Rubber Diode Overdrive

samhay · December 20, 2013, 01:15:26 PM

Here is another take on a Son of Screamer-type circuit using a somewhat modified version of the ETI soft clipping stage I came across here: http://www.diystompboxes.com/smfforum/index.php?topic=102162.0

The clipping stage uses rubber diodes - thanks to teemuk for putting a name to a schematic - and is 1-knob version of the design I posted on the other thread. It works pretty well and might sound a little softer than more traditional diodes in the feedback path.

I'm building a vero version and will post the layout once it is verified.

Kipper4 · December 20, 2013, 01:46:49 PM

Looks intresting Sam
Thanks I might just have to build it and hear for myself.
The feedback loop in the op amp looks a little like the tonefreak severe except it uses mosfets rather than bjt's.
cheers dude

nice simple tonestack too.

jubal81 · December 20, 2013, 01:59:09 PM

Glad you're working this up. I'm looking forward to trying it out.

SISKO · December 20, 2013, 05:52:41 PM

are you excpecting to sound like a ts?

samhay · December 20, 2013, 06:14:38 PM

Quote from: SISKO on December 20, 2013, 05:52:41 PM
are you excpecting to sound like a ts?

The tone shaping is not quite as mid-centric and the clipping is a little different. If I wanted to indulge in some marketing I might claim it is warmer and more tube-like. Ultimately, it sounds, not unlike, most dual op-amp overdrives.

blackcorvo · December 22, 2013, 02:48:23 AM

Could I use MPSA42/92 for the transistors? I got a bunch of those here.

samhay · December 22, 2013, 02:55:32 AM

Quote from: blackcorvo on December 22, 2013, 02:48:23 AM
Could I use MPSA42/92 for the transistors? I got a bunch of those here.

They should work fine. I don't think it will be too picky about the transistors you use as long as they are BJTs. If you don't use a complimentary pair, you may get asymmetric clipping, which would not necessarily be a bad thing.

As an aside, you can also get asymmetric clipping by changing the value of R6 or R7.

samhay · December 22, 2013, 03:00:13 AM

Here is a vero layout, which is now verified.
Sorry, I was out of 1590Bs, so this is designed to go into a 1590A.

pinkjimiphoton · December 22, 2013, 11:56:02 AM

nice layout, sam!!!

u appreciate the beauty of a well-laid out vero over time.

may give it a shot.. but i don't get it... rubber diodes?

sajy_ho · December 22, 2013, 12:12:34 PM

God bless Arsenio Novo for his golden design! Be aware taht those NPN_PNP pair must be Beta matched for the best results. You may also look at this:http://www.hexeguitar.com/schem/NTv2.gif

samhay · December 22, 2013, 12:23:20 PM

Quote from: pinkjimiphoton on December 22, 2013, 11:56:02 AM
nice layout, sam!!!

u appreciate the beauty of a well-laid out vero over time.

may give it a shot.. but i don't get it... rubber diodes?

Thanks Jimi - that's praise indeed.
Take a look at this thread, where teemuk posted a link about rubber diodes and some other uses.
http://www.diystompboxes.com/smfforum/index.php?topic=102162.0
Not sure if it would be to your taste as it only does a moderate overdrive - no fuzz!

samhay · December 22, 2013, 12:26:30 PM

Quote from: sajy_ho on December 22, 2013, 12:12:34 PM
God bless Arsenio Novo for his golden design! Be aware taht those NPN_PNP pair must be Beta matched for the best results. You may also look at this:http://www.hexeguitar.com/schem/NTv2.gif

That looks like a cool design too. Edit - does clipping to ground, which is different, but the sound clips sound OK. The siteis very cool, with schematics and project files etc.

I used a complimentary pair, as I had them handy, but didn't match them. It looks reasonably symmetrically clipped on the scope.

R.G. · December 22, 2013, 12:33:26 PM

I was inspired to take a step sideways. I just finished simming a circuit that adapts the idea behind the Thomas Organ Vox amplifiers' adjustable limiter to a lower voltage and adjustable clipping hardness. It's adjustable for asymmetry as well. The clipping threshold can be voltage controlled easily - this being something Mark mentioned in another thread.

About the only limitation is that the signal needs to be about 2-3X bigger than you'd normally want for simple silicon diode clippers. This is possible to make smaller by using Schottky or germanium diodes for the internal clipping diodes, but that seems to make the clipping harsher than looks desirable.

The diodes are connected in series, and a cap from the signal to be clipped goes to the center connection of the diodes. The outside ends of the diodes are connected to the outputs of a pair of opamps. The first opamp is a follower connected to a voltage source that goes from ground (or Vbias in single-voltage setups) to the + power supply. This can be a pot or another signal voltage. The second opamp is an inverter with its (+) terminal connected to ground/Vbias and its input resistor to the first opamp output. The two opamp outputs form DC outputs symmetrical about ground/Vbias. The diodes conduct to these offset DC voltages. The clipping threshold can be set from the diode forward voltages to as high as the control voltage fed to the first opamp (given, of course, that the power supplies to the opamps are large enough and all the other minutae to let them act like opamps in the range they're driven).

The hardest clipping is with the control voltage at the ground/Vbias, getting higher clipping thresholds (i.e. less clipping) as the control voltage increases. This may or may not be a desirable amount of rounding, so including a softening resistor in series to the junction of the diodes may make this more attractive. A 2K to 5K pot lets you adjust from hard clipping edges to quite softly rounded edges when there is a 10K in series with the main signal path. The softening resistor can itself be made controllable with a low-resistance LDR and LED setup, or can just be a pot. It can also be made controllably asymmetrical, etc. in all variations by using a different pot (or LDR) in series with each diode independently.

I'll work up some circuits after the holidays.

samhay · December 22, 2013, 02:19:30 PM

R.G. - sounds interesting. Was this inspired by Mark's post about crossover distortion? If so, it sounds like it has come along nicely.

There are a few designs of this circuit with more features that can be found in the thread linked in the first post; It is straight forward to independently vary the gain and the BJT's base voltage. The design here lets you vary both at the same time, which at least gets you using all 3 lugs of the drive pot. I have been thinking of varying the base bias with a different voltage source - perhaps the signal pre- or post-filtering - but have not found anything especially musical yet.

Your comment about the size of the signal voltage reminded me to add that I was originally quite skeptical that this design would work with a 9V supply and spent some time comparing it with larger bipolar supplies. In this case, I think you can just get away with it. There is a hint of op-amp clipping going on when the drive is high, but it is unlikely to be any more than you get with a Zen Drive-style MOSFET clipping approach.

duck_arse · December 23, 2013, 08:17:40 AM

can I just say, except for the 2k2, 470nF and 47uF, it's nice to see you using the E1 range of values, sam.

Gus · December 23, 2013, 08:23:34 AM

Do a search for Musicman amplifier schematics.
Look at the preamp section of some of the amps

samhay · December 23, 2013, 02:37:00 PM

Quote from: duck_arse on December 23, 2013, 08:17:40 AM
can I just say, except for the 2k2, 470nF and 47uF, it's nice to see you using the E1 range of values, sam.

Thank you for noticing. I tend to work with in orders of magnitude, and there are few things I like more than a 10k resistor. That said, I did think about the frequency response and audition a fair few caps before settling on all the 100n's. However, very few of the component values are particularly crucial and I always encourage breadboard experimentation...

samhay · December 23, 2013, 02:40:52 PM

Quote from: Gus on December 23, 2013, 08:23:34 AM
Do a search for Musicman amplifier schematics.
Look at the preamp section of some of the amps

Yup - slacker posted one of their schematics here: http://www.diystompboxes.com/smfforum/index.php?topic=102162.0
Same idea, but with fixed BJT base bias points and series diodes to the collector in place of the 10k resistors I chose.

samhay · March 18, 2014, 09:39:34 AM

For anyone interested, here is another approach to using the rubber diodes as clippers. This one is similar to the arrangement in the Musicman amps and the gain is about 2-50, which is enough to get a moderate overdrive. I don't know if this sounds significantly different to the first version - I don't have that anymore - but this way it is a little easier to set the clipping levels of the rubber diodes.
As drawn, the clipping is a little asymmetric, but this can be made symmetrical as noted in the schem. For more gain, decrease R2 and vice versa.

Edit - this arrangement works nicely when using an inverting op-amp (as in the Musicman amps) as you can dial in, fairly precisely, the maximum clipping point. This would be interesting in a two-stage clipping circuit (think Blues Breaker etc). More to follow...

News:

Rubber Diode Overdrive