#od-3

Cutting bass before distortion -- and adding it back after -- is a great way to improve the perceived “clarity” / “tightness” / “definition” of the low end. What those terms really mean is that the low frequencies are relatively cleaner and less distorted than the high frequencies. Virtually all high-gain amps do this, sometimes to a really extreme degree. By contrast, I can’t think of any design that boosts bass before distortion and then cuts it after. And anyway, if you cut bass before distortion but leave the later stages unchanged, you get a net reduction in bass just as if you had cut it later, but you also get an improvement in distortion quality and musicality from reduced intermodulation products.

The OD-3 is a small-scale emulation of the overdrive you’d get from an old Fender amp, stage by stage. Before clipping, the first thing it does is a mid cut that mirrors a Fender tone stack’s “knobs at 12 o’clock” response, between Q10 and Q13 (green trace):

The toggle I added to this pedal switches between the stock response and the blue trace (-3dB @ 100Hz), by switching C30 from 1uF to 100nF. This is not a big change to hear, but is hugely helpful in matching the pedal’s contour to certain amps, and also when using the neck pickup. It could be made more extreme by lowering C30 to 82nF or 68nF, but then it’s also cutting more mids.

[You might think that such a steep mid cut would be a tone-killer, but it isn’t. Why? Because the pedal is using clipping stages that compress, and that clip softly and asymmetrically. So the low-order harmonics it produces make the mids plenty thick. This is exactly what happens when you crank a tweed Fender amp -- same mid cut right at the beginning of the signal path, and it’s the later stages compressing/distorting first.]

The pedal doesn’t really boost bass per sé after distortion, it just knocks off some of the ‘harsh’ upper mids and gently rolls down the treble. The green trace below is the stock response of all post-distortion stages (including the tone control, which you can see is mostly active from 2k-6k). If you really really wanted to cut bass after distortion, you could change C8 to 18nF and you’d get the red trace. But at that point, the pedal would probably sound like a more nasal version of the SD-1. Who knows, someone might love it.

Schematic is here.

Side note: the schematic shows 1N914s in two of the clipping positions. The actual diodes in the pedal I worked on are not 914s; they had a significantly higher turn-on threshold of about 0.62V. The 914s in my parts bin measured 0.51V by comparison.

Also note that due to the insertion loss of the pre-filter network, D9 and D12 do not actually clip in normal operation -- it would require about 2.2Vpp (750mV RMS) of signal at the input of the pedal to make them clip. Exceedingly few guitar pickups are capable of this; usually only a massively boosted signal from another pedal would exceed the threshold. It’s likely that the diodes are there to limit the signal into the discrete opamp stage, in case you decide to use a booster in front of the OD-3. Which will, of course, yield more tonal possibilities due to the extra clipping. Boost away.

In my opinion, scalability is a huge challenge for musical equipment – something might sound great in a full band mix at ear-shattering levels, but sound terrible playing solo in your living room. This would help explain why nobody agrees on how ‘good’ this or that circuit is – everyone’s application is different.

2. I have no idea how hot your pickups are. In the case of the OD-3, it’s obvious from looking at the schematic that the design goal was to mimic the dynamic response of a cranked amp (i.e. ‘touch responsiveness’). There are three separate places where distortion happens: first, a simple clipper at the input of the differential amp stage (Q13), second, a soft clipper in the diff-amp’s feedback loop (this is controlled by the drive pot), and third, a Muff-style clipper stage at Q15. Just like a non-master-volume amp, the last stage will clip first, and the first stage will clip last. The three stages roughly mimic preamp, phase inverter, and single-ended power stage. Notice also that there is a filter before all the clipping stages that simulates an amp’s passive tone stack.

The point being: a low-output pickup is going to interact with all of this VERY VERY differently from a high-output pickup. My guess is that a hot humbucker is going to clip the first stage at all times regardless of gain level, which is going to make it impossible to control the color of the overdrive by altering pick attack (which is the essence of touch-responsiveness anyway). So a hot pickup might defeat the purpose of the design.

3. The ‘clashing harmonics’ or ‘warble’ thing sounds to me like intermodulation distortion, which is a big challenge to designing musical distortion circuits. But since I’ve never hooked an OD-3 up to a scope or distortion analyzer I can’t be more definitive than a hunch about this. If intermodulation as an idea is new to you, I’d recommend reading R.G. Keen’s Musical Distortion Primer.

In fact, the issue is much bigger than just the OD-3 design. At the risk of going off on a rant, ‘tube sound’ has nothing to do with EQ as such, it’s more to do with overtone spectra and lower/fewer intermodulation products. You hear this most with higher major 3rds, which are typically not very sonorous on guitars anyway (see equal temperament). This is one of the reasons why no distortion pedal really sounds like a tube amp…which is less of an issue if you’re using the pedal as part of a system that includes a good amp that’s opened up enough to sound its best.