Hi, It's time to tear down this new Rygold Hco 4000 series scope. I'll leave my uh, our first impressions and noise measurements. uh video LinkedIn up here. if you haven't seen it, it's got a new Rye goal Asic in it and I believe it also has a new Rygold front end and of course it is a 12 bit Joby So uh, this same I think the same front end chipset in this is going to be used in new HDO 1000 series which is on the way anyway.

let's take it apart, shall we? A fan of the feet? Really? Oh, you betcha Beauty Oh, look at that! Oh geez, that was a bit horrific. Got medieval and it uses kind of like the de facto standard four uh, screw Arrangement here just to hold on the back case. And by the way, on my first impressions one: I Totally missed the fact that this has a battery pack slider on there. Hence, all this wanky shape in here is designed and like Clips up and here designed to have a, um, a battery pack slide in there.

but I I Don't know if that option is available yet, but it's there. It's obvious, so obviously there'll be a back piece of being there. Anyway, this will lift off and we've got the metal work. Oh, and that's oh yeah.

there you go. So that's actually screwed into the back of the metal work there. Go to a couple of RFI uh tabs there and you can see it's probably going to be one big single board construction. Oh no.

actually I don't know. Anyway, we'll find out. Is it one big single bottle? Oh, look at this dual fan jobby dual exhaust on this bad boy. No wonder it's loud.

Jeez. So anyway, airflow wise, it's all coming in this side. I Like how it goes down onto the board as well. you can see the heatsink in here is uh, angled in the right direction.

oh that that heatsinki stands all the way down into there. When we get that off, you'll see it. But anyway, the air comes in here and they've got the fins in the right direction for the airflow and also the power supply inside there. so the power supply is in the upper half here and then that all comes out the other side.

But yeah, as I said in the Uh in First Impressions it is rather annoyingly loud and whiny. and as is common, we have to get the nuts off here. There's no washer, no star washer under there. Before we can lift off the entire thing.

there we go. although yep, yep. I have to disconnect a few things I'll get back to you and we're in there you go. and wow.

look at the heatsink on the front end here. and also would this be the Adcs as well? So um, yeah, the new front end, the 12-bit front end AC is chewing some power here. That is like the heftiest heatsink in this thing. Um, and usually when you've got four channels like this there'll be four A6 uh here for the front end and then probably two.

um, Adc's in here and then the acquisition um Asic and then the whatever processor they're using to uh run the Uh Android operating system on this thing. Anyway, that is very impressive and nice. Big single board construction and I am liking the look of the mains assembly down in here. The Earth point there is very nice look at that and they've got all uh crimps as well.

It's neat and tidy though should be easy enough to upgrade the fans in there for quieter ones. So yeah, check out the Just multi-stranded That is just one big crimp terminal in there. They've got the multi-strands coming out here because I guess they wanted to like reuse the connector over here and obviously this is the battery contact board and we can actually get that out there. You can see they've got a nice little plastic interface there with uh, metal threaded inserts so that's really nice.

So there there you go that is just um. it's basically just some mosfet switch in there for your battery so it chooses either the battery uh interface or it just comes from the power supply and that's it. Took me a few seconds to figure out what that board down there was, but it actually tells us uh AC triggering board. So they've got an update coupler there and so it just takes the mains output and the mains input here and just gives us an optocoupler output which then goes over this cable goes over to the main board.

So yeah, that's just for our line triggering. So they're going to do a lot of trouble there and we'll go through the board in some detail. but this is the LCD connection so we can take that off and then we can see it. But I'm going to just going to now remove all the heatsinks so that we can take some high-res photos and uh, go into it if you don't know I Always have high-res photos available on my EV blog Flickr account um linked over on Evblog.com Well, I Found an Rtex 7 under there and this one over here looks interesting.

We'll take a look at that and now let's see what's under the front end. Oh oh okay, there you go. Oh, isn't that nice die-cast case? Oh isn't that beautiful? Um, they've used uh, sill pads there and let's have a look. There's our four front ends.

Jeez, there's not much in it. Remember this is an 800 Meg bandwidth software upgradable front end. So this is 800 Meg So this is all their custom. I Do believe they've rolled their own custom front end, but I'll take some high-res photos and we'll go in there and check it out, but there's not much.

They've got two relays. uh, what brand are they not sure from this uh, point and up here we've just got uh yeah, they heatsinked all three of those chips. I'll tell you what. I'm pretty impressed.

Everything in here is metal threaded inserts for all of the uh, like holding down the main metal chassis here and you normally get self tappers for that kind of stuff. but anyway, that just easily popped out. Then we can see the front panel board here for the smart probe interface thing so that'd be going off on its own ribbon cable. I would be assuming then we have our Optical encoders because one of their marketing claims is that these are not uh wiper type uh ones that wear out.

These are Optical or photo encoders. You know they've got a lead and a photo transistor in there that you know detects the motion in either directions so there's no contacts to wear out inside these things. I'm not seeing a brand on it though, but um yeah, this is one of their uh what are their brags and it has been a a sticking point for several scope Brands um over the years and did Rigo actually cop some Flack over the years for it? I'm not sure. um but anyway yeah I think a few of the scope manufacturers have on the Eev blog forum and other places.

Um for you know their their pots wear out, you just use them so many times and but these are Optical So there's the interface and interestingly, they do have a large cut out in here. that would be for the Mixed star signal connector which is not there. It's not even populated on the board, so don't get your hopes up. Um, no, this is not a mixed signal.

Score doesn't have an arbitrary waveform gen, but interestingly, they do have uh, cutouts there for two extra B and C's that would be um, you know your ARB gen output and whatnot. So um, interestingly, what I thought was, um, it might have been the LCD connector is not. You can see it actually goes to the front panel board here. so they've got a ribbon and this cable actually going through.

uh, presumably for all of the uh contacts here. Um, that could be individual power going over to power. Uh, the active probes. and that power hypothesis makes sense because it goes over to here like this and there's um, like some switch mode uh, chips in there.

So yeah, looks like that's active uh probe power and the back of the interface board there. Check that out. They've got poly switch protection on all of the um, like there's like four for each one. That is a lot.

Um, I Guess they expect a lot of goose, you know, idiot. Engineers short out active probes all the time. It's interesting how they've gone to the effort to like emboss these out from the other side. I Guess that's to, uh, get a little bit more height for the connectors underneath.

Let's go through the main PCB here and I am capturing this in 4k so you will be able to see all the detail. But as I said, hi-res photos are available on Evlog.com If you want to have a squiz now, this is the main PCB here and if we compare it with the Rygold 5000 which was quite a few years ago but that was their new Phoenix chipset. I think it was at the time and they and they had like an eagle on there. um this one has like a I don't know, it's some sort of flying bird.

almost looks like a toucan or something. Um, but this is supposed to be the Centaur chipset so I don't know. um anyway, this is the original Rygold 5000. It was very uh, simplistic here and I don't believe I ever took these off because these were uh, adhesive glue.

so yeah, we couldn't actually see what was under these. even the front end I did take the the cans off anyway. I was able to get the heatsinks off these because these weren't uh, adhesive. So we've got a Xilinx RTX 7 here.

Um, so it's the main bad boy. So all their new Ultra Vision 3 stuff is inside the RTX 7 and that's the main memory there. There is no extra memory on the bottom I show you the bottom of the board but there's basically nothing um of note on there at all. Um, so yeah that RTX 7 is not cheap.

and if I'm right, digikey puts that at about 205 US Dollars Uh, 40 of quantity. So we'll have a look at the main processor over here in a minute. But anyway, we have our bird here. Um, somebody had fun on the PCB but this is really what we, uh, care about is the front end down here? So I actually take a closer up look at this.

Now as you can see, they're all identical. Um, all of these as I Don't think there's a single difference, uh, between them and they require substantial heat sinking. So this is the new Rygold developed Uh custom front end. but I believe this is the new Centaur chipsets upside down so all the electrons are going to fall out, but that's the RT 8847 or 4471.

uh, something like that. So um, yeah, a few hairy scaries on there. Um, so we've got two of those, so one of those obviously uh, shares the two channels and I Believe that's the case. you do I You know, turn on channel one and channel two and it halves the sample rate because you've got your single ADC here like this.

But if you turn on channel one and channel um, three like that or channel four for example, you'll get the full sample rate on two channels. most. uh, Scopes work like that and this in here which is also, uh, heatsink. This is actually you can tell by the component Arrangement down here that this is the PLL this is the clock generator PLL for this thing and that is a TI Jobby.

it's an Lmk0482 ultra low noise clock Jitter cleaner control cling Jitter cleaner. um with dual Loop uh PLL sounds. we've got Roomba function. Um, and yeah, it's just there you go.

Uh, femto zapping for all you uh, you know clock aficionados. you can go for your life in that. Anyway, this does have a external 10 Meg oscillator in I Don't know if it's this one down here. it's one of these.

Um, anyway, yeah, all this miscellaneous circuitry around here. this is for like, internal. uh, it's got 10 megahertz reference out. 10 megahertz external reference in as well.

but I'm not actually seeing the oscillator there though. so I don't know what's doing there and are these two leads? are these two? I don't know I haven't powered it up without the back on it, but uh, they look like there's there's two leads there. I Mean we can zoom in on that that that that looks pretty leady, doesn't it? So this here is the Rygold 5000 front end like this and uh, as you can see, there's the BNC input. Then we've got our AC coupling.

uh, switch in. uh, relay here. we've just got one. uh Asic here.

whatever that is I don't know. could even be a discrete off the shelf uh chipset and then all of your uh, divider stuff around here and then a just a differential pair output buggering off there. but the new one is actually substantially different. Let's have a look at the front end now.

I've actually uh, taken the bottom. so this is the bottom side of the well, the front, the bottom side of the actual PCB As such, but it's the B and C it's the business side of it and this is, uh, the top here. but this, uh, 5000 series rygold front end here. Um, this is like a lower end scope and you get it for like sub a thousand dollars now.

So it's more fair to compare this one with the upcoming HDO uh 1000 which I'm getting in another week or two. Um, and we'll take a look at that. So I expect a simplistic front end like this. so it's fairer to compare it with the Rygold 7000 uh series which I've done tear down of that as well.

and here we go. It's not rotated. unfortunately. can I rotate? So this is the seven thousand.

You can see that we've got two relays here uh, which we didn't have on the 5000 and we've got the Um AC coupling uh, relay here. that's the um little Cosmo uh solar State Jobby there and uh, it looks like I think I don't know if I Um saw this in the previous one, but uh, it looks like this actually has a separate 50 ohm path like this and a separate one Meg ohm path. um I might have missed that in the previous teardown, but have a look. but if we compare that with the new Hco 4000, here it is.

It's um, relatively similar. We've got our two relays here. You'll note that they are exactly the same and it's interesting to note that a Chinese oscilloscope actually uses Japanese Fujitsu relays because the best relays are made in Japan all the best stuffs are made in Japan they're actually a Fujitsu uh Joby There you go Ultra Miniature uh relay. Um, they're not shielded or anything like that, but they do actually specify uh, you know, high frequency characteristic here.

So yeah. Superior contact spring for high frequency characteristic so it complies with various standards. but they're not shielded. uh.

relays. They're not like high frequency coaxial relays or anything fancy like that. So this is a remember: this is an 800 megahertz front end when I was a boy I had to make front ends. Um, they didn't look like this.

Yeah, um, it's just absolutely incredible. Anyway, we've got the new Rygold Asic Here this is the RT 1642 um IQ So I there's no info on that at all. if somebody can get info on that I doubt rygola gonna give us anything I don't know I should ask. Maybe maybe they will.

You know they might give us a block diagram. They wouldn't give us more than the block diagram or thing, but this is Rygold's secret uh weapon here and this is of course this is not a 12-bit front end, but it would have the dynamic range and low noise capability because this is a low noise 12-bit Well, 12 bits is the converter which is further up. It's not in the front end, but the front end has to have the low noise uh, dynamic range for the Um you know to enable uh, the 12-bit functionality. But anyway, the uh so so the relays are the same.

So it seems like this does have a separate 50 Ohm path and a separate one Mega own path as uh, people were spec calculating on the EV blog. Forum You can see tiny little piddly traces there. They're really thin, thin as anyway if it goes through the relay like this. if you're AC or DC coupling doesn't matter, it goes through the relay and then it comes through like this and this is your AC path like that going into your dividery amplifier, differential, driver, front end chip.

But the 50 Ohm path actually is here and I have actually measured this. This point here is actually physically connected through to if we draw this, the relay has three. please forgive my mouse. but it has three contacts like this and this is the center pin and then it flips between there or down here.

Yeah, so that point is actually it's not actually connected over to here. it's actually physically connected through to just the just this actual input pin here like this. So I've measured that, but the 50 ohm, it looks like this flips it on. it goes through here I have measured that resist to there even though it doesn't say it on the top.

that is a 50 Ohm resistor and then it goes through here. Once again, contact over to here and this is your 50 ohm path. Here's another 50 Ohm resistor here and it goes up into there. So separate 50 Ohm and one Mega Ohm paths.

Interesting and once again, we've got all of our divider stuff like this. But this is Rygal's new secret weapon, which is their low noise front end. And as you saw in my previous video, this is not a 100 microvolt per division front end. It's only a one millivolt per division front end.

Uh, 100, 200 and 500 micro volts are software magnified, but you can do that because you've got the 12-bit converter. and anyway, uh, people over on the EV block. Forum We'll put the link down below. They have actually measured, uh, the noise and compared it with the siglet and uh, a Lacroix I think something like that.

And yeah, the Rygold does a pretty decent job. The front end is pretty decently. no low noise. um, especially for the uh cost.

So yeah, it's it's really good. but this is an entire front end. I Mean you know there's nothing doing over here. There's a whole bunch of bypassing and stuff.

uh, looks like we have a filter there because you can tell it's got the extra extra contacts in the middle, extra contacts in the middle there. you can see those. but apart from that, like there's nothing else doing here. um sorry I do have to.

My head's in the way. so let me move my head. um, floating. Dave head there we go.

But what? I didn't show you down here this. This image is flipped just to make it, um, the same way around. but this is a 4053 the Classic 4053 Jelly Bean 4000 series CMOS Analog switch. It's still used in everything.

This is a 272. There was another one. If you spotted up closely up on the main board, there's probably a whole bunch of these. The 272 is just A Here, it is.

Um, it's just a Precision um dual op amp. It's nothing you know. super special. So this would be doing the Uh bias function which this has which is actually different to the offset.

This actually I I Got that wrong in the previous video. I Just assumed that the bias in the front-end settings was the offset. but it's not the actual physical offset where you move the waveform up and down. That's different to the DC bias.

You can actually add a DC bias to the front end and I think I suspect that's what that's uh, doing there. Yeah, but there's nothing else um here doing at all. So it's that's an 800 Meg front end. There's not much costing I don't know what this Asics uh, cost of them? What sort of process they did that on I don't know if you know what, uh, sort of, you know process they would have used for that thing.

Obviously it's pretty high power because like, it needs a pretty decent heatsink as you saw. Now as for getting the signal out, you can see that there's actually two. There's a different way. There's um, actually two differential pairs coming out of here.

So these two here and these two here. So there's two differential pairs coming out. so I don't know what the deal is and I can't see those on the bottom of the board. so I think they're actually going through that.

This is what this via stitch in here is for I suspect Um, so yeah, that's obviously I don't know, it's buggering. Off to the ADC what is clearly Rygold's uh, four gig sample per second. ADC So this is their Centaur chipset here. um that they you know claim and the you know, the Ultravision three technology or whatever that's just been running the RTX 7 Fpga So this is the bottom of the board here As you can see like there's not much doing.

you can see all the Matched length traces. We've got the Wiggle Wiggle Wiggle years in here. Check those out! So what's going on here is when you see both pairs like that, take a snake. It means that they're matching the entire length of this pair with all the other pairs.

They're length matching. But when you see a wiggle wiggle wiggle year in just one of the traces like that and down here as well, what they're doing there is matching one the one side of the differential pair with the other side of the differential pair. they're just matching between the two. So there's two different types of length matchings and you can mix and match those two.

They want to ensure that the obviously this is coming out of the ADC. They want to ensure that data coming from the ADC is exactly the same matched timing going from both channels over to the Fpga here. Yeah, it's really nothing else on there. It's not very exciting is it? So we want to look at the processor.

Now here it is is a rock chip rk3399. Hadn't heard of this before. Turns out it's actually um, quite old. I've got a data sheet of 2018 here.

Um, and it's an Arm processor. It's running the Android operating system I think I showed that in the previous video. So yeah, it's got Cortex A72, quad-core Cortex A53 with separate Neon Co-processor Uh yeah, it's got um, H2c4 265 decoders, 10-bit jobbies, um, 1080p, 30 frames per second, Jpeg encoder decoder, pre image processors, and stuff like that embedded. 3D GPU when we don't need that.

But yeah, there you go for those playing along at home. Um, it's got cryptography, extensions and stuff. But yeah, I don't know, it's just. they presumably chose it I Don't know because it's cheap, or they have experience with the Eco system or whatever.

You could choose any Arm-based processor here, but this one's you know it's it's at least four years old. It's not something new and mysteriously there are two buttons up here I wonder what they do. They're not marked and they're populated. so that's interesting.

But as I mentioned before, this is the power supply up here. by the looks of it. Or at least part of that. um for the connector that goes off to the active probes on the front end.

That's like mostly. Um Power there they had all those wires going over. don't know why. Um, they're just separate fused ones I don't know and I don't know if this had uh HDMI output directs you.

Did it. Did it. Yes. Display interface one HDMI port.

There you go. So I'm not sure what that one's doing. Let's look it up. Yeah, I'm not finding any ready info on that.

So like you can see that some of the pairs go direct from the rock chip over the HDMI driver on there, but others come from the 4C. So I don't know what's doing there anyway? Um, here's our touch sensor for the touch screen and this is our LCD ribbon cable. You can see those going on physically over onto the LCD over there and basically that's coming directly from the rock chip over here. I don't know how much memory is associated with that.

You can decode the Micron part number over there as you can do for the Fpga as well. For the Micron memory, we've got a real-time battery. uh, backup? Yeah. so apart from like your auxiliary ins and outs here, um, there's nothing doing.

There is a third unpopulated uh USB over here. So I don't know. But yeah, obviously like this board doesn't even have the options for the of what you saw with the connector cutouts on the front panel. there's no option for mixed signal uh, waveform, gen or anything like that.

So nothing doing there. Um, at all. really? Uh, and one of these inputs over here was extra external trigger. Was that external trigger at the top? Anyway, we have a very nice populated uh JTAG over there for us, that's excellent if you want to hack this thing.

Is there any like serial? Oh yeah, there you go. That could be a uart interface. Geez, the the real Mouse operations really laggy on my 4K when I'm capturing my four straight K screen doesn't do this on the 1080. But anyway, this is the power input.

It's just I Think it's just 12 volts in. um, for the whole thing really. And um, and then you've got, you know? Yeah, yeah. Look, there's obviously like there's 0.9 volts here.

Is it? Yeah, there's 0.9 volts here. There's you know. there's separate voltage for the CPU there's 3.3 volts there. There's another.

uh CPU jobby over here. Vdd Center here I Assume that's the supply for, um, the high speed. Uh, split? um. transmission line termination.

So that would be what that's for. There's another analog. VGC Management analog: VCC There's one volt over here. there's 2.5 volts over here, there's 1.8 volts here.

there's another 1.8 volt generator here. There's like it's crazy. In fact, what we don't see here inside the front end, we don't actually see a low noise. Supply So uh, this looks switching.

like what's going on here not seeing any major inductories. So of course you wouldn't have a switching Supply power in your ultra low noise front end here. You're just not going to do that. but I'm oh maybe five, five, five point two.

would they be? Uh, low noise? They might be powering the front end perhaps. but I would have expected to see one for each and I didn't see it on the bottom. There is a three pin jobby there. but I don't think that's doing it.

So yeah, they must be supplying them outside, so that's that's surprising. Didn't expect that. There you go. That's it for the Um, Tear down the Rygold HTO 4000 So uh yeah, this is a it's a serious bit of Kit As I said like, the performance of the front end seems pretty good.

Like it's not industry leading or anything, but for the price point, um, it's pretty good. Now for the HCL 1000 series, upcoming Uh should be that should be on the Uh on the plane in another week or two. Um, so we'll be able to tear down that. But as I said, I wouldn't expect uh, the Dual relay front end because this lower bandwidth it's not 800 Meg but I suspect it might you be because it is a 12 bit.

Once again, it's 12 bit. so it's going to be using the new Centaur chipset and I suspect it will use the front end 800 megahertz capable. Obviously, who knows it might even go higher than that. We don't know.

Um, but yeah. I expect it to use the exact same chip. but as you uh so or in the Rygal 5000 I expected to eliminate because it won't have 50 ohm right? so it won't it won't need the relays, it'll probably just eliminate both of those and it'll just have the Uh AC DC input akadaka and um yeah, Bob's your uncle but where is the power supply for each each of the front ends? I'd like ultra low noise I Would have expected like this bad these bad boys to have a a low noise um, linear rig on each one of them I know maybe it's built in Anyway, it'll be interesting to compare to this with the HTO 1000 a much cheaper one which starts at 6.99 this one starts at 26.99 I think it is. Um, so it's significantly, uh, more expensive.

Um yeah. I don't know if they're like cheaped out on the Uh processor over here the RTX 7. You know you'll find that in any you know. top end oscilloscope.

uh, these days something like that. So I don't think they're neces like they haven't really skimmed there I guess um and they've developed their own custom Um, front end and new center chipset here. Oris Center Like both of these combined or something that might be, you know that might be the thing. But yeah, it's like it's amazing how simple the front ends 8 800 Meg Front end? Come on and it seems to be a pretty decent front end low noise, 12-bit capable front end.

one millivolt per division? Um, yeah, really quite amazing stuff. and this will have a software bandwidth limiters in there as well. I suspect Um, so yeah, there's probably like an I Squared C bus that comes into it or something that actually commands sends the commands uh to it because there's no separate Uh PGA programmable gain amplifier. It's all in here.

There's no separate differential uh driver. so it's got a programmable gain amplifier you know with with the attenuator uh, system and stuff. and it's got the differential driver output. It's probably got adjustable bandwidth limiters in there.

20 meg, 200, Meg the 400 and 800. Uh, Meg Um, because I I think they'd be implementing those in the front end and not actually like digitally inside the Fpga. but uh yeah. Anyway, you can tell that from the Um shape.

I mean uh, everyone over on the EV blog forums analyzing the uh, shape of the noise curve and everything. Um, and you can actually tell a lot from the shape of the noise curve. It's rather interesting. Anyway, that's really cool.

so if you like that, please give it a big a thumbs up. And as always you can discuss down below. I'll link the EV blog Forum down below where people are discussing this bad boy and uh, comparing the noise and analyzing, doing performance analysis and all sorts of stuff. So if you're interested in getting one of these, um, and you're you know curious to know how good like the 12-bit performance and front end is.

There's people over there doing tests and comparisons and stuff are really neat and I'm impressed by the construction of this thing too. it's you know it's really good and and it right us into have engineered this pretty well. so I'm quite happy with it and thanks to all the patrons who help out, pay for or the stuff that I do here um this is my full-time job and they help pay for it so that's always LinkedIn down below and is very much appreciated as is the EV blog story. If you want to uh, support you can uh buy like a multimeter on the EV blog store clamp meter coming soon by the way.

So I hope you enjoyed that and found it useful. Give it a big thumbs up comment because that adds to the metrics and it you know it really helps beat the algorithm. Catch you next time.