A more manageable fuzzlab build
A long time ago, I built a Fuzz Face that replaced a lot of the fixed resistors with pots. I called it the fuzzLab jr. It was fun, but in the end, it really didn't sound very good, nor was it terribly useful. I put it away and promptly forgot about it. Over the last few months, I've been working on the Armageddon Processor, and it has been exciting and frustrating all the same time. Exciting because it can make sounds I've never been able to obtain before. Frustrating because it is so complex.
So I decided (as I often do because I'm such a spazz) to start yet another new project in the midst of all the other stuff. This one is the fuzzLab Jr. as it should have been originally.
A lot of knobs on a fuzz face is hardly new is it? In fact, between the ZVex Fuzz Factory and the many-knob-fuzz-face that is a rite of passage for DIY'ers, it has been done to death. So why make another one?
Well, I think I've come up with some novel things that are simultaneously strange and musically useful. I've always been intrigued by the fuzz factory but disappointed in its lack of bass response and difficulty in getting good fuzz tones along with the stuttering/fart/glitch sounds.
My goal was to create a fuzz box with a great deal of controllability, but not at the expense of making yet another stuttering/fart/glitch box. The solution in the end was to start with a silicon tranny fuzz face, change the circuit in key places, and add tone stack of almost prehistoric design. The result is ten knobs of fuzzy goodness that I'm quite happy with.
The preamp section around Q1 is taken from the electro harmonix LPB-1 booster, an old and trusted single-transistor boost. I added this in the front to offer even more drive. Next it goes into a two-transistor fuzz face design. The key differences are the addition of a Comp switch. This shunts a 47nf capacitor between the collector and base which results is a bass-heavy compressed sound. With Comp engaged, you lose some volume so you'll want to turn the volume pot a little higher, but the results are pretty cool.
There is also a switch called feedback that can switch between the collector of Q3 and the base of either Q2 or Q1. When you switch in the feedback loop to Q2, it sounds like a wildly oscillating fuzz, but a bit more interesting because the VR4 pot is not directly connected, rather it goes through another 47nf cap which forms a filter. So the same pot controls both amount of feedback and acts as a filter control. When you switch the feedback loop to the base of Q1, something entirely different happens--it shuttles negative feedback to the beginning of the circuit which not only lowers gain, but completely alters the frequency response. There are also pots for controlling the voltage to the Q2/Q3 pair, a Bias pot to further tweak the voltage to Q3, a Choke pot between the Q2 base and Q3 emitter, and the standard fuzz face 'Fuzz' pot.
The whole thing ends up in a crude tone stack that you would more commonly see in a tube amp. From an electrical engineering standpoint, this tone stack should 'suck balls' because of the wild impedance mismatch. But it is actually not only very active, but very interactive with the overall circuit. So instead of sucking balls, it eats balls. You see the subtle marketing difference?
Does the combination work? I sure think so. I've been playing with this off and on for about a month and am still discovering new sounds.
I wanted to come up with a design that was simple and would belie the purpose of the innards. The 1790NS enclosure is a bit biggish, but gives me ample room for the 10 knobs and two switches. I thought a nice touch would be to use an amp-type jewel light instead of the traditional LED bezel.
It was wood sides. Heh. Wood.
PCB rev 1
Layin' it out
More wires than an illegal NSA wiretap
So how did it work out? It was an interesting project but even with the mods, it didn't add anything terribly new to the tricked-out fuzz game. I'm happy with how it turned out appearance-wise, but although it is a fun build, it's a lot of parts cost for minimal return.