Some BD-2 tricks

[Mark Hammer]

The very recently-released Keeley Muse drive, tailored to Andy Timmons' needs, is a sort of extension of the mods that Robert Keeley has been doing on Boss BD-2 pedals for some time.  I have absolutely no idea what those mods are, and with the puny speakers on my monitor probably couldn't discern what they are.  But seeing the Youtubes for the Muse got me curious about the BD-2.  So, I made one, using the layout here: http://effectslayouts.blogspot.com/2017/01/boss-blues-driver-2.html  I should note that I have absolutely no idea what the standard/custom switch does on the Waza version of the pedal.

I noticed that the stock unit uses a dual-ganged 250k pot to adjust the gain of the two discrete op-amp stages simultaneously.  I didn't have one of those pots on hand, but thought it might prove interesting to use two independent gain pots.  Preliminary results are interesting.  Using Ge diodes instead of Si type will certainly lower the amplitude of stage 1 signal hitting stage 2, and the Youtube demos suggest that.  Two independent gain pots will also allow for  achieving overall combined gain with less impact of stage 1.

A second experiment proved dramatic.  There are 4 clipping diodes on the output of stage 1, and I gather Robert changes those to germanium for one one of the mods/option on the Muse.  Diming the gain on stage 2 but lowering the gain on stage 1 yields the result you'd expect: less harmonic content. But just to the "left" of those diodes there is a passive tone-shaping network, that is a little more complex than the midscoop one sees on the Superfuzz and similar. 

Like all such tone-shaping networks, it does what it does using selective bleedoff of frequency content.  I learned many years ago that one could beef up the signal on any amp using a Fender tonestack by simply lifting, or at least increasing the resistance of, the connection of the tonestack to ground.  Less bleedoff = hotter signal reaching subsequent stages.  In this case, R11 (15k) was the candidate for lifting.  I unsoldered one end and sure enough, it supercharged the pedal.  Now, no longer a warm overdrive, it became a raging monster.  Easing back the gain of stage 1 yielded something more familiar to a BD-2 user, but still more intense.  I only finished powering up the completed circuit about 90 minutes ago, so I'll continue experimenting tomorrow.  But for now, the final product will have two gain pots and a "network lift" toggle for extra drive.  To be continued.

[Elijah-Baley]

It's very interesting.
One day I'll build the BD-2, too. I just a bit concern about the mid scoop character in this pedal, right because the part around R11 you mentioned, I guess. That should be a sort of fixed eq amp style that give to the pedal that typical scoop. Of course we can modified it, maybe just like you did.

I follow this thread, and maybe later I'll suggest you something about it, I guess I have some mods noted somewhere.

[Elektrojänis]

That tone shaping network looks a whole a lot like a Fender style tone stack. Values are a bit off, but mostly close. If I read it correctly, Treble is at Minimum (resistor value a bit higher than the typical pot), bass at max (or over, as the resistor value is again higher than the pot would be), and mids at max (or bit over again).

It might be usefull to feed the part values to the online tone stack calculator and try tweaking things. I did, and the difference to stock Fender tone stack with the same settings is not big.

[Mark Hammer]

Note that there is a gyrator circuit connected to the op-amp output stage.  I'll try playing with that today; first by lifting it to get a sense of how it changes the stock circuit's sound.

[bean]

AFAIK, the Keeley mod on the 4 diodes is swap D1-D3 for 1N4002 and jumper D4. But, he may have done more than one. I'm really interested to try the resistor lift.

Another mod I've tried is to swap out different values for C14 to control bass content without messing the gyrator.

[aion]

Another mod I've tried is to swap out different values for C14 to control bass content without messing the gyrator.

Adding a 25k pot in series with R20 is great for this as well. I learned that trick from the Xotic Soul Driven, whose back half is otherwise taken straight from a BD-2. It has the benefit of going back to stock when you turn it all the way down since it just adds resistance to what's already there.

The Machine Head Galaxie Mod is another really interesting one, definitely the most extensive mod I've ever seen to any pedal, but completely revoices it, lowers the gain and fixes a lot of the long-standing complaints people have with the circuit. I put together all the details on the mod in the Sapphire build doc, pg. 7 (though be aware it's keyed to the Sapphire's part numbering rather than the original BD-2 schematic): https://aionfx.com/app/files/docs/sapphire_documentation.pdf

[Mark Hammer]

Thanks for that.  I used J201s for my own build, but it turned out that I had a quartet of thru-hole 2SK184s sitting in a parts drawer, so I may end up installing those instead.

I'm installing an on-off-on toggle to select three values of bleed-off resistance for R11: 100k, 40k and 15k.  I installed 100k to the board, and use the toggle to select parallel resistances of 68k or 18k to achieve the described values.

I'm also going to be installing a toggle to lift C12 for a warmer tone.

[Elektrojänis]

Another mod I've tried is to swap out different values for C14 to control bass content without messing the gyrator.

You might try changing R10 for bass control too. You could even make it a pot. Difference to C14 or R20 (as suggested by aion) is that R10 controls the bass content before the clipping diodes. C14 and R20 would be totally post clipping. The first discrete opamp might distort too so R10 might not be before all the distortion, but earlier in the chain.

[Mark Hammer]

All worth trying out.  Me, I'm setting my ceiling on 4 pots and three toggles:
Gain 1
Gain 2
Tone
Volume
Boost toggle
Bite toggle
Clipping toggle
Given that this module is being added to a rack of about 40 other drives, I have no urge to make it do everything conceivable.  A little more than stock, but that's enough for me. 

[Elektrojänis]

That's plenty of controls. I guess the usual problem with diy is some sort of feature creep. It's just too easy to make everything a front panel control, and then you end up twidling with the knobs more than playing the guitar. It's not wrong and it can be fun or even musical, but sometimes it's good to set it, forget it and play the guitar.

[Mark Hammer]

I agree.  I generally aim for the most variation with the fewest knobs and switches.  Although, I have to reserve judgment until I try it out with a proper amp, rather than the little solid-state amp I use for testing basic functionality of circuits, the diode-bypass switch and lifting/engaging of C12 don't do anything particularly obvious.  Varying the bleed resistance of the stage 1 tonestack is far more impactful.  So, pending further testing, I may just leave it at the stated 4 controls and single toggle.

[Rob Strand]

The Machine Head Galaxie Mod is another really interesting one, definitely the most extensive mod I've ever seen to any pedal, but completely revoices it, lowers the gain and fixes a lot of the long-standing complaints people have with the circuit. I put together all the details on the mod in the Sapphire build doc, pg. 7 (though be aware it's keyed to the Sapphire's part numbering rather than the original BD-2 schematic): https://aionfx.com/app/files/docs/sapphire_documentation.pdf

I haven't tried it since it's quite a lot of changes and perhaps a few not adding a lot.

IMHO, this simple revoicing mod makes the tone more even and little less zingy.  Kind of shifts away from the exaggerated BD2 tone but not too much.  It's a very simple mod - no added parts.  (Lift one end of the cap and connect it to ground.)


As a warning if you add a simple switch for modded and not modded it will pop.  You really need a second cap and some de-pop resistors for that.  Then of course you are free to play with the cap value  .

FWIW, I believe the correct value for C17 (on the original schematic) is 6n8 not the 5n6 shown on some of the redrawn schematics.

[Mark Hammer]

Interesting.  That may be why I didn't hear a huge difference by lifting that cap.  A 6n8 value lets more upper mids pass through.  I imagine a suitable pop-free mod would be to put a higher value resistor (e.g., 470k) in series with the cap, and use a toggle to bridge the added resistor.

[Rob Strand]

Interesting.  That may be why I didn't hear a huge difference by lifting that cap.  A 6n8 value lets more upper mids pass through.  I imagine a suitable pop-free mod would be to put a higher value resistor (e.g., 470k) in series with the cap, and use a toggle to bridge the added resistor.

Just to be clear the idea is not only to lift the cap but to also connect it to ground.

Yes you would need to add some resistors in series with each cap; 2M2 would do it.

Normally the caps C17, C19 (original schem) form a capacitive voltage divider at high frequencies and R29 allows the low frequencies to pass.  The net result is a low-pass shelving filter.   I don't know why they  did it that way.  The series combination of caps C17 and C19 put a large capacitance across the opamp output - which is never a good idea.

The mod changes the filter from a low-pass shelving filter to a low pass filter.   It's subtle but noticeable.   Enough to take the sting out of the highs without losing the basic character of the pedal.

The galaxy mods are a bit heavy handed, it's really a different pedal afterwards.   I found it was difficult to compare lows from the clip,

[Mark Hammer]

Ahhh....  got it.  Thanks for that.
The 6n8 goes to ground in parallel with the 5n6,forming a single pole lowpass, rolling off beginning around 2.3khz.

Going by the video, the galaxie mods make it a more usable pedal at highest gains, but dull the mild coloration one would get, and likely seek, at lower gains.  I'll note that increasing the bleed resistance in the fixed tonestack enhances the lower midrange of that first stage.  In tandem with the tone pot, one can get the "woolier" sounds the galaxie mods aim for.

[Rob Strand]

The 6n8 goes to ground in parallel with the 5n6,forming a single pole lowpass, rolling off beginning around 2.3khz.

That's the idea.  It's more like 3.6kHz because of the loading effect of the tone pot and volume pot.   A good approximation would be to use 10k (tone pot) in parallel with the 5.6k, or 3.6k ohm, for the resistance. As expected 5.6n+6.8n = 12.4n for the capacitance.   So the filter cut-off is f3 = 1/(2*pi*R*C) = 3.6kHz.
I haven't looked at this for a while but regarding the pop:  If you wire the cap back to Vref instead of ground it will pop a lot less when switched.   Funny how you see things when they are posted in front of you!

[matt_garman]

If you guys haven't seen this already, you might be interested in the 4-part BD-2 analysis Chuck D Bones did over on the PedalPCB Forum:

• Part 1
  
• Part 2
  
• Part 3
  
• Part 4
  

[Mark Hammer]

So here's a quick and dirty video demonstrating the suggested mods.  It's built in the 2-1/2"x4" format I've used for my modular system.  The equipment is not the greatest, and the testing amp has limited bandwidth, so I hope any differences between settings are audible.  As you might be able to surmise, the topmost toggle is bypass-up/engage-down.  Clearly, altering the value of the 15k bleed resistor is quite audible.  Details are provided with the YOutube.
Normally, I'd say "Enjoy!", but in this instance I'll just say "Tolerate!".