Passive clipping (like Stewert Mac's Black Ice)

[allium_sativum]

When building my new Strat I decided to try a passive clipping circuit.  I looked at products like the Black Ice Overdrive (from Stewart Mac) and the Dirty Rice (has-sound.com).  I found some information on the web but not much, so here is what I learned and a few sound clips.     

So what is a passive clipping circuit and how does it work?  In this case it is a diode wired to act like a switch, when ever the voltage is high enough the diode turns on, any signal higher then this voltage is bled to ground.  When the voltage drops below the required voltage the diode shuts off and the signal acts as if the diode is not there.  If the voltage is not high enough the diode never comes on.  With one diode you will get asymmetrical clipping of the signal, if you add a second diode in the opposite orientation you will clip the signal symmetrically. 

Because most diodes require .70 volts to turn on and the hottest pickup I know of is the DiMarzio X2N with a rated output of .500 volts most diodes will not work.  To make this work you will need high output pickups and a Schottky diode with a low forward voltage like the BAT46 from STMicroelectronics. 

To connect this circuit you will add 1 or 2 diodes like you would a tone capacitor (i.e. one end to the pot, the other to ground).  If you use 2 diodes make sure that the strips on the diodes are opposite each other.  I would also use a 250k audio tapered pot, a linear taper will drop the voltage to fast and will be hard to control.  A 500k pot only worked for about 1/3 of a turn on most setups I tried.   

I also learned that the impedance of the amp or effects peddle you are plugged into effects how well the clipping works.   I was plugged into a Crate solid state amp and it did not work well at all.  Most tube amps should work fine.

I also learned 2 diodes make any other tone controls on the guitar useless.  With one diode the tone controls only work on the unclipped part of the signal.

For leads the output is less clipped.  The more the strings vibrating the more voltage output from a pickup.  1 string = lower voltage.

So what does it sound like?

Here are some clips.  They were recorded on a Line6 Tone Port UX2, all amp models and effects were shut off.   I used the middle (Duncan Cool Rails) and the bridge (Duncan Custom Custom) untapped in series.   The clipping is set to max.  No settings were changed other then the diodes between samples.

www.allium-sativum.com/graphics/Guitar/PowerChordsClean.wav
www.allium-sativum.com/graphics/Guitar/PowerChords1Diode.wav
www.allium-sativum.com/graphics/Guitar/PowerChords2Diodes.wav
www.allium-sativum.com/graphics/Guitar/Lead2Diodes.wav

 

[-CB-]

You can keep the tone control "active" with two diodes.  If you put the diodes ( and pot) on the signal from the pickup (as the tone control should be) then you'll get both, however, the tone control will likely negate the effect of the diodes, as its bleeding signal to ground, and not making it available for clipping.

I'd rather use a modified bluesbreaker pedal, or.. just use a small amp cranked up a wee bit

 

[Alfang]

Allium      very good post- thanks.  Keep your experiments comming 

I just might use that on one of my pickup testing guitars  it's already got a bunch of switches for other things, I'll just add another switch for the diodes

 

[allium_sativum]

You can keep the tone control "active" with two diodes.  If you put the diodes ( and pot) on the signal from the pickup (as the tone control should be) then you'll get both, however, the tone control will likely negate the effect of the diodes, as its bleeding signal to ground, and not making it available for clipping.

I'd rather use a modified bluesbreaker pedal, or.. just use a small amp cranked up a wee bit

I wired the guitar so the output goes to the volume control pot.  From the input lug on the volume pot I go to a switch so I can turn on and off the diodes.  The I connected from the input lug on the switch to the input on the tone.  Setting the tone control for maximum effect, all the frequencies that are bled off through the tone circuit come back when the diodes are cut in.  I tried with both a standard tone and a passive mid cut.

I am pretty sure this is because you have 4 different impedance values between the hot output of the pickups and ground.  (volume pot, tone pot and cap, diode pot and diode, and the amp/FX input)  When the diode has a low enough resistance in front of it to turn on, its impedance should be very low.  All the other impedances are higher for most of the audio range, so if you treat the circuit like a voltage divider then the diodes get most of the voltage and the tone circuit is not going to do much.  By the time you set the pot on the diode to have enough resistance that it is not dominant in the voltage divider, for most pickups the voltage will be to low to turn on the diodes. 

The reason the tone works while I have only one diode connected is because when the voltage is negative, the diode is switched off and it no longer part of the voltage divider.  Thus the tone work for the negative half of the signal only.

I think I agree, that a pedal or a cranked amp would be better for most users.  I just though I would try it out, and share what I learned.

 

[-CB-]

Can you make me a picture of what you did... ?  I'd like to see how you did it, because it shouldn't be so.

 

[allium_sativum]

Can you make me a picture of what you did... ?  I'd like to see how you did it, because it shouldn't be so.

 

[Alfang]

That schematic looks fine, dont you think a smaller pot on the clipping circuit would be better? 
I would think that with the voltage required to forward bias the diodes, just a small turn on that 250k pot would turn off the clipping?

 

[-CB-]

Nah, there's not enough current there to do that Alf.  Same as a plain ol volume control, you've got not much current... so 250k, 500k or even 1meg works. 

 

[allium_sativum]

That schematic looks fine, don't you think a smaller pot on the clipping circuit would be better? 
I would think that with the voltage required to forward bias the diodes, just a small turn on that 250k pot would turn off the clipping?

Depending on the output of the pickups I would agree that a smaller pot would give more control (but would do nothing to make the tone control work with the clipping).  I started with a 500k pot, because I had one lying around.  I was sure that 500k was too big, it was functional for about 1/8 turn to 1/3 turn depending on pickups selected.   With the neck and mid selected in parallel with coil taps (see switch wiring here, http://www.seymourduncan.com/support/schematics/3lil_hum_1v_2t_super5way.html) the clipping worked for about 1/8 turn. With the neck and mid selected no coil tap (see switch wiring here,   http://www.seymourduncan.com/support/schematics/3lil_hum_1v_super5way.html, note that there is a ground connection missing for this to work as labeled) I got about 1/3 turn.   When I added a 500k resistor from the lug with the diodes to the wiper of the clipping pot (this get you a 250k logarithmic taper tapered pot, but not an audio taper) the clipping worked for about 1/3 a turn for the neck and mid with coil taps and for about 2/3 turn for neck and mid with no coil taps.   Because with an audio taper at half a turn you should get 10% of the total value of the pot, somewhere between a 150k and 250k audio pot seems to be about right with the setup.

It should be noted I do have a switch to disconnect the ground wire to un-tap the auto taps  :icon_smile:  The neck/mid and mid/bridge with the coil tap should be the lowest output of this wiring,  the neck/mid and mid/bridge with out the coil tap should be the highest output.

I estimate that the output of the Cool Rails in the neck and mid to be .350 peak volts, and the Custom Custom in bridge to be about .370 peak volts when playing an E chord in first position.  When playing Lead I estimate about .290 form the cool rails and .310 from the Custom Custom.  I used a cheap volt/ohm meter and corrected from RMS to peak voltage, but I know these numbers are only a ball park estimate.  The RMS on that volt/ohm meter is assuming a pure Sin wave, and it is not.  I do not have the equipment at home to get a better measurement.  If I can I will barrow some equipment form work, I will get a better measurement.

 

[-CB-]

Can you make me a picture of what you did... ?  I'd like to see how you did it, because it shouldn't be so.

A 250k ohm audio taper pot is the suggested pot.  The wiring schematic you have is fine, maybe you have a ground open someplace, as the tone control should work, regardless of black ice operation.

The BI will take everything over (for argument sake) .25v and pass it to ground.  The remaining signal, that which is under .25 volts will be subject to the tone control.  Only thing I can think of is your grounds are not unified from volume, BI , tone, and output.

 

[CD]

Cool to hear the clips. It's definately not my sort of thing though.

 

[DiMitriR33]

this is a complecated thing. a tone control is a variable crossover or low pass filter. the cross over point is when the circuit impedence and the capacitor impedance are similar. a proper tone control in a guitar wont usually cut the fundemental frequency only the harmonics.
with the clipping circuit the impedance is a constant variable with voltage and therefore so is the crossover frequency.
it may be that the tone is working for the portion of the signal that is not clipped but in the portion that is clipped the high end roll off is at some higher frequency.
that combined with the harmonics created by the clipping may make the signal still sound bright although the tone is still technically working for some part of the signal. if you have an extra pole on the clipping switch try to wire in an extra cap in parallel with the tone cap, or just use a larger cap. that might clean up or even cut out notes in the upper part of the fret board. anything will be a compromise at best.

or you could do what i did and put in a batery and a buffer or some kind, there are many transistors and opamps that will work, be crative. the guitar isn't finished yet and i'm not sure i'll even put it in but the whole circuit will fit on the back side of a humbucker.

 

[allium_sativum]

I have started running some spectrum analysis on output with different settings.  It does look like the tone control does have some effect on the sound, but the effects are not highly noticeable audibly.   I will post sound files and graphs when I get them cleaned up. 

For my next project I will be adding a preamp.  (Giving a 2003 BC Rich Mocking Bird the 1970s circuit, well at least close)

 

[DiMitriR33]

hey allium_sativum,
you said it worked with only one diode right? why not use it that way? a popular mod for overdrive pedals is to put two diodes in series for one direction and just one in the other direction. this gives asymetrical clipping less distortion and more output. this is how the boss sd-1 is setup. it is thought to be pleasing to the ear. i'd think one diode would give a similar effect.

 

[allium_sativum]

hey allium_sativum,
you said it worked with only one diode right? why not use it that way? a popular mod for overdrive pedals is to put two diodes in series for one direction and just one in the other direction. this gives asymmetrical clipping less distortion and more output. this is how the boss sd-1 is setup. it is thought to be pleasing to the ear. i'd think one diode would give a similar effect.

It did work with one diode.  I now have a switch to go from one diode to two diodes.  One Diode should create even order harmonics, 2 diodes should create odd order harmonics.  I wanted the ability to switch between them. 

After a hours of looking at spectrum analysis for the output, and finding very inconsistent results.  I final found the problem, the tone capacitor was bad.   I replaced the capacitor and everything works fine.  I have no idea why the problem only should up severely when both diodes where in the circuit.  The problem was there when the diodes where not in the circuit, but the tone did have some effect.

The real point of starting this thread was to share what I had learned, not to trouble shoot the circuit.  Even if I had not found a problem, the tone not working with 2 diodes was not a big problem for me.  This guitar was about building something different and unique, if it did not work to my satisfaction I was going to change it, but I am happy with the results.    I will post the full wiring I ended up with if I get time.