Klarinet Archive - Posting 000040.txt from 1997/05

From: Jonathan Cohler <cohler@-----.net>
Subj: Re: Air : amount,speed,presure etc.
Date: Fri, 2 May 1997 18:16:56 -0400

Yossi wrote:

>Does science ( or we the wind-players) know what air parameters influence
>what sound parameters ?
>
>I mean:
>
>Is it the air speed which makes the ff ? or is it the amount of air
>?

This is actually a somewhat complicated question, and the answers given so
far on the list are not really accurate.

First, the quick answer is that perceived loudness of a sound is a function
of the amplitudes and frequencies of all of the components of the sound as
well as the quality of the listener's hearing.

Some very simple and practical consequences of this are that high notes
sound louder than low notes (at the same amplitude). This is why one must
instantaneously reduce the blowing pressure when one makes a slur from a
low note to a high note if one wants to make a seemless and even connection
wherein both notes sound equal.

Here is a bit of the physics of how this all works.

People often confuse sound level and loudness, which are two totally
separate things. Sound level is a measure of the intensity of sound
pressure waves in the air. It is measured on a scale called decibels and
is defined such that 0 decibels (dB) corresponds to a pressure of 20
micropascals or millionths of a newton per square meter. In practice, this
corresponds to an amount of sound that is just barely audible by the
"average" ear in a "quiet" environment when listening to a frequency of
3000 Hz.

Loudness is what a listener perceives. Therefore the scale for measuring
loudness, called "phons", was developed based on averaging empirical
results from tests of people with "normal" hearing. We humans can hear
best between 2000 and 4000 Hz (because the ear canal has a resonance at
approximately 3000 Hz), and our hearing gets less sensitive at both high
and low frequencies.

Therefore a loudness of 20 phons at 1000Hz corresponds to a sound level of
20dB, but 60dB at 16000Hz and 67dB at 31Hz. In other words, the sound
level has to be much more intense at the high and low frequency extremes to
be heard as the same loudness.

Now we add the complication of the cochlear nerve and how we hear. To be
as brief as possible, different areas of the cochlear nerve respond to
different frequency ranges (called critical bands). In each of these
frequency ranges there are "hair cells" that when excited transmit sound to
the brain.

Here's where it gets a bit tricky:

A sound whose components lie all within one critical band uses the same set
of hair cells to tell the brain how loud it is. Therefore, the loudness of
this sound will depend only on the sound level.

When a sound has components in multiple critical bands, however, more hair
cells get in the action and the loudness is equal to the sum of the
loudnesses in each critical band.

What this means in practice on the clarinet is that since most of the
overtones of a tone on the clarinet lie in separate critical bands, as we
add overtones (by blowing harder) the sound gets louder. And,
interestingly, *beyond a mezzo-piano, most of the increase in loudness is
due to the addition of partials and NOT to the increase in amplitude of the
fundamental*.

Therefore, as one gets louder on the clarinet, the tone color also changes
as the higher partials come into play.

By the way, hearing changes as one ages and therefore so does loudness and
tone quality perception. Older folks gradually lose sensitivity at higher
frequencies.

If you're interested in more information on this subject, a very good
reference is "Music, Speech, Audio" by William J. Strong and George R.
Plitnik, published by Soundprint in Provo, Utah (1992).

-----------------------
Jonathan Cohler
cohler@-----.net