Two fun circuits, 2nd harmonic clipper and multi-harmonic clipper

[Gus]

I had built an opamp distortion based on the 2nd harmonic clipping section in the late 90's early 2000's IIRC.  I might have lost/misplaced my notebook with the original schematic and circuit board because of hurricane Irene in 2011 can't seem to find the notebook or circuit.

IIRC In the late 90's or early 2000's I saw a circuit fragment in the magazine "Wireless World" of a 2nd harmonic generator circuit used in organs, The circuit around Q1(C1, C2, R1, R2, R3 and Q1) is based off the fragment note no power supply to the transistor.  The fragment is powered by the circuit in front of it is is a very clever design.  If anyone has the magazine I saw the fragment in please post the issue year month and page.

Sensitive to input level I simed at 1KHz and .07V P to P.  Green input.  Blue at the in of the clipper.  Red at the out of the clipper

The following are sims based off the circuit fragment(from memory) using transistors.



Added a PNP to clip the other side



Circuits not built yet but they should work
Have Fun

[pappasmurfsharem]

Any explanation of what this would sound like?

[Quackzed]

thats neat. it looks like the first transistor is biased close to ground and due to the highish gain of it its hitting the ground rail pretty good there and getting clipped. then its inverted and sent to the next transistor which is biased closer to the positive rail at max gain, this is i think where that already clipped side of the wave hits the positive rail and folds over creating that batman looking waveform. thats my understanding anyhow, i'd be curious to know what you make of it as to how its doing the foldover, and if my explanation is close or any where close? i've been messing with these type npn stages and trying to learn how they work...
as for how it would sound, and i apologize if i'm cutting in here, i'd guess when you play softly it would have a sweet overdrivey tone with a little sweet 2nd harmonic complexity to it, and when you play harder it would have an edgyer tone with more prominent octave up sound mixed in, and a bit more high end edgyness as well. kind of a mellow octave to edgy heavyer octave, depending on how hard you hit it.
the second schem would be similar except rather than second it would have 3rd harmonic overtones to it'd sound more compressed and distortioney rather than overdrivey but with similar dynamic character to the first schem.. ? i think? 

[GFR]

You could call it "the Batman" fuzz

[Gus]

Lets look at the first circuit
Q2 is setup as an amplifier you can find in textbooks
However note R9 and C3,  R9 value can be changed to adjust the gain of the first section.  R9 can be a gain control 0 to 5k etc.

C1, C2 R1, R2, R3 and Q1
For a negative going signal the output is loaded by R1 and R2 and then the signal goes "thru" R3 and the output is at C1 R11 node.  R3, R11, C2 form an AC voltage divider this is important.

Now here is the fun part with a positive going signal of enough amplitude to turn on the transistor via the R1, R2 voltage divider string thing change.  The transistor turns on and the output node C2, R11 is connected to ground thru the transistor (R3 and Q2 form a voltage divider)
As the signal drops in amplitude the transistor turns off.

Adding Q3 in the 2nd schematic clips the negative going side.

IIRC in an organ the circuit has a known input level, with a guitar bass etc application you will want to control the gain of the circuit before the 2nd harmonic circuit.  So a compressor would be good to add if all you want was 2nd harmonic.  IIRC the opamp circuit  I built with the Q1 Q3 section was very sensitive to level and and control adjustments.

The posted circuits are using gain stages that load the pickup because the clipping stage adds edges(horns) and edges are higher order harmonics.

The 100K at the output can be a 100K volume control.  If you remove or make R9 large enough the gain of the first stage is about 10K / 2.2K.

[Quackzed]

ok. thanks! i think i get it. it's not that q1 is folding the wave at higher signal swings, but rather that q1 needs a decent positive going signal swing to turn it on, and when its on its inverting the wave and supplying the 'foldover' looking part of the wave... 
thanks again gus!

[mac]

Adding Q3 in the 2nd schematic clips the negative going side.

Can Q3 be replaced with a diode?

mac

[Quackzed]

i'd guess that using a diode instead would result in a batman with flat shoulders rather than bat shoulderpads. 

[duck_arse]

I've seen something of mine making that top (single batman) waveform, maybe a fuzzface. I'll have to look it up .....

[Kesh]

i posted something about this kind of transistor off-biasing to get an octave up harmonic in the SPICE forum of this site

http://www.diystompboxes.com/smfforum/index.php?topic=99112.0

[Gus]

Kesh

What I am posting about is the Q1(Q3) section Q2 is a gain stage.  Look at the waveforms green is in to the gain stage, Blue is at the input of the the Q1 stage, red is at the output.  Note the only power to the Q1(Q3) stage comes from the signal output of the gain stage.  Q1(Q3) is acting as a switch and being turned on and off by the R1, R2 voltage divider string.

Waveforms are somewhat the same, the way they are developed is different.

[Mike Burgundy]

wait, what? that looks suspiciously like analog foldback distortion!

[Gus]

wait, what? that looks suspiciously like analog foldback distortion!

What is analog foldback distortion?

[Kesh]

Kesh

What I am posting about is the Q1(Q3) section Q2 is a gain stage.  Look at the waveforms green is in to the gain stage, Blue is at the input of the the Q1 stage, red is at the output.  Note the only power to the Q1(Q3) stage comes from the signal output of the gain stage.  Q1(Q3) is acting as a switch and being turned on and off by the R1, R2 voltage divider string.

Waveforms are somewhat the same, the way they are developed is different.

Yes, I know.

Q1 in your first circuit basically does the same as the Q in mine. Both forward bias the base collector junction over part of the waveform, and so part of the waveform is inverted compared to the rest of the waveform (actually as it's a common emitter circuit, i should say part of the waveform remains uninverted compared to the rest). Not that I'm claiming it as original. It's pretty old apparently.

[Kesh]

oh, and over the inverted part you have lost all the transistor's amplification, it's basically being bypassed, just a diode voltage drop, so impedance shoots up. this isn't really an issue with your circuit as you have the gain stage prior and have collector and base from its output.

[Mike Burgundy]

I encountered foldback distortion in our guitarists new Source Audio SoundBlox pedal - instead of hard-clipping, when the waveform hits the threshold and inverts *from that point*, so a sine top will look like an "M" with sharp tops and rounded hollow. This is something only possible in DSP to my knowledge. Certainly sounds very strange, weirdly organic, and very interesting.

[joelindsey]

This is really interesting. I have a Gibson G101 combo organ from the mid 60s that uses some kind of octave up circuit for the highest 4' octave. I presume it was to cut costs, as transistors were much more expensive in the mid 60s than they are nowadays. They must have used this with some sort of high-pass filter afterwards.


edit... nevermind, it was transformers. still cool though!

[brett]

Hi
am I correct in thinking that a Fuzzrite produces some similar results?
My stuff is all packed away in the shed, so I can't be sure.
Anyone have some fuzzrite traces, or can tell from the schematic, or has a spice of it?
cheers

[duck_arse]

I was getting the attached waveforms from a modified 3V maestro fuzz I made about 7 years ago.



whether imageshack gave me a good link or not, I never know.

it's a funny thing, but way back then I captioned the scope shot with "holy fuzzbox, batman, what's going on here?" in the operators manual.


(why can I never find anything on this site?)   

[deadastronaut]

You could call it "the Batman" fuzz