On a good day, the limits of human hearing are said to be 20kHz. As we age, that goes down in frequency.
So why, you might ask, do we expect our electronics and loudspeakers to produce frequencies well above our ability to hear them? In the case of PS Audio amplifiers, we strive for 50kHz and above (and our loudspeakers definitely exceed 20kHz).
How’s this all work?
When a stringed instrument (for example) is plucked, it produces a fundamental frequency (the note) and a series of overtones that are multiples of the fundamental frequency (setting the timbre and tone of the instrument). While many of these overtones are outside the range of human hearing, they still affect the audible frequencies that we can hear by something known as active interference.
What’s active interference?
If we’re discussing the output of a loudspeaker, ultrasonic frequencies can air-modulate the audible range to produce non-linear effects. Some are desirable (like a sense of air or openness) and some are not-so-desirable (harsh, odd-order additions).
In an amplifier passing an audio signal that contains both audible frequencies and ultrasonic frequencies, the non-linear characteristics of the amplifier can cause the ultrasonic frequencies to interact with the main signal producing new and unwanted tones. This kind of distortion is typically intermodulation distortion that occurs when two or more frequencies beat together generating a third (or more) new unwanted frequencies.
Intermodulation (IM) is almost always unpleasant to the ear/brain while our standard distortion, Harmonic Distortion, is more tolerable (and sometimes even welcomed if the generated harmonics are even order).
In the same way ultraviolet light (that we cannot see) impacts our skin in ways we can see (sunburn), those frequencies we cannot hear impact what we can hear.
The trick is to make sure the entire package of audible and inaudible frequencies plays nice with each other.