Tecnologia/Technology

AERODYNAMICS AND AEROACOUSTICS SURVEY FOR A

LOW SPEED SUBSONIC JET OPERATING AT MACH 0.25
A. R. Proenaa,

ABSTRACT

O. de Almeida ,
and R. H. Selfb
a

Universidade Federal de Uberlndia

Faculade de Engenharia Mecnica
Campus Santa Mnica
Uberlndia, Minas Gerais, Brasil
andsproenca@gmail.com
b

University of Southampton

Institute of Sound and Vibr. Research

Highfield, Southampton, SO17 1BJ
Hampshire, UnitedKingdom.

Received: October 29, 2014

Revised: November 30, 2014
Accepted: December 30, 2014

The purpose of this work is to study and characterize, in laboratory, the

aerodynamics of a free jet operating at subsonic regime and identify its
acoustic signature. This study aims to analyze the fundamental role of
turbulent flow structures in the total noise produced at different Mach
numbers. This work is focused at low speed subsonic jets operating at Mach
number 0.25. The research is done by analyzing the data obtained in
experiments using Pitot tube, hot-wire anemometer and acoustic
measurements. This work also describes the experimental procedures for
each step of analysis, as well as the characteristics of jet noise facility. The
data from measurements with Pitot tube is used to study the mean velocity
profiles. The average properties are also analyzed with anemometry system,
likewise used to study the turbulent intensity of eleven axial lines, ranging
from the center line to the edge of the nozzle (lipline). These results are
compared with the literature and is verified the accuracy of hot-wire
anemometers for turbulent intensities lower than 15%. A database with the
sound pressure level as a function of frequency is constructed from
experiments serving as data for further numerical analysis to solve this
problem.
Keywords: subsonic jets, aeroacoustics, aerodynamics, hot-wire
anemometry, pitot tube

NOMENCLATURE

U
Uo

normalized velocity
mean velocity at a specified location, m/s

U max maximum velocity at the nozzle exit, m/s

INTRODUCTION
The noise produced by an aircraft can be heard
as a whole, however it is generated by different parts
and by the interaction of external flow and these
components. According to the aircraft performance,
during each phase of flight, one region or equipment
should contribute more or less to the total noise. In
other words, the aircraft on the ground, while taxing,
on run-up from the jet exhaust, during the take-off,
underneath to departure and arrival paths, over-flying
while in route and landing, produces different noise
not only in terms of amplitude but also in its
composition Figure 1.

Figure 1. Aircraft and noise sources.

Regarding engine noise specifically, many

efforts have been made to understand the noise
produced by such sources and its sections separately.
The turbines and other mechanical noise are the
mainly responsible for the noise produced during the
take-off and climb and in a general point of view the
major sources of aircraft noise. During the approach
the engine noise is also considerable; however, due to
the advances in engine technology, the airframe is
somewhat noisier during this phase. The engine noise
also can be divided into two general categories:
internally generated noise or turbo machinery noise
and externally generated noise or jet noise. The turbo
machinery noise can also be divided from its
components, comprised of fan noise, compressor
noise, combustion or core noise and turbine noise.
Although high bypass-ratio turbofans have
considerable fan noise, the jet noise is responsible for
the majority of engine noise (Almeida, 2009). This
subject is extremely important and is the main
motivation of this work.
The aerodynamic noise is also important and is
caused by the airflow around the aircraft fuselage and
control surfaces. Likewise the first case presented the
noise increases as a function of velocity. The air
density either is an important factor, which within
low altitudes the noise is louder than high altitudes.
Also as the engine noise, different aircraft
components can produce different sources of noise.
Thus, main wings, tail plane, slats, trailing edge flaps,
fuselage, landing gear and wheel bays are the most
important individual noise producers, in special the

Engenharia Trmica (Thermal Engineering), Vol. 13 No. 2 December 2014 p. 33-40

33

Tecnologia/Technology

Proena, et al. Aerodynamics and Aeroacoustics

Total
Fan inlet
Fan exhaust
Core
Turbine
Jet
Airframe

Total
Fan inlet
Fan exhaust
Core
Turbine
Jet
Airframe

high-lift devices and the landing gear (Smith, 1989).

Figure 2 brings information about the noise during
the take-off and approach according to the sources
related to the airframe and engine.

Figure 2. Noise component breakdown at take-off

and approach (Almeida, 2009).
The aircraft systems like the cockpit, cabine
pressurization conditioning systems, Auxiliary Power
Unit (APU) and others also increase the total noise,
although its components are not too much expressive
as the others stated before.
This work is related to jet study and
consequently applied to basic of jet engines or also
called gas turbine engines. The jet engines are present
in most of passenger and military aircraft nowadays.
They can be classified according to type of
compressor (centrifugal, axial and centrifugal-axial
flow) and also the path the air takes through the
engine and how power produced.
The term jet noise is usually taken to mean
the noise of jet-powered aircraft. However, strictly
speaking, it is about only those sources associated
with the mixing process between the exhaust flow of
the engine and the atmosphere, as was said before
during the division in jet and turbo machinery noise.
This mixing noise is the only component generated
during subsonic operations. In super critical jet there
is also a secondary noise source which appears from
shock-waves. The present work looks into subsonic
regimes.
Ways to Study the Problem
Probably the most ancient way of investigating
a subjectthat we turn our attention herein is to
perform controlled experiments. In this case, real
scale experiments often become prohibited since the
costs involved are too high. In a world where things
change so frequently and new products are expected,
experiments with such costs are not interesting to
manufacturers. Scale experiments or reduced models
are very important in this context and the knowledge
about the data and its extrapolations are really useful.
By considering semi-empirical or empirical methods,
34

it can be said that these methods are fast, reliable and

have a long application in industry. However, one of
the greatest problems in these methods is the
restriction or applied considerations used to derivate
them and also the envelope of operating conditions to
be applied. As tailored tools for some specific jet
noise predictions, many times they are restrictive to
certain geometries, velocities, temperature and
pressure ranges. Nevertheless, it is also observed that
companies and organizations are interested in
predicting the noise from jets using methods related
to particular cases, and, sometimes, for just only one
family of aeronautical engines.
Part of the present work was developed jointly
with Embraer S. A., a Brazilian aircraft manufacturer,
by means of two projects: AeronaveSilenciosa: Uma
InvestigaoemAeroacstica, which was funded by
the So Paulo Research Foundation (FAPESP) and
EMBRAER S.A and Desenvolvimento de
soluesaprimoradas,
atravs
de
ensaiosaeroacsticos,
para
o problema
de
rudoexterno de aeronaves - DSA, developed by
Research Increment and Industrial Improvement
Foundation (FIPAI) and EMBRAER S.A., the first
and second phases of the EMBRAERs project,
respectively.
This work is also part of a cooperation project
between the University of Uberlandia (Aeronautical
Engineering Course) and the University of
Southampton, specifically the Institute of Sound and
Vibration Research. The jet aerodynamic and
acoustic results were provided in the ISVRs
laboratory called DOAK Lab. The test matrix was
built up in beginning of August, 2011, and the
measurements were made with the supervision of the
Fluid Dynamic and Acoustics Group FDAG. The
laboratory in question is an anechoic chamber built to
subsonic and supersonic acoustic measurements, this
work also was done to extend the applications of the
DOAK Lab to aerodynamic measurements, what will
be shown in details in sequel.
EXPERIMENTAL METHODOLOGY
The present work proposes a comprehensive
study about the physics of a subsonic jet by
aerodynamic and acoustic surveys aiming to better
understanding the mechanisms of generation and
propagation of noise in aircraft engines.
First of all, it is given a brief description of the
Doak Laboratory, the Rolls Royce UTC facility
located at ISVR, University of Souhthampton.
Information about the flow control and the data
acquisition system will be given in the sequence.
The ISVRs Doak laboratory is a 15m x 7 x 5m
anechoic chamber fully anechoic down to 400 Hz.
The four walls, ceiling and the floor are covered with
wedge type absorbent material. A non-forced exhaust
system is composed by a rectangular collector section
allowing air to pass through into a small secondary

Engenharia Trmica (Thermal Engineering), Vol. 13 No. 2 December 2014 p. 33-40

Tecnologia/Technology

acoustic chamber. It contains two silent cold jet rigs

of 2 and 4 pipe diameter. The larger of the rigs is
located on the centerline of the chamber and is used
primarily for the investigation of full-scale bleed
valves, having an in-line silencer. On the other hand,
the smaller rig utilizes a labyrinth pot silencer and
can be moved along the chamber. They are fed from
two high pressure compressed air (20 bar) from two
storage tanks and the range of velocity available for
testing is Mach 0.2 to 1. Therefore, single jets
measurements can be performed on flow regimes
characteristic of civil aircraft. Both polar and a
transversable azimuthal array of microphones can be
used to give a complete three-dimensional sound
field. Next figure (Fig. 3) shows a picture of the
anechoic chamber when the 4 rig was being used to
bleed valve investigation.

Proena, et al. Aerodynamics and Aeroacoustics

The present work was carried out with the 2

rig. The labyrinth silencer was placed in the middle
of the chamber, just after the larger rig. The cold,
subsonic jet was operated from the nozzle showed
above. In order to run aerodynamics measurements a
traverse system was placed inside the anechoic room.
Details of these changes can be found in the next
figure and compared to the previous one.

Figure 5. General view of Doak Laboratory with

traverse system to hot-wire anemometer and Pitot
tube measurements.
Figure 3. Anechoic chamber at DOAK Laboratory
during antransversable azimuthal array of
microphones experiment.
There are some nozzle options and the one used
in this work was a 38.1mm exit-diameter, convergent,
conical nozzle used for most of the tests done at the
Doak Laboratory. This nozzle was chosen because its
aerodynamic and acoustic characteristics were welldocumented in the Noise Test Facility (NTF) at
QinetiQ, Farnborough, UK.

Acoustic data is acquired using GRAS Type

40BF microphones and digitized using a National
Instruments NI PCI-4472 acquisition card with a
102.4kHz sample rate, and 24-bit resolution. The
time domain data is then transformed to the
frequency domain using a 4096 point Hamming
window-averaged Fourier transform. All subsequent
data processing is done using narrowband spectra.
Microphones are calibrated using a B&K Type 4220
pistonphone calibrator. Measurements of Overall
Sound Pressure Level are achieved by numerically
integrating the narrowband spectra with respect to
frequency using a trapezium rule method across the
entire range of narrowband frequencies. The
narrowband data may also be transformed into onethird octave band spectra using idealised third-octave
filters consistent with ANSI S1.1-1986.
Aerodynamic data is acquired using Druck
PDCR 820 pressure sensors and digitised using the
National Instruments data acquisition hardware
described above. The pressure sensors are calibrated
using a Druck Type DPI 610 pressure calibrator.
Aerodynamics Measurements

Figure 4. Sketch and picture of 38 mm reference

nozzle.

The measurements of the

profiles were performed using a
hot-wire anemometer is used
compare with Pitot tube) and
measurements.

Engenharia Trmica (Thermal Engineering), Vol. 13 No. 2 December 2014 p. 33-40

mean flow velocity

Pitot tube, while the
for mean flow (to
turbulence intensity

35

Tecnologia/Technology

Proena, et al. Aerodynamics and Aeroacoustics

Figure 6. Pitot tube of the micromanometer and Pitot

tube built during the experiments by Mat Gruber.
The Pitot tube is used to measure mean flow
velocity profiles and the spreading of the jet.
Furthermore, it was used as a reference velocity
measurement to calibrate the hot-wire sensors. Two
different transducer systems, coupled to the Pitot
tube, were used in these experiments: a micromanometer for Pitot tube, allowing velocities up to
180m/s and a Druk PDCR 820 transducer. Both data
is digitalized using the National Instruments data
acquisition board.
Hot-wire anemometry is the main measuring
system applied in this work. Single hot-wire
anemometers are the most common sensors applied
in flow measurements, for several reasons: reduced
size, price relatively low, high frequency response,
simple to use. One of the limitations is that it has to
be used for low turbulence intensities (up to 10%).
Various sensors had been used in this work. Each
probe has its specification for resistance and
operation resistance given by the manufacturer. Next
figure shows one of the single hot-wire anemometer
used and the configuration with the temperature
probe.

The velocity distributions were acquired along

the jet axis to different radial positions and the
workspace of mean velocity profile experiments is
demonstrated in the Fig. 8. The nozzle centre is
situated in 0, vertical axis, y, and represents radial
variations in relation the radius of the nozzle and x
is the jet axis. Red points symbolize where the data
were acquired. Only for single hot-wire probes, just
the points inside the blue rectangle were acquired.
Thus, for each Mach number analyzed, 963 points
were recorded to Pitot tube and triple hot-film
measurements, whilst 583 to single hot-wire.

Figure 8. Acquisition points along the region of the

jet for aerodynamic measurements.
Acoustics Measurements
This section shows the equipment used to
perform the acoustic measurements. It is mainly
composed by microphones, preamplifiers and
amplifiers. Far-field noise measurements were
acquired using polar microphones covering 6 angles
as shown in the Tab. 1.
Table 1. Far-field microphone location.
Mic.
NI
Mic.
number channel angle
Mic. type
1
3
40
GRAS 40BF
2
4
50
GRAS 40BF
3
5
60
GRAS 40BF
4
6
75
GRAS 40BF
5
7
90
GRAS 40BF
6
8
110
GRAS 40BF

Figure 7. Single hot-wire sensor.

36

Measurements were done with GRAS Type

40BF microphones with a frequency range of 10 Hz
to 100 kHz and dynamic range of 40 dB to 174 dB
(reference 20 Pa). This is a precision condenser
microphone for general purpose measurements in
open acoustic fields and it is externally polarized. As
a free-field microphone, it is for measuring the sound
pressure which existed before it was placed in the
sound field pointing towards the sound source, and its
disturbing effects are negligible for most application
frequencies due the small dimensions. The normal
sensitivity of the microphones used is 4 mV/Pa and it
was checked before each experiment day.

Engenharia Trmica (Thermal Engineering), Vol. 13 No. 2 December 2014 p. 33-40

Tecnologia/Technology

Proena, et al. Aerodynamics and Aeroacoustics

The acoustic measurements were performed

only to the far-field. Six different observer angles
[40], [50], [60], [75], [90] e [110] were applied to
acquire the noise signature.
EXPERIMENTAL RESULTS
This sub-section considers the mean velocity
profiles acquired by the Pitot tube and the single
sensor hot-wire anemometer. The results are disposed
in axial velocity profiles, radial velocity profiles and
a jet characterization of the whole region measured.

Aerodynamics Results
The aerodynamic experiments were carried out
originally to compose the axial profiles. The next
figures present the results for velocity distribution.
Figures 9 to 10 are extracted from Pitot tube
measurements. Firstly, Fig. 9 shows the normalized
mean velocity from the centerline to lipline at Mach
number 0.25, distributed along the jet axis domain,
from 0 to 13 diameters. Each line is normalized
independently, according to the Equation:

U0
U max

core length (Abramovich, 1963). According to the

Fig. 9, the potential core ends between 4 to 4.5
diameters downstream the jet axis.
The behavior of C2 and C3 is unexpected.
Since the potential core has a conical shape, the
greater the distance from the centerline, the higher
velocity decay. It is observed from the other profiles,
from C4 to Clip. Although all measurements
were corrected because the flow variation, the data
from these two lines are very discrepant, since it
occurred an instantaneous flow velocity increasing,
probably related to the rise of pressure in the
compressor for an unknown reason.
Keep moving in the radial position over the
nozzle region, it can be seen a new trend in the axial
profiles. Next figure shows this trend from S1 to
S20, again for Mach 0.25.

(1)

where U 0 is the velocity measured in meters per

second and U max is the maximum velocity measured
in meters per second along that line.

Figure 9. Velocity distribution along the jet axis for

different radial positions (Pitot tube, Mach 0.25).
The lines Cline and C1 represents the
measurements along the axial line in radial positions
0 and 1.05 mm from the center of the nozzle. The
velocity is expected to be maximum and
unchangeable until 4 to 5 diameters from the exit
nozzle along the centerline, since it is the potential

Figure 10. Velocity distribution along the jet axis for

different radial positions (Pitot tube, Mach 0.25).
The trend is clearly the expected. In the regions
close to the nozzle, there is a sudden increase in the
velocity as the sensor moves in the axial position. As
the measurements get in lines far from the nozzle, the
profile velocity is smaller and takes its maximum
velocity at higher axial distances. The trend shows
clearly the shear layer between the jet and the
stationary flow (ambient), located for each point
where the velocity profile reaches the maximum
value. The last results (especially after S17) the jet
flow is practically all mixed with the ambient and the
value of u velocity component is close to zero.
Now the same sequence is shown to
measurements using a hot-wire sensor. Comparing
the quality of the pitot and hot-wire techniques for
the measurements discussed, it is clear that the pitot
data present curves much smoother than the
anemometer. The reason is that the latter is more
sensitive to the flow variation. Furthermore, the hotwire calibration curve, a fourth order polynomial, is
more complex than the linear curve used by the
pressure transducer. As the results had to be corrected
to the unsteady pressure imposed by the compressor
system, the accumulated errors are most evident in

Engenharia Trmica (Thermal Engineering), Vol. 13 No. 2 December 2014 p. 33-40

37

Tecnologia/Technology

Proena, et al. Aerodynamics and Aeroacoustics

the anemometry system.

previous sub-sections.
Figure 13 shows the velocity distribution from
the exit nozzle to thirteen diameters downstream the
jet axis, also covering two diameters from the center
of the nozzle. It shows the results for the Mach 0.25.
These data were acquired by the pitot tube.
It is possible to see the potential core and the
shear layer behavior along the imposed velocity
regimes. The slight growth of the potential core with
the velocity increases is evident, as well as the shear
layer spreading, with the small thickness close to the
nozzle.

Figure 11. Velocity distribution along the jet axis for

different radial positions (Single hot-wire, Mach
0.25).
In order to check the trend and the accuracy of
the data showed above, the axial profiles obtained
with each system are confronted. The centerline for
Mach conditions 0.25 is also compared to other
experimental references. Figure 12 presents this
comparison. It can be seen that at Mach number 0.25
the axial velocity profiles from the pitot and the hotwire system are very similar. Far away around eleven
diameters from the exit nozzle, the hot-wire results
shown some inconsistency. It is believed that the
cause is again the flow variation imposed by the
compressor system. The work of Morris and Zaman
(2010) shows only the results for Mach 0.25, alleging
the fragility of the modern hot-wires probes with high
frequency response to high and unsteady flows. The
trend is confirmed, despite a presence of an offset.

Figure 12. Velocity distribution along the jet axis for

centerline at Mach 0.25. Hot-wire; Pitot tube;
Morris and Zaman (2010).
Jet Characterization
This sub-section is used to present a general
view of the jets velocity distribution. All lines
measured are considered and, such as in the previous
sub-section, the data is mirrored to better characterize
the jet. Although it seems to be merely illustrative, it
helps to summarize the information given on the
38

Figure 13. Jet characterization with pitot tube data for

Mach 0.25.
Acoustics Results
The power spectral density - PSD is a way to
show the strength of energy variations as a function
of frequency. The usual unit of PSD is energy per
frequency, and the integral of the PSD over a given
frequency range computes the average power in the
signal. This spectrum is computed directly by a Fast
Fourier Transform method - FFT or by computing an
autocorrelation function and transforming it.
In the present work, the Welchs method is used
to calculate the PSD of the signal acquired by the
anemometer. This method uses a FFT to obtain an
estimation of the PSD based on time averaging over
short modified periodograms (Welch, 1967).
Welch based on Barlletts method, which
consists in splitting the original points into a number
n of data segments of length L, compute the
periodogram by the discrete Fourier transform, and
then compute the squared magnitude of the result
dividing by the L. Finally, the result of the
periodograms is time-averaged for the n data
segments. The difference is that the on the Welchs
method, the system is divided into overlapping
segments, which are windowed in the time domain
before apply the discrete Fourier transform. It is
considered an improvement on the standard
periodogram and Bartletts methods, since it reduces
noise in the estimated power spectra when it reduces
the frequency resolution.
The processing is done by the Matlab pwelch
function, which returns the two-sided Welch PSD
estimation. It was used a Hanning window and a
sample frequency of 50 kHz (which is transformed in
25 kHz).
Still taking into account the fact that it is only

Engenharia Trmica (Thermal Engineering), Vol. 13 No. 2 December 2014 p. 33-40

Tecnologia/Technology

Proena, et al. Aerodynamics and Aeroacoustics

one component velocity, some results to approximate

the power spectrum density and compares with the
power spectrum of turbulence are performed.
The Fig. 14 shows the result for some points
along the C1, and C9 lines, for Mach 0.25. The
points inside the potential core have less energy than
the points in the mixing layer. The low frequencies
are responsible by the energy injection, whilst it is
seen that the high frequencies are dissipating this
energy.

(a)

(b)
Figure 14. Power spectral density in measurements at
Mach number 0.25.x/Dj=0, 2, 4, 6, 8, 10 and 12. (a)
C1; (b) C9.
The SPL pattern is now presented in observer
angle comparisons. Due the directivity of the sound
and the effects overprint, the sound sources are
convected downstream by the mean flow and the
maximum noise is radiated at observer angles
between 30 and 50.

CONCLUDING REMARKS
The present work deals with the jet
aerodynamics and acoustics through experimental
techniques, exploring the basic theory related to low
speed subsonic jets. The main goal was to study
fundamental of this kind of flow configuration, set up
and run measurements from different systems and
compare the results with the literature reviewed. It
represents a particular achievement to the
Aeroacoustics
Group
recently
founded
at
MFLAB/UFU.
The results are discussed in a methodical order.
Firstly, it is presented the aerodynamic work, which
compounds most of the efforts in terms of
experiments, post-processing and analyses. It shows
mean velocity profiles acquired by the pitot tube and
hot-wire anemometer sensor. The experiments were
carried out to several axial lines, and by combining
these points it is also possible estimate the radial
profiles, as well as a characterization of the subsonic
jets regions. Through the anemometer data, it is also
possible analyses the unsteadiness velocity,
estimating the turbulence intensity in axial profiles
and the power spectral density of the points
measured. The profiles are compared with the
literature for Mach numbers 0.25 presenting a good
agreement.
Secondly, the acoustic results are shown to six
observer positions for a wide range of velocities. The
data is confronted to the theory related to jets, and it
shows a good agreement. The experiments done in
this section are extremely simple in the way that
many researchers are developing advanced
techniques to jet noise measurements, but it is
essential to the development of complex sets of
experiments and insights about the jet physics,
extremely important to the creation of semi-empirical
methods. A database is created from these
measurements and may apply in future to validate
numerical analyses which are underway.

ACKNOWLEDGEMENTS
The authors would like to thank the financial
support
provided
by
Coordenao
de
Aperfeioamento de Pessoal de Ensino Superior
(CAPES), Brazil, for the development of this work.
They also thanks for all the support received at the
Institute of Sound and Vibration for the conclusion of
his MsC thesis and the Fundao de Amparo
Pesquisa do Estado de Minas Gerais (FAPEMIG).
REFERENCES

Figure 15. Sound Pressure Level for Mach number

0.25 in six different angles (40, 50, 60, 75, 90 and
110).

Almeida, O., 2009, Aeroacoustics of DualStream Jets with Application to Turbofan Engines,
Doctoral Thesis,
Aeronautics Institute of
Technology, So Jos dos Campos.

Engenharia Trmica (Thermal Engineering), Vol. 13 No. 2 December 2014 p. 33-40

39

Tecnologia/Technology

Proena, et al. Aerodynamics and Aeroacoustics

Smith, M. J. T., 1989, Aircraft Noise,

Cambridge Aerospace Series.
Abramovich, G. N., 1963, The Theory of
Turbulent Jets, The MIT Press Classics.
Morris, P. J., Zaman, and K. B. M. Q., 2010,
Velocity Measurements in Jets with Application to
Noisesource Modeling, Journal of Sound and
Vibration, Vol. 329, pp. 394-414.
Welch, P. D., 1967, The Use of Fast Fourier
Transform for the Estimation of Power Spectra: A
Method Based on Time Averaging Over Short,
Modified Periodograms, IEEE Transactions on Audio
Electroacoustics, AU-15, pp. 70-73.

40

Engenharia Trmica (Thermal Engineering), Vol. 13 No. 2 December 2014 p. 33-40