From Tape Op: Issue No. 15

But how in the world are you gonna get the world’s best input transistors, where the signal first comes in, with the very unique requirements that those transistors have,” into a tiny silicon chip? “That’s a very unique set of parameters to make the input transistors. Meanwhile here are the two output transistors, meant to handle lots of power and whatever, radically different requirements. They’re made in a substantially different way. Well how do you get the world’s best input transistors and the world’s best output transistors on the same 16th of an inch square chip of silicon? It’s just… you can’t do it. There are going to be compromises. It’s like trying to have one vehicle that’s both a Porsche 944 or whatever, and an 18-wheeler, over-the-road. So there are going to be limitations in monolithic op-amps, and oh, by the way, you need some real resistors on there as well, and some capacitors. So you certainly have the potential with a discrete op-amp of having much better performance. You can have much higher output current, which then allows you to use much lower-value resistors in the surrounding circuit, which then can give you even lower noise. And just one thing leads to another, and you can get much better performance. The trade-off being it takes up a lot more room, and it’s 50 dollars versus maybe 50 cents. But we’re not talking about building clock radios for the side of your bed, where who cares what it sounds like. This, in theory, is part of a business where people are pouring their heart and soul into the music that they’re recording. And so the difference between a 50-cent op-amp and a 50-dollar amp is not nearly as important to them as it might be to someone shopping for a clock radio for their kitchen or something. They don’t want to spend more than 9.95 for the whole radio, much less… You can do good pre-amps with monolithic op-amps, [but] I think judging by the reaction from my customers, this discrete design offers something superior. But again, your mileage may vary and your personal preferences are yours.”

“And then vacuum tubes, that’s a whole other area, of course. You can get 12AX7 tubes, or ECC83s, whatever, from 10 different sources out there. There’s lots of companies that make them now, and each one has its own sound qualities, so how do you deal with that. There are usually coupling capacitors in tube mic pre-amps. I suppose there’s a way to do without them, but usually you need them, so there’s that potential compromise. And one thing about [the JT-16-B] input transformer, which is a real important point, I think, is that it has a very low impedance ratio. 150-ohm primary, but just a 600 ohm secondary, which is an ideal match for the 990 op-amp. The 990 likes to look back and see low impedance coming from whatever is before it. So that transformer matches perfectly. Well vacuum tubes get their best noise performance when they look back and have a high source impedance.” At this point Hardy pointed to a Jensen JE-115K-E high-ratio transformer. “That’s the kind of transformer that you would typically find in a tube circuit as well, with a high impedance ratio to match better with the requirements of the tube. The trade-off being you get more voltage gain out of that transformer, you get about 20 dB of voltage gain, compared to 5.6 dB out of this, but [the JT-16-B] will perform in a much more linear fashion. If you compare the specs of this transformer to that transformer, you’ll find that this has wider bandwidth, the distortion is lower, the phase response is more linear, group delay, however you want to look at it, better specs—simply because the lower the impedance ratio. Laws of physics at work here - the lower the ratio, the better it will perform. So most tube circuits, I don’t want to speak for all of them, but typically tube circuits require a high-ratio transformer. So you’ve got the compromise of a high-ratio transformer. You’ve got the variability of all kinds of different tubes out there. You’ve got coupling capacitors in the signal path, so there are some of the potential problems. Again it becomes a matter of execution and how well somebody does a particular design. But all things being equal, that transformer will perform better. And no capacitors should sound better than capacitors. Again, I don’t wanna step on anybody’s toes too badly.”

“When I was designing all of these things, and really particularly with the M-1 I guess, because that’s a ready-to-go product, you don’t have to invent your own boxes and power supplies and stuff. So it’s the first really mainstream kind of product. I knew that it was a great op-amp and a great transformer, and with no capacitors in the signal path it ought to be beneficial somehow. There are so many great designs that people come up with that just fall flat on their face for whatever reason. It may even be a great design but there’s something about the packaging or the look or there’s a vibe there. You know, we’re dealing with human beings here; you just never know. I’ve had a lot of people compliment me on the metering of the M-1 for example. That seemed so important to me that, when you’ve got a signal of some unknown quantity coming into the preamp, this is sort of a great unknown. You really need to know, well how am I doing here? What kind of levels are we dealing with?”

Since when Hardy designed the M-1 he was unsure in what genre it would find its niche, the broad range of styles for which the people use the preamp has pleased him. “One of my earliest customers is a guy who was in Southern California, now he’s in Nashville, last I knew - Michael Wagener from Double Trouble Productions. He’s doing what I would call, and forgive me if it’s the wrong terminology, but heavy metal kind of stuff. You wouldn’t want to be anywhere near the guitar amps without serious hearing protection, that just real high level kind of stuff. There’s that kind of stuff going on, there are people doing spoken-word, voiceover kind of stuff, talking books. But there are all kinds of M-1s down in the Nashville area. There are people doing classical work, people doing jazz work. You name it. Whatever there is to do, it seems like the M-1 does an excellent job. It doesn’t have a certain sweet spot and then everywhere else it kind of falls apart. It just seems to do well whether it’s at the lowest gain settings or the highest gain settings or whatever. Again, I had my fingers crossed hoping people would kind of like it, and it just seems like it fits in everywhere, for which I’m grateful.”

After Hardy designed the M-1 preamp, “Deane Jensen asked me if I’d be willing to build his concept of a preamp, but using my packaging as the basis for it.” The joint collaboration resulted in the Jensen Twin Servo mic preamp. “On the Jensen there are two 990 op-amps per channel, whereas the M-1 uses just one. For two or three years, I sold that directly to Deane Jensen and then he would market that as his own preamp. But sadly Deane died after a couple of years of marketing [it]. And after a few years the folks at Jensen finally said to me, the people who took over after Deane passed away, they finally decided, well hey, you’re the mic preamp guy, why don’t you sell the Jensen twin-servo mic preamp? We’re up to our necks in transformer orders. We do transformers; you do mic pre-amps. So we’ll sell transformers. I certainly tip my cap to Deane Jensen. He was extremely helpful over the years. I certainly miss him, and I think the industry as a whole misses Deane Jensen and his contributions.”

Hardy designs his pre-amps on old Hewlett Packard 300 series computers. “A guy I used to be in a band with back in the early ‘70s, at some point in his life decided to write a CADD program.” Hardy wouldn’t name the ex-bandmate, because he abandoned the program over ten years ago. The anonymous programmer started the CADD software in the late 70s, before personal computers, so he wrote it in the native Hewlett Packard Pascal environment. Today, Hardy runs it, and maintains it himself too, on several Hewlett Packard machines, which he can buy used for affordable prices. He showed me some of his current designs in the CADD program, ranging from re-worked versions of his old replacement cards with better grounding patterns to a redesign for one of the bedrooms in his house.

Hardy finds running a small audio company challenging but worth the effort. “It’s frustrating and there are so many things that divert me from designing stuff. I’m behind on things I’d like to get designed. But I certainly like it a lot - I mean, I don’t think I would trade it. I was in enough bands where there were insane people in the bands generally screwing things up and causing chaos, and working for other people at various points, that I just decided I’ve had enough. I’m gonna just do my own thing here and call it the John Hardy Company. And that way I don’t have to answer to anybody, I don’t have to drive to work because I work here, and I like it a lot.” In addition to selling Hardy’s preamps, the company also distributes a German direct box, the AMB Tube-Buffered Direct-Injection Box.

He has a small staff helping him. “Just a few people, part time people that do assembly work. And I let them do as much as I can, but when it comes to the final testing and quality control and everything, that’s what I do.” His employees install the cards and the chassis, attach the knobs, and cover the unused areas with blank panels. “But then I come in and start neatening things up and give it a good inspection and some last little bits of soldering I have to do. And then power it up and make sure that everything is working right, calibrate and test it. Stick it in a box and ship it out, hopefully to a customer that will live happily ever after and love it for years.”

“As I’m designing more and more pre-amps, at least until I see some reason to change, I will be using those same basic ingredients. I’ll have different versions of it, for example [a preamp with] a variable high-pass filter. I’ve had various people say, ‘Well the M-1 is great, I wish I had a mixer, it’s like an M-1 with pan-pots, maybe and a high-pass filter’ - So a high-pass filter. I’ll probably add a switchable resistive pad on some models, whatever I come up with in the future. Offer more choice, but still based on the same basic principles, the same transformer, same op-amp. Variations on a theme, yeah - if it ain’t broke, don’t fix it.”