I Think that, uh, the next thing we'll have a look at is some real microphone designs. Uh, some mine, some Legacy designs. In order to do this, we'll probably want to have a look at Phantom Powering Phantom Powering of microphones has been used for 60 years or more and it's a well-known technique for running power to a microphone down the same wise as you're uh, putting the audio back up. Now let's do it.

Yeah, over on this side we have the uh, the mixer or the microphone preamp or whatever it is that's going to receive the microphone signal down. Over on this side we have the microphone itself and in between we have a twisted peir. It's got to be twisted peir? Yes with Shield Yep, With typically XLR connectors? Yes, any brand preferences. Anol Anol.

Yep, Okay, why do we go to all of this effort of using twisted pair with screen instead of just plain coax or even just twisted pair? Uh, Well, first of all, the screen keeps out electrostatic noise. Electrostatic noise is any noise which is capacitively coupled from, say, a high voltage wire down here. a Mains wire into these conductors. so the screen keeps out the electrostatic stuff.

The fact that they are balanced Twisted pairs keeps out any of the Uh noise that can be coupled by magnetic fields. 50 HZ Transformers Adjacent Transformers Uh current carrying conductors. that kind of thing. So typically what we've got over here in its simplest format might be say, a dynamic microphone which is a coil.

uh, and over here we've got Effect Ly, an instrumentation amplifier whose gain we vary and that might wrap around there as a uh, again, a faraday Shield an electrostatic screen and that's grounded over there so that any capacitive current that's induced into there has somewhere to drain to Mhm. That's a mistake that uh, some of the Americans tend to make in the 110 volt gear. They kind of actually forget to connect that ground to real ground as a result of which, any Mains leakage entering there makes the shell of the microphone effectively electrically hot. So grab hold of a microphone, grab hold of a grounded microphone stand, and see the shaking.

It's all part of. yeah, yeah, all right Now, the whole purpose of an instrumentation amplifier is that it will respond to differential variations in the voltage between those two conductors. It's a differential amp. Yep, but it won't respond to Common mode changes.

That means that we can induce quite a lot of common mode noise on that, and it's completely ignored at the output over there. That's all fine and good while you're using a dynamic mic. Uh, all falls over when you want to use a circuit over here that draws some power. This is where we adopt Phantom Powering Whereby, in its original Incarnation Going back six, seven decades, however long, we would put a transformer in there Mhm instead of having our solid state differential amplifier that would typically then go off to a valve stage Center tap that connect that via resistor to a supply voltage.

Over here, we'd have a Transformer from which we could extract a voltage. So between there and there, we've got a DC voltage that powers the circuitry which is in turn driving that Transformer Okay, the DC path because it's flowing through both halves of that Transformer gives a net zero magnetic or magnetization of the core. So how do you put it? The the core and this Transformer are not being stressed by any DC magnetization. Same thing applies over there.

m the uh, how do you put it? The standard P48 Phantom powering uses a 48v supply a 3.3k resistor. excuse me? Uh, which limits your how do you put it your short circuit current over here to something like about I Think it's 14 16 milliamps. Something about there mostly though you might be using about 8 or 10 milliamps consumption there. Mhm if you're pulling 10 milliamps here.

Well, that causes 33 volts drop across there, that 48 volts drops to about uh, what? 15 volts? and so you're getting 15 volts at 10 milliamps there. Yep, that's enough to power a fair bit of analog circuitry. Absolutely. Plus, maybe light up some idiot leads or something like that if you really want to.

Uh, a variant on that if you don't like Transformers And let's face it, these days, who does because they're expensive and their bandwidth limited. and the only kinds of people who really like Transformers are the ones who also like tubes. Okay, let's take that away. Instead of using a center Center tap transform and a pair of 3 K3 resistors, we'll feed each leg from something around about double that resistance.

If you want to round off, let's call it 6 K8 each from the 48 48 volt Supply Mhm. So far, so good. We're still feeding them balanced. Okay, we've loaded the line by a total of well, what 13k.

but he big deal because we've generally got something with fairly low source and Pie down there capacitor. couple off that into our differential amplifier Beauty That's what we do down that end. Okay, what do we do up here? We can do a few things. The easiest one is to pick up from both of those some DC voltage okay which we can then come over and we can do things like uh, Zena regulate those with reference to that ground there and then use that voltage there to run our circuitry which we then use to drive back into these fellas here.

mhm capacitively coupled. Now, if you've got a balanced Source over here y then Beauty you don't have to go any further now. I a differential driver? Yeah, and incidentally, an example of that might be a very simple fet circuit consisting of resistor J fet and resistor. Any voltage that you put in there appears there m and in antiphase there, so you can quite happily couple those straight over to there.

Nice. Not so nice because uh, the source impedance there is low The Source impedance. There is equal to that resistor there, so we haven't maintained a balanced he and it makes the thing susceptible to noise. Got it? One of the cleverest circuits I think was come up with by the company Shopes Microphone Company s C H O E PS I'm not certain about fact here, but I think this is who came up the circuit whereby you get one of those and uh, now oh, this is going to be tricky to draw.

This is going to be tricky to draw. Uh, okay, ground. Okay, over here. Uh, basically what we're doing is feeding into the emitter of a PMP transistor and into the emitter of a PMP transistor.

Uh, those collectors go into the thing that forms our positive Supply which is actually used there. Those uh bases there. uh, starting to look like a differential? Yeah, yeah. bias down to the collectors with high value resistors typically about 100K and we capacit a couple of signals onto those bases.

Yep, Oh, that's an elegant circuit diagram. Now that's just a different. It's like a differential front end on a opend. It is rather.

uh. The only unusual features are the fact that this Zena here has kind of locked the collector voltage, which in turn locks the base voltage at being maybe you know, half a volt higher depending on the bias currents flowing up here into those bases through those resistors, it's very close. The emitters are only about a volt and a bit away from the collectors. Yep, and fixes those at pretty much that voltage.

Nice low output impedance because they're emitter followers. Exactly. These resistors act as the low resistors for those emitters. It's It's elegant.

It's nice. I Like it. Yeah, and uh, is it still used H half a Brazilian Chinese Studio Uh. condenser microphones can't be wrong.

They're all using this this kind of topology mostly. uh, and it doesn't have a lot of downsides. Uh, one of the few downsides sides is that uh, I'm just trying to think if for some reason you get a short from one of those lines to ground. yeah, uh, ah, yes, you've got enough uh, capacitance there.

that, and that's typically sitting at 15 volts, you've reverse Avalanche those transistors by shorting that to ground. You can basically cure that by putting a couple of uh, uh, reverse bias diods there if you're sensible enough to do so. and that fixes that problem. The other problem involves this phase splitter over the front here, which, uh, on a good day.

From there to there, you get a gain of about7 or8 It's not a particularly good follower. J J Fets They're a follower, but they're not as solid as a Bipolar, so typically, uh, from there to there, it'll have a gain of about 7. uh, and therefore from there to there it'll have a gain of about minus. 7 mhm.

Total gain from there to there about 1.4 1.5 Okay, so far Beauty All good. The only problem is uh okay, you've got this very small Source capacitance feeding that I've run out of board space here. Ohy thump Okay, that's being fed by yeah, maybe 50 puff depending on the fat that you've selected there. It's going to have capacitance here and it's going to have capacitance there.

typical order of magnitude. Let's call that 50 puff and let's call that about, uh, say 10 puff. It's been a bit generous. Yeah, um uh.

Noles these are these aren't tiny tin FS are next level up in Junction size Etc uh reason because you're after fairly low noise. that might be 10 puff and that might be say about 30 puff. Okay, now here's where Mr Miller com the scene. Okay, from that point there to that point there we said we had a gain of minus.

7 mhm. So the total voltage from there to there if you like is 1.7 times the input voltage. That 10 puff looks like a about a 17 puff mhm cap down to there. On the other hand, that 30 puff cap? Well, we've got uh, one unit of voltage there, 7 there3 across it.

3 * 30 is about nine puff. So basically our 50 puff uh, capsule is loaded by an additional 26 puff of capacitance. That's huge. Ah, it's fairly large.

Well, 26 puff against 50 puff you're looking at probably 3db down 3 Tob attenuation compared to the open circuit voltage there. So it's a bit of a downside. Uh, the other downsides are limited linearity. not so much due to that, but due to the fact that these can only swing so much.

Remember I was saying that they're only you know, a vol a bit away from there so it can only swing down so much and that can only swing down so much. can swing up as far as you feel like. yeah, so limited output headro on that kind of circuit, but still immensely successful. Developed I think probably 3 four decades ago.

you know, back back in the dawn of semiconductors. Almost bloody good. circuit. Excellent.

Okay, but let's improve that a little. Okay, okay, we're still Phantom powering over there I'll just redo my terminations and that is still a typical configuration used today. Absolutely by everyone. Yep, yep, Everything from your little cheap uh Macky or Beringer right mixes through to your monsters right and through to uh, some of your multi thousand single Channel Studio preamps.

They'll be doing that. Uh, once you go beyond your C of ,000 into your $10,000 plus stuff, you're probably looking at a Transformer again because that's the wanker Market that you're dealing for. Yeah, okay, um I'm going to get shot down the Flames for that. That's all right.

I'm going to hell. we're all going. Oh, now on that last circuit. The Sh circuit uh I did point out the fact that the impedances there were both low and balanced.

Yes, Okay, so we get all that benefit of uh. noise rejection. Common mode? No. Comm common mode noise rejection in the system.

Did the trans pair have to be matched over here? Uh, not brilliantly. But yeah, 10 or 20% will do the job for you. It's mainly more about DC bias conditions than about anything else, right? Okay, let's once again thly matched. No, it doesn't matter.

Uh, not hugely. It helps, right? But it's not a decider. Let's once again, extract power off those with a couple of resistors. Whack that into a doy great big electrolytic.

Etc Absolutely. Whatever regulation you feel like, and again, we're coupling signal into those. The main criteria that we're after is to get the impedance on those two matched. Yep, Now what happens if we've got, say, a lovely single-ended M Hang on.

What about the mismatch on the caps on the Caps Yes, it does come into the SCH of things. You have to make those caps large enough. Yep, that uh uh. over the frequency range of interest.

and especially down to the magic 50 HZ Mark Mhm that their contribution to. Let's just say that that's the impedance of those two. Or yeah, the overall system. Uh, over here you might be looking at system impedances of well say 6 K8 Say you want that to be maintained down to it will below 60 HZ before they decide to start absolutely going up at different break points.

Yep, so you want their you want their reactants to be low enough at 50 HZ compared to the system reactant that a bit of mismatch there doesn't hurt and it only really takes effect down on the sub hurts. Yep, region nice and easy. So you'd be typically looking at using electrolytics. yes for that.

Hence why I Said cuz they're you know, going to be plus - 20% or something horrible like plus 50? Yeah, yeah. so if they're going to be plus - 50% just put them down at such a low frequency that plusus 50 doesn't matter, doesn't matter. Yep, so you've come up with this circuit. You've got a lovely preamp there at single-ended Mhm.

What do you do That's real easy. We've gotten as far as here. We've developed the world's brilliant single-ended preamp. What the hell do we do? Okay, how about we? I Don't know? Well, why did you design it single-ended to begin with? Why didn't you do a Def Amp? uh, differential outut.

Basically because Def Amps are going to be intrinsically always noisier than a good single ended. Oh, there's another video. Yes, let's couple that on onto there via resistor so that we get a defined output impedance. This has typically got a near zero output impedance, but uh, just for the hell of it, we might give that, you know, maybe 50 ohms or 100 ohms or 200 ohms or something like that.

And let's simply match that impedance mhm and connect it to ground. And all of a sudden, we've created a microphone which only uses a single-ended amplifier, but still has completely balanced impedances. So we get all the benefits of system noise reduction and system immunity from induced noise. I Yes, you do just with Simplicity right? Because we don't need a balanced apology amplifier anymore.

Got it? Let's have a look at a balanced apology far. Let's do it. and.