troubleshooting diy distortion

[shinychrome0]

i am trying to build this opamp based distortion pedal, and i'm a little confused. Does anyone see any problems with the schematic? I get no distortion or boost. The only knob that works is the volume knob, which is at the end of the signal chain, so apparently something is getting through. Why is there no boost in volume, no distortion, and no effect from the tone knob?

The schematic is attached.

[Albert Kreuzer]

First, C4 (1pF) is too small. Try 10µF or bigger. That should take care of the boost. Second, for the tone knob, try a 100k pot with a 100nF (0.1µF) cap. Maybe 0.22µF.

Hope this helps.

Cheers,
Albert

[shinychrome0]

ok so i have boost and distortion now (it would be great if someone explained why that capacitor controls boost) but it sounds terrible. How can i change the signal so that the bass frequencies are not distorted?

[Mark Hammer]

All op-amps behave like a car whose gas pedal is welded to the floor. That is, they are designed to go full blast, and the only means at your disposal to control their speed is use of the brakes, hand and foot.

In the case of op-amps, the "brakes" consist of the resistance in the feedback path, and in the case of the example shown, the resistance to ground from the inverting ("-") pin. the more feedback is applied from the output back to that input, the harder you are squeezing on the "brakes" and reducing the gain of the chip from maximum to something less than max.

Of course, how much feedback gets applied is a function of both what sort of resistance you place in the way,and how much of that feedback signal you get rid of (that part is a bit like poking a hole in the fuel line so that your engine has less to work with). The feedback resistor and the resistor to ground from the inverting pin, behave like a volume pot. As the "ground leg" gets much smaller relative to the feedback resistor, it gets rid of much of the feedback signal, just the way a volume control/pot would. IN fact, you can adjust the gain of what you have drawn using either changes to the value of the 1k ground resistor, OR changes in the feedback resistance, OR both.

Capacitors also play a role in determining gain. If there is a capacitor in the feedback loop/path, that provides what essentially amounts to a zero-ohm path for the high frequencies, meanng that lots of negative feedback is applied and gain for those frequencies is kept to a minimum. No surprise that the bigger the feedback cap value, the more treble is rolled off in the output signal.

For the ground leg, it works differently. remember that caps behave like a frequency-dependent resistor in some ways. So, if the only way for feedback signal to get dumped to ground is through a cap, then if the cap is a small value, only high frequency content will get dumped, and lower-frequency feedback will be conserve. To beat our control-by-brakes analogy to death with a stick, think of it like applying the brakes for straightaways, but lifting your foot for hairpin turns.

So, to have the same gain applied across the entire audio spectrum, without havig to worry about either stray DC or high-frequency (radi) oscillations contaminating things, you'll want to select caps that roll things off around 15khz at the top end (few guitars and amps ever even get out there much), and around 100hz at the low end (same rationale).

The formula to use is: Freq = 1 / (2*pi*R*C) [where R is in megohms and C is in microfarads]. So, with a 1k ground resistor, a 10uf cap gives us a low-end rolloff of 16hz. Going up to 1uf instead bumps the rolloff to 160hz, which isn't too bad. With a 1pf cap value your gain doesn't really kick in until well over 150mhz. If your pickups or ears can detect that, you have my respect. Small wonder you didn't hear any impact of the gain control.

For the feedback cap, you have to think in terms of the maximum resistance, since the pot resistance will vary. With a maximum resistance of 500k, a 47pf cap will provide a rolloff around 6.8khz. That is lower than our 15khz maximum, but the rolloff is shallow, and the clipping adds lots of treble, so even 6.8khz will seem inordinately bright. Happily, you have your treble-cut control to fix that. Perhaps equally important, your max gain - as shown - is 500x. If you amplify the input noise of whatever is feeding your circuit by 500 times, that's a whole lot of hiss in the output. So taking steps to rein in how much gain is applied to the top end is a smart move.

[Bloodbath]

Another great explanation by Mr. Hammer. Awesome.

[shinychrome0]

Ok great. That all makes sense. My only other question is what can i do to make sure the bass frequencies do not distort? I know pedals like the tube screamer and basically every other overdrive pedal do not push the entire audio spectrum through the clipping diodes.

[52 Bill]

My only other question is what can i do to make sure the bass frequencies do not distort? I know pedals like the tube screamer and basically every other overdrive pedal do not push the entire audio spectrum through the clipping diodes.

I think that Mr. Hammer already answered your question:

"The formula to use is: Freq = 1 / (2*pi*R*C) [where R is in megohms and C is in microfarads]. So, with a 1k ground resistor, a 10uf cap gives us a low-end rolloff of 16hz. Going up to 1uf instead bumps the rolloff to 160hz, which isn't too bad."

Adjust the value of the cap until the low freq. rolloff suits your needs.