In this post, I’ll delve into the effects distortion has on sound.
What is Distortion?
Briefly put, distortion refers to any modification of a sound wave. This distortion is commonly achieved through excessive increases in volume, resulting in the introduction of new frequencies which did not previously exist in the original sound wave. These new frequencies can either be harmonic overtones (odd/even overtones) or inharmonic, and therefore not musically related to the original sound.
My Choice of Distortion
To help demonstrate sound distortion, I chose to use Ableton’s built-in audio effect, Amp. Amp is a digital amplifier emulation designed in collaboration with Softube. It was designed to emulate what Ableton calls “timeless tones of seven classic guitar amps.” The seven Amp presets include a ‘Clean’, ‘Rock’ and ‘Bass’ emulation among others. The ‘Clean’ preset is modeled on a 1960 amp used widely by British guitarists of that decade. The ‘Rock’ preset is an emulation of another 60’s amp, a “classic 45 watt amp, perhaps the best known rock amp of all time.” Finally the ‘Bass’ preset is an emulation modeled on what Ableton describes as a “rare PA from the 70’s, which produces good low end fuzz at high volumes.”
I used Ableton’s Operator to produce a 100Hz Sine Wave (G1 on a musical keyboard).
I then used a combination of Ableton’s Spectrum Analyser and Voxengo’s SPAN to take a closer look at the frequencies the distortion was adding to my pure 100Hz tone.
Finally, I used the three presets ‘clean,’ ‘rock,’ and ‘bass’ as mentioned above. All presets were used with their default settings, and the dry/wet ratio was set at a constant 50% for all.
It’s interesting to note at this point, that the sine wave Operator produces is not totally pure (as you might expect from a pure wave) and there exist a clutter of harmonics, especially above the 1 kHz mark. In context, this isn’t of concern though, as these would be inaudible.
Distortion Type 1
Using the ‘Clean’ preset, it’s clear that the distortion has introduced audible harmonics, other than the fundamental 100Hz frequency.
The most prominent of these harmonics, is around the 290Hz mark, or D3 on a keyboard. The harmonics added seem to be uniform. The initial jump from 100Hz to 290Hz, follows a second jump to around 490 Hz and one further jump to around 690 Hz. Starting at the point of 100Hz as a reference of 1, these jumps appear to move from 3, to 5, to 7, to 9 etc, and loosely resemble the odd harmonic structure of a square wave. This relationship is far clearer using Voxengo’s SPAN.
Indeed if you run a square wave through Operator at the same pitch of G1 without distortion, the first 2 major harmonics are at 290Hz and 490Hz. So by adding this distortion, I have in effect created a pseudo square wave.
Distortion Type 2
Returning to my original sine wave and engaging the ‘rock’ preset revealed this spectrum below.
The same frequencies are introduced at around 300 Hz, 500 Hz and 700 Hz as in the previous example, however, this time, there are noticeable frequencies added in between those spaces. Those frequencies exist most notably at 200, 400, 600, 800 and 1000 Hz. These harmonics are all even harmonics from the reference of 100Hz and in addition to the odd harmonics from the first example, combine to create what has begun to resemble a sawtooth wave. If we change the sine wave into a sawtooth wave within Operator, we see a similar spectral display; the odd and even harmonics after 100Hz feature prominently and continue uniformly up through the frequency range.
The major difference between the two distortion emulations so far other than the harmonics produced, is the intensity or loudness of those harmonics. The odd harmonics were present in the first example but after about the 4th of 5th harmonic, didn’t even register on SPAN’s analyser. Contrast this with the analysis using SPAN below and you can clearly see the prominence of the additional harmonics, right through to 20kHz.
Comparing the timbre of these two sounds, the first sound is far thinner than the harmonically-dense sound created using the rock preset.
Distortion Type 3
Finally, using the ‘bass’ preset, we again see the presence of both odd and even harmonics, with the odd harmonics notably peaking through more than the even.
Although the harmonic structure looks similar to the 2nd example, the timbre of the sound is considerably different. The reasoning for this is again best illustrated using SPAN…
The harmonics exist throughout the spectrum, but the density and intensity of these harmonics pales in comparison to the 2nd example, especially in the higher frequency bands. Thus the tone is far deeper/darker, which makes sense considering Ableton named the preset ‘bass.’
That will be all for today; hopefully I helped in some small way illuminate the dark-ish world of distortion.
Davis, G., & Davis, G. D. (1989). The sound reinforcement handbook. Hal Leonard Corporation.
Newell, P. R. (2008). Recording studio design. Taylor & Francis.
White, G., & Louie, G. J. (2005). The audio dictionary: revised and expanded. University of Washington Press.