Wang X. Vehicle Noise and Vibration Refinement
Подождите немного. Документ загружается.
Advanced experimental techniques in vehicle noise 203
© Woodhead Publishing Limited, 2010
noise pollution and desired high-level music reproduction, e.g. with
headphones.
When two signals are received in parallel one of them may be masked
by the other, depending on the level and frequency difference. Figure 9.6
shows the situation for a one-third octave band noise at 1 kHz with a level
of 90 dB. Each tone that is lower than the spectral masking curve cannot
be perceived subjectively.
The concept of loudness and spectral masking helps one to interpret
objective data in order to fi nd critical frequency ranges and whether tonal
noises are annoying or just inaudible. Other psychoacoustic metrics are
sharpness, roughness and fl uctuation strength. They are defi ned only for
artifi cial noises and in general are not standardised (see Table 9.5).
Besides the psychoacoustic metrics there exist several additional metrics
such as tonality and impulsiveness that are intended to classify sounds. One
10
100
90
80
70
60
50
40
30
20
10
0
–10
100 1000 10000
Frequency (Hz)
Sound pressure level (dB)
Masking area
5 PhonGD
100000
E
x
c
i
t
a
t
i
o
n
9.6 Spectral masking according to ISO 532B (free sound fi eld).
Table 9.5 Overview of sound quality metrics
Metric Unit Description
Loudness sone, phon 1 sone = 40 phon for 1 kHz tone of
40 dB
Sharpness acum 1 acum for noise of 60 dB, 1 bark
wide, centred around 1 kHz
Roughness asper 1 asper for 1 kHz tone of 60 dB, 100%
amplitude modulated at 70 Hz
Fluctuation strength vacil 1 vacil for 1 kHz tone of 60 dB, 100%
amplitude modulated at 4 Hz
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
204 Vehicle noise and vibration refi nement
© Woodhead Publishing Limited, 2010
of these additional metrics is Articulation Index (AI), which originates
from the telecommunications industry (Fig. 9.7). The AI concentrates on
the amplitude (30 dB) and frequency range (200–6300 Hz) of speech. It
applies a weighting scheme similar to the above A-weighting but with an
additional 9 dB per octave suppression at frequencies of >1 kHz. An addi-
tional spectral weighting is applied by adding different percentage values
per dB with the highest scores in the 800–5000 Hz frequency range
(maximum around 1800 Hz). It is often used to describe acoustic comfort
in the interior compartment (or the difference between two conditions)
with a single number value.
In each one-third octave band a certain sound pressure level is assigned
a percentage value. The AI is the sum of all percentage values for a given
one-third octave spectrum. If the background or masking noise is louder
than the upper limit, it is assumed that nothing can be understood at all
(=0% AI). If the masking noise is below the lower limit, it is assumed that
everything can be understood (=100% AI). When applied to vehicle noise
it is desired to extend the listening range to higher and lower levels in order
to better distinguish an excellent car from good cars, e.g. under extreme
operating conditions such as cobblestone excitation. In such circumstances
the AI can exceed 100% or be negative. To distinguish both variants the
extended AI is often abbreviated to EAI. Note that AI and EAI are among
the few metrics for which a high value is desired (higher is better).
With the advent of the diesel powertrain, impulsiveness and modulation
become more and more important. Modulations occur, for example, if the
combustion noise from each cylinder is not identical. For objective analysis
of modulation a quasi-stationary signal is required (e.g. constant engine
rpm for several seconds). Depending on the modulation frequency the
noise is perceived, e.g., as rough (modulation frequency 15–300 Hz) or
fl uctuating (modulation frequency <20 Hz).
15
Often the noise sensation cannot be fully described with a single standard
objective measurement. Then the noise of different test objects or different
test conditions is recorded and subjectively rated. In order to get a more
realistic reproduction an artifi cial head is used with two microphones placed
at the ear locations (binaural head) and the noise is played back via high-
quality headphones.
Such a recording can also be used to verify an assumed problem fre-
quency or frequency range. The recorded sound can be fi ltered (e.g. noise
from 1300 to 1400 Hz can be suppressed by 10 dB by a notch fi lter) and the
original and fi ltered sound can be compared subjectively. This not only
gives a hint which frequency band must be improved but also how much
the noise must be reduced to achieve a desired improvement.
The subjective rating of a noise is often expressed in complex perception
categories such as powerful, aggressive, refi ned, etc. If all ratings are
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
© Woodhead Publishing Limited, 2010
100 200 400 500 800 1000 2000 40005000 6300 8000 100002500 31501250 1600630250 315125 160
Frequency (Hz)
Measurement Al values (%)
100
98
96
94
92
–1.200
–1.133
–1.067
–1.000
–0.933
–0.867
–0.800
–0.733
–0.667
–0.600
–0.533
–0.467
–0.400
–0.333
–0.267
–0.200
–0.133
–0.067
0.000
0.067
0.133
0.200
0.267
0.333
0.400
0.467
0.533
0.600
0.667
0.733
0.800
0.867
0.933
1.000
1.067
1.133
1.200
1.267
1.333
1.400
1.467
–0.600
–2.067
–1.933
–1.800
–1.667
–1.533
–1.400
–1.267
–1.133
–1.000
–0.867
–0.733
–0.600
–0.467
–0.333
–0.200
–0.067
0.067
0.200
0.333
0.467
0.600
0.733
0.867
1.000
1.133
1.267
1.400
1.533
1.667
1.800
1.933
2.067
2.200
2.333
2.467
2.600
2.733
2.867
3.000
3.133
3.267
–0.600
–3.142
–2.925
–2.708
–2.492
–2.275
–2.058
–1.842
–1.625
–1.408
–1.192
–0.975
–0.758
–0.542
–0.325
–0.108
0.108
0.325
0.542
0.758
0.975
1.192
1.408
1.625
1.842
2.058
2.275
2.492
2.708
2.925
3.142
3.358
3.575
3.792
4.008
4.225
4.442
4.658
4.875
5.092
5.308
5.525
–0.325
–3.825
–3.542
–3.258
–2.975
–2.692
–2.408
–2.125
–1.842
–1.558
–1.275
–0.992
–0.708
–0.425
–0.142
0.142
0.425
0.708
0.992
1.275
1.558
1.842
2.125
2.408
2.692
2.975
3.258
3.542
3.825
4.108
4.392
4.675
4.958
5.242
5.525
5.808
6.092
6.375
6.658
6.942
7.225
7.508
0.425
–3.750
–3.450
–3.150
–2.850
–2.550
–2.250
–1.950
–1.650
–1.350
–1.050
–0.750
–0.450
–0.150
0.150
0.450
0.750
1.050
1.350
1.650
1.950
2.250
2.550
2.850
3.150
3.450
3.750
4.050
4.350
4.650
4.950
5.250
5.550
5.850
6.150
6.450
6.750
7.050
7.350
7.650
7.950
8.250
1.350
–4.375
–4.025
–3.675
–3.325
–2.975
–2.625
–2.275
–1.925
–1.575
–1.225
–0.875
–0.525
–0.175
0.175
0.525
0.875
1.225
1.575
1.925
2.275
2.625
2.975
3.325
3.675
4.025
4.375
4.725
5.075
5.425
5.775
6.125
6.475
6.825
7.175
7.525
7.875
8.225
8.575
8.925
9.275
9.625
2.275
–5.417
–4.983
–4.550
–4.117
–3.683
–3.250
–2.817
–2.383
–1.950
–1.517
–1.083
–0.650
–0.217
0.217
0.650
1.083
1.517
1.950
2.383
2.817
3.250
3.683
4.117
4.550
4.983
5.417
5.850
6.283
6.717
7.150
7.583
8.017
8.450
8.883
9.317
9.750
10.183
10.617
11.050
11.483
11.917
3.683
–6.283
–5.800
–5.317
–4.833
–4.350
–3.867
–3.383
–2.900
–2.417
–1.933
–1.450
–0.967
–0.483
0.000
0.483
0.967
1.450
1.933
2.417
2.900
3.383
3.867
4.350
4.833
5.317
5.800
6.283
6.767
7.250
7.733
8.217
8.700
9.183
9.667
10.150
10.633
11.117
11.600
12.083
12.567
13.050
4.833
–7.933
–7.367
–6.800
–6.233
–5.667
–5.100
–4.533
–3.967
–3.400
–2.833
–2.267
–1.700
–1.133
–0.567
0.000
0.567
1.133
1.700
2.267
2.833
3.400
3.967
4.533
5.100
5.667
6.233
6.800
7.367
7.933
8.500
9.067
9.633
10.200
10.767
11.333
11.900
12.467
13.033
13.600
14.167
14.733
6.233
–11.500
–10.733
–9.967
–9.200
–8.433
–7.667
–6.900
–6.133
–5.367
–4.600
–3.833
–3.067
–2.300
–1.533
–0.767
0.000
0.767
1.533
2.300
3.067
3.833
4.600
5.367
6.133
6.900
7.667
8.433
9.200
9.967
10.733
11.500
12.267
13.033
13.800
14.567
15.333
16.100
16.867
17.633
18.400
19.167
9.200
–12.100
–11.367
–10.633
–9.900
–9.167
–8.433
–7.700
–6.967
–6.233
–5.500
–4.767
–4.033
–3.300
–2.567
–1.833
–1.100
–0.367
0.367
1.100
1.833
2.567
3.300
4.033
4.767
5.500
6.233
6.967
7.700
8.433
9.167
9.900
10.633
11.367
12.100
12.833
13.567
14.300
15.033
15.767
16.500
17.233
9.167
–11.083
–10.450
–9.817
–9.183
–8.550
–7.917
–7.283
–6.650
–6.017
–5.383
–4.750
–4.117
–3.483
–2.850
–2.217
–1.583
–0.950
–0.317
0.317
0.950
1.583
2.217
2.850
3.483
4.117
4.750
5.383
6.017
6.650
7.283
7.917
8.550
9.183
9.817
10.450
11.083
11.717
12.350
12.983
13.617
14.250
8.550
–11.100
–10.500
–9.900
–9.300
–8.700
–8.100
–7.500
–6.900
–6.300
–5.700
–5.100
–4.500
–3.900
–3.300
–2.700
–2.100
–1.500
–0.900
–0.300
0.300
0.900
1.500
2.100
2.700
3.300
3.900
4.500
5.100
5.700
6.300
6.900
7.500
8.100
8.700
9.300
9.900
10.500
11.100
11.700
12.300
12.900
9.300
–10.333
–9.817
–9.300
–8.783
–8.267
–7.750
–7.233
–6.717
–6.200
–5.683
–5.167
–4.650
–4.133
–3.617
–3.100
–2.583
–2.067
–1.550
–1.033
–0.517
0.000
0.517
1.033
1.550
2.067
2.583
3.100
3.617
4.133
4.650
5.167
5.683
6.200
6.717
7.233
7.750
8.267
8.850
9.300
9.817
10.333
8.850
–9.167
–8.750
–8.333
–7.917
–7.500
–7.083
–6.667
–6.250
–5.833
–5.417
–5.000
–4.483
–4.167
–3.750
–3.333
–2.917
–2.500
–2.083
–1.667
–1.250
–0.833
–0.417
0.000
0.417
0.833
1.250
1.667
2.083
2.500
2.917
3.333
3.750
4.167
4.583
5.000
5.417
5.833
6.250
6.667
7.083
8.5508.550
–4.083
–3.917
–3.750
–3.583
–3.417
–3.250
–3.083
–2.917
–2.750
–2.583
–2.417
–2.250
–2.083
–1.917
–1.750
–1.583
–1.417
–1.250
–1.083
–0.917
–0.750
–0.583
–0.417
–0.250
–0.083
0.083
0.250
0.417
0.583
0.750
0.917
1.083
1.250
1.417
1.583
1.750
1.917
2.083
2.250
2.417
2.583
90
88
86
84
82
80
78
76
74
72
70
68
66
64
62
60
58
56
54
52
50
48
46
44
42
40
38
36
34
32
30
20
28
26
24
22
EAl values (%)
Loudness (dB(lin))
9.7 Articulation index according to Leo L. Beranek.
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
206 Vehicle noise and vibration refi nement
© Woodhead Publishing Limited, 2010
displayed in a spider diagram, different sensoric profi les can be compared
(see Fig. 9.8).
If a group of similar sounds are subjectively rated in given categories by
a representative jury and analysed with a variety of objective parameters
(e.g. dB(A), AI, spectrum, modulation spectrum, etc.), it is in general pos-
sible to correlate the subjective ratings with a combination of objective
parameters, e.g.:
subjective rating factor objective parameter=×
(
)
∑
ii
j
i
(9.14)
If a suffi cient correlation is found it is possible to assign a subjective rating
to a similar sound or to defi ne the objective parameters of a target sound.
As soon as the group of similar sounds is changed signifi cantly the correla-
tion exercise must be repeated.
9.6 Ultrasound diagnostic techniques
In the fi eld of vehicle noise, ultrasound devices are mainly used for leakage
detection in sound package or sealing applications. An ultrasonic source
(in general battery powered) is placed on one side of a barrier and the
ultrasonic receiver (also in general battery powered) on the other side. This
is a classical noise reduction setup. Moving from audible noise to ultrasonic
noise (typically a warbled tone around 40 kHz) has the advantage that the
result is independent of audible background noise, thus allowing ultrasonic
leakage tests in workshops or even in a production plant.
The intensity of the received ultrasonic sound is displayed on a scale.
Additionally the ultrasonic signal is transferred into the audible range by
heterodyning and can be heard, e.g., through headphones. Both items help
to localise leaks by looking for locations with high ultrasonic intensity.
Refined
Sporty
Powerful
Rough
Balanced
Harmonic
Bright
Sharp
Loud
Aggressive
Vehicle 1
Vehicle 2
9.8 Description of interior noise by means of sensoric profi le.
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
Advanced experimental techniques in vehicle noise 207
© Woodhead Publishing Limited, 2010
However, ultrasonic sound behaves slightly differently from audible
sound. Due to its shorter wavelength an ultrasonic wave may pass through
a hole which would have been too small for an audible wave in the typical
frequency range up to 10 kHz. Small cavities may amplify ultrasonic waves
by resonance effects which are not acting on audible waves with higher
wavelengths and vice versa. So interpreting the intensity in detail is diffi cult
and the quantitative results with ultrasonic detection may differ from those
with audible noise excitation.
A sound source must be small compared to the radiated wavelength in
order to radiate omnidirectionally. As ultrasonic waves have very small
wavelengths this requirement is in general not fulfi lled for an ultrasonic
source. Therefore the absolute intensity of the received signal may depend
on the exact location and orientation of the ultrasonic source.
Therefore, ultrasonic leakage devices are easy to use even under noisy
conditions and give quick and intuitive information on relative ultrasonic
intensity changes that in general correlate well with intensity changes of
audible noise. However, due to the different wavelength and spectral
content the effect of measures in general is not comparable.
9.7 Advanced material testing techniques
In the fi eld of vehicle acoustics it is very important to understand the effect
of sound absorbing and sound deadening materials. The sound absorption
of small, fl at parts for normal sound incidence can be evaluated with a
standing wave tube according to ISO 10534 (also called a Kundt tube: see
Fig. 9.9). If the diameter of the tube is small compared to the radiated
wavelength, plane wave propagation can be ensured. The classical test
procedure is to generate a tone with a loudspeaker on one side of the closed
tube and to evaluate the standing wave pattern near the surface of the test
material backed up with a solid wall (see Fig. 9.10).
From the ratio P
max
/P
min
the absorption coeffi cient α can be derived:
α = 1 − r
2
(9.15)
Test
sample
Microphone
Loudspeaker
X = 0
9.9 Scheme of Kundt tube.
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
208 Vehicle noise and vibration refi nement
© Woodhead Publishing Limited, 2010
where
r
PP
PP
=
()
−
()
+
max min
max min
1
1
(9.16)
In order to cover the whole frequency range typically two tubes are required,
one with 10 cm diameter for the frequency range from 50 to 1600 Hz and
one with 3 cm diameter for 500 to 6300 Hz. In order to avoid the time-
consuming manual scanning of the standing wave pattern at successive
frequencies, modern Kundt tubes use broadband noise excitation and
several microphones at fi xed locations together with adapted evaluation
software.
For locally acting porous absorbers the tube results are relevant and
random incidence results can also be estimated. For plate resonators the
results are not representative if the mounting conditions differ from real-
world mounting. The advantage of the tube method is that only small test
samples are required.
In general the noise comes from all sides. This can be simulated by using
a reverberation room with a diffuse sound fi eld according to ISO 354 or
ASTM C423. In a fi rst step the reverberation time T
60_empty
of the empty
room is measured. Although no obvious absorbing material is in the room,
the reverberation time is not infi nitely long, so some parts of the reverbera-
tion room are not perfectly refl ecting and absorbing the sound energy. A
simple model of the room assumes that the effect of the distributed absorp-
tion is represented by a lumped element, a perfect (100%) absorber with
area A
equiv
, also called equivalent absorption area.
In a second step the desired material is added to the room and a new
reverberation time T
60_full
is measured, so the total effect can be explained
by a new equivalent absorption area. In order to ensure a proper modal
Distance (m)
0.0 0.1 0.2
P
max
= P
1
+ P
2
P
1
– P
2
= P
min
0.3 0.4 0.5
2.0
1.5
1.0
0.5
0.0
Pressure (Pa)
9.10 Standing wave pattern in Kundt tube (1000 Hz, r = 0.7).
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
Advanced experimental techniques in vehicle noise 209
© Woodhead Publishing Limited, 2010
density the volume of the reverberation room must, e.g., be 200 m
3
for a
cut-off frequency of 100 Hz.
The difference between both conditions is the difference of both equiva-
lent absorption areas:
T
V
A
60
0 163
=
.
equiv
(9.17)
ΔAV
TT
full empty
_full _empty
−
=−
⎛
⎝
⎜
⎞
⎠
⎟
0 163
11
60 60
.
(9.18)
α = ΔA/S (9.19)
where T
60
= reverberation time = time for 60 dB noise decay (s)
V = reverberation room volume (m
3
)
A = amount of 100% absorbing area equivalent to absorption
effect (m
2
)
α = absorption coeffi cient (%)
S = surface of absorbing material (m
2
).
In general, 10–12 m
2
of material (or an equivalent number of complex
parts like seats or head liners) is needed to ensure a signifi cant change of
the reverberation time. During material development this is a signifi cant
disadvantage of the test method. Moreover the measured absorption coef-
fi cient depends signifi cantly on the design of the reverberation room, the
amount and location of diffusers, the microphone and source location, the
estimation of the decay time, etc.
16–18
Therefore the sound package supplier RIETER (formerly Keller or
Interkeller) tried to combine the best of two worlds in a practical package,
and created the Alpha Cabin (see Fig. 9.11), which became a de facto stan-
9.11 Alpha Cabin.
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
210 Vehicle noise and vibration refi nement
© Woodhead Publishing Limited, 2010
dard in the fi eld of absorption measurement, although there is no offi cial
standard such as ISO. The Alpha Cabin is a small reverberation chamber
on wheels with a volume of 6.44 m
3
(smaller by a factor of π
3
than a standard
reverberation room) and a small entrance door, that allows one to evaluate
the absorption properties of a sample from 400 Hz (π times higher than for
a standard reverberation room) to 10,000 Hz. This provides more realistic
diffuse fi eld excitation but allows smaller samples with sizes of approxi-
mately 1.2 m
2
(smaller by a factor of π
2
than for a standard reverberation
room), so for complex parts often only one sample is required. The Alpha
Cabin comes complete with loudspeaker excitation, microphone including
positioning system, and data acquisition and post-processing system, thus
ensuring consistent results all over the world. Therefore the Alpha Cabin
is also used for production quality control.
Another important material property is the viscoelastic damping of
sound-deadening material, which is extensively used on body panels, e.g.
in the fi rewall and fl oor region. The damping factor is estimated indirectly
according to the Oberst method (see Fig. 9.12):
19
1. In the fi rst step a metal strip (e.g. 220 mm long, 12 mm wide and 1 mm
thick) is fi xed on one side and excited on the free side with a magnetic
transducer. At a reference point the vibration response is measured.
Excit.
LDV
Excit. = excitation with magnetic transducer
LDV = laser doppler vibrometer
LDV
Excit.
Metal
Metal
Damping
9.12 Oberst bar.
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
Advanced experimental techniques in vehicle noise 211
© Woodhead Publishing Limited, 2010
The frequency response function (FRF) shows several peaks (high
excursion per force input) at the mechanical resonances.
2. In the second step, damping material is applied to the same strip. The
thickness of the applied material is measured at several locations (and
optionally levelled) and the width is made identical to the strip width
by grinding. Finally the FRF is measured again.
3. For both FRFs curve fi tting is applied to extract the exact resonance
frequencies and damping. The shift of the resonance frequency is the
effect of the density, the dimensions and the elastic modulus of the
material. The reduction of the peak and Q-factor at the nth resonance
frequency is the effect of the damping.
The damping depends on the temperature, so all FRF measurements are
performed in a climate chamber with several hours of thermal settling.
Typical temperature ranges are −20 to +60°C in steps of 20°C. As the
change in frequency and peak amplitude is only small, high accuracy is
needed to give reliable results. In general the damping may also vary with
frequency, therefore ISO 6721-3 requires one to interpolate the evaluated
damping at the resulting resonance frequencies for a target frequency of
200 Hz.
19
In the fl oor region often a combination of different materials is used.
Damping material is applied to the sheet metal, and on top comes a layer
of foam or other porous material covered with a layer of heavy mass and/
or carpet. The effect of the sound-deadening material can be measured by
using the supporting structure as a wall between two isolated reverberation
rooms. Either noise or vibration excitation can be used. The structure is
excited in one room and the response is measured in the other room, once
with and once without the added material. For a complete fl oor insulation
the test setup is pretty complex and a moulded part must be manufactured,
which is very expensive in the prototype phase. For practical reasons a
smaller test setup is desired that works with fl at material. Again it was
RIETER that proposed a device called ‘Apamat’ for structure-borne exci-
tation with steel balls, and a device called ‘Small Cabin’ for airborne excita-
tion. The supporting structure in both cases is a plane sheet of metal (84 cm
× 84 cm) of desired thickness. On top of that the desired material setup can
be positioned. The lower and upper chambers are different for both devices.
Measurement with a plane sheet is used for reference. Both devices again
became a de facto industrial standard.
In many fi elds of vehicle development CAE tools have to be used. If
materials other than metal have to be modelled correctly, their properties
must be known or measured in detail. For porous absorbers the properties
can be represented by a Biot–Allard model.
20,21
This requires the inputs
shown in Table 9.6.
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2
212 Vehicle noise and vibration refi nement
© Woodhead Publishing Limited, 2010
The full characterisation of porous materials needs to be measured using
a variety of special equipment which is very complex. Each supplier of such
material must develop its own tools to acquire each of these data, as no
standardised devices and processes are available. This further complicates
the comparison of different materials from different suppliers by CAE
tools. Again, RIETER fi lls the gap and supplies a complete set of tools
under the product name ELWIS.
Besides common FEM–CAE models special software programs are
available such as SISAB or WinFlag that are able to estimate the effect of
multi-layer systems (absorption and damping).
22,23
9.8 Advanced tachometer reference
tracking techniques
For motor vehicles some noise components are related to the ignition fre-
quency of the engine or other rotational devices such as the fan, alternator,
etc. For combustion engines most harmonic responses are a multiple of the
rotation frequency. The rotation frequency is called the fi rst engine order;
the ignition frequency of a four-stroke, four-cylinder engine is the second
order (one ignition after each two revolutions for a four-stroke engine
multiplied by the number of cylinders that ignite one after the other). If the
rpm of the device is known, these related parts (= engine orders) can be
analysed separately.
For quasi-stationary tests on engine test rigs the estimation of the rpm is
not a problem, but for fast-varying engine speeds (e.g. second gear wide
open throttle (WOT) acceleration) this is more complex. Even more
complex is the estimation of rpm fl uctuations, e.g. within one ignition cycle.
This requires multiple information concerning the current crankshaft angle
per revolution, e.g. each 5 degrees. To achieve a resolution of 5° at 6000 rpm
Table 9.6 Parameters for the Biot–Allard model of porous absorbers
Parameter Type Symbol Unit
Composite porosity acoustic
ϕ
(–)
Air fl ow resistivity (AFR) acoustic
σ
(Pa⋅s/m
2
)
Composite tortuosity acoustic
τ
(–)
Composite pore shape factor acoustic c (–)
Viscous characteristic length acoustic
Λ
(m)
Thermal characteristic length acoustic
Λ′
(m)
Young’s modulus structural E (Pa)
Damping loss factor (DLF) structural
η
(–)
Poisson’s ratio (for foam materials)
(range: 0.15–0.48)
structural
ν
(–)
Copyrighted Material downloaded from Woodhead Publishing Online
Delivered by http://woodhead.metapress.com
ETH Zuerich (307-97-768)
Sunday, August 28, 2011 12:05:51 AM
IP Address: 129.132.208.2