Why do I write about my audio hobby? I sat on this piece for a while, because it is perhaps not so productive to publicly expose my oddball enthusiasms. Arguably, the whole thing is a waste of time — and indeed, as you will read below, I myself decided that I was rather too involved, and wanted to cut back as I too have better things to do. But, you could say that about most any hobby I suppose. Nevertheless, I seem to feel an obligation to tell others what I have discovered, both people already involved in this subculture, and perhaps those that might be interested. It’s strange that this goofy thing actually has funny aspects and complications to involve the adult mind.
It seems to me that creative people have creative hobbies. By “creative people” I don’t mean artists necessarily, but also entrepreneurs or fund managers — people who aren’t just cogs in someone else’s machine doing what they are told, with no particular aspirations beyond that (or even understanding of what I am referring to), but who want to do something in the world that is theirs alone. Or, even if they make a living in a typical sort of way, find an outlet for their creativity in their free time. Likewise, noncreative people, or rather, people whose natural creativity has been suppressed, do not have creative hobbies. Their “hobbies” tend to involve buying things others have made, which is also what 98% of “audiophiles” do, who are not designers and builders like I am. As you have no doubt noticed, I have not just been a designer and builder, but also a person who tends to design and build particularly florid and elaborate things, because, if you’re going to take the time to do this anyway, why not go big? Thus, my fun and games here serves as a sort of template for creative hobbies in general, and also, creative activities in general, in the wider world as a whole. A person whose creativity has been suppressed can perhaps help it to blossom again by adopting a creative hobby. I suggest Legos — especially the old-style generic rectangular blocks. Brat Pitt and David Beckham have said they enjoy playing with Legos as adults.
Today’s tale is mostly about the aspects and complications.
Although this endeavor is not really a major focus of mine at this time, nevertheless a lot has happened since my last update a year ago. The most interesting thing is that I completed my horn speaker project, which was underway early last year. Further steps involved designing some subwoofers, which I later had custom-built by a local craftsman. Then, I had to buy all the associated amplifiers and digital crossovers. Since I already had a small tube amplifier that could serve as a tweeter amp — a cute and actually rather well-made little item from China that actually cost $200 including shipping — I decided to try replacing my speaker-level tweeter crossover with the digital crossover and multi-amping the tweeter. The digital crossover allowed both steep fourth-order crossover slopes and time alignment, plus control over the level, all of which I found made a substantial improvement. I added a bit of digital equalization at the source level, and a bit more in the digital crossover.
Thus, we ended up with a rather complicated system, with three amplifiers, an analog line-level crossover plus a digital crossover, plus two layers of (modest) digital equalization. Definitely not something sold in stores.
And I immediately had the urge to sell it.
This is not because the result was disappointing. Not at all — it was awesome, and exceeded all my hopes. Almost every time I turned it on, I thought “I didn’t remember that it actually sounds this good …” The sense of tone in the midrange was intense, burning, addictive, riveting, edge-of-your-seat type stuff — the kind of effect that I have only found with SE DHT amps directly wired to compression drivers. The dynamics were thrilling. It was … what a wild and crazy stereo system should be, a magic carpet ride of delicacy and power, subtlety and searing emotion and drama. Notice that I don’t say “accuracy” or “imaging” or “neutrality”, because this system is not accurate or neutral, doesn’t “image” much at all, and besides it doesn’t matter. I say this having owned and listened to a lot of other stuff including Altec A5 systems with 288B drivers and 1005 horns, multiamped with single-ended triodes and tube line level crossovers; Fertin field-coil wideband cone drivers on open baffle, with multi-amped open-baffle bass and supertweeters; RCA MI-9584 drivers on MI-9595 horns; the RCA “Ubangi” basshorns loaded with Altec 515s, and a lot of other stuff that many people think is pretty good, and is indeed pretty good. They were stepping stones.
Read a more detailed description of this speaker system here.
Why did I want to sell it? There were a few reasons. First, because I was enjoying the speaker-building process, and had a bit of momentum by then which might have translated into new projects. Second, because I wanted someone else to hear what I was hearing … to understand what was possible in this stupid audio hobby, and for not really very much money either. I wanted someone to leapfrog across the desert of useless opinion and boring gear-of-the-month hype and perhaps bring something special into their life.
Eventually, I settled down. After the initial momentum of building-enthusiasm — the momentum that was needed to get this done in the first place — faded, I decided I didn’t really want to keep on building things, but rather enjoy the fruits of my labor for a while. Also, it really did sound good, and it was all set up and running, so why not just turn it on and enjoy it for a while, like a year, or maybe ten years?
But, then another thing came up. I found that these speakers didn’t fit my lifestyle too well. An ambitious device like this just begs to be used. And, that would mean bothering the neighbors more than I would really like to. I live in a rural area, but in a village rather than a country hobby-farm. Plus, it really is big, and doesn’t fit my space well. But, more than that, I found that I just didn’t want to think about these things anymore, and that I would somehow have to carry this burden just by owning this thing, even if the building process was done. Like a fine restored vintage automobile, a system like this becomes a thing that you do. And, I mostly like doing other things these days.
In short, I discovered, after five years of listening to headphones before setting up this system, that headphones fit my interests better.
Also, there was another thing: the completion of this system represented a long cycle of endeavor going back to late 1999, when I bought my first vintage large-format compression driver on EBay, of putting together a really great superefficient horn system to use with directly-heated single-ended triodes. My RCA “Ubangi” basshorn-based system of 2008 was very good but was never really “done” before I had to sell it. There were still a dozen bugs to work out. This system was not really “tweaked,” but it did represent a realization of its potential and my vision for it — no halfway hacks like my initial use of a single cheapie subwoofer for bass duties, for example. Everything was finally built-out the way it should be.
I felt like I had somehow completed what I had set out to do, going back as far as 1999, and didn’t really have to go on unless I felt like it.
So I did sell it. I put the whole thing up for sale, with multiple amplifiers and crossovers and cabling and everything. A complete turnkey system. All of my efforts over lo these years, and all you have to do is just buy it readymade — which you can’t do in stores. The only other way to get to a place like this is to walk the path yourself, which would take years, and you might never get there. Most never do. The price was little more than the cost of the parts. $3800 including everything but the Welborne 2A3 main power amps. This is peanuts. Plus, I’d sell you the Welborne amps too, for $1200 which is about the new cost of the tubes alone, if you didn’t already have some amps.
I suppose this sounds like a lot of money to some people, but it is really a bargain.
But, nobody wanted to buy it.
I realized that one reason something like this is not sold in stores is because a typical plug-and-play audiophool really is not suited for owning something like this, even in its completed state. I say “completed” only by my standards — the standards of someone who can design and build both electronics and speaker systems. I realized that a lot of people are sort-of serious about audio but they have no idea that a “first-order highpass passive line-level crossover” means a capacitor and a resistor, or how you would make one even if you did know that, which happens to have a cutoff frequency of 150 hertz and is compatible both with your signal source and amplifier input stage. Even if you did know all that, you still have to built it, which means firing up a soldering iron and drilling holes in boxes, which a lot of internet-surfing knowitalls are allergic to. For me, this is like making toast. The third-order passive line-level crossover in this setup, for example, has a pair of 9.6 henry signal chokes, which I had custom-wound by Jack Elliano of Electraprint. Yes, I called a man in Las Vegas and had him specially wind some wire around a core of magnetic steel, to my specifications, because you can’t buy something like this even from a giant electronic parts store like Digi-Key. (That was back in 2006, which was a big year for these things.)
The crossover is completed with some paper-in-oil capacitors that were made in the Soviet Union, and were sent to me from a man in Ukraine, via EBay, before hostilities there of course. This is nothing special among builders, but it is pretty special if you stop and think about what it actually is.
Anyway, keeping something like this up and running is a bit like a 1960s British roadster. OK, not nearly that bad, but sometimes there are a few little problems, you have to do a little diagnosis, and perhaps some soldering.
I have a friend in the area who is kinda into this stuff, and actually owns a 300B single-ended amplifier that was custom-made for him by Blackie Pagano in New York City. I think the design is basically the Reichert Flesh and Blood. I thought these speakers would be a big step up from his Cain and Cain fullrange cone driver speakers — and he could probably sell those for more than I was asking for these, and pocket some money too. But, he wasn’t interested. Maybe it was a little like asking a California weekend beachbreak surfer if they want to paddle out at Pipeline at the North Shore. Yes, but … no. Maybe a speaker like this is really best for someone who builds their own amplifiers, like a 45 SET driven by a 10Y interstage coupled, powered by a 250V stack of sealed-lead-acid batteries, actively biased with another -50V battery bias supply, and which includes 12 pounds of filament chokes — an amplifier that I actually built, specifically to drive these horns, in early 2007. It produced about 1.5 watts of peak output power. The vintage bakelite panel meters that I put on the battery B+ supply were really cute.
Eventually, I did sell this speaker system, in parts. What a waste. But, the people who would be interested in something like this are builders themselves, and builders are interested in building. Even then, in that tiny subculture, most people have no appreciation for these things. The most important part, the crazy big midrange horns and vintage wideband compression drivers, were purchased by a Vietnamese man in New Jersey who owns a rather nice nail salon, and didn’t speak good English. Anyway, he already had a lot of speaker parts himself, and will have no problem adding subwoofers and tweeters and crossovers and helper amplifiers to his own heart’s content, from his own large inventory of such things. He told me that he intends to replace the dented sheet-aluminum quarter-sized horn mouths with some custom flares made of fiberglass, with full-size mouths, which would be a whopping 44″ in diameter. That is so cool. I hope he actually does it, and I think he is crazy enough to carry that football all the way into the endzone.
So, all of this helped me realize that I was no longer interested in owning speakers. I had my chance and made my decisions. I could change my mind, and maybe in five years, if I live in a different place, I’ll be buying or building something again. But, for now, I see no speakers in my future, except the little wooden shoeboxes now set up in the kitchen. I am happy about this.
This allowed me to sell all my amplifier-related stuff, and also my vintage gear like compression drivers, 416A woofers, bullet tweeters, and a lot of other stuff that I thought I would like to play with someday. Most of it went to EBay, and almost all the buyers had unpronounceable Asian names.
Along the way I sold my turntable, which I bought when I was fifteen. It was a Micro Seiki DQ-3, pretty nice stuff for a teenager. Over the years, the electronics inherent in the quartz-locked direct-drive turntables of that era sort of decayed, and the deck developed some bad speed issues. In short, it was broken. Nevertheless, I was able to sell it for more than my new purchase price, because people today know the Micro Seiki stuff was the pinnacle of ambitious 1970s-1980s Japanese turntable design. Plus, it has the spiffy MA-707 tonearm.
This reduces me to headphones, and a digital source. The headphones are the same Sennheiser HD600 cans that I’ve been using for ages. You might remember that I tried some other, recent options like the Sennheiser HD800, at roughly four times the price, and found that I preferred the classic HD600 by a large margin. In the past I also owned a heavily-modified Denon D2000 from Lawton Audio, which was also supposed to be oh-so-much-better than the HD600, but which I could never get to sound right. The digital source is the same Metrum Octave DAC from last year. Audirvana has been supplemented by uncompressed 16/44 streaming from Tidal (tidalhifi.com), with the help of Amarra, which improves things by a shocking amount compared with using Tidal alone.
Naturally, I wanted to nail down my headphone setup. My 01A-71A headphone amp, which I originally built in 2001, was begging for a rebuild, which I hadn’t got to in the midst of all this speaker stuff. I can’t even keep track of how many times this was rebuilt. Oddly enough, after versions with direct coupling and odd biasing schemes, I ended up with a simple self-bias cap-coupled design. The eighty-year-old 01A tubes (handmade by American craftsmen in the Art Deco era) tend to drift a lot and have various levels of wear. I wanted something that was stable and reliable, and was willing to give up a little nth-degree of refinement to get that. I also installed a new rear panel to keep all the cords off the top sheet, which is ugly.
I also ended up rebuilding the SLA battery B+ supply previously mentioned, putting it in a much smaller box. It now has 150V of B+ but four separate 7Ah batteries for use as independent filament supplies. I put together an eight-pole Speakon cable for an umbilical, plus there’s another BNC cable to handle B+. (DHTs are complicated.) Between the amp and the battery box, this rebuild was pretty comprehensive and approached the difficulty of building everything from scratch. This is the fifth version of the battery box I think, since its original incarnation in 2001. I also put together a little battery-powered amp using ten 2SK170 jfets in parallel, which sounds pretty darn good, even compared to the 71A amp. I call it the “Beast of 10 jfets,” and think of it as sort of a Pass SIT-1 for headphones. It instantly obsoleted my other solid-state headphone offerings around here. And, even with its batteries, it fits in a 7.5″ square Hammond 1590F die-cast aluminum box.
But, the big discovery this year was that I could listen to most of my digital music, about 80%, using the outputs of the DAC direct to headphones. The Metrum DAC actually has no output stage. Like the old Kusunoki TDA1543 design (which I point-to-point wired in 2001), the output comes directly off the chips, with I/V conversion by a simple resistor. So, I am going direct from the R2R ladder in the chips to my headphones with no intervening electronics. (I don’t think the DAC even has an output cap, but uses a balanced arrangement.)
This sounds really good. Better than any of my amplifiers. Some recordings need additional amplification, but most don’t.
And so, after all this, chronicled in part above, I am listening to something close to nothing at all. Not one single stage of amplification. I could live happily with this DAC-direct setup, plus the jfet amp when I need a little more on some recordings, and put the 71A amp away altogether. But, I just like the Christmas-tree aspect of the DHTs, especially those 1920s-era 01As with their thoriated tungsten filaments, and all-battery-powered all-DHT amps are kinda special, and it does sound better too.
Actually, for a lot of recordings, I need up to 12 db of attenuation even off the unamplified DAC outputs. (I find that I have to attenuate by roughly the degree to which the recording is mastered too hot.) I put together some 600-ohm autoformer volume controls for that purpose, wound by Dave Slagle, which also happens to be one of those things not sold in stores in a plug-and-play form, and which finalizes that little cycle as well.
This is the part of the story where the drama ends and I live happily ever after. Great music reproduction is like great food or great wine, or great books or great clothes or great furniture. Many people are perfectly happy to live without it, but you can bring it into your life if you want, and it might be nice if you did.
