Thermoacoustic Oscillations

Donald Fahey
Wave Motion & Optics
Spring 2006
Prof. Peter Timbie



Abstract
In a time of heightened environmental concerns and increased importance
on efficiency, thermoacoustic standing waves and oscillations provide a
possible solution to many contemporary problems. Elementary standing
wave theory is discussed, as well as the connection between
thermodynamics and acoustical oscillations. Additionally, the benefits of
Sondhauss oscillations and devices for their production are detailed for
standing waves. Finally, insight into applications of thermoacoustic
resonant tubes, such as engines and refrigerators, as well as other exciting
aspects of the topic are covered.








2

Thermoacoustic Oscillations
Sound is of the most natural and familiar phenomena that we experience on a day
to day basis. Over the course of the past century, people have learned to harness sound as
a means of benefiting society. Ultrasonic surgery, noise-canceling systems, and SONAR
are all excellent examples of such ingenuity. In the increasingly environmentally
conscious world we live in, the idea of replacing other needs with less harmful acoustic
counterparts is an attractive one. Refrigeration, electricity, and mechanical energy are all
examples of applications that sound is well-suited for, and they introduce the notion of a
tie between acoustics and thermodynamics. Discussed here are the basic concepts behind
thermoacoustic processes, how they have developed, and in what direction their research
is leading us.
The study of thermoacoustics has a rich and interesting history. Lord Rayleigh
was the first to give a thoroughly qualitative description of thermoacoustic effects. In his
work The Theory of Sound, published in 1887, he discussed the ability to generate
temperature differences using acoustic oscillations. The subject remained largely
untouched for over eighty years, until 1969, when Rott began a series of publications that
signaled a revival in thermoacoustic research.
Of the first topics considered, perhaps the most important was the Taconis
oscillation. This phenomenon was discovered when the open end of a gas-filled tube was
immersed in liquid nitrogen and cooled to a cryogenic temperature. Curiously, when the
tube was removed from the coolant, it began to vibrate and sing loudly. In his
3
landmark analysis of this effect, Rott paved the way to the quantitative study of
thermoacoustics.
To begin, we should briefly review the concept of standing waves in columns of
air. Since we are dealing with longitudinal pressure waves in a gas, in order to obtain an
equation for our waves we must solve
2
2
2
2
t x

where
dV
d
V

= ,
which gives
) sin( ) cos( kx t B kx t A + = .
To produce standing waves, two waves (
1
and
2
) must be traveling in opposite
directions with equal amplitudes, the sum of which is
) cos( )] sin( ) cos( [ 2 kx t B t A + = .
This results in nodes of zero displacement spaced every one-half wavelength, and
antinodes of maximum displacement located halfway between the nodes. The number of
nodes within the tube depends on the harmonic of the standing wave and the boundary
conditions. In a tube with one closed end, the boundary conditions dictate a maximum
displacement at the open end, and zero displacement at the closed end. Differing from a
tube with two open ends, it follows that only odd numbered harmonics can form, with the
first (n=1) defined as the fundamental harmonic. In this case, we have

2

n kL = where

2
= k and

2
= v , so
L
vn
2

= ,

4
where L is the length of the tube, and n is the harmonic number. Here we see that the
frequency is dependent on the harmonic number, velocity (in the medium), and tube
length. As one would expect, decreasing the length of the tube decreases the wavelength
at the fundamental harmonic, and thus equates to a higher frequency (for constant
velocity of sound in a gas). Furthermore,

n
L
n
4
= for n = 1, 3, 5,

This becomes useful when considering applications of vibrating air columns, especially
in the area of thermoacoustics.
Thermoacoustics refers to the production of acoustic waves by way of heat
transfer (or the converse, that is, sonically induced heat gradients). This is usually done

Figure 1. The Rijke Tube on the left has two open ends, and a heating element near the base. The
Sondhuass Tube has one closed end, which is heated to produce odd-numbered harmonic standing
waves.

5
by way of a resonance tube, either with two open ends, or one end open and one end
closed. In the case of two open ends, pressure waves are referred to as Rijke oscillations,
and have been studied thoroughly for two centuries. Of more interest to us are
oscillations in a tube with one closed end, known as Sondhauss oscillations. Although
not discussed quantitatively until the latter half of the 20
th
Century, these tend to be more
efficient and industrially applicable waves.
As stated before, Sondhauss oscillations occur exclusively in tubes that have one
or both ends closed. Heat is applied to the closed end of the tube. As the temperature
increases, heat is converted to kinetic energy in the gas molecules at the closed end of the
tube. The gas molecules accelerate towards the cooler end of the tube, thereby creating
an area of relative low pressure in the heated end. As these molecules cool off, other gas
molecules accelerate towards the hot end to fill in the area of low pressure. These
molecules are then heated, and the process begins again. This completes one cycle of a
thermal oscillation, and is analogous to the Stirling Cycle.
The acceleration and deceleration of the gas molecules with respect to time maps
to a sinusoid. Essentially, the result is a self-sustained series of longitudinal sinusoidal
air pressure oscillations. Within the tube, the pressure wave fronts are approximately
planar, whereas minimal toroidal behavior occurs in the vicinity of the mouth of the tube.
The predominant waveform outside of the tubes mouth is spherical. It comes as no
surprise that these traits are equivalent to those of sound waves, and in most cases audible
sound is a direct byproduct of the entire process.
The term thermoacoustic oscillation refers to the thermal origins of sound
produced by pressure oscillations. Experimentally, its incredibly simple to build an
6
apparatus capable of producing such oscillations. Clearly, a pipe or tube-like cavity
shape is necessary. A smooth, straight cylindrical interior avoids unwanted eddying or
nonlinear pressure variations that would greatly complicate analysis. Additionally, a
minimum rate of heat application to the closed end of the tube is required. In the
thermoacoustic revival of the 1970s this was about 30 watts per cubic inch.
Perhaps less obvious is the importance in placement of the heating device. The
rate of heat transfer is directly proportional to acoustical velocity. Applying the principle
of superposition, it is to our advantage to place the maximum heat gradient at the point of
maximum gas molecule velocity. In standard air column oscillations, such as in a flute or
Coke bottle, one switches between harmonics by varying the rate at which air pressure is
introduced to the system. That is to say, if one blows harder, higher order harmonics
occur. Similarly, at minimal rates of heat application, the fundamental harmonic is
excited. For the tubes commonly used in laboratory experiments, these frequencies are at
the lower end of the audible range. To adjust for this, a greater temperature difference is
applied, often times resulting in the second harmonic (n = 3). The maximum gas
molecule velocity in such systems appears approximately one third of the tubes length
from the closed end, therefore, this is where the heating element should be placed for the
most efficient arrangement.
One of the first major modifications to the basic Sondhauss tube design was the
introduction of a bundle of small pyrex tubes to the system. These tubes are positioned
right next to the heater on the side closest to the open end, and serve three specific
purposes. First, they provide a small contribution to the system by acting as a regenerator.
In these circumstances, the regenerator works by having each small tube perform the
7
function of bottleneck for the traffic of gas molecules trying to pass through. Under
pressure, the gas is forced through, gaining velocity in the process to accommodate for
the smaller cross-section. Next, the bundle operates as an insulator between the hot and
cold regions of the tube. This creates a steeper temperature gradient, thereby improving
the systems efficiency. Lastly, the bundle acts as an acoustic impedance, creating a
phase shift. It fulfills the aforementioned stipulation that gas motion, pressure, and
temperature waves all be temporally in phase. Other less notable changes have been
made to modern Sondhauss tubes, although their effects on the basic process are minimal,
and are usually designed to fit the needs of a specific experiment.


Figure 2. A modified Sondhauss Resonating Tube. Note the position of the glass tube bundle and
heating element.

Throughout the 1980s and 90s, many researchers furthered our understanding of
Sondhauss resonant effects by improving thermoacoustic efficiency through the
introduction of regenerators, couples, and modified tube structures. Of particular interest
to present research was Hoflers invention of a standing-wave thermoacoustic refrigerator.
Here, the time-phased superposition of gas motion, temperature, and heat transfer
8
oscillations in the sound waves work efficiently at absorbing heat from loads at low
temperatures, and transferring it to a sink at higher temperatures. This refrigeration
process is a hotly studied research area today, and holds many potential applications in
the power, appliance, and space industries.
Similar to Hoflers elegant refrigerator is the harnessing of Sondhauss oscillations
as a means to power an engine. There are two basic types if acoustic engines: standing
wave and traveling wave. Those that use standing waves (contrary to the analogy made
above) are natural engines, and require modifications to the walls of the tubes interior.
In 1979, Ceperley proved that the desired Stirling engine is of the traveling wave class,
triggering the development of theory in traveling thermoacoustics. Although such
engines rival the efficiency of those found in standard automobiles, they are far too bulky,
and for the most part lack the ability of being incorporated into other technologies. Still,
primitive versions of thermoacoustic Stirling engines have already been implemented in
refrigerators for some modern-day spacecraft.
Undoubtedly, Hoflers idea of a refrigerator with no moving parts, along with
Swift and others development of noncombustion engines are reasons for excitement in
the realm of thermoacoustics. Today, we are on an upstroke of research and results, and
each year new findings bring us closer to replacing existing technologies with greener
and more efficient ones. Soon we may all be cooling our houses with thermoacoustic
refrigerators, and powering our cars on old Don Henley tapes.



9


Bibliography

Backhaus, S., G. W. Swift. New Varieties of Thermoacoustic Engines, 9
th
International
Congress on Sound and Vibration, July 2002

Feldman, K.T., R. L. Carter. A study of heat driven pressure oscillations in a gas,
Journal of Heat Transfer, v 92, n 3, Aug. 1970, p 536-41.

Galiullin. R. G., I.P Revva, G. G. Khalimov. Thermoacoustic effect in a resonant
semiopen tube, Journal of Engineering Physics, v 43, n 4, Oct. 1982, p 1126-32.

Hecht, E. Optics. 3
rd
Edition, Addison-Wesley, 1998.

Swift, G. W. Thermoacoustic Engines, Journal of the Acoustical Society of America, v
84, n 4, Oct. 1988, p 1148-80.

Swift, G. W. What is thermoacoustics? Los Alamos National Laboratory Website,
http://www.lanl.gov/thermoacoustics/ehistory.pdf

Wheatley, J., T. Hofler, G. W. Swift, A. Migliori. Understanding some simple
phenomena in thermoacoustics with application to acoustical heat engines,
American Journal of Physics, v 53, n 2, Feb. 1985, p 147-62.