Hi In my previous video, I did a teardown of the new array gold Es 1050 - Z There it is. it's still in bits and an amazingly low-cost oscilloscope for 4 channels 399 bucks. It's absolutely incredible. So I was curious to know how they have actually re-engineered the for the input analog front-end channels on these to lower the price point and get four channels for the price of two.

And as we saw in the previous teardown which if you haven't seen, I'll link it in down below. check it out first. I Noted that they had gone for a day entirely discrete transistor based front-end Pretty much so. hey, I figured how are they doing it and also how are they doing the bandwidth limit in on this thing? Because the model.

whether it's 50 mega, Hertz 70 megahertz or 100 mega Hertz it's just a software configurable thing. So how they changing that and living in the bandwidth inside the scope? There's only one way to find out. Reverse-engineer it. Let's go and the first thing again I Want to do is take a photo of both sides of the board So I've got my camera set up on the tripod here and I'm actually using a high F-stop here so that I get a really deep depth of field so that high components are out of focus and I'm making sure I'm focusing directly on like the smaller start surface-mount component on the board.

so I'm going to keep that So hence using the high F-stop value is going to give me a long shutter speed in this case is like half a second or something. So I can't handhold this thing gotta set up on the tripod. and I'm going to set the frame just right. and I'm going to try and keep the same frame for both our shots.

when I flip the board over just so that I can get pretty close to the correct scale factor inside the camera. You don't have to do that, you might older you can and might ultimately have to scale the images in software later. But anyway, it's nice if you can just get it first go and also just make sure you've got decent lighting. The reason I've got it at a big angle like this is because as if I had it flat like that straight up like that, I would get sad Oh in from the cans up here due to my overhead light.

So you really want some decent light. you know, if you've got one of those that light tense or something like that, then you can get really good shots. and what you also want to do is get a torch and maybe not a point source like this. but I'm going to use one of my Up light boxes or something like that and put it behind so that you can see the tracers through the board and hopefully you can actually see if there's any inner traces happening in there.

So anyway, if you've got a nice even light source, you can light up the whole thing at once, and that can help on multi-layer boards. But hey, if you've got like an internal ground plane I mean this one is a reasonably flood field, then yes, not a lot you can do about that and then what you got your images loaded in, then you can do various art processing techniques on them too. In this case, I've converted to black and white here, which sometimes helps, and then I've converted, added a filter to find all the edges like this, and then you can further reduce the colour. So when you go to print these sorts of things on overhead transparencies, for example, you pretty much are left with just the traces and the pads and components and outlines of the components.

and that's what you want to do when you. It's handier to do that when you're using the overhead transparencies. comparing the top and bottom layers. Okay, what I've chosen to do here as the first passes.

Actually get a negative of this board image here and it looks pretty funky when you've done a negative and I've scaled them to exactly the same size, done some micro rotations and things like that. I think I'm getting fairly close so I'll print out one as a reference and then do because you'll need one as a reference and then you print out the other one and then you can just micro scale that if you have to. So I think I'll print this one out on my black-and-white printer and see how that works out and then what you do is print each layer out onto overhead transparency like this so that you can see through it and you line them up and if you scaled correctly today they will overlay like that and all the he is. You basically use the vias as alignment markers on there and too easy.

You can now see and follow signals through top and bottom of the board and then you can ever put a white page on the bottom or insert it if you just want to do one side and you just go on like that so that makes it real easy to play with. And the good thing about transparency is like this is that you can now come along with some white board markers like this or highlighters and you can highlight all the traces are one by one as you do them in different colors so you can have all your ground all in green and or you you know your positive rail in red and all that sort of stuff so you can you know really make sure you don't miss anything. I Try and show you some of that layer alignment up close. If you have a look at those three holes over there they're nice ones to sort of line up.

You can see that there's holes on the top side of that and they just line up perfectly and then over this side we can get get some there and we can just line those up brilliantly in one corner and the next. But often as I said, you are printing out like the first one and then do some micro adjustments on the second one. It often pays just to print it out on paper first. Don't waste your overhead transparencies because often you're not going to get it quite right on the first pass.

Depends on how you good at how good you are at doing this sort of stuff in your edit programs. I'm not that crash shot, but I managed to eyeball it no problems. So now comes the fun part of it: tracing out your circuit and you pretty much only kind of need a basic multimeter just to measure and confirm some resistances and maybe try measure some capacitors in circuit although that's usually not easy to do, but ideally what you want for measuring in circuit resistances. So unless you want to de solder the parts or lift one end of the part not easy on SMD parts, for example, as opposed to the old-fashioned not through-hole type where you could lift one leg on the end of it pretty easily, then what you want ideally is a multimeter with a low-voltage Ohms function and neither.

these multimeters here, for example, have it, but I'll show you how to check to see if it's got a low-voltage functionality on the Raziel since range and the reason this is important is because you're measuring in circuit. If the output voltage of your meter here is too high, then you risk turning on PN junctions in your circuit and that can upset your reading. So you want as low an output voltage as possible. and some old old meters particularly back in the day I'm talking the 80s something like that 1980s it was very popular have a button on there for low voltage Ohms function, but it seems pretty rare these days now.

I Can actually check this Bremen meter here for example BM to 5/7 Nice little sort of hundred-dollar class multimeter by the way, if you're looking for one to really quite a nice meter. anyway, what we can do is measure C as a second meter to measure the output voltage here on our Ohms range and then change the range to see what output voltage we're getting and look on the twenty Meg range here. Alright, I think it goes up to sixty thousand counts or whatever. We're only outputting point two six volts so that's not enough to turn on a pick a typical Silicon P.m.

Junction so we can change our change our range here and there we go. Half a volt that's getting towards something that would start turning on a PN Junction but still, it's not bad. So and you just go through and check all the ranges to see Basically, if it's under half a volt, you're probably doing okay. It should ordinarily be around.

You know, if you getting like point three volts or under, then it's pretty good. So the maximum we're getting out of that is five a half a volt. So that's not a bad meter for tracing out that circuit. But if we went the other direction and tried to use this Agilent u 1272 IE Once again, an excellent meter, but we're still only getting out half a volt.

But if we change our ranges okay, that's not bad. It's not bad, but we'll look down at the Ohms range. we get three point two volts. Ollie Crap, that's enough to even on the kill Ohms range.

Look three point two volts. There you go. So that's not the best for using out for taking in circuit resistance measurements. It's going to switch on PN junctions anyway if you really want to make sure and you are measuring in circuit measure one way that get a reading on your owns range and then swap the leads over and read it again just to see that the value is repeatable.

And if that value is repeatable in both directions then you know you could be pretty certain that your meter is not turning on any in circuit diode junctions. But not hundred percent guaranteed. but it's a good quick test anyway. That's just a little in-circuit measuring tip so we're ready to trace this Circuit down.

and this is the painstaking part and pretty much resign yourself to the fact that you're going to miss something. But anyway, we can at least get a good first pass on this thing. So we've got ourselves our the pin outs and and get printouts of the data sheets and all that sort of stuff. So I've written down some pin outs of the most common parts on here that I didn't know or didn't want to goof up from memory.

and then we've got our transparencies ready to go like this: We've got our multicolored highlighter pens, our whiteboard markers, and we've got ourselves a pencil. Pencils important and remember: always have a rubber on your pencil and the next thing we're going to want to do is search for these pesky SMD transistor codes and they can be a real pain in the butt. So I just type in SMD transistor codes into Google and well look the first four hits here. I've got various of the SMD codebook which allows you to like the first character of the code.

The bases are all that sort of stuff. I've got a search one for example so I can bring in my picture on my board. Look here we've got that seven AET on a whole bunch of these transistors here so we can type in 780 and see what we get. SMD Search Boom! it's an Mm Bt 13 I know for you Standard 3904 NPN transistor.

No problems whatsoever and then they've got entire catalogs like this all the way around here. and then we've got a there's one on the digi-key website and micro commercial components Corp SMD marking. And unfortunately, the issue is is that there's not a huge amount of standardization on these codes, so even with the same manufacturer, they can actually use the same code for different parts. And it's just it gets a bit messy.

so it's not an exact science this. but yeah, it's not too hard to at least get a first ballpark of the codes. And here's an example of where you can get confused over exactly what apart might be. In this case.

we've got two parts on the back side of the board that are labeled 1b. There are SOT 23 and it can either be a standard double to double to NPN transistor as you familiar with here bipolar or it could be this one here which is an IRL M/l 2803 and this is an N-channel mosfet. So it could ever be a bipolar device or regular. you know, just a regular switching transistor double to double to or it could be this power MOSFET here.

Hmm. and of course, the only way to actually find that out is to just draw up your circuit and then look and analyze your second go. Well, Does it make sense for it be to be a bipolar transistor here? or does it make sense for it to have a little power? MOSFET in this particular position? So yeah, we just don't know at this state. So you just draw it in as a generic symbol, make a note, and then you know, fill in the blanks later.

And of course, the way it starts something like this oscilloscope because it's got basically a single input down here on the BNC and it's going to have an output over here and that's pretty much it. And circuits are always drawn from inputs on the left hand side, outputs on the right hand side. That's just the common convention. so you would start with your input here.

there's our input Center pin for the BNC going through a resistor here, going into our relay. there. we'll get our pin out for our relay and then we just start drawing it step by step and then highlight in both the top and bottom sheets here as we go in multiple colors if you need to, and then every now and then you'll get to a point in the circuit here where you like I couldn't see where that one I went to. So I originally had a question mark there because that it went down into a middle layer and I couldn't see it.

It wasn't on any of my transparent overlays, but once I drew the rest of it here I many I Realized well these two bases must be coupled here. So sure enough, I measured the two and they are shorted out so that one that being straight across there like that beauty. and likewise here I've got another point. The resistor on the base of these two coupled transistors I don't know where that goes.

it went down to the bottom to the middle rate. wasn't on my transparency layers. So once again I busted out I knew it's hit. Pretty sure bet it's going to be the negative.

Rail down in there and sure enough it is. And just remember that if you're using this transparency technique, these transistors here on the bottom and all our active devices will be a mirror image of what they are on the top. So if we've got the top here and we've got ourselves, well, let's have a look at the photo overlay. It's a bit clearer here.

If we've got a this pin here, is the base emitter and collector of this transistor? the same transistor on the bottom here because this is actually a mirror image photo. This one is not the base. this one's the base. this one's the emitter and this one's the collector.

So it just. it's often hard to actually remember that when you're doing this. you can often, you know, just have a little brain fart and forget that and goof up the schematic. so it's different if you prefer the physical technique of having the border like this, and then just flipping it over and up trying to trace things directly like that because then when you flip it over, you have the correct orientation as per your the pin out in your datasheet.

You don't have to mentally flip things and Murphy will of course ensure that you end up with a Via that drops through to the inner layer, which you can't see on your top and bottom plots here like this. So you get out your continuity tester. This is where a fast continuity tester comes in and you put it on the point you want for example, and then you can drag it along. I see pins and every other point in the circuit.

And yes, it is a systematic approach Pretty much I Mean if you've already reverse engineered half the circuit, you might be able to sort of guess where it goes next, depending on it's a function in the circuit, the the Vo and net that you have, but you know basically it's It's a systematic search for where that thing goes. And yes is tedious and this sort of stuff does take time. So yet, multi-layer boards can be a real pain. And yes, it can be even more annoying when your net is on this side of the board and you think it goes to the other side.

or you've checked everything on one side. So you're going to go like this and get the tongue at the right angle, apply just the right amount of pressure so that you pierce any oxide coating on the solder joint that's another trap and then get on the other side and start probing. It's just taken forever and then the next thing you've got to watch out for is traces under chips which you can't see like this. The T or V 2 7, 4 quad Op-amp here now.

I originally didn't trace this one. I was too busy I got to the input to the Op-amp and I was too busy tracing the FET amplifier around here and just got carried away and extended that out anyway. I've come back to here and I started tracing it out and the inverting terminal down here at Pin 2 drops down to a Via down in here and let's have a look at that. You can see that it okay, it drops down on the bottom side.

so we go down to the bottom side here and it goes through a capacitor like that. So I drew it. you know? So I drew it as I saw it. but of course that doesn't make any sense.

You've got to have some sort of negative feedback happening here. So you look at the top side again and you go. well. it's is it going to an internal layer and then going out? Well, it could be I've already found traces on the internal layers, but check this out.

Check out these resistors here. These look like classic feedback resistors for the Op-amp and you'll notice that there's a trace going off underneath there. So aha, does that one go off under there under the chip to that the backside of that pin which you can't see well. You get your multimeter out and you buzz it and it turns out yep, it does holes right on the money.

So yeah, it's just got to watch out for those things. Use a bit of intuition when it comes to these sort of things. You don't know that that can't possibly be right, and you know that you have to find those resistors somewhere else and they're always going to be close by. and once again, you end up getting stuck on ones like this.

I Mean here's our input switching relay. Here's our main input: AC Coupling cap and we've got a resistor here which is a 4.7 Meg and it's going off to a Via there, which just goes nowhere like it. Well, of course it goes somewhere. it goes into an inner layer, but we can't see traces anywhere else on the thing.

is it going off this way? That way you know who knows what it's going to. This is where we, you know we had no clue until we've done a good lot of the circuit. Now we can have a look at the circuit and see where it can logically lead to. And here's the circuit that we've got so far.

Please excuse the crudity of the model I Didn't have time to build it to scale or to paint it. Now we've got our input over here. of course there. We've got an input attenuator here, which is then in which you can bypass with these two relays here.

I will relay contacts that's actually the same physical relay on the board. It's the big large one you can see there down there. we go and what have we got? Here's our AC coupling cap in here. so we've got a path going down here I'll explain this later, but we've got our AC coupling cap here and we've got some clamping diodes and here's this mystery: 4.7 Meg resistor.

It's just going off to Lala land I I Didn't bother tracing it back then, but where does it go now that we've got the rest of the circuit? Well, I couldn't find the output of this Op-amp here either. It didn't make sense. it didn't go anywhere. so you know what the hell's going on, so it's got to be going somewhere.

The output of the Op-amp and I couldn't trace that one either. and I figured well look this part of the circuit here because we've got AC coming through here and DC coupling through this path selectable here. our virus solid state relay here. well this must be the DC path and then over here I figured out that we had some and E squared pot over here the adsr the Ad 5207 and that's just buffering that and that's feeding in.

So this must be the offset control for the channel. the DC offset to shift the waveform up and down, the vertical vertical position control on the front panel. So the output of that has to be going back into here and offsetting the signal before it gets into our FET amplifier over here. So by deduction, this point here must connect to this point over here.

And sure enough, I buzzed it out after all this time and bingo, that's where it went. So if we have a look back at our overlay, there's our 4.7 Meg resistor. And here's pin one of our chip all the way over here. the output of the op out there.

So this drops down here like this and it must go under. Well yeah, it probably goes under because I couldn't see it through these gaps in here. When you push, shine light through, it couldn't see it. so it's probably running under there like that around there and up to up to pin 1.

Yep, up to pin 1 over here like that. and who? After all that work, we're finally finished. Well as finished as I want to be. To figure out how this thing works, this front end works, and how they're doing the bandwidth selection and yet this is pretty damn ugly.

So I've redrawn it a bit nicer. Here we go. let's take a look at this sucker. It's drawn in Dave cat of course.

So let's start out here. Here's our being simple: Got a 75 ohm resistor and then a selectable attenuator in here so you can bypass it. There's a common relay there, just bypasses the whole lot. There's a little trimmer cap in there, and well, a bit of compensation across the input resistor here to smooth out the response.

and well, nothing fancy at all. And then it's AC coupled and goes into our FET input amplifier. And this is a very standard arrangement here. We've got a J FET on the input here and a low impedance and either follow our output and that goes off to the defense which I've got on a separate sheet here and we just got some bias resistors here.

It goes down to the negative rail and also, you'll see that the input here was clamped by above Ninety Nine Diode. It might look a bit weird because I've got the ground up the top here. It's actually negative reference. So we've got a Zener diode here, clamping it at some voltage below the rail so you don't want the input, any input transients to go straight onto the rail.

you want them to be clamped to your Zener diode and then you've got a 2k protection off to the rest of you Rayo and that's pretty easy. so that's a you'll find this configuration. It pretty much standard in tons of oscilloscopes way back to the older analog scope days. Very very common and this part here is rather interesting because or this amp is always AC coupled.

so it's only amplifying the high-frequency stuff. It can't amplify the DC stuff directly from the input here. To do that, it's tapped off right at the output to the switch here. and this is our AC DC coupling selection here.

Like sometimes, it's done like old analog scopes is done right in the input. Here they will have like a big AC coupling cap in here somewhere in which you can short out. But this is done differently because we need to bias the position of our waveform inside our front-end amp here before our vertical position control. So all this section here basically passes the DC stuff and does offset as well.

So if you're measuring DC on your scope for example, and you've got DC selected and it's bypassing this AC coupling cap here, then the signal is not going through here. Of course, because that's AC coupled, it's got to go through here and then up to here. And then that allows us to add in another DC signal here for our vertical position control. and they're doing that using a Analog Devices ad 5207 a squared pot.

You'll notice the question Marquee I didn't try Oh, I Couldn't trace where that pin went and no, it didn't go down to ground, it's gone somewhere else and I we just went or whatever, it doesn't affect the functionality of the circuit anyway. and I will likewise here with the question mark. If you see question marks anywhere means: I couldn't readily trace them and I just gave up I Can put some more hours into it and try and find a bit anyway. And then we've got a couple of muxes here.

Oh, I didn't label those at 7 4 Hz at 405 3 We use a couple of these in the Rygel front end and look, they're put in an 8k - you can select an 8k - resistor in series with that a squared part. and yeah, so they're just getting various art settings for that and you can put in another 2k resistor as well. And then they've got some sort of amp here. I Couldn't figure out where it went to anyway.

Doesn't matter, that adds in a DC signal into here and allows us to shift and position that waveform up and down before it gets into the ADC here. Now one interesting thing to note this Op-amp which is at LV 274. by the way, it's only like a like a low bandwidth precision low-power Op-amp so it's not the full bandwidth. If you're wondering why it's you know it.

They can get away with like a 3 megahertz bandwidth. Op Ampere is because all the AC stuff is going directly into the FET here. so this is only affecting the DC shift offset so you don't need a high bandwidth Op-amp here. although in the Ds2 thousand ones, we'll take a look at the schematic.

They did actually use an 8 megahertz art Bandwidth Op Amp here instead of this sub 3 megahertz one here. But anyway, you'll notice if you were keen that this is open loop arm. Well, it's not ok because it wouldn't work as an amplifier so it's got to be closed loop. but I couldn't find where this resistor.

Well, I found that this resistor went to the vertical position control here, but there's no feedback from here. I Mean it's obvious that this Op-amp here has to be in this feedback loop here, so it has to tap off here somewhere. but damned if I could find I'm going to probably have to have another shot at it. and if we have a look at the old Das 1052 II I think we'll find it's much simpler than this one and you'll see that it does actually feedback.

And here's the schematic for the D S 1052 II The older one, not the 1052 is Zed this new one. it was drawn by a Helene so thank you very much. ie. So here we go here.

they are side by side. We've got our input attenuator here so it basically are exactly the same thing happening with the bypass relay there. I've just drawn a little bit expanded. He's done it like this so a little bit different and look as I said, a different Op-amp here.

they've got the Ad 85 1-0 I've drawn mine sort of slightly separated out. That's just how I decide to do it I wasn't referencing this one at all, so everyone draws things up slightly differently. And this uses an eighty eighty five, one zero. and it is like an eight megahertz band with one, but it's basically the same thing.

Here's the AC coupling cap we had down here, which has been bypassed by the well in this case. it's a solid-state relay not sure what it is inside the 1052. II there's the part number if you want to go look it up and oh yeah, we've got the offset amp Here is the same for Meg seven resistor going into the J FET here. the same clamping arrangement, except they clamp it to the rails where they got a Zener here, but basically exactly the same thing.

and then what else are we got? Here we go: Alwah amp. Our fit amp is almost identical, almost identical. They've got another, they've got a resistor in here, whereas these the emitters tied to the collector here. but it doesn't matter.

and they've got an output series resistor here. they didn't have it in this one or I couldn't find it. So it's a slightly more compact configuration here. And by the way, they've got some, you know, fairly decent filter in here.

They've got a two-stage filter for this supply and then I've got a diode between these two, so I'm not sure if this is parent something else I didn't actually follow it off, so it could be. Anyway, they've got some mark clamping between the rails there and this open loop configuration of this DC offset amp here that I was talking about and how it should ultimately be referenced back to here. Well look, if you have a look on the 1050 a schematic. bingo here it is.

Look the inverting terminal of the DC offset amp there goes through an 806 K resistor directly to the output here as I thought it must be and very curiously, look, they've got that same value 806 K resistor here and I just had a look at that to verify and know it's not actually connected over to here like that. Of course you wouldn't, you know, have your output of your Op-amp on there, so you know. but they've got exactly the same value resistor exactly the same connected to the inverting terminal over here, but this one goes off to the vertical position control whereas the 1052 Ii just has the Yo channel one position just adding into there at the lower part of that resistor divider there. So yeah, it's You know they've substantially changed things, but anyway, there's got to Altima be some feedback from here coming back and getting through whether or not it comes up through here, through the E square pot and everything else it could be doing that I mean that one there I checked that one's not connected to there.

so I don't know what you know I don't know exactly what's a going on there. but anyway, it's got to come back. otherwise that thing to be open-loop and it would work at all or it'll work as an excellent comparator. So anyway, all of that is essentially exactly the same as what we've got here, except the big difference we're going to see next.

look this amp. here we go. this on mine. it goes off to the next page which we'll take a look at next but on this one it goes into a well.

a rather expensive if your try to save cost an Ad 837 programmable Gain ampere and then we've got a differential driver. Once again, that's another analog. They know it's a national are part LMH Six Double V 2 and these things cost money right? You know, Even if you they're not manufacturing, you know a hundred million of these scopes so they're not going to get them. Rock-bottom price and manufacturing.

You know, tens of thousands of these scopes. So the price of these chips actually matters. So they've done away with these two chips as we'll see and replaced it with a complete discrete transistor solution in this design. And if you remember from our teardown video, that was the big surprise and take away from the teardown was that it used an all discrete transistor solution instead of these chips which we had before.

So that's how they've really ran, engineered and lowered the price of this 10 54 Z and probably the reason why they can afford to put 4 channels in here whereas before they could only afford to put in 2. So this is what I Really wanted to see how they've implemented this discrete transistor solution and how they're implementing the bandwidth filter in between the models. So let's take a look at it. We've basically got a very state.

It looks a bit complicated, but if you ignore that, okay, that doesn't exist there. Okay, then you've got a pretty standard AR diff arrangement here. Here's our input from our amplifier on the from the J fat and low impedance emitter follower on the previous side here and it's a pretty standard differential configuration I Couldn't figure out another question mark couldn't figure out where that came from. So we've got our differential output here and this comes around and it goes straight into the ADC of course, straight through.

But then they've got these switchable filters hanging off here. They're switching in different value capacitors from each one of the differential lines down to the negative rail. and they've got four transistors which I Didn't know where they go off to, but they're presumably go off to like the micro controller, the digital control so that they can switch these capacitors in and out and they're matched up here. of course.

So if you're going to switch on this one, you would switch on this one as well and that would have an 8 20 yard Par cap from each differential line down to the negative rail. And likewise, you can switch in the 560 here. And of course, when you've got two different values like this, you can actually have 4 different configurations. You can have none on at all, so they're not having any effect on the line that just passes straight through, so that would be full bandwidth.

or you could on the 560 puffs caps here and that would decrease your bandwidth again by a small amount. And then you could switch in your eight and then disable that one and switching your 820 puffs here and that have yet another bandwidth. And then if you really want to do, you could switch on all four transistors and have them in parallel, and that would give you your greatest bandwidth reduction. So there's four different selectable bandwidths there, and they're doing that on the differential line.

Very interesting, so it looks like they've put a bit of thought into this. and the D is 1050 to Z of course, is only a recent model so, but it looks like that They've planned it our way back when they originally designed this thing because they've put in four different band width art configurations here, so presumably are you turn them all off and that's a hundred megahertz? Or maybe they've got the five sixty puffs on for the hundred megahertz or whatever. And then they turn the eight twenty puffs on to give you the 70 megahertz bandwidth model. and then they might turn both on or four on there to give you the 50 megahertz D is 1052 Z So I think that's how they're doing the bandwidth selection.

And of course that's all going to be under a software control as well. So when they program the theme, they program the model number at the factory and it gives it your software control bandwidth. But what's going on under this lens cap here? Well, let's take a look at that, shall we? Basically, they're duplicating the exact configuration again. So imagine that's now gone, right? That's now gone.

and we're And we're looking at exactly the same thing because the input comes in here and drives both bases there. But they have selectable control over here. Once again, the base of these are transistors. These bias transistors down the bottom have the go into HC 405 3 so they can switch select one or the other.

And what's the difference between the two? Well, the only thing I could find is look, this has a 200 ohm series resistor. this has a 680 ohm. They both have 1k twos in there so they are different. So what's happening here as I Believe that is the bandwidth selection for the 20 megahertz bandwidth filter and they're doing that in the differential amplifier itself.

Oh, and by the way, I haven't drawn it in. but just as an aside from the differential output here, they were actually tapping off. two of those was going into a TLO seven - that's at the TLO seven to S4. they've got some PMP BCA five sixes here and I couldn't figure out the feedback out configuration there.

but anyway, they're just obviously some sort of drivers. Nothing to do with the bandwidth configuration. Anyway, I haven't gone that far. This is what I really needed to know.

This was the money shot. Well so there you go. A little attempt here at reverse engineering that new Rai gold Es 1050 to Z and I found some interesting stuff in there and that's what. I Laughter.

This wasn't a complete reverse engineering effort to do absolutely the whole board. I Really just wanted to find out what was going on in that discrete amplifier. our front end there and there might be errors in this I haven't taken it, haven't simulated it, any of that sort of stuff. That'd be the next step to make sure I haven't even sanity check that haven't double-checked it, done whatever.

So if you do see any obvious errors in here, please men let me know and I can correct them. But yeah, we found some interesting stuff how they're doing the bandwidth limiting in there. so I hope you enjoyed that little look at just one technique for reverse engineering a board like this: Sarah Everyone's got their own way of doing it and depends on the board. you know you might do it differently, but this was actually a bit of a pain in the arse.

being a multi-layer board, quite a few traces going off where I couldn't see them and obviously if you had, if you're lucky enough to have like an x-ray machine or something, that'll be really handy to do stuff like that. But anyway, so this did. If you think that this is like an hour or twos work, think again. A lot of our I put a lot of hours into actually just getting this far.

It was lots of red herrings and a little you know, dead end traps and stuff like that and just really kind of annoying and tedious work to do. But hey, if you want to reverse engineer something like this, this is what you have to do. And if you really wanted to be a hundred percent sure, you'd have to go through and check it or get someone else to check it. and then you've got a simulator to make sure it all works and you've got the correct you know configuration.

You haven't left anything out and I guarantee there's an error or two in there. But yeah, I found out what I wanted to find out and that's the main thing. So as always I'll link in all the data sheets and everything for this these are chips I'll scanning these little Dave had drawings and you can have a look at those and please if you see any errors, let me know. If you've got any comments, please leave them down below or on the Eevblog forum.

and don't forget if you liked the video, please give it a big thumbs up because that helps a lot. Really does. With all the you to be searched stuff and things like that it keeps me up the top. so thanks Catch you next time you you.