"that" Overdrive [documentation]
Posted: 12 Jun 2021, 13:34
Here's my latest creation, that Overdrive! No, not that one, this one! 
You might know I like to be very straightforward or to clean up other designs usually, this began as a test to see if I could be more flamboyant with my designs. Also I wanted to take on the difficult challenge of making an overdrive pedal which doesn't borrow heavily from known designs. In the end my minimalist approach is still there, and this isn't a physically big circuit and the values are all common: despite extensive breadboard testing, I didn't have to give up on that.
We all know which is the predator in this category, and we know it's green now, I wanted to refresh on that a bit. For example, we know the pre gain low shelf is very important to keep a clear distortion, and it's very important to me, but what if we could make that rolloff steeper, and maybe that way being able to push it lower? The most obvious solutions would be a tiny input cap or an active high pass filter at the input, but I'm not a fan of either approach, either because of interactivity or because of having the noise of an active stage before all of my gain. Instead I thought, what if we feed back more of the low frequencies? what if instead of the usual shelving lowpass formed by the feedback network in an inverting op amp we try to make it second order?
Same thing for the post-gain lowpass. First order passive filters are ubiquitous, and they work ok. Even the green pedal, with its active tone control, in practice doesn't go far off at all from the response of a cut-only first order lowpass, once you combine the active stage with the fixed lowpass before it. A second order filter here is relatively unexplored, and can be used among the other things to preserve more of the nice lower treble while still keeping the bees in the nest with the steeper rolloff.
This is the final iteration. I've messed with some things while getting ideas, like parallel resistors+capacitors in the feedback for more of a custom-fit frequency curve (but for the grounded leg that would have given me more bass not less!), switchable clipping, fixed frequency gyrator mid control... In the end I tend always to simplify though. For the feedback I settled for two parallel networks, one traditional with variable gain, one a fixed second order lowpass, which then get mixed together. The response of the gain stage looks like this (stepped gain pot, you can see the bass cut is sharper than with a traditional green pedal):
But I know glassy and clean bass isn't for everyone, so I added a bass control which in practice isolates the additional network, leaving you with something more familiar (again green for reference):
Finally there's the S-K lowpass. A trick here is to switch between the usual buffer and a gain stage, which increases the final Q. That's what the resonance switch does. The small peaking near the cutoff is very very effective in giving you more sounds. The boosted Q value has been chosen to be still low and usable but different enough. The 100KC dual pot is a bit unfortunate, but other values can be used with scaling, and A taper if wired in reverse. Here's a sweep of both modes:
Build pic:
Vero:
Demos:
You might know I like to be very straightforward or to clean up other designs usually, this began as a test to see if I could be more flamboyant with my designs. Also I wanted to take on the difficult challenge of making an overdrive pedal which doesn't borrow heavily from known designs. In the end my minimalist approach is still there, and this isn't a physically big circuit and the values are all common: despite extensive breadboard testing, I didn't have to give up on that.
We all know which is the predator in this category, and we know it's green
now, I wanted to refresh on that a bit. For example, we know the pre gain low shelf is very important to keep a clear distortion, and it's very important to me, but what if we could make that rolloff steeper, and maybe that way being able to push it lower? The most obvious solutions would be a tiny input cap or an active high pass filter at the input, but I'm not a fan of either approach, either because of interactivity or because of having the noise of an active stage before all of my gain. Instead I thought, what if we feed back more of the low frequencies? what if instead of the usual shelving lowpass formed by the feedback network in an inverting op amp we try to make it second order?Same thing for the post-gain lowpass. First order passive filters are ubiquitous, and they work ok. Even the green pedal, with its active tone control, in practice doesn't go far off at all from the response of a cut-only first order lowpass, once you combine the active stage with the fixed lowpass before it. A second order filter here is relatively unexplored, and can be used among the other things to preserve more of the nice lower treble while still keeping the bees in the nest with the steeper rolloff.
