Turbulent K Model of Ute-Like Musical Instrument Sound Production
Turbulent K Model of Ute-Like Musical Instrument Sound Production
Rolf Bader
Summary. The sound production of flute-like musical instruments like the transvers
flute is governed by the coupling between the mouth or embouchure hole in which the
flute player blows and the flute tube. Here the flute tubes eigenfrequencies forces
the self-sustained oscillation of the generator region at the blowing hole into the
tubes resonance frequencies. This paper supposes an explanation for this behaviour.
Experiments show a very small amount of energy supplied by the players blowing
actually getting into the tube of about 3.5%. So most of the air flow is blown into the
room outside the flute. The modelling of the flutes presented here shows a turbulent
description of the process as consistent with the experimental findings. The flute
tube, which forces the flow in its direction leads to a large directional change of the
flow, which results in a strong turbulent viscous damping of the system. So there is
strong evidence, that in the nonlinear coupled flute system of blowing and tube the
tube forces the blowing system in the tubes eigenfrequencies because the tubes air
column is much less damped than the generator region at the soundhole.
Fig. 1. Schematic view of a flute-like instrument. The air jet out of the players
mouth hits the embouchure lip or cut. The embouchure wall height is enlarged
compared with the tubes thickness. The tube is closed at the left and open on the
right end (with a boundary condition of zero pressure). The cavity on the left end
of the tube plays an important role in the tuning of the overblown pitches of the
flute (which is not a topic of the present paper).
If the correct pressure is not applied with the flute, no sound is produced
[Coltman 1968] [Coltman 1969]. This is because the impedance Z of the flute
defined via the blowing pressure p needed to achieve a certain flow v like
p
Z= . (1)
v
So if high pressure is needed to get a certain flow, the tube ’resists’, it has
a high impedance. This impedance of the flute is complex, means there can
occur phase shifts between pressure and flow.
Figure 2 shows the impedance behaviour of the flute-like instruments [Colt-
man 1968] with respect to blowing pressure. As the spiral gets larger, the
pressure increases. It is measured with an artificial driving mechanism at the
open end of the flute, applying different sinusodial frequencies to the tube.
The applied pressure then corresponds to a measured flow in the tube. The
air inside the tube is damped out via an acoustic resistance in the tube (i.e.
small glass capillaries) so that the air column in the tube can not go into
eigenfrequencies. The impedance has a real and an imaginary part. The real
part corresponds to the resistive (positive) or generative (negative) behaviour
of the tube. The imaginary part tell us about the phase shift between pres-
sure and flow. So the only possible operating point of the flute is the largest
negative point on the real axes. Here, the generation is highest. There is no
phase shift between pressure and velocity. So the curve shows, that there is
just one possible pressure region to generate the desired pitch, the player must
blow with this precise pressure to produce the sound. Of course this region
changes a bit for each pitch, so the flutist must change the blowing pressure
while playing melodies.
But despite this generation process, in contrast with organ flute pipes,
where the blowing around a lip produces a self-sustained oscillation [Dequand