Distortion Circuit Types - General and Technical Discussion

[JRM]

If one check the distortion pedals available in the market it's amazing: on Thomann.de one get 434 items with prices ranging from €18.90 of the Danelectro D3 FAB Metal to the €399.00 of the Alairex H.A.L.O.; on Musician Friend one can choose from 466 pedals ranging from $11.99 to $399.00 (Dano and Alairex again although much cheaper than here in Europe as 1 EUR=1.27 USD!). And this are the mainstream effect companies as some "boutique" builders don't sell through this big stores.
Sure that in this mix there are a lot of Boosts, Overdrives, Fuzz and, off course, real Distortion pedals. In the end of the day I think that apart from clean boosters, all this kind of pedals/circuits could be considered as Distortion. Well that could be the first line of discussion:

• 1. From what value of gain can a circuit be considered a distortion circuit?

Reading this is quite helping in terms of understanding the several design options: http://www.geofex.com/effxfaq/distn101.htm
But...
From 15 alternatives from which 3 are for amps, not easily applicable to pedals, one reach more than 400 variations so, one can conclude that there are mainly a few basic designs with a bunch of versions (with some of them just real copies of others).

• 2. So, which are the basic designs?
• Tube Screamer
• Fuzz Face
• Big Muff pi (original)

[anchovie]

• 1. From what value of gain can a circuit be considered a distortion circuit?

It's a distortion circuit if it distorts. Gain is just the amplification factor of something - whether it distorts or not is down to a combination of input signal level, gain and headroom.

Your "basic designs" list will most likely start arguments over at what point a circuit becomes something original rather than a derivative of something else. Expect other arguments over what the differences between overdrive, fuzz and distortion are. Were you hoping to find something conclusive? 

[Mark Hammer]

"Gain" and "distortion" are two separate, but connected, things.  There is perhaps no better illustration than the "Black Ice" distortion gadgets, which literally have NO gain, but provide distortion in response to hard picking.



Gain can certain help to bring any signal closer to the point of clipping, but then, as so many pedals amply illustrate, selective use of diodes can keep the same gain and move the clipping point further away, such that there is less clipping for the same gain.

[amptramp]

I see the following distinctions:

1. Amplifiers with a non-linear response throughout the entire signal range.  Because of the curvature of the output/input (i.e. variable gain) they always distort at every signal level but the distortion becomes greater with amplitude as you use more of the gain variation.  The Fuzz Face is the most familiar of this type.  I would consider this to be a pure distortion rather than a fuzz in spite of the name.

2. Amplifiers with clipping such as the DOD250 which is a gain stage with antiparallel diodes at the output.  This behaves as a linear amplifier until the output reaches the clipping level which may be symmetric (mainly odd harmonics) or asymmetric (even and odd harmonics).  This is a fuzz.

3. Tube screamer circuit which is a non-inverting stage with feedback clipping (which may be symmetric or asymmetric) but since there is a gain of unity for the input signal, there isn't a guillotine-like clipping of the output but a clipped highly amplified signal riding on the original input.  This is electrically similar to a DOD250 with resistors in series with the diodes.  This is somewhat between a distortion and a fuzz.

4. In some types of distortion based on single-ended amplifiers like the Big Muff Pi, there is capacitive coupling to the diodes so that low frequencies have a higher clipping threshold than high frequencies, reducing the muddiness of the signal.  This is also somewhat between a distortion and a fuzz.

5. CMOS clipping or comparator clipping would be considered a fuzz.  It takes the input and assigns a fixed positive value for any signal above zero and a fixed negative value for any signal below zero.

If you can imagine a waveform being the summation of a high-frequency component and a larger low-frequency component, a DOD250 clipper output sits at a fixed maximum value for inputs that exceed a certain level.  Thus, the high frequencies are removed when the low-frequency plus the high-frequency signal is at this level.  This leads to a muddiness as the high frequencies disappear for the duration of the extremes of the low frequency signal.  The use of resistive or capacitive coupling to the clipping diodes tends to remove some of this effect by allowing some of the original signal to ride through and this may be broadband with resistive coupling or favouring low frequencies with capacitive coupling.

I define a distortion as being non-linear but without discontinuities in the output/input gain function.  Anything with a discontinuity is a fuzz and has a region of linear operation for signals below the clipping level.

[bool]

The DOD has a "black ice" at the output ... in perspective.

Or the black ice is diodes without the 741 ... (or xyz semiconductor circuit) etc.

[JRM]

• 1. From what value of gain can a circuit be considered a distortion circuit?

It's a distortion circuit if it distorts. Gain is just the amplification factor of something - whether it distorts or not is down to a combination of input signal level, gain and headroom.

Your "basic designs" list will most likely start arguments over at what point a circuit becomes something original rather than a derivative of something else. Expect other arguments over what the differences between overdrive, fuzz and distortion are. Were you hoping to find something conclusive? 

I was thinking of the gain of a circuit that takes the signal from a guitar and "loads" it to a clean amp.
Maybe I didn't express myself correctly (English is not my native language). More important than to know that pedal X or Z are derivatives of Z or original pedals is to understand what makes their distortion different from the huge competition. Maybe is because they use a different type of diodes on the op-amp feedback loop type of circuit (or the Tube Screamer type, if someone prefer).
I'm not hoping to find nothing but to discuss and learn.

[antonis]

I was thinking of the gain of a circuit that takes the signal from a guitar and "loads" it to a clean amp.

English is not my native language too so, if I understand what you mean, you are talking for a PRE-amplifier..

[merlinb]

Because of the curvature of the output/input (i.e. variable gain) they always distort at every signal level but the distortion becomes greater with amplitude as you use more of the gain variation.  The Fuzz Face is the most familiar of this type.  I would consider this to be a pure distortion rather than a fuzz in spite of the name.

Anything with a discontinuity is a fuzz and has a region of linear operation for signals below the clipping level.

But the FuzzFace does clip- that's how it makes the fuzz!

I think I would divide distortion into three types:
1) Simple harmonic distortion caused by a device operating over a large (slightly non-linear) range, but not clipping;
2) Amplifiers that are driven to clipping;
3) Diode clipping, which does not necessarily require any gain or amplification at all.

[midwayfair]

Just for fun ....

Anything that alters the waveform is distortion.

Tremolo, phasing, etc. -- all distortion.



I'll see myself out.

[Bill Mountain]

Because of the curvature of the output/input (i.e. variable gain) they always distort at every signal level but the distortion becomes greater with amplitude as you use more of the gain variation.  The Fuzz Face is the most familiar of this type.  I would consider this to be a pure distortion rather than a fuzz in spite of the name.

Anything with a discontinuity is a fuzz and has a region of linear operation for signals below the clipping level.

But the FuzzFace does clip- that's how it makes the fuzz!

I think I would divide distortion into three types:
1) Simple harmonic distortion caused by a device operating over a large (slightly non-linear) range, but not clipping;
2) Amplifiers that are driven to clipping;
3) Diode clipping, which does not necessarily require any gain or amplification at all.

Can you give an example of number one?

[Thecomedian]

Classifications are purely about how they sound, with a little anachronism.

Just for example. There might be crossover, but in general, its the sound each pedal makes that determines what the maker calls it. If someone boxed a fuzz inside a pedal and called it "ultra distortion" and somebody here got their hands on it and started playing it, they'd say "hey, this is a fuzz!".

[R.G.]

Can you give an example of number one?

Vacuum triodes before clipping;
Carbon Comp resistors with signals at or above 70V peak to peak across them;
JFETS with fully bypassed source resistors:
Bipolar devices with fully bypassed emitter resistors;
Diodes with signals less than about 50mV operated in the "knee" where they start to conduct; also LEDs, diode connected bipolars, and for bigger signals diode connected JFETs and MOSFETs
Electrolytic caps operated under circumstances where they have significant signal voltage across the cap;
Ferromagnetic core inductors operated near saturation.

That is - practically anything operated without linearizing feedback to smooth out the device's natural curves. Film resistors, wire and bulk foil resistors, non-electrolytic caps, electro caps with only small signal voltages across them, inductors well below saturation or air core are examples of the few things that don't have soft nonlinearity.

[Michael Zartarian]

for the op -

there are many basic types of distortion circuits, but a huge majority of overdrive and distortion circuits are based off the tubescreamer and RAT topologies, respectively. these both use opamps and clipping diodes.  i would venture to guess 75% of the distortion pedals on the market use the op amp technique because it doen't rely nearly as much on mojo components or getting things 'just right' like we are willing to do in DIY land.

i just wrote an article about opamp distortion techniques here:
http://bit.ly/Zrclle

its a very basic overview that is meant for a less technical audience than many of the experts here.  i hope it is helpful in answering at least some of your question -

(my first post - i cant believe it since I've read this forum and it's earlier incarnations since before the dotcom bubble)

Mike