Low End Shunt for Distortion Circuit

[Bob M.]

I have a lap steel that I sometimes play thru a homemade distortion loosely based on a MXR Dist + or a DOD 250. I've modded this distortion circuit so many times to get it right for this instrument that it probably doesn't bear much resemblance to what I started with, except for the IC. Even though I play thru this distortion only about a a quarter or third of the time, I'm really happy with the sound with one exception: There are some really big low end artifacts that creep in from time to time and I'd like to develop a way of rolling off the very lows when using the distortion. I would think I'd like to be rolling off or shunting to ground everything below 120Hz or 150Hz or maybe even 200Hz.
I think a 6db per octave might work but I might need a 12db per octave filter. I do run lots of lows on my amp but I'm unwilling to change this because the clean sound (when not using the distortion, most of the time) is really awesome just as it is now. I'm not sure if I'll filter the input or the output of the distortion but I'll experiment with this. Can someone point me in the right direction about how to filter out this very low end?

Thanks,

Bob M.

[Chuck H]

A capacitor in series with the signal, prior to the clipping stages, is the obvious solution. But can this be implemented on your pedal without affecting the clean tone? That's really all there is to it. If the pedal will true bypass then a series cap of a selected value (or a smaller value cap than is currently in use) should trim more bottom end.

[pdf64]

Just to expand on Chuck's point, see FDP - Forum how the 'Yngwie' circuit has 10 times lower input cap (and half the value of output cap) compared to the stock 250.
So playing with those 2 cap values, and perhaps adding a third coupling cap between the clipping diodes and output volume control, may help you to get the result you want.
If you want to nail it without tinkering, a couple of graphic eq pedals, one either side of the distortion pedal, would probably do it.
Pete.

[Mark Hammer]

Varying bandwidth via cap choice at different points is tantamount to sticking an EQ before or after a distortion circuit: they each have different impacts.

If one reduces low-end bandwidth after clipping, then you get the harmonic content resulting from the clipping of the low end, but not the low end that instigated it. If you attenuate bass prior to clipping, as the venerable Tube Screamer does, then you eliminate both the bass and the greater clipping that tends to be produced by higher-amplitude bass content. That results in a "smoother" tone.

The basic Dist+/DOD250/YGM308 topology affords a few places to intervene. There are input and output caps, as others have noted. However the most pivotal intervention, in my view, is the cap along the ground leg of the op-amp, which can be used in a very strategic manner.

Normally, one will see a 4k7 fixed resistor, a 47nf cap, and a gain pot of 500k or even 1M (generally reverse log). When the gain is minimized, you get full bandwidth with that combination of components. When the gain is maxed, however, the 4k7/47nf combo yields a low-end 6db/oct rolloff of 720hz.

Personally, I think the values were selected to a) introduce more "bite" as gain is increased, by attenuating the lows, and b) move any hum arising from use of lots of gain applied to non-HB pickups further into the background (or at least not making it quite as obvious).

That's how it is usually done, but it's not how you have to do it. Remember that the gain can be altered by means of the feedback resistance, the ground leg, OR BOTH. So, you could make the ground leg 4k7 and 68nf or 100nf, to provide a fixed rolloff around 500hz or 340hz, respectively, and use the feedback resistance value to alter the gain. Depending on your musical tastes, that might be a 10k resistor and 1M pot, or perhaps even a 47k and 500k pot if, let's say, your tastes focus primarily on mild overdrive.

When the feedback resistance is used to vary gain, any capacitance in the feedback loop will have an impact on the high end rolloff as feedback resistance is varied. So, in our example, if we used a 47k+500k arrangement (yielding gains ranging from 11x to 117x), and a 100pf feedback cap in parallel, our treble rolloff would start around 33khz at min gain, and trim back to 2.9khz at max gain (which would be a nice warm growl).

Again, remamber that all of this is taking place before the signal hits the diodes. And the clipping you hear is a function of what is being clipped. Make the source signal "warmer", and the resulting distortion will not be as harsh-sounding.

Note that "warmth" can be impacted on at two point in the basic circuit. One is the value of the feedback cap, and the other is the value of the cap to ground in parallel with the diodes. The stock value of .001uf does precious little other than keep ultra-sonic crap out ot the amp (rolloff starting just under 16khz). A 2200pf cap will bring that down to 7.2khz, and a 3300pf cap brings it down to 4.8khz.

[Bob M.]

All very good suggestions. I have worked alot with input and output series caps to shape the tone that I'm after and it's pretty much there now. Can't really do pre or post graphic eq's on this setup nor am I able to do true bypass for complicated reasons. What I was hoping to do was get something along the line of 'plug and play' PA mixers that have a bass rolloff/attenuation at the input. These help alot to control the 'mud', especially with vocals. I was hoping to be able to adapt this to my circuit somehow.

Thanks for your input,

Bob M.

[guitician]

To keep your current response but roll off more bass you could add another active component, such as a FET voltage follower, to add another point of series coupling capacitance.