Sound of harmonic distortion

One of the basic and most common ways to objectify the sound qualities of an audio device was always its harmonic distortion. Harmonic distortion can be given by its total level (THD), or by the plot that shows its actual spectral content. But it remains a controversial topic. Some claim it is audible yet at a very low level, but others say it is inaudible even when it reaches quite high amounts. Sometimes, harmonic distortion is discussed with regard to its more or less audible components, and their more or less “dis-harmonic” effects, subjectively speaking. But no theory about it appeared widely accepted so far. So whether you believe it is the be-all and end-all of audio equipment quality, or nothing to really care about, you can easily find supporting views.

That is the starting point here. But for now, I will skip discussing further this topic, and instead, I will cut a long story as short and effective as possible, and move it to a rather practical level.

So, below you will find several examples of the distorted sinewave. You will first find pure sinewave, and then the same sinewave distorted by different harmonic components. I used 500 Hz basic frequency instead of the usual 1 kHz, so high order harmonics remained inside the audio band.

Harmonics audibility will depend somewhat on your listening setup, and on your hearing abilities and skills. With a decent audio setup and average hearing, you may hear 5% second harmonic, but you will hardly hear it at 1%. Harmonics’ audibility will increase as their order increases, and probably there will be no matter if it is odd or even order. So fourth and fifth harmonics 1% definitely won’t pass unnoticed. Then it takes to go to lower levels. Thus seventh harmonic (according to some, it is the critical one) is clearly audible at 1%, but it will get almost inaudible at 0.1%, and probably below the threshold when lowered further to 0.05%. The same 0.1% and 0.05% examples are also given for the fifteenth harmonic, which probably should go below 0.05% to get inaudible.

But again, it is up to you to listen to the examples, and check what you hear, and get your own sense of this.

(If your web browser or audio hardware does not support 24-bit playback, please scroll down.)

As for my bottom line, it is rather one of a general kind:

Any attempt to get across “objective” knowledge is faced with the impossibility to avoid own perspective. The point of view is, and remains, a fact of view. Still, this doesn’t make any knowledge impossible or invalid. Nor do all views are equal. The problem with the point of view is not in having one, but in being unaware of it, or not acting accordingly. And being aware of it also helps possibly correcting it, and moving to a better position.

The other topics, like “Does harmonic distortion have smearing effects even when it is not directly audible?”, I will leave for some other day.

Added on 25.05.2015:

In addition to the above examples, which are 24-bit files that play nicely in web browsers such as Chrome, here are the same files reduced down to 16 bits. Unfortunately, not all browsers support all bit depths, for instance, playback in Firefox is still limited to 16 bit. The bit depth of the files below is reduced by using dither and noise shaping, which are very useful and effective in this case, so I believe the conclusions will be the same with both the above and these examples.

Scroll to Top