Hi Well I completely come a guts are on the previous video that where I measured the noise on this ride and Rd6 double O Six power supply here and I was measuring much higher than spec here. I was measuring in the order of you know, half a volt peak-to-peak noise and it looked pretty horrible. Well it turns out that dumbass Dave didn't engage his brain and actually measured this wrong. So let me show you what I did wrong and how to measure this properly.

So here's the signal here that I was measuring at 6 amps and as you can see peak to peak just over 500 millivolts there and you can see that there is some ripple on there so there's some lower frequency. Ripple The ripples only in the order of two divisions you know are 40 millivolts. Actually it's not that bad, but I completely come Agartha on measuring this peak to peak stuff. Why and how? Uh-huh It's an interesting question and it's a real trap for young players including Dumbass Dave when he doesn't have his brain engaged because well, I knew this but I just love, just completely forgot.

Come I got sir. So the problem lies. as you might have guessed in my probing technique Channel 2 there I've just got a regular like a 50 Ohm Rg-58 coax B and C and I've got one of these are B and C to banana plug adapters and I'm plugging this directly into my power supply. So this is actually it's using our coax like.

This is actually a pretty good way to get nice high frequency bandwidth measurement. Have I done a separate video on that? Not sure. Yeah, I might have to, but it, while using coax, is a brilliant way to get a high frequency bandwidth of probing stuff. especially low signal measurements because often you don't want to use your traditional Times 10 probe here which country to its name actually divides your signal by 10.

So if you've got, if you're trying to measure a you know, 10 millivolt noise, the last thing you want to do is be dividing it by 10 to give you like only one millivolt. So then it's much harder to measure on your oscilloscope. You're you know you're really down in the low end noise of your oscilloscope there. So yeah, you want to avoid the Times 10 probe if possible, and using a BNC is one way to do that.

but I completely goofed it. While this is fine for measuring the low frequency ripple stuff, it's no good for the high frequency content. although in theory it should be what have I done wrong. Ah, put your thinking cap on and try and figure it out before the next clip.

And no, it has absolutely nothing to do with the fact that these leads are just like flapping around in the breeze. They're unshielded for a reasonable length here. That's not the problem, now it's just a regular. Our coax like this is of course going to measure zero ohms for the conductor in there.

Well, basically right, it's it's just a wire that goes directly through. But and this will be important in a minute. Trust me. Now, this coax of course is a shielded coaxial cable.

You've got the center conductor and then you've got the outer braid as well. And this is a transmission line as far as high frequency content it goes. And the stuff we were seen on the Oscilloscope, that high frequency ringing content. That's all about high frequency content.

And when viewing a high frequency content like that, you need a proper transmission line like a coaxial cable. and you can see that this is our Rg-58 cu cable, which you know they're pretty industry standard cable. It's not bad coax and it's got relatively high bandwidth, so it's not a problem with the coax in itself. But the part where we've come a gutsy here is that because it's a transmission line, it has to be a have a matched impedance in the system.

otherwise you're going to get mismatch. and as a result, what we're going to see on our Scylla scope is any ringing or high frequency content like this. because we've got a mismatched transmission line mismatched impedances. it's going to effectively, in this case amplify the the or what's actually happened in here.

This switching power supply does actually have noise as we'll see in a minute we'll do a proper comparison with some proper probing, but in this particular case it's going to amplify that and give us a much different result. In this case, like half an order of magnitude different result than what we're expecting. So yeah, it can really play a big role matching your coax cable. Now of course, this is a 50 ohm impedance coaxial cable.

so what we need to do is actually match the impedance and there's two ways to do this. The traditional method is if your oscilloscope had 50 ohm input impedance termination, you could just enable that and Bob's your uncle, but you can come a gutsy with this very easily when measuring power supplies like this. Now, we're only measuring five volts here, but five squared V squared on our to calculate power 25 divided by 50 Ohms. That's going to give you half a watt, so you need at least a half of what rated 50 ohm load in this particular case to actually do that.

So I happen to have one here. This is good old HP number. Adilyn Tor A Keysight Rubbish. This is a 50 ohm one watt terminator, but it's a like an inline one and they're just cool.

So if we stick that in like that. Bingo. Look at this. There you go.

There's our signal. We're now talking a hundred and forty-three millivolts peak to peak instead of 500. So that is our true signal that we're getting. not the one that we'll see in before.

Don't But there's another way to do this, especially if you're measuring higher voltage power supplies like this. So you can just do this with a 50 ohm resistor. You don't have to put it at the end of the coax cable here. You can actually put it in the front of the coax cable like this.

So I've got a 50 ohm resistor. It's actually 51. Good enough for Australia hanging off here. So let's plug our coax directly back in like we were before and we get our horrible, you know, 500 milli volt signal.

But instead of probing it from here, let me prove it from the other side of this 50 ohm resistor. Bingo! Getting 200 milli volts peak-to-peak because we no longer have the 50 ohm termination load here. So yeah, but anyway, you can put in just a 50 ohm term, right? It doesn't matter whether it's at the start of the coax or at the end like that, you've just got a terminate it properly. But I Know you're saying days, stop around with this coax cable stuff.

Just use your oscilloscope probe. Well, the dirt. Of course we can just use our oscilloscope probe. so let's try that now.

I've got that plugged into Channel 1. Let's hook that up I'm not putting it on the other side or resistor I'm just using that as a connection point and Bingo there it is there. There's our same signal. we're 50 millivolts per division now and measurement around about 210 milli volts peak-to-peak So that's with our probe in times 10 like this.

So it's actually dropping that signal down to effectively 5 millivolts instead of 50 millivolts per division. And that's okay for large amplitudes like we get in here. but for much smaller levels and noise that you're trying to measure, you don't want to use the X 10 position like this. So anyway, sit up, let's come off.

There we go. There's our actual our signal using a properly compensated X 10 probe like this. And yes, I have done the compensation adjustment on here. Which is important because let me show you, if you don't compensate your probe properly in Times 10 mode, well look, you're really come a gutter and measure the wrong signal amplitude.

Look at that. See, we can actually get the incorrect level. so make sure you compensate your probes correctly in times 10 mode. And of course you compensate your probe using your compensation adjustment on the front there and you can see it's peaking and go down like that.

But you'll notice that if I switch that to X 1 compensation does absolutely nothing because it's not a thing in X 1 mode. So there you go, get your tongue at the right angle. Probe is compensated so now you're ready to measure your signal. In this case, Ci knows that we want to measure or the high frequency content.

You've now got a high bandwidth compensated probe, but aha, there's one thing. But if we go into the vertical menu here, you'll notice that I've had my bandwidth limit turned on. There it is. bandwidth.

This is the 20 megahertz bandwidth limit. I Could turn that off and look, it's going to be completely different. So why would I want the 20 megahertz bandwidth limit turned on? Well, it's basically due to convention. almost like as a de facto standard, noise is measured over a 20 megahertz bandwidth.

So if you go, look up data sheets for any power supply that specifies noise. For example, it's usually specified over a DC to 20 megahertz bandwidth. And this is why almost every scope practically every scope on the market has a 20 megahertz bandwidth limit like this. Don't ask me where exactly that came from, and why the industry standardized on 20 megahertz.

It just did. And that's why every scope has a 20 megahertz bandwidth limit. And when you're doing noise measurements like this, you should have the 20 megahertz bandwidth enabled because that's just by definition, like you don't have to do if you want to specify your noise over a different bandwidth like the 400 megahertz Oh 200 megahertz bandwidth of this scope. even though I'm only using a 100 megahertz probe, you um, then knock yourself out.

But that's not industry standard. Use 20 megahertz, please. But you might be thinking, and you should be thinking, but Dave This is coaxial cable. and it is.

And it's not 50 ohm terminators. So why can these probes work without the 50 ohm termination at either the front or the end of the cable like this? Well, that has to do with the particular design and construction of oscilloscope probes. Now I've done a whole video on this twice. Secret of times, one oscilloscope probes revealed and I highly recommend that you actually take a look at that video.

but if we actually measure the resistance just like we did before of an oscilloscope probe, let's put in times 1 mode like that. Okay, so we just measure the conductor through the middle. Of course we got zero before because in a regular coax, it's just a single strand cop I going right through the middle like that. but on an oscilloscope probe AHA 330 ohms.

And in that other video, I've actually taken apart a probe and I showed you why. Because it's a lossy coax, they do this deliberately inside the cable. They don't just use regular copper, they use a different like micro me type material and they give it a little bit of wiggle wiggle wiggle year in the middle and that actually creates a lossy coax. And that's why oscilloscope probes are specifically designed and matched for the one megahertz and capacity front end of oscilloscopes, whereas coaxial cables, Well, whilst you can get much better higher bandwidth from them than an oscilloscope probe because they a passive oscilloscope probe like the the best ones you can get.

So let's stop at 500 megahertz really so you can't get much higher than that using a like regular coax cable. but they're less forgiving because you have to terminate them properly. whereas your passive oscilloscope probes are specifically designed and constructed and matched for your oscilloscope. and that's difference.

So when in doubt, use your oscilloscope probe I didn't. Oh, it's just like lazy Dave and just plugged in the coax like this and well, yeah, we can come a gutter. So if I put both probes down there like that and bingo, look at this. They match pretty well don't they? They're both far 50 100 millivolts per division.

We can change that a little bit. There we go, 50 millivolts per division and well, you can't There's a little bit of a difference. The yellow one is the oscilloscope probe in times 10 mode can see. Its got like a bit more high-frequency content, but now we're sort of getting down to the details of the probe here.

like if I touch that if I fiddle around with that you might be able to see that signal change a bit. Look at that see Wiggle wiggle Wiggle. Yeah and you can. Yep, you can see that change so you know you stake down to like high-frequency probing techniques and stuff like that.

But it's not. You see that it's not actually changing the peak up here, it's just changing the more high-frequency detail. And that's not really the important stuff. When you're measuring the Peter peak noise and you know we're getting like 200 millivolts to 10, it's sort of.

You know, it's neither here nor there. So both of those probing techniques with the 50 ohm and in front like this and the coax is working. A treat. But this is in Times 10 mode.

What if we switch this two times? One probe here. Look at that. There we go. That's times one probe and you'll notice it's a little bit higher.

It's not that terrific and it's lost all that high frequency content in there that we'll see in in Times 10 mode. And that is because if you watch my times, one oscilloscope probe revealed video and you should I explain why a tines one probe has a lower bandwidth. In this case, it's usually about like five megahertz. Something like that.

It's not terrific. So yeah, it's actually lower than your 20 megahertz. But you know you're still going to get your peaks like that. you know it's going to be sort of like near enough.

This is not. You know, if you really want to do this with the utmost of precision, well, you know there's better way you know you've got to fiddle around. So there you have it. ie.

turned my goof into a hopefully informative video where you learn something about probing. and in this particular case, apologies - I'm Rd Tech Ryu deaiing there ride and our D6 double O 6 does not have huge amounts of noise. In fact, it's a little bit. you know I need to probe it a little bit better? maybe, but you know it's it's it's spec was a hundred millivolts our peak to peak I think and we're measuring about 200 at full load.

But anyway, if we drop this down to an amp for example. but something different, you know, hundred and 30 millivolts peak to peak RMS noise are only talking eight millivolts something like that. So it's not nearly as noisy as I made out in my review video because I goofed it so you know? yeah. I might still do a video investigating how to like take the edge off some of that ringing in there.

Maybe we need to put some internal ferrite beads across some of it. like the switching components like the MOSFET or or something like that. You know there's several ways to do that depending on you know, the best mitigation strategy for getting rid of that, but so yeah, sometimes they just don't engage my brain. and I should have picked that up in the video because I knew that.

but well, you know happens. anyway. I hope you enjoyed it. If you did, please give it a big thumbs up.

As always, discuss it down below and I might have to do a video on this soon. I'm available over on library TV or Lb Y dot TV So definitely go check that out. It's a decentralized video sharing platform and I've got almost 2700 subscribers. I Want to get over a thousand subs? You know it feels like the early days of YouTube back in the garage and you know, like, let's get a thousand subs on LBI by.

Anyway, it's quite a nice star platform. It's up-and-coming It's a decentralized Ark crypto based thing. Definitely check it out. I'll leave that in down below as well.

Anyway, they've enjoyed it. Catch you next time.