The builder

I build fully analog tube guitar gear, in my workshop. One piece at a time. The first thing I'm shipping is a preamp. This is the story of how it started and why it's worth the time.

How it started

For years I've played through tube amplifiers — a Fender '65 Twin Reverb, a Vox AC15, a Mesa Boogie Mark V and so on. Tried digital too — currently own a Line 6 Helix Floor. Two shelves of guitar pedals. Each one was the answer to a different kind of question.

Then I got a Mesa Boogie TriAxis, and that one bugged me. Beautiful piece of gear — the lead voicings were everything I wanted, and until then I hadn't even thought about breaking the head into two separate components — preamplifier and power amplifier. But after using it for a while, it felt like the natural move — two pieces of gear, each for a specific purpose.

The EQ section, though, never gave me the control I needed for the way I actually use it — compared to the Mark V I also own. And from time to time I found myself wanting finer resolution on the controls than the design offered. But it was great, groundbreaking — finally I could switch between the sounds I love without turning knobs and go from Metallica to Tool, Mono, Clapton, or almost anything I liked with a tap of the footswitch. But the EQ issue continued to exist, so I started the search…

I'd begun spending time searching for EQ pedals or rack units that could handle the levels a tube FX loop puts out, ideally with preset support — and the news wasn't good. I was frustrated with the supply-voltage limits in the pedal world. 9V, 18V — even the best pedals on the market are working inside a voltage range that constrains what the signal can do, particularly with high-gain or transient-heavy material. I knew there was headroom being left on the table everywhere in the chain.

So I decided to design a better EQ for it. Just for myself, for one rack, to solve one problem — but the design gave me an excuse to finally build for the headroom I'd been wanting all along.

That project grew. Once I had a recallable analog EQ working on the bench, I started looking at the rest of the signal chain and realised the same approach could work everywhere — every parameter that's usually a trim pot or a fixed value could be made recallable without compromising the analog signal path. I searched every option I could find — LDR technology that's now dying off, MOSFET switching, photoresistor banks, motorised pots. None of them gave me what I needed without compromising the audio or digital in some way.

What started as an EQ for a TriAxis was turning into a full three-channel preamp with total recall. BiFlux is that preamp.

I'm not a touring musician. I'm a bedroom player who loves guitar and pays close attention to what I'm hearing. The original problem still drives the design. I'm not building this for a market I imagined. I'm building it for the kind of player I am, with the specific frustrations I had, and the people it'll resonate with are people who've had the same frustrations.

Why I'm the one building it

Two backgrounds matter for this product, and I happen to have both.

I've been a ham radio operator since 1996, YO2LLS. That's nearly thirty years of building electronics by hand — including tube HF power amplifiers running 1kW and up, which is the kind of project that does not forgive mistakes. Tubes, high voltage, RF, custom transformers, the discipline of getting a design right because the alternative is expensive or dangerous. The analog tube chain in BiFlux uses everything I learned in that work. Tube circuits aren't a new domain for me — they're where I started.

I'm also a software engineer by trade — twenty-plus years of it, including co-owning and running elapseit, a project management platform we've operated for years. That career matters here in two ways, one practical and one substantive.

The practical part: the day job funds the development. BiFlux doesn't need to make money on day one to exist. The pace is set by the engineering, not by a runway. And I decided not to cut corners where it counts. (Two displays of different sizes on the front panel because the eye needs different things in different places. C0G capacitors in every frequency-defining position of the equaliser. Thin-film resistors throughout the signal path. The kind of choices that nobody sees in a spec sheet but everyone hears.)

The substantive part is more important. The hard problem in this preamp isn't the analog — I've spent decades on the analog side. We know exactly why a Mesa sounds like a Mesa, why a Fender sounds like a Fender. Duplicating those sounds isn't the hard part. The hard problem isn't even the digital. Precision digital potentiometers and ladder switching matrices have existed for years in the high-end audiophile world (I'm something of one myself — and I think all guitar players are), but they haven't made it into the guitar world, mostly because of cost and complexity. The hard problem is the integration: making digital control of analog circuits feel, to the player, like there's no digital layer at all.

That's where the software engineering background does its work. A recallable preamp is only useful if the recall doesn't get in the way of playing. You can't ask the player to navigate menus to change a gain setting. You can't make them remember which preset slot has which sound, or what gain setting was in use. You can't break the muscle memory of reaching for a knob and turning it. Every interaction has to feel exactly like the analog amp it's replacing — and behind the panel, every interaction has to read and write digital state correctly, instantly, reliably, with no glitches or surprises.

The thing that's frustrated me for years across every piece of recallable gear I own — Strymon, Source Audio, Universal Audio, the Boss DM-101, all of it — is the same problem: load a preset, and you have no idea where the controls actually are. The screen shows you a number. The knob is wherever you last left it, with no direct connection to the loaded preset. You're flying blind. The relationship between what you see and what you hear is broken. BiFlux exists in part because that pattern bothered me enough that I had to solve it.

UX that disappears. State management that doesn't lose user input. Systems that recover gracefully when something goes wrong. Building this preamp wasn't a leap from one domain to another. It was the same engineering discipline applied to a different output medium.

The other thing the combined background gave me was the ability to see this opportunity at all. Most analog builders don't have the digital chops to attempt total recall done properly. Most digital builders don't care about analog signal paths enough to do them right. There's a narrow Venn overlap of people who do both, narrower still those who've spent decades building tube circuits for unrelated reasons, and narrower still those who play guitar attentively enough to know what's missing from existing options. I happen to be in that intersection. So I'm building it. And it will be a dream come true.

What makes it different to use

Every preamp parameter — channel, gain, EQ, presence, master, the things you reach for — has a real physical control on the front panel. Turn the gain knob, the gain changes. Same as any analog preamp ever made.

The difference is that every one of those controls is also recallable. Save a preset, and when you recall it the controls show you exactly where they're set — actual visible positions, on the front panel, instantly. Not "the preset is loaded, here's a number on a screen." You can see at a glance what's happening with the sound, exactly as you can on an analog amp where the knob positions are the state. You can A/B between sounds and watch the controls move, hearing what changed and seeing what changed at the same time. (I tried motorised pots first — analog, beautiful in principle, working. But too slow on a fast preset change, and the mechanical assembly is too prone to failure over years of use. I had to find a different way.)

This is the design centre of the whole product. Not just "analog with recall." Analog with recall where the recall doesn't change how you interact with the preamp. The player's mental model stays the same as it would be with a vintage Marshall or Fender or Mesa. Everything underneath is digital state and analog signal path, working together to preserve that mental model.

I have not seen another product do this the way BiFlux does it. There are recallable analog preamps where the controls don't move and you have to read a screen. There are digital products that model analog and have moving controls but no real tubes in the signal path. There are analog preamps with no recall at all. BiFlux is the combination I wanted and couldn't find — and once you've used it, the alternatives feel like compromises.

Why now

The components had to exist. The time had to exist — the day job stabilised into something I could sustain alongside a second project. The market had to be ready — the modelling-amp saturation point, plus a real cohort of musicians who've started asking out loud for what BiFlux makes. And I had to be ready, which took longer than I'd like to admit.

I don't expect to sell hundreds of anything. A small number of each thing I build, to people who care about the specific things this gear does.

What you can do from here

If this resonates, join the list at the bottom of this page. I send build notes as the work progresses — actual notes from the workshop, not marketing. As BiFlux gets closer to shipping, those notes will go deeper into the design choices behind it — the things that make this preamp what it is, one decision at a time. Occasional, when there's something worth saying. No noise, never shared.

The first preamp units ship at the beginning of 2027 — that's the target. But the work doesn't stop there :)

---