Death by Audio - Octave Clang [traced]
picking up an old thread here i know!
I've made the below in kicad based on the trace - thought it would be handy for anyone hitting this thread like i did. (i've added the power input with protection) if anyone spots a mistake just shout!
Cheers
Darren
I've made the below in kicad based on the trace - thought it would be handy for anyone hitting this thread like i did. (i've added the power input with protection) if anyone spots a mistake just shout!
Cheers
Darren
Another necrothread on this one. I came across this effect in the EQD "Show Us Your Junk" episode with DBA[1] and attempted a build using the schematic at PCBGuitarMania[2].
I was having trouble getting the circuit to make any sound at all, then to make anything resembling music... until I swapped out the sketchy amazon-sourced 741 I was using for one with a TI logo (likely from stompboxparts). In the video Oliver talks about using a specific batch of 741s from Radioshack, and it does indeed seem to be a circuit that's sensitive to the chosen op-amp. Across three different varieties of 741 in my drawer, only the TI-branded part produced a decent sound (...along with the occasional radio interference Oliver alludes to in the video).
Notes:
- The circuit does also seem sensitive to the diodes chosen. I tried a few different samples of 1n4148, 1n914, 1n90p, and some supposed 1n34As that looked a little too similar to the rest. They all allowed the circuit to make noise, but I didn't get the intended sound till I tried the more "vintage" looking 1N34As obtained from stompboxparts (https://stompboxparts.com/semiconductor ... ium-diode/). Maybe another issue of imitation products for the others I tested, but the range of forward voltages do cover the range that these 1n34as sit within... 
Edit: returning to it I could not reproduce the diode dependence. Swapped in standard 4148s and it’s working just fine. Not sure what was going on the other day.
- With the (working) TI LM741 I'm using, the circuit does not seem sensitive at all to grounding the offset pins (neither 1 nor 5) or pulling them to ground through a resistor. I've left both floating to good effect.
- Similarly, leaving the non-inverting input floating (surprisingly) does lead to the intended behavior. Tying it to ground (through a resistor or directly) does not have an effect.
- I played around with tone pot values, and settled on an A100K to focus the in on the more usable range before the pedal quiets down beyond the point of usability.
- I initially spent a bunch of time trying to get closer to the specified resistance values, but found them to be less critical. I didn't have a 910k resistor for the input stage, but a 10M paralleled with a 1M gets close enough. Going from 1M towards 910k does seem to make a slight difference when the tone knob is turned up, but not as much as, well, having the right op-amp.
- On the transformer wiring, there's more flexibility here than I initially appreciated when trying to get the circuit working:
1. The important thing is that the middle tap of the output side is grounded, and at least one tap on the input side is grounded.
2. Having the signal flow be Secondary->Primary yields what seems to be in the intent of the effect. The direction of current doesn't matter much on either end of it.
3. Swapping the flow from primary->secondary as indicated in the PCBGuitarMania schematic also works and does tame the effect a bit, but the octave effect is much more subtle/elusive.
4. On the other hand... to make the effect even louder and give it an even more obvious octave effect, I found it useful to use the center tap on the secondary, instead the whole winding - so signal goes from one end of the secondary to the center (or vice versa). The more obvious octave effect makes it come out more readily on lower pitch notes / lower strings, and might perhaps make the effect more usable on a bass (I don't have one to test)?
- If I build a PCB for this, it'd be cool to include four switchable modes for the full range - octave off (but with hard clipping diodes I guess?), primary->secondary, secondary->primary full winding, secondary->primary center tap.
Overall a nice effect - harsh and unpredictable in exactly the way that you'd expect, but a surprising range of sounds and interactivity with the instrument. E.g. with the tone knob turned all the way down, I found it suddenly gets spitty like a gated fuzz.
[1] https://www.youtube.com/watch?v=_RslAoTyIwk
[2] https://pcbguitarmania.com/product/death-by-octaver/
I was having trouble getting the circuit to make any sound at all, then to make anything resembling music... until I swapped out the sketchy amazon-sourced 741 I was using for one with a TI logo (likely from stompboxparts). In the video Oliver talks about using a specific batch of 741s from Radioshack, and it does indeed seem to be a circuit that's sensitive to the chosen op-amp. Across three different varieties of 741 in my drawer, only the TI-branded part produced a decent sound (...along with the occasional radio interference Oliver alludes to in the video).
Notes:
- With the (working) TI LM741 I'm using, the circuit does not seem sensitive at all to grounding the offset pins (neither 1 nor 5) or pulling them to ground through a resistor. I've left both floating to good effect.
- Similarly, leaving the non-inverting input floating (surprisingly) does lead to the intended behavior. Tying it to ground (through a resistor or directly) does not have an effect.
- I played around with tone pot values, and settled on an A100K to focus the in on the more usable range before the pedal quiets down beyond the point of usability.
- I initially spent a bunch of time trying to get closer to the specified resistance values, but found them to be less critical. I didn't have a 910k resistor for the input stage, but a 10M paralleled with a 1M gets close enough. Going from 1M towards 910k does seem to make a slight difference when the tone knob is turned up, but not as much as, well, having the right op-amp.
- On the transformer wiring, there's more flexibility here than I initially appreciated when trying to get the circuit working:
1. The important thing is that the middle tap of the output side is grounded, and at least one tap on the input side is grounded.
2. Having the signal flow be Secondary->Primary yields what seems to be in the intent of the effect. The direction of current doesn't matter much on either end of it.
3. Swapping the flow from primary->secondary as indicated in the PCBGuitarMania schematic also works and does tame the effect a bit, but the octave effect is much more subtle/elusive.
4. On the other hand... to make the effect even louder and give it an even more obvious octave effect, I found it useful to use the center tap on the secondary, instead the whole winding - so signal goes from one end of the secondary to the center (or vice versa). The more obvious octave effect makes it come out more readily on lower pitch notes / lower strings, and might perhaps make the effect more usable on a bass (I don't have one to test)?
- If I build a PCB for this, it'd be cool to include four switchable modes for the full range - octave off (but with hard clipping diodes I guess?), primary->secondary, secondary->primary full winding, secondary->primary center tap.
Overall a nice effect - harsh and unpredictable in exactly the way that you'd expect, but a surprising range of sounds and interactivity with the instrument. E.g. with the tone knob turned all the way down, I found it suddenly gets spitty like a gated fuzz.
[1] https://www.youtube.com/watch?v=_RslAoTyIwk
[2] https://pcbguitarmania.com/product/death-by-octaver/