A Study on Vibration Characteristics of Vehicle Muffler Depending on Excitation Sources

Yoo-jong Lee, Doo-seuk Choi, Key-sun Kim

A Study on Vibration Characteristics of Vehicle Muffler Depending

on Excitation Sources
1

Yoo-jong Lee, *2 Doo-seuk Choi, 3 Key-sun Kim

Dept. of Mechanical Engineering, Graduate School, Kongju National
University, [email protected]
*2, Corresponding Author
Div.of Automotive & Mechanical Engineering, College of Engineering,
Kongju National University, [email protected]
3
Div.of Automotive & Mechanical Engineering, College of Engineering, Kongju National
University, [email protected]
1, First Author

Abstract
In this study, the engine vibration that affects the exhaust system and road vibration transferred to
the muffler through the vehicle body was measured and analyzed. Also the dynamics of the exhaust
system, which may generate noise in the passenger compartment using a vibration measurement device
were measured, and the changes depending on engine rotation speeds were verified. The resonance of
the muffler was obtained from a frequency response function (FRF) after sine sweep excitation,
followed by muffler vibration measurements using the direct body excitation method under simulated
road conditions and mode shape analysis for each condition.
Through these results, the effects on the muffler of the vibrations transferred from the road and those
created by the engine was verified.

Keywords: Exhaust manifold, Vibration, Direct body excitation, muffler, FFT analysis
1. Introduction
A vehicle exhaust system emits exhaust gas generated from a vehicle engine. It also reduces noise
generated by the combustion of fuel in the engine. It is installed at the bottom part of a vehicle and
consists of pipes extending from the muffler to the outlet. Such a structure is not advantageous in terms
of vibration prevention. Its structure should be designed in line with the vehicle bottom layout, which
poses challenges when its design needs to be changed. The two major influencing factors for exhaust
system vibration are the vibrations transferred from the engine and road excitation. For example, the
vibration caused by road irregularities is transferred to the exhaust system through its mounting points.
Such vibrations have adverse effects on the ride comfort and the noise generated by buzz, squeak, and
rattle (BSR) perceived in the passenger compartment [1].
Recently, Yuan [2] and Shen [3] tested, through numerical analyses, the effects of vibration
generated in association with engine exhaust gas emission. Additionally, Kim [4] conducted a study on
muffler-vibration-generated noises. Many intensive studies have recently been conducted to reduce
vibration in vehicle exhaust systems [5-7].
Vibration of the vehicle exhaust system can be reduced by designing the exhaust system resonance
frequency to be different from the natural frequencies of other vehicle parts. Another method is
reducing the input excitation transferred from the engine by installing bellows or changing the
positions of components.
Moreover, the relationship between the frequencies of muffler resonance and cylinder firing, a
vibration caused by combustion in a four-stroke engine, is the most important part of an exhaust system
design. Such resonance frequencies can be measured using sine sweep excitation or an impact
hammering test.
In this study, the method of direct body excitation, unlike the conventional method for exerting
excitation to the wheels was chosen. This method has the advantages of reducing the effects of the
suspension system and simplifying the vibration model, thus facilitating the analyses of the behaviors
of vehicle parts. The vibration characteristics of a muffler installed in a vehicle using direct body
excitation were identified and the effects of the vibration of the muffler tip on the exhaust system using
mode shape changes were analyzed. Additionally, the relative motion between the muffler and the
vehicle body by varying the engine speed from idle to 1500, 2000, 2500, and 3000 rpm was analyzed.

International Journal of Digital Content Technology and its Applications(JDCTA)

Volume7, Number13, Sep 2013

496

A Study on Vibration Characteristics of Vehicle Muffler Depending on Excitation Sources

Yoo-jong Lee, Doo-seuk Choi, Key-sun Kim

2. Experimental Method
A muffler can exert considerable influence on the vibration of the exhaust system. To determine
these effects, the vibration transferred from the muffler to the vehicle body was analyzed, and the
vibrations of the muffler and the body were measured, and attempted to extract the correlation
between them.
The muffler used in our experiments had a hanger positioned at the muffler tip, as shown in Figure
1, and connected to the body. Figure 1(a) shows the vibration measurement positions where 5
acceleration sensors were installed; they were installed on the muffler (13), the hanger (4), and the
body where the hanger is connected (5). We used accelerometers capable of measuring in three axial
directions.

5
4

C-pillar

A-pillar

3-axis accelerometer,
(a) Sensor position

(b) test excitation position

Figure 1. Test install

Figure 1(b) shows the direct body excitation method we used to measure the muffler resonance
frequency. We induced excitation in the vertical direction (z axis) by connecting the vibration exciter
directly to the bottom part of the monocoque-type vehicle body, with A-pillar on the right-hand side,
and the C-pillar on the left. Excitation was implemented at sine sweep frequencies ranging from 1 to
100 Hz. Additionally, a random power spectral density (PSD) excitation was applied at frequencies of
8120 Hz. These profiles were established by measuring real-situation road-to-vehicle vibrations. In
order to identify the noise sources in the passenger compartment, we used excitation profiles already
validated and currently in use.
First, we ran the engine to determine the vibrations propagating from the engine. The test conditions
varied between 0 and 3000 rpm, i.e., zero load, idle (800), 1500, 2000, 2500, and 3000 rpm using the
gas pedal. Table 1 outlines the measured engine speeds. In consideration of errors due to possible
differences in gas pedal angle, measurements were performed three times at each engine speed, and
their mean values were used for the analyses. The frequencies were determined by measuring the speed
of the engines crankshaft axis using a tachometer. Data were collected using a 15-channel signal
measurement device (LMS, USA). Data analyses were performed by analyzing frequencies and
frequency response function (FRF), etc. using the Test Lab Signature Process program.

Table 1. Engine speed conditions (unit: rpm)

800
1500
2000
2500
3000

1st

2nd

3rd

Avg.

826
1496
2072
2679
3166

810
1618
20066
2683
3120

805
1627
2087
2520
3028

814
1580
2075
2627
3105

497

A Study on Vibration Characteristics of Vehicle Muffler Depending on Excitation Sources

Yoo-jong Lee, Doo-seuk Choi, Key-sun Kim

3. Experiment Results
Figure 2 shows the sine sweep test results, i.e., the body and muffler vibrations in three axial
directions. Each graph represents the vibration level at each test frequency over time. Figures 2(a), (c),
and (e), which represent body vibrations, reveal that strong vibrations occurred in the range of 2050
Hz. In particular, the z axis showed the greatest vibration amplitudes, whereas the vibrations in the xand y-axial directions did not have a significant impact on the body. The body-manifested vibrations
were transferred from various parts, but did not incur any body-associated resonance in the range of 1
100 Hz.
Figures 2(b), (d), (f), which represent muffler vibrations, reveal that strong vibrations occurred in the
range of 1030 Hz, and resonance occurred at 16.6 Hz. In particular, vibrations were equally large in
all three axial directions owing to the flexible characteristic of the hanger. The low-frequency range
was included in order to avert the occurrence of resonance under the impact of firing frequency during
engine idle time.

(a) Body vibration z-axis

(b) Muffler vibration z-axis

(c) Body vibration y-axis

(d) Muffler vibration y-axis

(e) Body vibration, x axis

(f) Muffler vibration, x axis

Figure 2. Sine sweep vibration colormap

498

A Study on Vibration Characteristics of Vehicle Muffler Depending on Excitation Sources

Yoo-jong Lee, Doo-seuk Choi, Key-sun Kim

Figure 3. Graphs comparing RMS values of vibration in the frequency range of 1100 Hz (25-s
intervals)
Figure 3 shows the root mean square (RMS) values of vibrations generated in the main frequency
range (1100 Hz) in the fast Fourier transform (FFT) frequency domain of 25 s in which the resonance
occurred. The magnitude of vibration is shown in decreasing order of body, hanger, and muffler for
each direction. From this, it can be inferred that the vibrations generated by the resonance of the
muffler are absorbed through the hanger rubber such that the vibration transferred to the body is not
significant.
Figure 4 shows the random excitation test results and allows comparison between the muffler and
the body vibrations. In the case of the vertical direction (z axis) as shown in (a), the max frequency
occurred at 11 Hz, and vibrations also occurred at 16 Hz. The vibration levels of the muffler and body
were 0.51g and 0.19g, respectively (g = 9.8 m/s2), demonstrating that the impact of road excitation
induced muffler vibration on the body vibration is insignificant. As shown in (b), vibrations also
occurred in the horizontal direction (y axis), albeit weaker (0.38g) than those in the vertical direction.
This is assumed to be attributable to the free space for horizontal vibration due to the mounting being
placed on the upper side of the muffler. However, the impact of vibration in the transversal direction (x
axis) was not significant compared to those in the other directions. This is because of the quasi absence
of body vibration except in the vertical direction. Despite the insignificant effect of the vibrations
transferred from the muffler to the body, the vibrations in the vertical direction cannot be ignored; it is
therefore necessary to perform a design-around to reduce vertical muffler vibrations.

(a) Vertical vibration (z axis)

(b) Horizontal vibration (y axis)

Figure 4. Random excitation PSD measurement

499

A Study on Vibration Characteristics of Vehicle Muffler Depending on Excitation Sources

Yoo-jong Lee, Doo-seuk Choi, Key-sun Kim

Figure 5. Sensors positioned on the muffler bottom for measurement

To determine the effect of engine excitation, we measured the vibrations at 5 positions by installing
sensors on the front (1), middle (4), and rear (7) of the muffler, the mount (m1), and the body, as
shown in Figure 5.
The purpose of measuring the engine-excitation-induced vibrations was to determine the major
engine-rpm-dependent vibration frequencies and to verify the effect of the vibrations other than
resonance on the vehicle body. An idle condition was imposed after starting and rpm stabilization, and
the measurements were performed in the order of idle, 1500 rpm, 2000 rpm, 2500 rpm, and 3000 rpm
for 20 s each in 3 trials. In order to determine the changes in vibration according to the flow of exhaust
gas, we divided the muffler into three sections based on the inlet and outlet.

(a) x axis

(b) y axis

(c) z axis
Figure 6. Changes in vibration in three axial directions for each engine speed.

500

A Study on Vibration Characteristics of Vehicle Muffler Depending on Excitation Sources

Yoo-jong Lee, Doo-seuk Choi, Key-sun Kim

The representative values for each engine speed (rpm) were obtained by calculating the average of
three measurements. The results of comparing these values are shown in Figure 6. While the vibration
increased with an increase in rpm, in the case of (c), which represents the z-axis direction, a larger
degree of stability manifested in the vicinity of 1500 rpm than in the idle state. In the x-axis (a), y-axis
(b), and z-axis (c) directions, strong muffler vibrations were observed at 3000, 2500, and 3000 rpm,
respectively. The body vibration increased as well, to a lesser degree, with the increase in the muffler
vibration in each direction, thus demonstrating that muffler vibration has only a slight effect on body
vibration. Nevertheless, the effect of the muffler on the body was insignificant in all three axial
directions.
The position-dependent changes in vibration at the bottom part of the muffler are illustrated in
Figure 7, which shows that muffler vibration varied according to the measurement position. In (a),
which represents the 800 rpm condition, the strongest vibration occurred at m1 (near the mount),
whereas the vibration levels according to the left and right sides hardly differed. In (b), which
represents the 1500 rpm condition, the vibration level was maintained in a stable state, and strong
vibrations occurred at 2000 rpm and above. In particular, the right side of the muffler showed a higher
vibration intensity than the left side, and the front side showed a higher intensity than the back side.
These results suggest that vibration modes were formed, resulting in different vibrations depending on
position. Based on this finding, we measured the vibration modes with respect to engine speed to
determine their relevant effects.

(a) 800 rpm (idle)

(c) 2000 rpm

(b) 1500 rpm

(d) 2500 rpm

Figure 7. Changes in vertical vibrations for each measurement position

Based on the measurement positions of a real muffler, measurements were performed at all 11 nodal
points of muffler mode shape, taking into account all 3 axial directions, and measurement results are
plotted in Figure 8. The respective modal frequencies were 24, 34, 74, 81, and 93 Hz, tending to appear
in the range below 100 Hz, and the muffler vibrated in the vector sum directions of the x, y, and z axes.
There was almost no change at the mounting point (m1). By observing the vibration directions, we
verified that the muffler vibration could not be represented by a value of one direction, and that instead
of a simple rigid-body-mode vibration, a bending-mode vibration occurred.

501

A Study on Vibration Characteristics of Vehicle Muffler Depending on Excitation Sources

Yoo-jong Lee, Doo-seuk Choi, Key-sun Kim

This can be explained by the fact that the position of the mount and the bending modes of the pipe
exert force in the xy axial direction. In (a) and (c), which represent the 800 and 2000 rpm conditions,
respectively, bending modes occurred along each axis. In (b), representing 1500 rpm, among the
measurement positions illustrated in Figure 5, the strongest vibrations occurred on the right side at
point 1. In (d), representing 2500 rpm, the strongest vibrations occurred on the left side at point 1. In
(e), representing the 3000 rpm condition, a torsional mode was verified to occur around the xy axis.

(a) 800 rpm (idle) 24 Hz

(b) 1500 rpm 34 Hz

(c) 2000 rpm 74 Hz

(d) 2500 rpm 81 Hz

(e) 3000 rpm 93 Hz

Figure 8. Time data, Maximum vibration mode shape

4. Conclusions
In the present study, the changes in vibration depending on road and engine excitations were verified,
which are likely to be influenced by a muffler mounted on a vehicle body. Additionally, the effects of
actual vibrations on the muffler and body were analyzed.
The conclusions obtained from these results are as follows:

502

A Study on Vibration Characteristics of Vehicle Muffler Depending on Excitation Sources

Yoo-jong Lee, Doo-seuk Choi, Key-sun Kim

1) The resonance frequency of the exhaust system was verified to be 16.6 Hz using sine sweep
excitation, and this frequency was confirmed to reflect the design for avoiding idle-state engine
excitation frequency (13 Hz).
2) Strong road-excitation-induced vibrations were seen at 11 Hz in the y and z directions, but their
effect on body vibration was insignificant.
3) The effect of the muffler vibrations transferred to the body via the hanger was insignificant;
however, given the fact that the increase in muffler vibration level entails a change in body
vibration, body vibration should be reflected in the muffler design.
4) Vibrations at the engine speed of 1500 rpm showed more stability than under other conditions,
and muffler vibrations at 2500 rpm and above showed larger variable breadths than under other
conditions.
5) Engine excitationinduced muffler vibrations were influenced not only by the z direction, but
also by the y and x directions.
6) The occurrence of muffler mode shapes led to differing vibration values depending on
measurement positions, and stronger muffler vibrations occurred in the idle state than when the
vertical vibrations at the mount point connected to the hanger were 1500 rpm and above.
7) Frequencies of the mode shapes manifested at 2494 Hz depending on engine speed; the muffler
vibrated in the xy oblique axis direction, and bending (at 800 and 2000 rpm) and torsional (at
3000 rpm) modes occurred.

5. References
[1] LI Zhongjie, WANG Shuzong, LIAN Yongqing, Design and Simulation Study of Special Engine
Exhaust Muffler, Journal of vibration and shock v.27 no.5, pp.166 - 168, 2008
[2] Yuan Zhaocheng, Ma Jiayi, Li Shengcheng, Fang Hua, Study on the Simulation Design Method
About Exhaust Muffler of Vehicle, Sound and vibration; ICSV14 2007 , pp.423, 2007.
[3] SHEN Yinggang, FAN Qianwang, WEN jiaqing, A Study on Acoustic Performance of the Axial
Perforated Pipe Exhaust Muffler Considering the Stream, Noise and vibration control v.28 no.1 ,
pp.108 - 110,114 , 2008.
[4] Jong Sung Sa, Han Gil Kim, NVH of Automotive, CMG, Korea, 2010.
[5] Won Ku Lee, Ji Hoon Kim, Jae Yong Choi, Jin Ho Cho, A Experimental Approach for Buzz,
Squeak and Rattle Noise of the Interior Assembly Part in the Vehicle, KSME 2010, vol.11, pp. 14541459, 2010.
[6] Jae Young Kang, Key Sun Kim, Seok Min Choi, Tae Jin Choi "Vibration Analysis of Vehicle Seat
Depending on Driving Condition", Transactions of KSAE, Vol. 18, No. 3, pp.110-115, 2010.
[7] Wang Jie, Dong-peng Yue, The modal analysis of automotive exhaust muffler based on PRO/E
and ANSYS, Advanced Computer Theory and Engineering (ICACTE), 2010 3rd International
Conference on v.6 , pp.V6-251 - V6-253 , 2010

503