Anderton Tube sound fuzz howling problems

[thom__stone]

Hey all, I'm very new to this whole pedal building business, infact this is my first ever pedal so excuse me if I miss pieces of vital information.

I decided to start with the Anderton Tube sound fuzz because it didn't look overly complicated (with my dad looking over my shoulder) and it seems to be a very popular pedal. When i put in my true bypass circuit (with a 6 pin stomp switch, a battery and no LED) I noticed that the volume jump from the bypass to the effect was HUGE! and after i mounted the circuit and components in the case it started to howl like a mother bitch on the lead setting. I used a PCB which Ive checked over and over, double checked the values on all my components, checked the capacitors are the right way round, the only thing I'm unsure of is the earthing. It still distorts or fuzz's and seems to work, I'm just not 100% sure of it. Ive also tried about 15 different IC's as i ordered a tube of 25. all of them seem to howl. is there an obvious mistake I'm making here?



here's the schematic, I'm pretty sure i followed it exactly when it comes to parts,

[bosleymusic.com]

Possible solutions:
Are you using shielded cable from the input/output jacks to the board?
You may want to increase the value of C1 and C2 to lower the cutoff frequency of the filtering.
Change R6 to a lower value.
Is your project in a metal box or a plastic one?
A plastic box will let in external RF / interference, etc... a grounded metal box will give you proper shielding.

[psychedelicfish]

Words like "howling" suggest that your circuit is oscillating. Try moving the input and output wires further apart, or as bosleymusic.com suggests, use shielded cable for the input and output wiring. You may find that the oscillation goes away when you turn down the "fuzz" or the "output" controls. If the input and output traces are close together, then your problem may lie there. Another thing to try is to increase the value of either C6 or R1.

The reason for it oscillating is that the signal from the output is fed back into the input somewhere. Each of the gain stages "flips" the signal upside down as it amplifies it, and if you feed the upside down signal back to the input it cancels out some of what is going in to the input, which reduces the gain (how much the stage amplifies). This type of feedback is called negative feedback. In the case of this circuit, there are two gain stages, one after the other, that both "flip" the signal upside down, meaning that the signal on the output of the second stage is the same way up as the signal going in to the first stage, only bigger. This reinforces the signal going in to the input, so the gain of the stage increases. This is called positive feedback.

The thing about positive feedback is that it is impossible to eliminate phase shifts between the output and the input. Phase shifting is where the signal is shifted over slightly, and it happens when the signal passes through any capacitance or any inductance. There is inductance in the wires/traces, and there is capacitance between wires. The capacitance between wires is significant to us because the signal is fed back into the input through it. This capacitance causes a phase shift, which means that every time the signal goes back into the input, it gets shifted over slightly, causing an oscillation. This is probably what's happening in your circuit. Either your input and output wires are too close (which you can fix by moving them further apart or using shielded wire, which "interrupts" the capacitance), or the signal is being passed back through the power supply (which you can fix by filtering the power supply more, so increasing C6).

I hope you (or someone else who can help you) understood some of that...

[poppyman]

You should double check that you have the unbuffered version of the IC. It should be written on the chip itself ("UBE").

[guitarkill]

You should double check that you have the unbuffered version of the IC. It should be written on the chip itself ("UBE").

+1.

Also, check your "lead dress". If you have a bunch of wires closely running in parallel, one wire can pick up stuff from the one adjacent to it and contribute to oscillation (what you call "howling").