Hey, I'm here in Chino, California at Manley Laboratories. Thank you so much for inviting us in here. Cheers. Showing us around. You have some awesome facilities here. All of your stuff is made here and I really respect that. Yeah, thank you. Made here in Southern California. I started out as a guitarist. That's what got me into audio. And so I kind of fell into this misconception that tubes are for distortion. Tubes are... In guitar amplifiers, they're definitely for distortion. And it's that distortion that we love the sound of with guitar amplifiers, right? And one of the important things about vacuum tube design is they run on very high voltages and they have very high head rooms. But when you want to push it hard, hard, hard into distortion, the tube clips very gracefully. And that's why it's so beautiful sounding in a guitar, as opposed to a solid state device that's running, say, on 15 or 18 volts or something. And then when you hit its limit, it just squares off and sounds horrible. Just super abrupt and horrible sounding. So that's one difference between vacuum tubes and solid state is you've got a much larger linear range to work with vacuum tubes. And then when you do push it to the limit, it's going to distort very gracefully and very gradually. As I've started to learn more about pro audio equipment, specifically your stuff, it's not always for that extreme saturation why you'd use a tube. Can you talk about why you use tubes in your gear? Say we're talking about a big power amplifier and we had some... We had dynamic range that wasn't all killed off by the world's compressors and limiters. And you have some very quick transients like the strike of a piano. If you've got all the signal is mainly like this, and then you have this big spiky transient, because you have all that headroom, you're not going to hit that piano and it goes "grrrh" Like if you hit a digital over in your A to D, you know about it. It's either good sound or completely not usable. Imagine if you're A to D, if you could push the limits like that and you kind of sort of wouldn't really hear it. That's what we're talking about with vacuum tubes. Take the Manley Variable Mu, right? You can purposely overdrive the input to get a little bit crunchier sound, not crunchy, just kind of fuller thicker sound as a color choice. And again, this is using gear on the extremes that we're talking about, but you can use it where you're just looking at the extremes a little bit, just to get a little bit more flavor and color into your sound on purpose. If you had to describe what that flavor is on a physical level, what is it? It's kind of thicker. Talk about the science behind it, this harmonic distortion is kind of what we're talking about. Is that right? Well, there's a lot of misconception in the world about that. If you're getting even order harmonics, structured distortion components versus the odds, and that would be to explain a little more, you have your primary frequency and then if it's a doubling or tripling of the frequency, it's the odds or the evens, right? And so the even components are generated usually from a single-ended circuit topology. As opposed to push-pull circuitry, which uses complementary pairs in transistor land, or can use a plus and a minus supply with tubes. Like the Variable Mu is a little push-pull circuit. Its harmonic structure is actually more odd order harmonics, not even. So it's again, if you put them up on a bench, say you put up a Manley Variable Mu, which is push-pull versus a Manley ELOP limiter, which is single-ended, you'll see the difference with the harmonic structure right away. So you'll see that, say you're doing 1k, you'll see 1k, and then the next harmonics will come up at 2, 4, 6 on an ELOP, say single-ended circuit, or 3, 5, 7 on your odds, right? And so they do sound different, but we commonly think of the even-order harmonics as kind of rich and luscious and beautiful sounding, as opposed to the odds. I sometimes fall into the trap of getting too technical, but at the end of the day, I remember hearing you say in a different interview that empirical data is important, but there's some emotional data that should be taken into account as well. Sure thing. You know, I'll grab a little prop right here to illustrate a story. Here's a little prop. It's a little transformer that we developed right here. We had a prior part for a little 50 watt amplifier using EL84 tubes. And in the 90s, it was my chance to work on that product, and I was creating a new stereo version of that monoblock with the EL84s, which would later become the Stingray, the Manley Stingray. I had remembered that there was a review in Stereophile, The reviewer said, "Well, it's a nice little amp, but it's got no bass." And I was like, "You know, why is that? It's not the tubes fault. The vacuum tubes give you all the frequencies." The limitations come from other areas, and especially through the transformer. So I said, "Let's look at that output transformer a little more." I mean, it is kind of small, isn't it? Physically kind of small. We measured what we had, and we measured the prior transformer we were using from an outside vendor, and got a handle on its measurements. And tonally, I knew what it sounded like from listening to it. And then it's like, "Okay, so if we're looking for more bass, where are we going to get that from? If we're looking for more power at low frequencies through this little transformer, where are we going to get that from? Well, we need to push up the inductance." So we changed the geometry of the wind, and I think we might have changed the metallurgy of the laminations, or whatever we calculated back then. And we got it on the bench and measured it, and it's like, "Oh, yeah, look at that. We're getting a lot more power there. Look, look, 20 hertz, 10 hertz. Cool. Look at that. Wow, that's a big improvement." And then we wound another one right here in this room, probably on that machine, and took the pair downstairs, put them in the amplifiers, put them on big knife switches or whatever, and listened to one versus the other. And it's like, "Okay, I know empirically we just saw the measurements. This new one, it measures better. It's got way better bass response, but it's really boring sounding. Why is that? It's like, "Well, let's back off." We could see it in the measurements. That one was, "I'm going to make up numbers. I don't remember the numbers." But say the first part was 50 Henrys, and then we went up to 130 Henrys, and then we came back down to 90 Henrys. Again, I'm making up the numbers, but whatever. We wound another set of parts, put it on the bench. Oh, yeah, see? It's saturating a little earlier at the lower frequencies, but let's not give up. Let's listen to it. And we listened to it, and what I mean about gradual saturation, it produced a sound out of this little baby amplifier that made it sound like a lot bigger amplifier, because there was this little bit of extra bloom or fatness in the bass in the lower frequencies. It just made it sound wonderful. I mean, kind of like a tone control, but we're talking about very subtle distortion products. It all contributes to the overall sound of the product, and it's our job to balance the empirical data that we can see on the measurement bench with what we hear and the emotion. And that's something that I remember distinctly that day when I heard that second prototype. I got goosebumps on my arm, and I was also kind of moving and tapping. It's like, that's emotional. I can't measure that. And it was involuntary on my body that that was happening, right? But that's what matters. So we're talking about transformers. You know, you can see the glow, the beautiful glow of a tube, especially on the Stingray. Transformers don't glow too much. No, there's a block inside, right? But help us understand what's going on inside. What is a transformer? Well, the transformer's job is to take the, say, in a tube amplifier, the primary side is hooked up to the vacuum tube anodes, the plates, where the high voltage lives. And its job is to take the signal from that point, and it might be working the impedance. It's kind of like resistance, but it's got other factors involved, but we call it impedance. The impedance level, maybe you're working at, I'll just make up a number, like 4,000 ohms or something. And then the transformer has to transform, has to take that 4,000 ohms and reduce it to, say, eight ohms for your speaker, or five ohms or whatever. So it has to transform that impedance down from thousands of ohms to something your speaker can deal with. And at the same time, it's going to not allow any of the DC voltages to get through it. That's its other big job, is DC cannot pass through here. And why is that? There's no physical link between the primary and the secondary. Whatever's happening on the primary side of the transformer is inducing a change in the secondary. So there's no physical path for that DC voltage to go through in plain terms. So that's the transformer's job. So DC will not pass through it, and it's transforming the working voltages or impedances or whatever from the primary to the second. Whatever happens on the front end is going to happen on the back end in proportion to how many turns are on the primary side and how many turns are on the secondary side. And the ratio between that is called the turns ratio of the transformer. And so that's very, very simple introduction to transformers. You got to start simply. This was then built right here in this room. But yeah, we've used these same machines to wind all of the transformers since the mid-90s when we set this room up. I know some people watching this are hearing transformers. They've heard of that before. They're not really sure what it is. And they're hearing winds, turns. They're talking about turns of wire. Can we show this machine? Yeah, let's look over here. Here's some machine. These Adams Maxwell machines are made in Southern California. Why do we wind them here in California? To have the consistency of production, this is the very same machine that's wound all these transformers over the last almost 30 years. What we do is we've got a piece of paper with its winding diagram. This transformer is very simple. And it declares 750 turns of whatever gauge wire that is. And they load it up onto this tensioner and bobbin here. And they pull the wire on onto the bobbin. It's going to attach the first winding. It's going to start at pin one and end at pin five. There you go. And then they will program the number of turns. They'll program the pitch, which is how fast it's going back and forth. And they'll set the limit from whatever that is, 0 to 5 eighths of an inch or whatever. They'll set that with these knobs here. So it's going back and forth. And as this turns, zzzz, and it's pulling the wire, and it's populating the wire back and forth like this. And then it loads up that turn. Then you stop. And it says finish at pin five. You tie it off at pin five. And then you start the next one. Some transformers, like this one, it's a very simple part. There's only four windings on it. Other transformers, we might have 19 windings on that are a lot more complicated. This is just a one-to-one transformer that's used for line level in a Manley VoxBox for instance, and its job is to give you, it's not needing to remove any DC voltages off the primary, its job is to take the output of the tube circuit, pass that through to an XLR connector in a balanced form, that's its job. So we took a single-ended signal from the output of the VoxBox and gave you a balanced output through a transformer. And again, between each layer of winding, they're going to put a layer of mylar tape, sometimes we use a thing called fish paper, which is a paper product, it's an insulation, sometimes it's a mylar tape, or there are different materials that are used inside the transformer to separate all the layers. But yeah, let's look at what happens after we've got the bobbin all wound, because we can't make anything transform quite yet without the laminations. Here's a big transformer. So once we've got this, and in this case they've taken the layers and they've combined in serial and in parallel the different windings, and they've combined them inside, and then they brought them out on these wire leads. So they've terminated the magnet wire inside and brought them out on conventional wire leads, because that's how we're going to load this into the chassis. But we're not done yet, right? So they've wrapped up the bobbin real pretty with this cloth tape, then they're going to take the laminations. Now you might have heard of toroidal transformers, and they're wound around a circular donut core. There's different core materials, there's another thing called an R-core, which is kind of a combination of an EI transformer and a toroid, sort of. But these are called EI transformers, and why is that? I wonder why? Why are they called EI transformers? I have no idea. It's really a mystery, huh? So how that works is when we're going to load in all the laminations, we're going to take our E's and we're going to load them into the tongue is here. We're going to load them in here by hand like this, and alternate. So we're going to have stacks of E's on one side or another, you know, like this. And then we're going to put this one in and that one in and just stack it all up. And then after that, we're going to turn the transform on the side and the spaces between the E's, we're going to slide the I's in so that it all solidifies. Now you ask, why don't you just take a big old block of an E and a big old block that's an I and just bolt the damn thing together? Well, that's because in magnetics, when you're putting a current through this, you'll get a current that gets set up in the lamination material. And if this were a big block of steel, you would get a lot of currents in there. They could have their own life, you know? So we use very skinny, isolated. These have a little coating on them, so they're not electrically coupling to his neighbors on the floor above and below him. And that's why is to reduce the eddy currents in the core material. That's why we don't use big bricks of steel there. We use real skinny little guys. So that's that, but that's why they're called EI transformers because the lamination, they're stamped out of a grain-oriented steel in these shapes. So you've said that, you know, the craftspeople we've met here, they've been doing this for 30 years, some of them on the same exact machine, done it on thousands of transformers. And let's not name any names, but that's not how every company is doing this, right? It's not the same machine. It might be hundreds of machines. Well, we're very rare that we even wind all our audio transformers in-house. Most people purchase them from a transformer supplier. Right. So the tolerances are much wider in that case. It certainly can be. And also the integrity of what we're making, like you said, it's been made by the same guy on the same machine with the same drawings, with the same materials. You know, we buy the same magnet wire we've always purchased. The metallurgy of all these laminations, like on the little laminations, you can order them with different amounts of nickel versus steel or so on. And, but yeah, you have to be able to trust your supply chain in the people, in the design, in the machines, and all of that to hold tolerance for all the parts. And what's great about doing all this here at the factory is that a part, that same transformer we made in 1996, we can make the exact same part today in 2023. So like if for some reason it stopped working, you could replace it with brand new parts. It's going to be the exact same part. There's no changing over time. Cool. Thank you again. I think for me, this really helped me visually understand how transformers work. It helped maybe dispel some of the myths out there. If you want to watch some more content we've done with EveAnna, go ahead and click the video that's on your screen now and we'll see you over there.
