which is coming from this Xantrex power supply and this is the fan noise you can hear here. Um, so yeah, I can just adjust the voltage I can go up to Uh 300 and it's a 300 volt 4 amp Supply I can go up to 311 no worries. and on Channel 3 here EV Blog: Hvp are 70 high voltage differential probe coming back in stock in some September they've had a part shortage.

Anyway, it's it's an awesome. It's probably one of the best in high voltage differential uh probes on the market designed and manufactured by Uh Sapphire and we're going to use that to actually probe the exact same point. not at the same time don't do it at the same time, but uh, at the same point as our optical fiber probe here just so that we can show. hopefully a difference between uh, the common mode rejection of an optical fiber probe and pretty much the best differential uh, high voltage differential Probe on the market.

and how this as good as it is, is not the tool to use to measure the high side with a 300 volt switch in at one megahertz which is what we're going to have here. All right, let's control our rodent short scope remotely. I Love this via the ethernet. The software uh in a web interface is just uh fantastic.

This is just working as a uh in a web browser here so we can get the full screen or we can turn on the front panel here, whichever we prefer. Let's go go for the front panel. Now what we've got here is uh, the green waveform here as you can see are 40 volts uh per division here so we can go up to 320 volts. This is our H Bridge output.

Okay so where I've limited the probe here so we can go in there and well we can look at this. It even shows the transparency. Look at that we can adjust the transparency. So with both of these channels, I've actually uh, bandwidth limited these uh to the probe bandwidth because well, there's no point going higher than that, you're just going to get increased noise.

So um, so our green waveform here I've just turned it on enough. It works with no, we still get the switching waveform here the yellow one with no uh voltage at all I can completely switch off my high voltage. Supply We still get the switching waveform. We're still driving the mosfet, but of course there's nothing there to switch.

Okay, so our output are here. so we're currently set a very low. So I've set it to uh, 40 volts. So it's just switching off and on.

Now we're generating a signal. here. we're generating a yeah, one megahertz, Uh, five volt, P2P 2.5 volt offset. So just a TTL signal in with a 50 duty cycle and we can play with that later.

So the Channel One here we go. Let's switch over and we can see the scale here. Uh, you can see that it's not much. It's like, you know, Six, seven volts.

Uh, Peak to Peak something like that. That's what you'd expect when you're driving the mosfet. But remember, this is the differential voltage against the gate and the source of that high side power transistor. The one which is currently elevated at.

well, it's jumping between 0 and 40 volts. Okay, so it's continually switching at one megahertz. You can see the schematic up above the waveform at is at the source of that high side power transistor. So we're measuring the gate Source voltage.

so the source is continually swinging and that is our problem. The greater the voltage, the more your Uh common mode rejection ratio needs to be. The higher your common mode rejection ratio of your probe, it needs to be in order to eliminate any effects of that switching output from that high side gate Source measurement here. So anyway, this is the classic Uh Gate Drive waveform.

This is exactly what we'd expect. The on period is uh, from here, over to here Yeah, you can see my cursor up over to here and that ringing that we get in here. You'll see this, uh, change a bit, but it's not. It's not a huge amount, but that will be due to that little extra length of coax that we had in there.

If I had the proper little um RF adapter in there, then that would likely almost certainly uh, go away and we get a nicer waveform. But anyway, when the output switch is here, we get a little bit of ringing on the output. That's because of the uh, that longer lead that we used on that high voltage R probe. Once again, if I probe that a bit better, we'd tighten up the measurement on there.

But um, yeah, for the purposes of today's experiment, that doesn't matter. All right. So what I'm going to do do now is I'm going to adjust that power supply from 40 volts all the way up to Uh 300 and just over 300 volts. and we're going to see if that yellow waveform the gate Source voltage actually changes if changing that Source voltage and switching between a much higher amplitude actually changes the measurement on that yellow gate Source high side power transistor.

So let's adjust it. We're at 40 volts at the moment and here we go Right, we're going to go up and look. it's basically that yellow waveform is not changing at all at all. There's a little bit of you know, switching stuff which is changing a little bit, but basically we still get the waveform.

Look, we're all the way up to 311 volts and it based on that yellow waveform hasn't changed at all. we're still getting proper signal measurement. Uh, Fidelity on that waveform. and that Source voltage is switching up and down at a frequency of one megahertz from zero to four 100 volts and that optically isolated probe is doing the job.

and we're not. Basically, Look at it. Look, there's no difference at all. I Can't like overestimate how difficult this measurement would be without this optical fiber probe.

In fact, let's try it. Let's hook up one of the best high voltage differential probes and see the difference. Oh no. I hit the stupid preset button and screwed all my settings.

Damn it. What I've done now is created a reference waveform that Channel One switching waveform that we have with the optical fiber probe I've saved that as a reference waveform. so that's the white one, so we'll be able to compare it to, uh, the high voltage differential probe and that's a very good use of uh, those reference waveforms for your scope. If you've never used them before, this is what they're good for.

You'll notice now that we're actually getting so much uh, ringing and crap on that uh, or different from that differential uh probe that, uh, like, we can't trigger off this. So I'm going to have to choose a different Source I'm going to have to choose the I'm going to have to choose the uh actual Um output waveform down here. Now we can see it. Look at that orange waveform.

Okay, we can increase the brightness here. We increase the intensity there. Look at this. the orange waveform.

Okay, it's the same. It's the same. This is the best home. One of the one of the best high voltage differential probes on the market.

and it just does not like that. And we're only switching like, um, what is it? 40 volts, less than four, like 30 odd volts or something like that. Let's let's actually turn it up and see what happens. So let's wind it up.

Watch that. uh, orange waveform. Okay, look look at how bad that's getting. Look at all the ringing.

Look at this. Look at this right. It is absolutely atrocious. It is awful.

Okay, and yeah, the probing's not absolutely ideal. but you're gonna get this. You can see that how much of a major problem that is. But let's actually go down okay and we'll go down in frequency.

and if we go down to one Kilohertz switching frequency, you'll see not really a problem, right? You don't like We once again we can go up to all the way up to 300 volts and you don't notice any change in that uh, waveform whatsoever. So the whole voltage differential probe is fine at one kilohertz, but as it goes up in frequency, your common mode rejection ratio drops. and then you get all those problems and the whole idea of uh, these using Gans And to get the high efficiency, you've got to use the high frequency so you can't use the high voltage differential probe to measure it. You've got to use an optical fiber Pro cooler.

So Back At One megahertz. Again, you can see that as we like as the voltage gets higher at one megahertz. Okay, you can't get any proper signal for Delhi. Look at it compared to the white reference waveform there that will get in with the fiber optic probe.

This shows that this is the only fiber optic. Pros for this application. they're the only game in town. Sure you could use some other isolated uh scope uh probe to do it, but then you cut.

The whole idea of doing stuff like this is that you use the one scope to get a time a reference, start measurements on the like multi-channel so you can measure um, the different stuff that's going on. And you can't use a high voltage differential Pro optical Fiber probe. It's the only way to do it for this application and for many other like you know, a high-end like physics experiment applications and stuff like that and power and all these power measurement applications that have high frequency high voltage switching stuff like this. Gotta use an optically isolator probe.

That's what your 100 DB of common mode rejection ratio does. It's just it's not a problem and we can play around with the Uh Sig gen here. So let's uh, adjust this duty cycle. Look, you know, like you've got no chance of measuring this with a differential probe.

There's just too much ringing, let alone at at higher voltages. All this stuff is just. it's just going to absolutely dominate. But if we go back to our fiber optic probe here and we adjust our duty cycle right, look look at how we can still see the signal.

Fidelity Look right. You can just see everything right. even at like one megahertz switching frequency. It's just like it's amazing.

and I switched off the external high voltage Supply completely. You can see that on the Green Wave on there and this is the Baseline I've actually got both the fiber optic probe and the differential probe actually plugged in at the same time. Don't recommend that at you know for a real measurement, but you can actually see the Baseline uh ring in there of the differential probe in the orange there so it you know it's it's not too bad. We can do better if we actually probed it.

uh, better. But as you saw before, once we turned on that high voltage and then it started switching, that common mode signal started to swing on the output, which caused the gate and source to just keep rising up and down. up and down. Then that's where your common my rejection ratio, your poor common mode rejection ratio.

And remember this one megahertz switching frequency. You don't Uh, look at the one megahertz common mode rejection mode figure. You actually have to look at. um, the higher harmonic frequencies.

That one megahertz is going to extend to tens of megahertz, Even hundreds of megahertz. So the pro bandwidth uh matters. And my high voltage differential probes only 70 Meg uh bandwidth by the way. And these isolated probes because they're not actually a differential uh Op amp architecture, they can actually be faster.

They're like an active Pro. They can go to Gigahertz or you know, even higher. So yeah, look, it's like that's the Baseline You saw how horrible it just went purely by having that output switch the gate and source. Just a huge difference.

Huge! So we'll quickly measure the noise here. and I've set up my scope here. You can see it in the top as Uh 15 bit. So I've got the HDR mode enabled here.

200 megahertz bandwidth limit so we're going to need 15 effective bits even though this is only a 12-bit Arc converter. This is a really schmicker scope for doing these sorts of measurements, so let's have a squeeze here. Um, and I Just want to show you one very nice feature: if I turn the scale down to one millivolt per division. I've turned off the times 10 things we don't have the Times 10 probe attached.

you'll notice that these little red markers down here see them. So in the bottom left corner there next to the vertical scale, they indicate that we're actually over ranging even though it doesn't look like we're overranging here. and I can actually, uh, turn up the intensity like this and you can see it still doesn't look like we're over ranging, but it detects any Peak above or below the ADC sample range and as long as you get one Peak there, it'll like set the red thing there Flicker at a rate that you can sit visually see it just to tell you that you're off scale. so I can actually go in there and adjust it to two millivolts per Division and we're gone now.

So I just I Just want to point that out: I did a separate video on this on the Uh second, Channel a little short and I just really appreciate nice touches like that. Beautiful. Whoever implemented that and I just wanted to go zero to five megahertz bandwidth here because I wanted to show you just these little uh, switching spikes in here at 315 kilohertz. So obviously, inside this thing, there's a switch in converter and we're just getting uh, the switching noise out of that, you can see the harmonics side there can I physically drag that level down.

Yes, I can. So we get the extra Peaks there. Oh, isn't that nice? They are so we can get the harmonics of that uh, switching frequency. Not a big deal I just it's there.

And I wanted to show you and the actual art noise here. We've got our AC noise here. I've done a whole video on that now you've got to choose that correctly. so it eliminates the DC offset here.

and we're getting 782 microvolts over the full 200 megahertz bandwidth. and I've now got a 200 Oh, I've now got a 333 megahertz span and you can see at about 200 megahertz, it starts to fall off because well, this thing only has a 200 megahertz bandwidth. so you expect, uh, the bandwidth to start falling off there. but that looks pretty good.

you know, at around about like 80 minus 80 dbm noise floor or something like that. So yeah, it's uh, not too shabby at all. But the reason you want this, of course. common mode rejection ratio over 100 DB which I'm not even sure I could measure 100 DB common mode rejection ratio here in the lab.

but anyway, it's it. It's huge. You've already seen the benefits. and I see very little drift in this thing.

Uh, either. It's pretty good. So I'll switch on the cow mode now and you'll see it. Cow Mode now.

Boom, There we go. That was it. It's quick. and then the offset button.

You can just move it. You know, if you've got a DC offset in there, you can trim it like it's looks like it's about like 0.2 millivolts per step or something. It's really quite small. so I'm adjusting that now.

and uh, you can see that the steps are really quite tiny. So yeah, it's really nice. So there you go. I Hope that's given you some insight into what.

um, these really? Advanced expensive probes? Uh, can do? Yeah, they're not cheap, but they're the only game in town if you want to actually develop these. Leading Edge High efficiency, high voltage, high frequency, high performance, uh switching products. And as I said, do like other physics research and there's all sorts of like physics experiments that you can do high voltage, uh, you know, isolated differential, uh stuff. And you need these fiber optic uh probes.

You need that massive common mode rejection ratio CMR Um, like 100 DB up to like, what is it? 50, 60 kilovolts or something like that For you know, physics research and you wouldn't be those using those sorts of voltages for uh, you know, regular Electronics stuff. So really exotic, uh research, but stuff like just designing a high efficiency power brick or you know, the newfangled EV Chargers that need massive high voltages, high currents, and you know, a high frequency switching super efficient uh converters. This is the only game in town, so thank you very much to mixig for uh, sending this in. This is absolutely brilliant and hopefully um, I've given you a good insight into um, how this measurement is simply not possible with existing uh, probing stuff.

It it really is the only game in town. It's remarkable anyway as I'll link in uh like all sorts of application, uh, nodes and data sheets and all sorts of things down below. As always, you can just discuss test gear like this over on the EV blog Forum Absolute best place to do it. and if you like the video, please give it a big thumbs up and you can discuss down below as usual.

Catch you next time! Thank you Foreign.