Hi. In a previous mailbag video, we took a quick look inside this. Xan Zu x2 you 808 valve based hi-fi amplifier on eBay for like a you know, 50 bucks or something like that. It's one of these things.

uses two white J fits with pilot lights here and you can see the little pilot lights in there. look at them. and of course yes, it's got the obligatory wanky blue LEDs in there to make a look or fancy pantsy. Now a few people asked, um, do the valves actually do anything in this thing I They actually hooked up a door or are they just you know, just there for show and it's just an absolute joy because they've just got a couple of regular audio grade op amps inside this thing as we saw last time.

But you know clearly the heaters are on. Look, the pilots are on. There we go. We can see them actually heat it up there.

so it's obviously doing something. and there was another video someone linked in and posted. Somebody else did a teardown of this thing and they were actually connected. but how are they connected? Does it Is it? Actually, How is it actually in the signal path? Because valves are usually high-voltage operating devices and we've got it.

Comes with a six volt plug pack. six volt DC plug back here in the back and we saw some maybe some boost converter circuitry in there and stuff like that. But anyway, I thought we'd do a little lot reverse engineering of this thing of the circuit in here and just see what it actually looks like. Let's go now.

as we saw last time, these are 6j nine valves and it just so happens our we do have the datasheet for these. These are from our Sylvania so we can get the pin outs for this puppy and trace out the circuit. I'll link in the datasheet down below and if we have a look at the actual board here, you can see to our audio grade op amps. Our classic in a double V 3 2 is nothing wrong there at all.

We got ourselves our vowel sockets and as we noted last time, looks like we have a couple of switch mode controllers here. Little tiny V peanuts shot 23 there. You can tell it's a switch mode converter because we got ourselves do. We got an inductor and we've got some capacitance here and A and a controlling element here, so clearly a switching converter.

We've got another one down here as well. So got to switching converters and what they're actually doing, what voltages they're generating, what for, etc. Won't know until we actually take a look at the circuit and look. I mean the valves are clearly connected in here in in some way, shape or form.

At least you know a good lot of the pins are I'm not sure if every pins actually connected, but yet, there's not a lot of complexity in that circuit at all, so we shouldn't have too many problems actually reverse engineering that One of the issues with reverse Engineer and I've done a video on this on how to reverse engineer. In that case, it was the Rygel Diaz 1054 analog oscilloscope front-end so I'll leave that one in if you haven't seen that. So I won't go into detailed explanations of how to reverse engineer this. Suffice it to say that it's It's much easier when you have a double sided board like this, and you know through-hole parts.

Fairly easy to trace everything, so shouldn't take me too long. Well, I thought I had the datasheet and here it is: It's the 6j 9 from Sylvania and this has 6 J9 written on it and it's a triple triode. Okay, fantastic. Basically essentially R3 GA fits with pilot lights in the one package, but it's in a 10 pin bulb and that didn't register until I Went to look at the pin out here: I Labeled things.

You know: there's the cathodes, the gates, the plates, the heaters, Everything's fine, But what? This one is only 9 pins. There is no 10th pin for the cathode, so I'm not sure if it's the same thing, but without the cathodes, making two of them essentially useless. Um, two of the triodes? useless or what? and no idea what the and as it turns out, I Look inside this thing and this pin here you can see is not actually connected to anything at all. So yeah, what? The nothing matches up trap for you players.

Even though this is a six J9, it's actually got six J9 space. Basin you know J Sort of like hanging on the end of it and this is apparently a replacement. and if you look at the eBay ad, they actually say that that valve is actually a replacement for it's not actually a genuine 6 J 9 from Sylvania which we had the datasheet for which didn't actually make any sense because there's a triple triode you know, designed for RF amps and VHF stuff. you know it didn't too complicated to high-frequency didn't seem to make sense, but it's actually a substitute for the EF at the E 180 F which they also mentioned in the ad as well in the eBay ad.

So you look up the datasheet for that and sure enough, it's A. It's a pentode designed as a widening for wideband application, and this is pretty much what you'd expect in something like this. A pentode is quite common in these types of you know now, vintage are you know, tube audio amps and we've actually got some better data. We've actually got some graphs, characteristic graphs, and stuff.

so beauty. Now let's take a quick look at the pentode if you haven't seen it before. If you don't know your valve stuff, it's one of the more complex valves are due to that many various elements in here, it's more complicated than a triode, for example, you might have heard of that. What? This is a pin toad? Why Pentode? Well, because it's got five different elements.

one, two, three, four five. There you go. so let's take a look at them. At the top, here is the plate, which is also called the anode and then we've got a couple of grids in here.

We've actually got three grids. The next grid down is what's called the suppressor grid and then the one below one. In the middle, there is the screen grid and the one at the bottom. Here is the control grid otherwise known as the gate.

So compared to a J FET which is essentially what there was really I they J fits with pilot lights, the pilot light being the heater down here, of course. So you hook up typically a six volt heater here and that just hits up the cathode which emits all the electrons and they flow through the various grids if they're allowed to. Based on the grid, voltages are to the plate up here to the anode. And yes, this is electron current flow as opposed to conventional current flow, which you're probably more familiar with circuit design, but essentially are.

What it comes down to is just like a RJ FET. This is basically our gate, our input here, and then we've got our cathode and our anode here. If we take a look at a little Dave cab during of a rudimentary I'm pentode amplifier, then basically what we've got is the gate here that you saw before. The gate here is essentially that's the control grid down here.

so that's basically the input or the gate of the Jay FET so to speak. And then we've got our cathode resistor down here, which sets the bias and things like that. We've got some bypassing on there, and then typically the suppressor grid is going to be strapped down to the cathode down here, and then your screen grid here is typically connected up to your positive voltage rail here, which we'll call the HT role, the high tension rail or you can call it or whatever's you know, Chad more traditionally used with valve amps and things like that. Anyway, that's just got some up by passing on that one, going to details of why all that sort of jazz is done, But then we've got our anode resistor up here, which then just a see is just a sea coupled off to give our output.

It doesn't have to be AC coupled, but it you know it likely is in a typical circuit. So that's basically you know. Hence why valves are effectively Chaifetz They're transistors. They're You know, it's just that they use old school.

they're filled with vacuums, and they use a little heater element. But essentially they're transistors. or more specifically, of course, field effect transistors because there's electric fields in here. Get it? That's what Jay FET stands for or Junction field effect transistor.

Same thing. Chief it with a pilot light Now without having actually started on the here reverse engineering of this board. I II Tracing out every single trace and seeing where everything's hooked up. Maybe we can just have a quick look of where the pentode is actually in this thing.

Is it used as the input preamp? Because we've basically got our input here Stereo? of course. Hence, we've got two Op amps and two pen toads up here. There's going to be one for each channel. So you know.

here's our input. and here's our headphone output. So is the pentode used as the headphone output driver parent light unit, power amplifier for the headphone, or is it used as the preamp input? Well, my guess is it's typically used as a preamp input, So that's what I think we'll find here. and if you check it out, you'll notice that our input here, there's a trace going off.

It goes over to our pot here. Okay, so it looks like our pot is like directly on our input. that's our volume control. Pot Sauce is effectively attenuating the input here.

and then you have a look here on the bottom that goes over here to these two cups. so they're AC coupling that and Bingo! These go over to the pen down here. so it's obviously used as an input pre amplifier. And of course, the first thing you want to do this is avoid Murthy and make sure you get the pin outs right.

This is actually the bottom view and the way I was able to figure that out. If you don't know, like the internal structure and stuff like that physically see it look, they've got a not connected pin here and it was fairly obvious to see which pin was not connected. So that's actually pin 1. It's a bottom view and when you flip that over there we go.

That becomes pin 1 that becomes pin 1. Make sure your Markham and you don't come a gutter and sometimes it's just easier to D soldier things to look under them. like you remove the chip. see you might be able to see some traces going under there.

That's very handy just otherwise you know you started a randomly you know buzzing out pins. You can see the other traits from there to there which you wouldn't have seen necessarily unless you took that out and you didn't know there was a resistor under there for example, hidden under the socket. so it's well worth taking out. There's the blue LED of course the Winkie blue light.

Hmm. and it goes without saying you only have to do it for one channel once you've traced out one channel here. the other channels kind of be identical so just don't bother. And during the process you end up with some sort of gibberish like this which you have to redraw because it doesn't really make much sense.

Hmm. So after a little bit of doodling, you end up with this. Here it is. Here's the Dave CAD reverse engineer and addition of this hi-fi headphone amplifier.

Now you'll notice here that the screen grid is actually connected up to the plate up the top here. And what this does is it actually converts it from a pin toad into a triode, so it's effectively working. You know you might as well have put a triode in there. Now you know our pros and cons between a pen toad and a triode in terms of a front end amplifier preamplifier like this.

But I Believe that I connecting it as a triode at least offers a lower noise. But there's a whole bunch of other downsides as well, which I won't necessarily go into. One of the disadvantages of that, apparently is that you're going to need a higher dry plate drive voltage up here. So yeah, I you know me.

But doing this also apparently gives it the triode sound in quote marks. Oh god, let's not go there. anyway. They quote this thing: the spec is like point double-o five percent distortion okay in the pass band.

And you know year vowels are supposed to do funky things sonically when your overdrive them and stuff like that. But who the hell's going to overdrive a headphone preamp? front end like this. It's just I don't know. It just seems like a complete wank.

Anyway, what else we've got here? The +6 volts power directly from the plug pack. It's going straight to the heater. and by the way, of course we've only got one channel here. There's going to be an identical channel because it's Daro for the other channel so only need to draw one here and then there's two.

DC Anyway, so this the filament heater voltage hit. not surprising because they said you know, valves, the six volt here, the voltages. That's why the Jesus six while plug pack didn't need it in extra parts for that. No worries whatsoever.

Those two DC to DC converter switch modes that we saw in here not that little five pin SOT 23 and the so8 there one generates the negative voltage which I think was this one down here. We've got three output caps hundred Mike sixteen volts. so there they are three there. We've got three for the plus and positive twelve volts.

So this is a this is the plus 12 volts switch mode and these two will be the input filter caps for those which I didn't draw and sure enough, they do have the suppressor grid connected down to the cathode down here. which is you know, fine and dandy as we saw in the original art. Dave CAD One over here, but you'll notice that it is different. They've got it on the high side here and they taken the output signal from the cathode instead of from the plate.

So if you compare that with a FET circuit or a even a BJT, So what happens if you tap off what is effectively the source or the emitter here on a regular transistor circuit, which you might be more familiar with? Well, it's an emitter follower, a source follower, or in this case a cathode follower. So basically, this thing is not a preamp. As such, it's just a buffer. It just takes our high impedance say input here and just buffers it.

and that's it. And this did confuse me for a second because I was kind of that and I had it in my head that there would be a preamp in here. But don't you know? Of course, because this is a line input. It's not like it's a microphone input and it needs a microphone preamp.

It's designed to take line level signals and give you your headphone out so you don't really need. Well, you don't need an Amp Lair preamp on the a preamplifier, You just need a buffer. That's pretty much it. In fact, they probably could have got away with the buffer and gone straight into the Op amp.

So hmm. as I said. Wayne Factor: You know, tubes look cool. They light up what's not to like and it's not even a complicated in a biased arrangement.

for a cathode. Follow: For a tube cathode follow. you can get you know much more complicated one. So it's as simplistic as you can possibly get.

It's almost as if you know, Yes, they have just thrown the tube in there for the sake of having a tube. And with the fairly generic a tube they've got here and the the simplicity of the circuitry, the fact that they using it as a buffer, it's like like why it's It's just complete weight factor. That's all it is. it's got to be.

But anyway, after that what they're doing the any double V 3 to the AC coupling, that and then we're basically it. This looks a bit convoluted. You might not have seen this before, but you know, let's just take the DC condition first. Okay, when you're trying to analyze circus taking the DC condition to figure out what's going on, it's not a bad way to do it.

So C12, a capacitor is going to be open circuit. Okay, so pretend they're not there, right? What have you got? Well, also, pretend that this Op-amp is not here. Okay, because it's just a voltage follower. Okay, so it's effectively doing nothing.

Okay, so you can take all that out of the circuit and just connect that through to there. What do you got? You've got a 10k resistor. Don't worry about this capacitor here. It does nothing.

At DC it's open, so you've got a unity gain amplifier there. There reason that they're put in the second Op-amp here is for extra drive capability. So it's the same signal. Exactly the same signal, but they're now buffering that with a second Op-amp and then driving the output of this Op-amp is driving through this 47 ohm resistor to the output and this one's driving through the 47 Ohm resistor.

They got the effectively parallel drive there from the two Op amps. Not necessarily uncommon, but in the AC condition. of course, then these capacitors going to matter I didn't measure the values of this so I don't know that necessarily Roloffs But of course you will start then getting some gain at AC because this R12 down here is a very low value for Ohm 7. It's effectively basically just shorts this out to ground.

So you know, depending on the frequency here, you're going to start getting a bit of gain in this things. So how flat the passband gain is in here? you know, presumably over the audio bandwidth? I Don't know. You'd have to get the exact values and measure the performance of it now actually I Posted a photo of this while I was a while I got the wrong datasheet I Posted it on Twitter and some people said that apparently this is this. six J, 9j or /r E 150 F valve is, you know, just a pretty crap quality valve.

It's just, you know, generic because it's popular because, well, there's lots of stock of it or whatever. and well, yeah, I don't know either way. but yeah, and then there they and there haven't hooked it up as a pentode. they've hooked it up as a triode.

So yeah, go figure. The performance of these things Probably me now just to prove that this is indeed a cathode follower. So the output is going to be the same as the input or actually not quite. It's the output here is always going to be slightly less, the gain is always going to be slightly less than one that has to do is various parameters in the pentode itself.

It has to do with all sorts of stuff to do with valve amps. and I certainly won't go into it. but you know the amplification factor mu comes into it and the output impedance. and you know everything else.

And interestingly, they give you the characteristics when it's wired as a triode, exactly as we've got here. Look, the G2 is connected to the anode exactly what we have here, so it's not this data sheets really nice. They give you those values for this particular configuration because it is going to change as opposed to here are the characteristics for just as it's used as a pentode. and the other thing here I mean the Num.

As we said right back at the start, the nominal operating voltage of Valve's is very high. It's like hundreds of volts and in this case what this one means is it's it's nominal anode supply operating voltage and 160 volts. Well, we've only got a plus minus twelve volt rail here, so that really changes the characteristics of this thing, but may not hugely matter in this sort of low signal level application. Although I think they are doing the right thing here.

Although I'm no expert on valve ants, that's to be certain. but the cathode resistor here is usually a rule of thumb is going to be it should be an order of magnitude lower than the load impedance which we've got here at. Well, you know, AC coupled 47k, so you know it There at least got the right order resistor value for the cathode resistor, although that's probably as high as you'd want to go. You know? Ideally they're probably the lower the better.

So let's actually power this thing up and I won't measure the performance of it Said: ain't get all excited, No, it's going point double-o 8% distortion static point double O 5 no and not going to happen. Um, but I've just basically feeding in a signal from my function G No here. I've just got half a volt peak-to-peak one Kilohertz sine wave. Nothing fancy, so let's switch this puppy on and see what happens.

The yellow trace is the input, so that's the control grid input. So right down here I'm measuring the control grid input, this yellow waveform and the and then we're measuring the cathode on channel 2 so that'll be the blue signal here. So let me switch this thing on and see what we get and we expect again, close to 1. No phase issues or anything like that.

no inversion. It's a fault it's a cathode follower, but we do expect the game to be slightly less than 1, so takes a while it's heating up TD Nap takes a while for the for the heater. Come on. you can do it.

You can do it here. We go away and boom there we go. Look at that, that's pretty close. and if we go to what channel - oops I switched off channel - if we go to channel - that's a problem with having you know the single control handle in all four channels.

you never, you know you've got to check which channel you're on before you hit the button. You can accidentally turn it off like that really is quite annoying. So yeah, I did it again. See so you got to make your channel to selected and then we can Center that and bam.

Anyway, that's just a little side rant. So there we go. We're looking at again, slightly less than one. There we go.

just a smidgen under. but you know that's that's pretty good. That's pretty good. That's following pretty well as you'd expect for a cathode follower and you'll notice if I actually switch this off okay, and then I you know, switch it back on after you know, a few seconds, A few tens of seconds, then it's going to pretty much come back straight away because that cathode is still hot In well, literally Hot.

so it's you know, so it doesn't need that ten seconds or so to warm up. But you'll find that the datasheet will actually tell you the warmup time. And sure enough, Bob's your uncle? Check it out. Cathode Heating time: Nominally twelve seconds, maximum eighteen seconds incuriously.

or note, they've actually specified a distortion here, which is interesting, but once again, I'm not hugely familiar with valve datasheet, so I'm not sure what particular configuration that actually refers to, but yeah, one point. six percent for a one Code: 1k low for a hundred milli volts. RMS Input: Hmm. December 1968 Geez, we hadn't even land on the moon then, so that's probably all this little one hung low brand hi-fi amp actually deserves.

I think But anyway, I Think that was rather interesting to have a look at that thing. And yes, the tubes do actually do something There you go there, cathode, Follow us. I Hope you learn a little bit about tubes there if you haven't seen them before. and no doubt all of the tube aficionados come out of the woodwork and rave on and all the audio fools will come out and I'll start raving on and they'll have a big flame war and it'll be hilarious.

Love it! So there you go. I Think this has gone for long enough? I've got 30 minutes worth of material. Ah goodness, got to go edit this thing before I head off on holidays. In fact, when I'm posting this I'm probably on a beach somewhere.

Yes. Anyway, if you like the video, please give it a big thumbs up. and if you want to discuss it, jump on over to the Eevee blog forum. YouTube Comments: All that sort of jazz, you know? Anyway, catch you next time.

Wait, hold on to your hat. I Thought: I would actually get a second opinion on this thing. And what better second opinion than one of the world's best audio designers, my mate. Doug Ford from Doug Ford Analog Design, who you've no doubt seen on the blog before and a former head designer at Rode Microphones former head designer at Jan's Audio, a couple of companies you might be familiar with.

and yes, he's designed amplifiers, mic preamps including two preamps as well. so he's one of the best in the world. What's his opinion? Well, I asked him quickly and he said basically I'm happy to say he came to pretty much the same conclusion I did in that it's it's basically a week really the you know there's no point having the tube in there. Yeah, it's going to Doug says most likely add some much second and third harmonic distortion into there, but ultimately it would perform better and it'll be cheaper and simpler if they just use the any double 5:32 opamp.

So yeah, what a wank! Thanks Dougie! And if you haven't seen Doug's our Microphone multi-part Microphone Design video series I'll link that in that down below. It's fantastic. It's like five parts or something. We sat down for hours at the whiteboard.

It's not just about microphone design, it's all sorts of design techniques and circuit topologies, and all sorts of weird and wonderful things. So definitely check that out. Catch you next time you you.