Bazz Me, Fuss You -- stop me if you think you've heard this one before [documentation]
- lykwydchykyn
- Breadboard Brother
So I breadboarded an Escobedo Push-Me-Pull-You the other day, and while it gives a decent octave up the fuzz was kind of weak (at least, on my breadboarded version).
So I says to myself, let's drop that gain control. Nuthin' good happening under max anyway.
Then I says to myself, what happens if we replace that feedback resistor with a diode, say a white LED? Well, what happens is that first stage becomes a bazz fuss and you get that glorious BF sound pushing (and pulling?) the octave section. Nice. I go ahead and ditch the resistor and cap and tie the emitter straight to ground.
Then I says to myself, could I do the same on the push/pull circuit, replacing the feedback resistors with white LEDs? Well, it turns out that yes, yes you can. And it sounds quite nice.
Once again I says to myself, since the octave is created by the roughly equal push/pull of the PNP/NPN pair, could I adjust the octave amount by adding a resistance between the output of the first stage and the base of the PNP? Well, yes, this also works quite nicely. With about 1M of resistance after the coupling cap, I essentially am left with a two-stage bazzfuss circuit.
Since this is all just working so nicely, I made a few other common tweaks to coupling caps, RF filtering, power filtering, etc. And here's the final schematic:
Note that the 22k resistor on the octave stage should be adjusted to bias the NPN collector to 4.5ish volts (well, 1/2 Vcc anyway). I used a S9012 and S9013 transistor in my build, but mostly because I just wanted to use up some that I got in a kit. You could probably use 2n3904/2n3906 or whatever.
Dunno if this has been done before, but I found it fairly pleasing. Especially given its simplicity and low parts count; I've built a lot of overly complicated octave up fuzzes that were less satisfying than this.
So I says to myself, let's drop that gain control. Nuthin' good happening under max anyway.
Then I says to myself, what happens if we replace that feedback resistor with a diode, say a white LED? Well, what happens is that first stage becomes a bazz fuss and you get that glorious BF sound pushing (and pulling?) the octave section. Nice. I go ahead and ditch the resistor and cap and tie the emitter straight to ground.
Then I says to myself, could I do the same on the push/pull circuit, replacing the feedback resistors with white LEDs? Well, it turns out that yes, yes you can. And it sounds quite nice.
Once again I says to myself, since the octave is created by the roughly equal push/pull of the PNP/NPN pair, could I adjust the octave amount by adding a resistance between the output of the first stage and the base of the PNP? Well, yes, this also works quite nicely. With about 1M of resistance after the coupling cap, I essentially am left with a two-stage bazzfuss circuit.
Since this is all just working so nicely, I made a few other common tweaks to coupling caps, RF filtering, power filtering, etc. And here's the final schematic:
Note that the 22k resistor on the octave stage should be adjusted to bias the NPN collector to 4.5ish volts (well, 1/2 Vcc anyway). I used a S9012 and S9013 transistor in my build, but mostly because I just wanted to use up some that I got in a kit. You could probably use 2n3904/2n3906 or whatever.
Dunno if this has been done before, but I found it fairly pleasing. Especially given its simplicity and low parts count; I've built a lot of overly complicated octave up fuzzes that were less satisfying than this.
- CheapPedalCollector
- Resistor Ronker
There's always multiple ways to tweak something and almost infinite possibilities of tweaking tone and response. Nice work, I'll breadboard it and see if I like it.
Information
- Posts: 4
- Joined: 26 Jan 2022, 21:29
This sounds fun. Hopefully a Vero is on its way
- lykwydchykyn
- Breadboard Brother
This layout works, though it's a bit awkward; made it for a particular scrap of vero I had on hand with a corner cutout:
EDIT: I should probably mention my color convention for the unterminated wires: Red is signal input, White is +9v, Black is ground, Yellow is signal output.
- lykwydchykyn
- Breadboard Brother
I should also mention about the coupling cap values: They aren't particularly magic. The original circuit had all 100n caps, and you could certainly do that if you want more low end. I partly wanted a tighter low end and partly wanted to see if I could get away with smaller caps (I'm using the dipped mylars, so the 100n are significantly larger than the 47n. If you want to use box caps it won't matter).
- lykwydchykyn
- Breadboard Brother
Also, a demo for those interested:
- CheapPedalCollector
- Resistor Ronker
Has anyone tried taking the output from the other side of the differential pair?
- lykwydchykyn
- Breadboard Brother
I did this on the breadboard. It sounded the same to me, though I believe it was out-of phase.CheapPedalCollector wrote: ↑20 Sep 2022, 21:41 Has anyone tried taking the output from the other side of the differential pair?
- CheapPedalCollector
- Resistor Ronker
Seems like it would be a lot louder coming out of the other side.
- lykwydchykyn
- Breadboard Brother
So, following up on this... I built another one today after trading away the last one. I was quite dissatisfied with the output volume, and your comment came to mind. So I moved the output cap to the other end of the transistor pair and, yep, a good bit more volume to be had there.CheapPedalCollector wrote: ↑24 Sep 2022, 07:33 Seems like it would be a lot louder coming out of the other side.
Also did a cleaner vero:
- lykwydchykyn
- Breadboard Brother
For sure, for sure; but I do love a simple circuit. I've taken a few cracks at building a fuzz with a blendable octave, and this is such an elegant and minimal way to get that. If I can get reasonable gain without throwing another stage on there, so much the better.
hello. I am also testing on a breadboard. I don't have Q1's mpsa13, so I created Darlington with 2n3904. Q2 is 2n3904 and Q3 is 2n3906.
All three LEDs are red LEDs. When I changed the LED of Q3 to 1n4148 as a trial, the sound became rough and powerful. It seems that the octave output is not much different. Is it because the forward voltage has changed?
The bias voltage is an octave around 4.7v when the red LED is on,
When Q3 is set to 1n4148
It will be around 6.8v.
All three LEDs are red LEDs. When I changed the LED of Q3 to 1n4148 as a trial, the sound became rough and powerful. It seems that the octave output is not much different. Is it because the forward voltage has changed?
The bias voltage is an octave around 4.7v when the red LED is on,
When Q3 is set to 1n4148
It will be around 6.8v.
- lykwydchykyn
- Breadboard Brother
I can't say I that I have an intuitive grasp of how this circuit operates, but I suspect the more differences you create between the voltages on Q2 and Q3, the less fundamental cancellation you get. I think there's a lot of experimentation to be done here with different diode combinations.
I've been developing another circuit based on this one but with two octave stages. I get all kinds of fun, synthy sounds out of it. I'll do a separate post on it sometime when I have my ducks in a row.
I've been developing another circuit based on this one but with two octave stages. I get all kinds of fun, synthy sounds out of it. I'll do a separate post on it sometime when I have my ducks in a row.
hello. I tested it with a circuit simulator. Q1 has led Q2 has 100k resistor and Q3 is 1n4148. I put 40hz in the input and adjusted it so that around 80hz came out from the output. The bias trim was around 1k and the octave sound was good. Q3 can produce an octave even with a 470k resistor, but the sound feels different. 1n4148 is rougher. I tested it on a breadboard and I like it.
Information
- Posts: 1
- Joined: 10 May 2023, 08:46
Sorry to bring up an older thread.
With all the Lizard Queen hype, I've just made one of these as didn't really need the extra gain stage/control. While everything works pretty well and sounds good, I have one weird problem - the octave control does absolutely nothing for 99% of the rotation using a 1M pot (either linear, log or anti-log), with a small kick of octave right at the end. I dropped it to 100kA (the same as the Lizard Queen) and I still have the same problem. I built it with a trimpot for setting bias, and Q2 is half supply voltage.
On a whim, I moved it down to a 1kB pot and everything works -perfectly-. I get a good range of pure fuzz to octave fuss, and all tones are usable in the middle ranges.
Any idea why the disparity? Does the original have the same issue as mine?
Thanks!
With all the Lizard Queen hype, I've just made one of these as didn't really need the extra gain stage/control. While everything works pretty well and sounds good, I have one weird problem - the octave control does absolutely nothing for 99% of the rotation using a 1M pot (either linear, log or anti-log), with a small kick of octave right at the end. I dropped it to 100kA (the same as the Lizard Queen) and I still have the same problem. I built it with a trimpot for setting bias, and Q2 is half supply voltage.
On a whim, I moved it down to a 1kB pot and everything works -perfectly-. I get a good range of pure fuzz to octave fuss, and all tones are usable in the middle ranges.
Any idea why the disparity? Does the original have the same issue as mine?
Thanks!
- lykwydchykyn
- Breadboard Brother
It probably has a lot to do with the transistors you use. I've built this circuit a few times with different transistors, and the octave knob almost always has to be adjusted a bit by ear. Maybe it was the S901x transistors that worked great with the 1M, or maybe it was some other factor at work. I can only suggest that you just go with your ear.
I've also tied the empty lug to ground with a large cap (in the µF range) which gives an interesting result as well.
I've also tied the empty lug to ground with a large cap (in the µF range) which gives an interesting result as well.
- lykwydchykyn
- Breadboard Brother
DrewSpriggs and I talked over on reddit a while back and determined that the difference in the octave pot value came down to placement. You can put this pot between the emitter of Q3 and R4 and it seems to work a bit more consistently. Makes the schematic more LizardQueen-esque, I suppose; you'll want to adjust it's value between 1k and 10k depending on your silicon bits, just see what makes the sweep go from octave to no octave nicely.
Here's an updated version of the schematic: I'll add a few more observations from playing with this circuit:
- It has a good amount of output to drive the tonestack of your choice. BMP is a good pairing.
- Low-Vf clippers to ground after the fuzz can give you some interesting sub-octave effects under the right conditions.
- Sounds killer on bass if you add a clean blend. Very synth-bassy.
Here's an updated version of the schematic: I'll add a few more observations from playing with this circuit:
- It has a good amount of output to drive the tonestack of your choice. BMP is a good pairing.
- Low-Vf clippers to ground after the fuzz can give you some interesting sub-octave effects under the right conditions.
- Sounds killer on bass if you add a clean blend. Very synth-bassy.