Ship in a Bottle/Glass Cannon fuzz -- schematic, layout, and demo

[midwayfair]

This one probably has some promise. The diode feedback thing isn't something I've seen before, and the tone control might interest some people -- it cuts bass and treble simultaneously.



This takes a tiny bit from a few different fuzzes, but mostly was me poking around on a breadboard trying to use a really limited range of component values without resorting to building something everyone already knows, like a Fuzz Face. I first tried Darlington bazz fuss (noisy and fizzy) to save on the large-value resistors (remember, I was making this with glass components ... values over 100K are super hard to get), but I ended up with something a bit closer to a generic 60's two-transistor cascaded fuzz.

Then I went poking around trying to figure out where germanium diodes could go, because more glass components rawr. I actually killed two transistors feeding DC right to the base with a grounded emitter. I did it twice a half hour apart. Man that was dumb! Anyway, at some point I got lucky and hooked up the collectors together. It really only works in this particular arrangement, with the anode on Q1. Suddenly I had a bunch of clipping that sounded slightly different from back-to-back diodes to ground, and the second transistor behaved a little differently too.

PRR did make one suggestion about the diode:

> the diode is doing some rectification, right?

If the diode is poled as shown, it can only pull Q2 collector *up*, not down.

[...]

Q2 is not explicitly biased. Its idle current is super low. Signal can cause much-more current, but not-much less current. With signal, average current increases, collector voltage drops.

I would suspect that D1 prevents it from dropping "too much".

There may be other ways to do the same thing. But if it works, be happy.

Interestingly, doing this with the diode makes the transistor non-functional if you have a feedback resistor from the base to collector. I didn't play around with the idea of hooking the base resistor up to +9V, but I guess I could experiment. As it stands, this MUST be built with germanium because leakage current is required for Q2. Q1 can be silicon. You should include R2 if you want to make this with a higher gain transistor in Q1.

Q1 biases near 5V, Q2 is higher, about 6.5V or 7V.

The tone control is a voltage divider doing two things at once. At max, C4 is shorted, so C3 and the volume pot set your frequency cutoff at ~159Hz. There is also 100 (plus 2.2K) resistance in series with C5, which results in very little treble cut. As it's rotated counter clockwise, the resistance between the output and C5 decreases, which cuts the treble, and C4 is progressively less shorted -- and because capacitors divide in series, when the pot is at its minimum, the output cap becomes about 50nF, and your frequency cutoff rises to ~318Hz.

I think an audio taper may be a better call -- the darker settings were really bunched up in my prototype. I'll change the schematic when I get around to it. C5 is probably larger than is really useful (maybe 47nF or even 22nF are better), and making C4 a smaller value would increase the range of the bass cut, but you probably don't want to make it too small.

The power section is pretty standard -- just some filtering and a protection diode (use a 5817, not the Ge I've got in my schematic ...).

Here's the demo:

And here's a layout (for real components, not glass ones):
https://dl.dropboxusercontent.com/u/9878279/Jon%20Patton%27s%20mini%20layouts/Ship%20in%20a%20Bottle%20%28Glass%20Cannon%29/Ship%20in%20a%20Bottle%20%28glass%20cannon%29%20fuzz.pdf

[Arcane Analog]

*slow clap*

Sounds awesome! I dig the all glass angle. I do not think I have ever heard of a pedal with all glass components. "Going mojo" may be expensive but it sure is fun.

[midwayfair]

*slow clap*

Sounds awesome! I dig the all glass angle. I do not think I have ever heard of a pedal with all glass components. "Going mojo" may be expensive but it sure is fun.

Heh, yeah, it was kind of fun, reminds me of building with the giant 600V caps I accidentally ordered from Smallbear when I was first starting out.

Can't say I think they impart any special sound (and since the caps are actually ceramic, there's almost certainly some benefit to a change there), and TBH the OC140s may not be leaky enough for Q2 to function optimally.

[Jdansti]

Sounds great, Jon!