Hi Sir! In my previous video about our two layer versus four layer and we had a look at using some H field AMC probes to measure the radiation on PCBs and they're very handy bits of kit and the one I've got is the tech box EMC probe and they've very nice. Little said about 3h field or magnetic field probes and one electric field probe along with a 20 DB wideband amplifier for the thing just to amplify the signal. You can connect these directly to a scope without an amplifier, but it's like pretty low level stuff. So external amplifier is nice and this is a nice bit of kit.

Of course it comes with the full performance sight characteristic curves. Here's the coupling loss measurement and they're just great. but the problem is they're like over three hundred US dollars on Amazon So I Thought we'd have a go at making our own H field probe for 10 bucks. Let's go.

I've got a piece of rigid coax and a little pre amplifier here. This was basically our ten dollars on eBay delivered. This is a rigid coax. it's RG 402 can see it's actually got a 400 percent rigid outer shield on it.

it's 50 ohm coax already pre terminated with SMA connectors and you can buy just the coax on its own, but might as well buy them pre terminator like this and these are like five dollars delivered. Oh no, sorry, three dollars delivered on eBay and you can get these little pre amplifier boards as well. they come in. many companies actually make these and sell these on eBay but you know you can get them for as little as like six dollars each and delivered on eBay Like you can get ones like in cases and stuff.

this one's just like you got a feed in your own external five, you know? I Think it's six to 12 volts this one. but yeah, this is a 30 DB gain one as opposed to 20 DB gain in the tech box one. but we can just start knocking a 10 DB attenuator to make them equivalent. But ten bucks can we turn these into our own? Do It Yourself Hey Phil probe.

Well let's there we go and there's just a closer look at the rigid coax. for those who haven't seen it before and it is quite stiff. it really is very difficult to bend these but of course you can bend and form them which is very nice for like I'm sneaking through a product and like if you open any like you know, top end spectrum analyzer or something like that, you'll often find these rigid coax is going everywhere. They're just, uh, they're just fantastic.

So we'll try this. you could. You know you could do it with regular coax, but this rigid stuff is very nice because it just forms a nice rigid structure for the probe whereas coax is just gonna like flap around in the breeze. Now of course you can actually make a hate-filled probe just out of a piece of copper wire on the end of a BNC connector but end in a kind of work, you know? and it actually does work.

But I Thought we does. Splurge the extra three bucks and get this nice rigid coax which I gives us the nice art shielding so it shields out some of the a field as well because we're really after a H field our probe here and if we have a look at this h field probe from tech Box, it's all it's going to be. People think that there's many loops of wire in here. That's not so what this is is.

this is just going to be a just a PCB Likely a very nice out, you know, controlled impedance like Rogers dielectric or or something like that, but it's basically got the SMA on the end. There might be some ferrite syn here, but I doubt it. I think it's just the PCB straight through. So anyway it's just a PCB with a 50 ohm transmission line embedded in there and then it that so the ground on either side and it's going to continue around here like this.

probably with Vir stitching on either side and all the way around. and then they're going to short the other end to the ground over here. so it's actually a shorted loop and they're gonna have a little break in the shield there. And like I said, you can do the same thing with just a B and C You literally take the output of the bends.

They make a loop out of it sometime ever around one or a square one and then just short it back to ground and that's your EMC probe. and they're doing exactly the same thing here. and that's what we're going to do with our rigid coax as well. Now, some designs actually have the break over here, but then it's a non symmetrical design, so I'm sure this one is actually going to have the break in the middle.

It it just gives you a symmetrical shielding across the loop like this. Now, the difference between a round one and a square one is a bit academic really, But if you wanted to get down to individual traces, then a square one with the end that's flat can often be better because you can actually put it right next to a straight trace like that. and then if you have a square one, it's going to couple better into the probe. potentially.

But we're just going to duplicate this round one today so that we can sort of maybe get an A/b comparison. So we just cut our rigid coax in half. Which means that we actually get two of these babies. and as you can see, it just looks like any other coax with the dial, a solid dielectric in the middle with a solid core conductor inside, and then the rigid outer, hundred-percent shielded outer.

Well, it's not braid, but it's a solid outer core which makes it stiff and rigid. hence the name Now because we're going to form this into a loop and then solder it on door, terminate it on one end, and then we need to put the slit in the in the middle of it like this. just before you do that. It's probably best just to practice how to get in there and cut out the shield.

We'll just give that a go. Probably best just to get in there a knife and you're going to have to get in there and somehow like just split it along one edge and then bingo, you should just be able to peel that off. Now as I said, a common way to do this is to simply just bend this back on itself like that and just strip a tiny lever like just a small amount exposed over to here and literally just solder the the center of that back over to the shield over here to form the shorter turn. You can do that, but as I said, then it's going to be non symmetrical.

So what we want to do is actually not have any exposed here. We actually just want to cut it, then solder the inner conductor on there and also the outer conductor as well, and then put our split across there. Now I'm going to actually try and match the diameter here as I said, but the diameter actually isn't really critical. I Just want to be able to do a reasonable a be comparison between these.

You can make it as small or as big as you want the light, hence why the set actually comes with these different sized heads because the larger the loop, the greater the sensitivity, the greater the pick up. But unfortunately, the less resolution you're going to have to be able to spot problems. So generally you'll go over your board with like a larger one first to try and find any potential issues. and then you might get a smaller one like this, which he actually does.

Then there's probably a tiny little hole in the thing. It works exactly the same. It's a smaller diameter you can go over and then you know, get a bit finer resolution to try and pinpoint the exact location of any radiated emission. And as I said, what we want to do is short out both the inner conductor and the outer shield to the shield over here.

so maybe leave a bit out and then bend it over it's at 90 degrees and then solder the whole lot and while heating this up I found that it actually tended to expand outwards. I Guess you do, You know, thermal expansion? I Just wanted to go back straight, so probably best just to hold the loop together while you do it. So there's our end result there, which kind of sort of matches the diameter of our tick box one. And just for good measure, you might want to throw a couple of clamp ferrets around there and maybe either heat shrink those on just to keep it tidy or something like that that just might take the edge off anything picked up by the shield.

And of course, don't forget to measure that it is actually the inner conductor is actually shorted out over here. Just measure that, there's a direct short on there. Now we can just get in there and cut. Start cutting the shield a little gap.

not huge. Just be careful you don't go too far and cut that inner conductor. That could ruin your day. So there you go.

I've got two little cuts around there. like into that now slightly off symmetry. She'll be right. and now we'll just cut that outer braid out of there.

Look at Bobby Dazzler right? So let's test this sucker. I'm going to get the Tech Box probe here as a reference the gigatron board that we did in the previous video. There we go. there's our response there and I've frozen that.

Let's actually plug in our new do-it-yourself probes if we get the same response now. I'm going to use the exact same art Tech box amplifier here. kind of plug it straight in. and yep, it seems to be working as a H-field probe.

Look at that. No worries, let's put it over the crystal again and get the same amplitude as before. Bingo. Look at that.

Ah Bobby Dazzler Let me get an average on that. Check it out. that is insanely close. That's just nuts.

Get real-time They're near identical. You can't really ask for much difference as small discrepancies in the diameter of this thing and their positioning and stuff like that. Now $10 do-it-yourself probe. Basically exactly the same performance is that? Take I'm not going to claim it's exactly the same performance, but you know it's pretty close over this sort of bandwidth and now address mode ID Decoder chip over here.

let's have a look. Let's get the tick box probe. Believe that's pretty close to the same height. Get tongue at the right angle.

It's near on Identical. You can see like the same Peaks up here, like say up at the high end. Here you can see it's getting exactly the same peaks. Everything else the performance is you know is near identical as you can get.

Really okay. let's try it over a much wider frequency range. Now up to one gig and see what's what. We'll get.

The crystal once again. yellow ones, the tech box and this one is our do-it-yourself jaw be more than good enough for any practical application of a H-field probe. And you have to remember that these probes don't have a do really a direct correlation to the far field radiated emissions that you're going to get in the test house. they're designed for like troubleshooting, finding any potential spikes before you get, and spend all the expense on a full art far-field compliance measurement.

But hey, we're using the Tech Box receiver. Let's try our five dollar Joby from ebay and see what happens. This is a 30 DB one, so it's got an extra 10 DB Again, so that might be an issue, but it works. A treat.

6 volt our power source drawing now 22 milliamps I Believe this particular ball can go from 0-60 12 volt. Actually, this is interesting. It is significantly higher amplitude with the lower gain 20 DB 1 here. So so that's a bit surprising, although the gain of these things does change with our frequency.

But yeah, I wouldn't have expected that much difference. That's pretty dramatic. What's going on there I Think you can buy this just on its own for like 200 bucks or something. But basically I'm Yeah, Well, let's do a teardown shall we? Well, that looks pretty schmick doesn't it? But yeah, is it worth the money? Well, you tell me because this one is a couple under bucks if you buy it on its own.

this one, it's five bucks delivered from ebay. They use a different part but the topology is absolutely same. 50 Ohms are controlled impedance line AC coupled input and output and then just feeding in this supply. In this case, just 5 volts being fed in and that's it.

via Via an inductor of course and Bob's your uncle. but yeah, they're basically same thing. it just comes down to which chip is used. It's K 17, 13, 43 so I'm gonna have to decode that job' give it a look, but there's no reason why you can't roll your own if you wanted exactly the same as this.

But yeah, there's not much in it and it's cheap-ass one here. Well I don't know what that job' is, but either way you look at it when you use the exact same amplifier. The performance is identical beauty but we're not done yet because this probe is all exposed. We've got the big ground shield around here and the last thing you want when you're probing around or is to shoot out any pins.

that could really ruin your day. So let's solve that problem. This commercial probe has like a rubber baby buggy bumper protected like vinyl rubbery coating over the PCB here. So what we're going to do is rubber coat this as well with Plasti Dip This is for like you know, automotive, your you know, plasticizer rims on your car or whatever I don't know and that should provide a nice I insulated layer.

There it is, it insulates electrical shock, vibration heats didn't sound, all sorts of stuff. hmm haven't used this before I Give it a burl, just stick that on and well I only had time to give it two coats and it's a little bit how you're doing, but it is like getting a rubbery coated like insulated. I think it needs probably a nothing good two coats. it at least it didn't really like sticking to that plastic over there so.

but of course you could use a heat shrink or maybe a like if you'd this some other dip solution or something you could dip it in. I Don't know how they actually do this coating on here. If anyone knows, please let us know or I don't know. everyone's not going to be happy unless I actually show you what's inside this thing.

So let's see if my guess is correct, that it is actually just a symmetrical split shield with a single controlled loop terminated on one side. just like we did here. And surprise surprise, not really there it is. There's the split in your shield like that, both top and bottom and yet that put a few vias around there.

Not a huge amount, but just enough to well stitch it together and you can probably just see inside there. In fact, it's going to. it's closer to this side than it is that side because they're actually doing this on a four layer board. You can see the vias around there like that.

and yeah, it's just a single 50 ohm controlled impedance trace in there. But as I said, you know this will be a controlled impedance, A dielectric no doubt. But that's that's all that's inside. This is just a shorted loop and be shorted on one side.

Hang on and yep, there you go. You can see the extra vias over there so it comes in and goes around the center of there all the way around with LBJ and terminates on that other side there. exactly like the one we just built, but we did it with coax instead of PCB and I might do a follow-up video. I'm actually you know laying out of board and actually doing one of these getting a PCB manufactured, but the performance is near on identical between these, just less.

the rubbers left a bit of a bit of residue on the board there. Hold on to your hat. hang on. This preamp does actually perform I increase.

I was operating a six volts before it said six to twelve volt range, but of course it's going to be voltage dependent and take it up to 12 volts and look at the response. This was the one before with the TLO Once the tech box and it looks like we could be that 10 DB higher. The wave shapes exactly the same. so let me whack in a 10 DB attenuator in there and I think we're gonna be on the money.

so I whacked a 10 DB attenuator on there and look at that where I like it I have to get the exact right height and hold your tongue at the right angle, but that's near on identical If I move it away and move it in. there you go. That preamp works just fine. There's nothing wrong with it, so don't believe those eBay ads.

It said 6 to 12 volts. Yeah, right. So there you have it. Three bucks for the coax, about six dollars for the board.

like under ten dollars delivered. and are you gonna throw in some heat-shrink or whatever. And as far as I can measure, it's basically the same performance as this $300 Tek box set here, of course I Consider this a part one of this video because we need to like measure the full performance, have to manufacture another one. get like the coupling response and all.

you know that sort of jazz. and and I've got some other amplifiers coming as well I Just ordered a couple of them, this one just happened to turn up first. You can buy ones with like a shield, enclosures and stuff like that, but it works fine. Ten bucks.

Practically identical performance. So I hope you like that and it's encourage you to go and build your own H fill Pros: because they are a really nice bit of kit. even if you're not going to send something out for pre-compliance just a like a sweep over your board, we don't have to do a separate video on their different usages of them and probably do another one making an E fill probe there. even a simpler than the H field.

Pros: But yeah, for 10 bucks, that's absolutely fantastic. So if you found that useful, please give it a big thumbs up. And as always you can discuss down below on in the Youtube comments or eevblog forum. Catch you next time.