Symmetrical or asymmetrical diodes for clipping?

[aron]

Just wondering if you like symmetry or asymmetry with your clipping diodes.

[WGTP]

How about both.  I seem to prefer symmetrical for lows because they are less wooley and a little more undercontrol, but for highs I like asymmetrical to add even order harmonics and sweeten/fatten them up.   

[soulsonic]

I was really surprised how cool different colored LED combinations can sound. I did it once just because I didn't have two of the same color, and it sounded great. Later on I realized that they don't clip at equal voltages, so it's slightly asymmetrical. I like the yellow and green combo.

I've also done the usual asymmetrical with two diodes on one side and one on the other - I like it.

Overall, I've ended up using asymmetrical clippers so much that it's kinda become my default style. The mismatched LED thing is a little less asymmetrical because they're not way off from each other, and I think I prefer that one the most.

[Gus]

for opamp types like the TS and rat, symmetry.  I let the amp generate its sound due to the preamp and output stages.
  You don't alway know how many inverting stages are in line.

[twangquack]

How about both.  I seem to prefer symmetrical for lows because they are less wooley and a little more undercontrol, but for highs I like asymmetrical to add even order harmonics and sweeten/fatten them up.   

Yeah, I'm not sure I like using using just symmetrical or asymmetrical. In the last few pedals using diodes for clipping (including variants such as reverse-junction MOSFETs, JFETs, LEDs, etc.) I like having at least one symmetrical arrangement for lead playing. But just-a-bit-asymmetrical is also cool. I changed out one switchable option (one side) of three 1N34A and replaced them with a yellow LED, so that the "both switches all the way UP" setting kicks in a red LED opposed to a yellow one ... I think there's an approx. 200mV difference in these two particular LEDs. Sounds great in this TS clone!

But an asymmetrical option, especially using the softer clipping stuff like 1N34A diodes and reverse-junction MOSFETs is great for rhythm playing, especially at low to moderate gain levels. I'm glad I bought some 1N5818 diodes (from Small Bear) to have a lower forward voltage silicon diode option, besides digging into my usual pile of 1N34A diodes, to both "direct" (reverse-junction MOSFET routine) and to create at least a moderate asymmetry in certain switchable options. I'm really enjoying one switchable option of a BS170 (reverse junction, using a 1N34A as traffic cop, so to speak) opposed to a red LED ... and with an LF353 as opamp. I tweaked the tone just right and was getting these great Billy Gibbons-style pinch harmonics!

Gus is right-on about the rats and the TS clones in that if you had to go with just one option (sym vs. asym), the symmetrical arrangement is probably the most versatile overall for these pedals. I like a certain amount of the slight wooliness of asym for some lead playing, usually at lower gain levels, but the sym arrangements with much higher gain settings are really in your face, and I like that!

soulsonic's point about the "mild asym" is also something important ... I'd like to try the yellow and green combo. The red and yellow combo I just added to a TS-808 clone is really great at higher gain levels and I'm finding myself playing at much higher drive levels than I used to, in the past (and don't my neighbors just love me for that) ...

[brett]

IMO it is really sweet to have asymetric clipping in your pedal for those times when the amp isn't clipping much (ie. most of the time with clean amps).
Is it inconsistent to like symetric clipping in pedals, and asymetric clipping in tube amps?  I think it is, a bit.

[WGTP]

Another thought.  Into a clean amp, asymmetrical for the extra harmonics.  Into an ovedriven tube amp, symmetrical.   Single coil, asymmetrical for fatness, Humbucker symmetrica for more definition.  :icon_cool

[mac]

maybe a "depends on the project/amp/setting/etc" round button
that would be my vote.

when i use asym diodes i like to have closer voltage drops on both directions, ie, 4148 (->) and 2xGerms (<-)

mac

[darron]

symmetrical mostly for me. i use humbuckers. try clipping with just the one diode to get an idea of how asymetrical sounds at its most extreme. it's got a much more attacking distortion with a fast decay. i do like that notion of asymmetrical for rhythm rather than lead, because this is sometimes where i just wants some powerful 'tuned noise'.

frequencies being dealt with does also play a bit part i think.

i know people out there might thing this thread is too subjective, but it's nice to get some opinions from people too.

[Mark Hammer]

Wish I could answer the survey, but I can't, because my choice/response is not in there: both.  It's important to understand that clipping, while ostensibly a fixed phenomenon (i.e., signal hits this forward voltage and the diodes conduct), is really a variable and graded phenomenon since only some parts/moments of the signal are ever at or above clipping threshold (we use guitars not oscillators).  How one picks, and signal conditioning of various types that alter what percentage of the time the signal will be at threshold, take care of the rest.  As such, for me injecting asymmetry into a clipping configuration simply adds one more parameter to vary.  It's not the same thing as asking do you like flanging or phasing, wah or auto-wah?  Symmetrical and asymmetrical clipping are not different from each other in that particular way.

[Gus]

Take a look at the output of a TS with different diode setups.  The circuit is Cap coupled so the signal moves around a "center" so how asymmetrical does it have to be to asymmetrical?

Now the real thing to do would the be feed a stock TS sine waves  at say  720Hz 360Hz and 1440Hz at different input levels and gain settings(low middle max) and diode settings( a 1 and 2  1n4148 setup).  They do a FFT and look at the harmonics.  Been meaning to do stuff like this just got to make time.

[mac]

It's important to understand that clipping, while ostensibly a fixed phenomenon (i.e., signal hits this forward voltage and the diodes conduct), is really a variable and graded phenomenon since only some parts/moments of the signal are ever at or above clipping threshold (we use guitars not oscillators)

To do list: LFO controlled threshold.

mac

[the recluse]

How about both.  I seem to prefer symmetrical for lows because they are less wooley and a little more undercontrol, but for highs I like asymmetrical to add even order harmonics and sweeten/fatten them up.   

Has anyone implemented this (is it even possible) in the same fashion as Anderton's Quadrafuzz where the signal is split into multiple differently eq'd bands and then clipped differently and mixed back together?  Even if it wasn't necessarily into as many bands as the QF, a Hi-Lo split with different clipping sounds like it might be cool in theory.

[mac]

The most simple arrangement I can imagine now is:
3 filters, high, middle and bass,
mxr+ like distortion for each band with different diodes,
mixing all into an opamp buffer.

mac

[Mark Hammer]

How about both.  I seem to prefer symmetrical for lows because they are less wooley and a little more undercontrol, but for highs I like asymmetrical to add even order harmonics and sweeten/fatten them up.   

Has anyone implemented this (is it even possible) in the same fashion as Anderton's Quadrafuzz where the signal is split into multiple differently eq'd bands and then clipped differently and mixed back together?  Even if it wasn't necessarily into as many bands as the QF, a Hi-Lo split with different clipping sounds like it might be cool in theory.

One of the quiks one always has to take into account is that the overall average amplitude of higher pitch notes is smaller than that of lower pitched notes.  Indeed, this is the very basis of the tone-shaping in the Tube Screamer and so many other similar pedals aiming for a "consistent, smooth" clip.  If bass notes are generally louder than treble ones, and louder notes get clipped more than softer ones, then you can even out the clipping by taming the level of bass notes on the front end.

Extending this a little further, if the range of notes/frequencies seen by each "clipping module" is restricted by a steep bandpass filter then the range of levels seen by the clipping circuit is more consistent.  In theory, at least, if you applied symmetrical to one and asymmetrical to another band, you would be more likely to hear the "pure" effects of that type of clipping.

Even so, what one hears from symmetrical/asymmetrical arrangements is still dependent on picking and level setting, and there are many cases where what you hear from each would be minimally different, or would provide minimal returns for the effort invested.

[Johan]

I like to build up several stages, inverting, low gain, completely assymetrical ( just one diod ).
this way, if you look at the output with an oscilloscope, it looks symetrical, but since every stage brings its own harmonics and every stage after adds to that, it givs a completly different feel than single stage clippers...so miniumum two inverting assymetrical stages giving a symmetrical output...
j

[mac]

Even so, what one hears from symmetrical/asymmetrical arrangements is still dependent on picking and level setting, and there are many cases where what you hear from each would be minimally different, or would provide minimal returns for the effort invested.

What if the amount of distortion applied to each filter is different?

mac

[JDoyle]

Even so, what one hears from symmetrical/asymmetrical arrangements is still dependent on picking and level setting, and there are many cases where what you hear from each would be minimally different, or would provide minimal returns for the effort invested.

This is true in the case of a diode 'clamp' like in the Dist+ and the Rat, but in the case of a TS or other 'absolute value' feedback clipping circuits, the gain of the circuit drives the signal to nearly always clip and picking strength doesn't matter as much.

Maybe this is a distinction that needs to be in the poll? asym/sym for feedback circuits and asym/sym for diode to ground circuits?

I know that Aron prefers the dynamics and feel in the diode to ground arrangement, while I prefer the compression and feel of the feedback type circuit... I'm sure this is a very subjective choice to each builder...

Just a thought,

Jay Doyle

[Mark Hammer]

Remember that the diodes set an absolute threshold for clipping.  Once the signal is above XX millivolts or volts, the diode/s conduct.  How close the signal is to that critical voltage is a function of:
- the actual absolute threshold set by the number and type of diodes
- how much gain is applied to the signal along the way
- how hot the input signal is to begin with
- any passive signal loss along the way

If, for example, notes produced on the first 5 frets of your two lowest strings are generally at a lower amplitude than notes produced on frets 5-through 12 of your two highest strings, then if the filtering produces the same degree of attenuation or boost for each of those ranges, AND the gain in any subsequent clipping stage is exactly the same for both bands, you will get more clipping for the lower band than for the higher one.  You can offset that, however, by increasing the gain for the higher band.  What is key is maintaining the same overall average amplitude, relative to the clipping threshold.  This is a part of what I have blathered about for years now as "proximity to clip" - the difference between any part of the signal and the clipping threshold.

Part of what likely limited both the popularity and sales of the Quadrafuzz was the fact that, for all of Craig Anderton's genius, it ignored the amplitude differences between strings/notes and provided no means of adjusting or precompensating the sensitivity of each clipping band, only adjustment of their mix/blend after the clipping.  The result was that it probably never sounding aggressive enough for some folks.

So, if you were to split the guitar signal up into two or more bands, and feed each band to a different clipping stage, you would need to provide some means to adjust how much that stage clips by adjusting how close the filter output signals are to the clipping point.  Use of a 2+1 diode complement sets a higher threshold than a 1+1 complement does.  If it's 2+1 LEDs, that difference is much greater than a trio-vs-duo of silicon or germanium diodes.  Of course how you set it depends on what you want of the asymmetry.  Some folks want clipping from both half-cycles of the signal, just more from one side than the other.  Others may want to have clipping from only one side of the signal and none from the other.  Something like that would of course be achieved by using one or more diodes going in one direction and none in the other direction.  I realize many are more familiar with the idea of 2+1 vs 1+1, but 2+0 is also asymmetrical clipping.

Actually, now that I think of it, here's an idea.  Imagine you have two successive noninverting op-amps, with diodes in the feedback loop and gain set by a variable resistance to ground from the inverting pin.  So, the diode location of a TS, but the gain adjustment of a Dist+/DOD250.  Op-amp A has a pair of series Si diodes (1N914/4148) going in one direction and none in the other direction.  Opamp B has a similar pair of diodes, but going in the opposite direction.  The gain-setting resistance for each stage (and gain goes up as that resistance gets smaller) goes to a single pot whose wiper goes to ground.  If you turn the pot anywhere off the midpoint (assuming it's a linear pot), the gain of one opamp gets higher and the gain of the other gets lower.  So, you can set their gains in complementary fashion to produce the same "proximity to clip" for each half-cycle of the signal, or make one stage more sensitive and "clippy" than the other.

Of course since making them equal means aiming for less than maximum gain from either one, you'd need an additional front end to provide boost.  And since gain is multiplicative  (e.g., a gain of 100 in the first op-amp, and a gain of 100 in the second means that the 2nd pair of diodes would see a signal theoretically amplified 10,000 times), you'd need to provide some attenuation between stages, and keep stage B's gain modest.

This is true in the case of a diode 'clamp' like in the Dist+ and the Rat, but in the case of a TS or other 'absolute value' feedback clipping circuits, the gain of the circuit drives the signal to nearly always clip and picking strength doesn't matter as much.

Maybe this is a distinction that needs to be in the poll? asym/sym for feedback circuits and asym/sym for diode to ground circuits?

Now this is exactly why I'm glad you're back.

So, if I am following you correctly, because forward-conducting diodes in the feedback loop dynamically adjust the gain, which in turn changes what sort of feedback signal there IS to feed back, there is a sort of self-regulation in place, not unlike the postgain-feedback arrangement of something like a Dynacomp.  In contrast, with a Dist+ type pair of diodes to ground, the gain is the gain, and the threshold is the threshold, and the two do not interact.  If so, then maybe those folks who are futzing around with this or that diode complement in their Tube Screamers ought to have a switch to select between a pair of diodes in the feedback loop or a pair of diodes in parallel with the cap to ground after than op-amp's output.  Conceivably that would switch between less and more spongy/compressed feels.

[WGTP]

I guess you could have 2 different diodes with the same clipping threshold voltage, but different turn on rates (I think it is what it is called).  For example a soft mosfet and a SI.  If the mosfet measures as a diode at 1.4v, that would be about 2 1N4818's.  On Jack's CD, he indicates that different diodes have different harmonic content in their clipping effect.   I think this would be similar to various degrees of asymmetry.