DIY Variable Crossover - how?

[CamH132]

Hey guys,
I'm currently in the processes of designing a multi band distortion pedal where the lows and highs are split with a variable crossover and then effected by their own individual clipping stages before being mixed back together in the output stage. Up until this stage I've just built basic distortion / compression pedals so I've got a few questions about how i should go about this...

Creating a fixed crossover seems easy enough but i'm wondering if it's possible and how to go about using a pot to control the frequency at which the signals are split?

Also, i would like to have another single pot to control the mix between the two bands in the output stage - is this as simple as wiring the output of the high band to lug 1 and low band to lug 3 on a pot?

Cheers

[Brisance]

dual gang pot, with each as a variable resistor in an RC.

[mth5044]

This:

http://sound.westhost.com/project148.htm

And the second half of this:

http://www.runoffgroove.com/splitter-blend.html

You can get rid of the inverter on the top return of you don't need it and make that stage identical to the bottom.

[CamH132]

Ah nice, thanks guys.

The dual gang pot in an rc network seems nice and easy - don't know why i didn't think of that. Are there any real advantages (i assume a possible flatter frequency response?) when using the state variable electronic crossover linked by matt over the simple dual gang pot / rc network setup?

[Mark Hammer]

If I were you, I wouldn't mess around with manipulating the crossover point.  I suppose you can always dicker around with different ones during the breadboarding and design.  BUt when it comes to a functioning pedal, you'll have more than enough parameters to be preoccupied with.  My advice is, find the frequency-dividing points that you like, and stick with them.  Then you can devote your attention to adjusting their respective gains, and blending the bands.  That's more than enough to futz around with.

[GibsonGM]

Any rules of thumb that you've discovered over the years, Mark?  I've been kinda curious about this kind of 'differential shaping' (for lack of a better term) for a while but haven't gone too far into it.  My trips into paralleling the distortion have been limited to playing with the WHOLE split signal, then recombining - not as glamorous as I'd hoped for. 

Say I want to use a diode clipper on my highs, but more of a traditional FF arrangement on lows....what would good starting cutoff points be?

1kHz and above = "highs", and 1kHz and below for the "Lows", maybe?  Don't know if there's a 'holy grail' area...thx....

[Mark Hammer]

http://www.diystompboxes.com/smfforum/index.php?topic=51810.0

[blackieNYC]

I started with a big muff pi filter.  It has a dip in the mid range, so I ended up tinkering with the values.  I was only going for a very mild overdrive, but after trying a bunch of different filter points, I ended up going back to the BMP tone split.
I want to try this again with a much heavier distortion. I expect that crossover point would change.

Mark, how does the flexidrive do for a contemporary metal sound?  I would think slightly less distorted lows might give you a tighter and more percussive chug.

[Mark Hammer]

I would think one would want at least a 2-pole filter for frequency-splitting.  Here's the final version I settled on.  You'll note that the clipping diodes for the upper band are placed in the feedback loop, a la TS-9, and not to ground as in the earlier version.  You'll also note the added lowpass filtering on the upper band's output.  I figured that would soften the highs a bit.  Some folks might want to increase R4 a bit to reduce drive, and also use separate 100k pots to ground from R8 and R16, so that drive could be set independently, instead of the reciprocal drive control you see here.  I was aiming to see just how much flexibility I could get with just 3 controls, and a quad op-amp.  The x-over is around 880hz.

Couldn't tell you how it sounds for "metal".  Haven't plugged it in for a loooonnnnggg time.  It did get a lot of different sounds, though.  I have the board, and picked up a 1590B this morning.  I may re-install it in something this weekend.

Someone did send me a layout for it.  I wish I could remember who so I could thank them publicly.  In any event, the OP is cordially invited to give it a whirl and see if it does what he was aiming for, whether stock or modified.  I've done the legwork.  No reason why others shouldn't take advantage of it.

[GibsonGM]

Awesome design, Mark - thanks for showing!    Funny how those 'little ideas' end up growing, and growing, huh?   

[Mark Hammer]

See if you can find a schematic for the Yamaha MBD-100 Multi band Distortion.  It's a bit like a mini-Quadrafuzz, with 3 bands clipping separately.

http://www.youtube.com/watch?v=6KLfR-G3H38

[Brisance]

Mark Hammer:
that pedal sounds incredible! I was contemplating on distorting different bands separately to better preserve chords, now I want to do it even more!

[Mark Hammer]

Well now you CAN!  Note that this schematic can be simplified a bit, if using TB instead of electronic switching.  The low section (path shown in the top of the drawing) is a lowpass, where the other two paths are bandpass type. The "Balance" conrol adjusts blend of lows and mids, with the highs unmodified.  I assume they are addressed by the tone control further downstream.  It may well be simpler and more flexible to simply have an arrangement like the Quadrafuzz with a level control for each band feeding a common mixer section.  I've attached a reduced version of the Quadrafuzz (Tri-Fuzz?) for reference.