MARCH 1970

TECHNICAL NOTES

571

Batdorf, S. B., "A Simplified Method of Elastic Stability

Analysis for Thin Cylindrical Shells/' Rept. 874, 1947, NACA,
pp. 285-309.
9
Yamaki, N. and Kodama, S., "Buckling of Circular Cylindrical Shells under Axial Compression," to be published in the
Reports of the Institute of High Speed Mechanics, Tohoku
Univ., Japan.
10
Baruch, M., Harari, O., and Singer, J., "Addendum to TAE
Report 76, Low Buckling Loads of Axially Compressed Conical
Shells," March 1969, Technion Research and Development
Foundation, Haifa, Israel.

Downloaded by UNIVERSITY OF TEXAS LIBRARIES - AUSTIN on December 16, 2015 | http://arc.aiaa.org | DOI: 10.2514/3.5710

ROBERT F. MCALEVY III*

AND
ALEX PAVLAKf

Stevens Institute of Technology,

Hoboken, N. J.
Introduction

N 1916 Hartmann1 discovered that under appropriate

conditions intense noise emanated from, and pressure
oscillations were produced in, shallow circular-cylindrical
cavities of uniform cross-sectional area placed in sonic airjets.
Now known as the Hartmann whistle, this device has been
subjected to extensive investigation directed towards improvement of its acoustic efficiency2 (Fig. la).
In 1954 Sprenger3 discovered that under appropriate conditions a very slender Hartmenn whistle, with a length-todiameter ratio, l/d, of 34 in contrast to a nominal l/d of about
4, exhibited intense heating at the closed end. Now known
as the resonance tube, (Figure Ib), this device is a less
efficient noise generator than the Hartmann whistle, but is of
interest due to its heating characteristics. Sprenger recorded
average resonance tube end-wall temperatures as high as
840F with clean air and 1830F when solder particles were
accidentally introduced into the driving airstream. The
generation of temperatures several times greater than the freestream stagnation temperature by a simple passive device
appears to be unique.
It is believed that the underlying aerodynamic resonance
phenomenon might contribute, under certain conditions, to
enhanced meteorite "pitting" and ablation of heat shields on
atmospheric re-entry vehicles, as well as thermally induced
structural failure in a number of devices of technological
interest, e.g., pneumatic systems. Further, it represents a
potential explosion hazard, as has been recognized in the
past.4 Simple laboratory experiments illustrate graphically
this aspect. For example, a small slug of wood placed in a
resonance tube will ignite and burn vigorously in a few
seconds.3-4 Indeed, there has been an attempt to harness
this explosive potential in a passive ignition system for hydrogen-oxygen rocket engines.5
All previous investigations of the phenomenon known to the
authors were restricted to resonance tubes of uniform cross
Received September 8, 1969; revision received October 27,
1969. Mechanical Engineering Department funds partially
supported this work. The authors would like to acknowledge
the help of A. Cerkanowicz who developed much of the resonance tube testing facility and discussed with them on several occasions the subject work.
* Director, Combustion Laboratory. Member AIAA.
I Graduate Student; now at MSD, General Electric, King of
Prussia, Pa.

WHISTLE

AIR JET

CYLINDRICAL

Fig. 1

Tapered Resonance Tubes:

Some Experiments

HARTMANN

TAPERED

RESONANCE TUBE

RESONANCE TUBE

Schematic of Hartmann whistle and resonance

tubes.

section along the entire length. By contrast, the subject

study dealt primarily with tubes that were unique by virtue
of having cross-sectional area distributions which decreased
continuously from the open mouth to zero area at the closed
end, i.e., tapered tubes (Fig. Ic). In practical situations that
hold explosion hazards, such as incipient failure of pneumatic
system seals, it is believed that tapered resonance tubes
approximate the actual geometries better than do the conventional tubes previously studied.
Aerodynamic Phenomena of the Uniform Cross
Section Resonance Tube

Instability has been found to be excited by sonic and supersonic airjets6 and by subsonic and supersonic wind-tunnel
flows. The most widely used method of excitation for experimental investigation has been sonic airjets.
The first realistic description of the fully developed flow
within conventional resonance tubes was produced by Thompson.6 By using a correctly expanded supersonic airjet he
produced experimentally well-defined boundary conditions at
the mouth, which enabled him to construct a wave diagram
for the internal flowfield. This was well supported by
experimentally observed pressure histories and flow patterns
within the tube.
Thompson and others observed that the conventional tubes
resonated at approximately the acoustic (simple organ pipe)
frequency based on the freestream stagnation temperature.
During part of the cycle some of the airjet gas is ingested,
and much of the same gas is expelled during the remainder of
the cycle. Thus, a slug of indigenous gas is trapped near the
end wall for many cycles.
On every cycle a system of shock waves and expansion
waves of nonuniform strength transit the indigenous slug.
As the local entropy increase during compression is greater
than the local decrease during expansion, it follows that the
net thermal effect of a cycle is to produce an incremental increase of local temperature. Since resonance tubes operate
at several hundred cps, even a small temperature increment
per cycle can result in rapid heat-up of the indigenous gas;
in extreme cases reaching temperatures far in excess of 1000F
within several seconds. No valid means of predicting resonance tube thermal effects are known to the authors. For
example, not even the maximum temperature theoretically
possible in the adiabatic situation, i.e., no heat loss to the
surroundings from the tube, can be realistically predicted.8
Description of the Experiment

A variety of combinations of airjet velocity, ratio of airjet

nozzle diameter to resonance tube diameter, resonance tube
l/d, and separation between airjet nozzle and tube mouth have
been found to produce the aerodynamic resonance of interest.3 As the thermal effects were of prime interest in the
present study, a configuration was selected that maximized
it. That is, an airjet, produced by expanding from stagnation
conditions of 70 psig and 70F through a converging nozzle of
Y in. diam to the atmosphere, was used for driving axially
aligned tubes of $ in. diam area located I j in. away. The
tubes were all about 10j in. long. (Experimental details are

Fig. 2 Shadowgraphs of 2-dimensional tapered resonance

tube operation (vertical shadow cast by external structural
member).

Downloaded by UNIVERSITY OF TEXAS LIBRARIES - AUSTIN on December 16, 2015 | http://arc.aiaa.org | DOI: 10.2514/3.5710

VOL. 8, NO. 3

AIAA JOURNAL

572

presented in Ref. 9.) A rectangular cross section was employed for photographic observations in both the tapered,
i.e., wedge, and conventional configurations. Also, a circular cross section was employed in both the tapered, i.e.,
conical, and conventional configurations used in studies of
thermal effects.
The data obtained during this operation included: pressure
histories obtained by means of rapid-response-time piezoelectric transducers; experimental wave diagrams produced
from an amalgam of individual shadowgraphs of the internal
flow taken at successive times in the resonance period (a
variable delay circuit between the pressure transducer output
and shadowgraph spark-light permitted efficient gathering of
these data), spectral analysis of noise generated, temperature
histories by means of thermocouples and rapid-response-time,
thin-film, resistance thermometers, and ignition characteristics of some organic materials inserted in the tube.
Experimental Results

The following is a summary of the experimental results:

1) Conditions required to produce resonance in tapered
tubes were approximately the same as in conventional tubes,
and the exterior flowfields were very similar.
2) Tapered tubes produced slightly higher noise levels than
conventional tubes of equal length (142 db vs 136 db) but they
resonated at a 50% greater frequency.
3) Shadowgraphs taken during the operation of the wedgeshaped tube indicate that the leading shock wave formed during the inflow portion of the cycle was curved (Fig. 2a); and
also that turbulence was generated by passage of the exiting
shock wave (Fig. 2b).
4) For the wedge-shaped tube, Fig. 3 shows that the inward running shock wave (depicted in Fig. 2a) increases in
velocity with decreasing tube area, but that the shock wave
that runs out the tube (Fig. 2b) remains at constant velocity
even though the tube area increases. (Expansion waves are
present during tube operation, but are not depicted in Fig. 3.)

Fig. 3 Tapered-tube experimental shock-wave diagram:

fraction of total resonance period vs fraction of tube
length.

5) For tubes of uniform cross section, the endwall surface

temperature history was similar to the endwall pressure
history.
6) The mean temperatures near the endwall of both
conical and conventional resonance tubes (constructed from
heavy-wall, in. f o.d., Pyrex tubing), detected by means of ^
in. diam. thermocouples inserted through their mouths were:
1050F for the conventional tube when the end was protected
from the cooling action of the airjet blowing along the outside of the tube, 1340F before the unprotected conical tube
softened, expanded, and burst. (Pyrex softens at 12001300F.10)
7) Solid blocks of wood and nitrate-ester rocket propellants were easily ignited within conventional resonance
tubes. Other organic materials, in the form of liquid or loose
powder, were blown out of conventional tubes before ignition
could take place. However, tapered tubes tend to trap
powders at the tip. And polystyrene powder, which was
found to ignite readily and burn in the open atmosphere, was
found to ignite readily and burn spectacularly in a tapered
resonance tube. These results are presented in detail in Ref.

Conclusions

Similarity in over-all behavior of resonance characteristics

for all tubes tested, both conventional and tapered, demonstrates that internal geometry is not an important factor in
either initiating or sustaining the phenomenon. As had been
suggested previously,6 events that take place near the tube
mouth probably are of critical importance.
Both the observed curvature of the inward propagating
shock, and the remarkable independence of outward propagating shock velocity on changing cross section area (Fig.
3), suggest that one-dimensional models are probably too
simplistic to describe the process realistically.
Thermal and other effects are intensified by tapering
resonance tubes, and consequently, the explosion hazard
associated with naturally occurring geometries is greater than
that indicated by experience with conventional resonance
tubes. Ignition of various materials in resonance tubes
demonstrates the reality of this hazard.
References
1

Hartmann, J., "On the Production of Acoustic Waves By

Means Of An Air-Jet Of A Velocity Exceeding That Of Sound,"
The Philosophical Magazine, Vol. 11,1931, pp. 926-948.
2
Hartmann, J., "Construction, Performance, And Design
Of The Acoustic AirJet Generator," Journal of Scientific Instruments, Vol. 16,1939, pp. 140-149.
3
Sprenger, M., "Ueber Thermische Effekte Bei Rezonanzrohen," Mitteilungen aus dem Institute fur Aerodynamik au der
E.T.H. Zurich,Vo\. 21,1954, pp. 18-35.
4
Cerkanowicz, A. E., "Resonance Tube Study," Research
Rept. ME 400 and ME 401, Nov. 1966, Stevens Institute of
Technology, Hoboken, N. J.
5
Conrad, E. E. and Pavli, A. J., "Resonance-Tube Igniter
For Hydrogen-Oxygen Rocket Engines," TM X-1460, June
1967. NASA.
6
Thompson, P. A., "Jet Driven Resonance Tube," AIAA
Journal, Vol. 2, No. 7, July 1964, pp. 1230-1233.
7
Vrebalovich, T., "Resonance Tubes In A Supersonic Flowfield," TR-32-378 1962, Jet Propulsion Lab., California Institute of Technology.
8
Shapiro, A. H., "Shock Waves and Dissipation in a Resonance Tube," Journal of the Aerospace Sciences, Vol. 26, 1959,
pp. 684-685.
9
Pavlak, A., "Tapered Resonance Tubes," ME thesis, June
1968. Stevens Institute of Technology, Hoboken, N. J.
10
Baumeister, T., ed., Marks Standard Handbook for Mechanical Engineers, 7th ed., McGraw-Hill, New York, 1967, pp.
6-188.