Bulletin of the Transilvania University of Braşov

Series I: Engineering Sciences • Vol. 5 (54) No. 1 - 2012

DIESEL ENGINE TURBOCHARGERS:

ANALYSIS AND TESTING

C.C. BORICEAN1 Gh.A. RADU1

Abstract: Turbochargers have been and still are a necessity when we talk
about internal combustion engines downsizing. The turbocharger is the
engine part which produces multiple vibrational and noise phenomena
during functioning. Regarding the optimization related to noise and vibration
produced by turbochargers, in this paperwork, there were accomplished
several software simulations and laboratory tests. The simulations were
accomplished in order to highlight the main natural frequencies that could
occur during the turbocharger functioning. For validating the data sets
obtained by simulations there were performed test rig tests focused on
rotordynamic stability of turbochargers.

Key words: turbocharger, rotordynamics, natural frequencies, modal analysis.

1. Introduction vibrations than the turbochargers with

classical hydrodynamic bearings. The noise
From the various methods used for and vibrational phenomena produced at the
supercharging internal combustion engines level of turbochargers rotors, which are the
the most used one, involving fabrication only elements in motion, are transmitted
costs, which are relatively low in further to the housing, where there are
comparison with other methods and also amplified, resulting in a complex vibrational
due to high adaptability on various engines, phenomena, in some cases the generated
is the turbocharger. Modern solutions of vibration could be captured by human ear.
turbochargers use, in order to sustain the
turbocharger shaft, hybrid rolling bearings, 2. Objectives
while classical turbochargers use
hydrodynamic bearings. The main The main objectives of this paperwork is
advantage presented by using hybrid rolling focused on highlighting some aspects
bearings, consists on a higher durability related to turbocharger rotordynamic
unlike the case of hydrodynamic bearings. stability and also on highlighting aspects
These rolling bearings involve also some related to vibrational phenomena that
disadvantages which are related to higher occur at the level of turbocharger entire
noise produced during functioning, than the assembly. For accomplishing the main
hydrodynamic bearings. Also some tests objectives there were involved software
highlighted the fact that turbochargers rotors simulation methods and also test rig tests
sustained by rolling bearings produce higher accomplished with specialised vibration

1
Dept. of Automotive and Mechanical Engineering, Transilvania University of Braşov.
2 Bulletin of the Transilvania University of Braşov • Series I • Vol. 5 (54) No. 1 - 2012

monitoring equipment. For accomplishing rotor stability (amplitude) is established on

software simulations there was used a value which doesn’t coincide with the
Matlab-Simulink software solution and central geometric axis of the rotor.
also 3D modelling techniques accomplished At a stiffness k = 1e6 N/m the rotational
using Catia V5 software in order to use the movement amplitude can be observed in
generated geometry in simulations (modal Figure 2.
analysis). The experimental research was
accomplished using two categories of
methods: static and dynamic. The static
methods are related to accomplishing
laboratory modal analysis and dynamic
methods refer to validation of some
vibrational phenomena on the turbocharger
specialised and approved test rig.

3. Simulation of the Dynamic Behaviour Fig. 2. The y3 (displacement of bearing 2

of the Turbocharger Rotor on y direction) - displacement for
k = 1e6 N/m
Been the main source of noise and
vibration the turbocharger rotor needs a We could observe that also in the case of
special treatment. In the first step there was Figure 1 the rotor is stabilizing its motion
accomplished a mathematical model of the on an axis which doesn’t coincide with the
turbocharger rotor sustained in central geometric axis. In order to highlight
hydrodynamic bearings, model which is the influence that the stiffness has over the
capable on highlighting the rotational motion of the rotor the obtained simulation
movement amplitudes of the rotor. The data is summarized in Table 1.
mathematical model [2], [4], [10] was
designed in such ways that the rotor Displacement for bearings Table 1
stability could be controlled by modifying Displacement Displacement Stiffness
some parameters related to: eccentricity; for bearing 1 for bearing 2 value k
mass; stiffness and damping [6]. Using Y1[m] Y2 [m] [N/m]
Matlab software there were accomplished 11.8e-6 11.4e-6 k=0
several simulations for different rotational 11.7e-6 11.1e-6 k=1
speeds and bearing stiffness [3], [9]. It 11.9e-6 11.8e-6 k = 1e1
could be observed in Figure 1. That at a 11.95e-6 11.6e-6 k = 1e2
value of k = 0 N/m of bearing stiffness, the 12e-6 11.65e-6 k = 1e3
11.97e-6 11.35e-6 k = 1e4
11.3e-6 11.5e-6 k = 1e5
8.1e-6 12.5e-6 k = 1e6

4. Modal Analysis of the Turbocharger

For highlighting the main natural

frequencies that could appear during the
turbocharger functioning it was accomplished
Fig. 1. The y2 (displacement of bearing 1 on the 3D modeling of the turbocharger rotor
y direction) - displacement for k = 0 N/m involving also the bearings, in order to use
Boricean, C.C., et al.: Diesel Engine Turbochargers: Analysis and Testing 3

the modelled parts for accomplishing For accomplishing the modal analysis
software modal analysis [1], [7], [11]. The there was used the vibration Pulse 12
main natural frequencies obtained for the platform supplied by Bruel & Kjaer.
rotor part are as follows: The used platform and transducer and
impact hammer connection are presented
Mode shapes and frequencies Table 2 in Figure 5.
Mode shape Frequency [Hz]
1 409.3
2 410.46
3 666.44
4 669.89
5 858.38
6 921.37
7 5003.1
8 5022
9 5880.1
10 5912.4
Fig. 5. Acquisition platform

The considered chanel connection are as

follows:
• for channel 1 the impact hammer;
• for channel 2 accelerometer no. 1;
• for channel 3 accelerometer no. 3.
The accelerometers used are 4507 B types
and the impact hammer is 8206-003 type.
Fig. 3. Rotor mode shape no.4 The method used in modal analysing was the
impact hammer method. For a correct test
The natural frequencies obtained for the data sets there was performed accelerometer
turbocharger bearing, are as follows: calibration using the 4294 type calibrator
supplied by the same producer. The natural
frequencies obtained for the turbocharger
rotor are highlighted in Figure 6 and the
exact values are presented in Table 3.

Fig. 4. Bearing mode shape

The first mode shape is characterized by

the natural frequency of 17820 Hz for the
outer ring and 2934 Hz for the entire bearing.
In order to validate the results obtained
during the simulations, it was accomplished
a modal analysis test using specialized Fig. 6. Frequency response function for
laboratory equipment. turbocharger rotor
4 Bulletin of the Transilvania University of Braşov • Series I • Vol. 5 (54) No. 1 - 2012

Acceleration versus frequency Table 3

Damping
Frequency Acceleration coefficient
No
[Hz] [(m/s2)/N] ζ [%]
1 76 228 m 0.059
2 105 209 m 0.484
3 140 196 m 0.287
4 227 146 m 0.341
5 269 194 m 0.157
6 298 235 m 0.155
7 346 169 m 0.167
8 394 241 m 0.131 Fig. 8. Transducer placement and impact
9 432 215 m 0.1
point
10 477 358 m 0.099

The configuration and placement for the

transducers and the impact point for the
hammer were maintained in the same
position as used in the software simulation.
The used configuration is presented in
Figure 7.

Fig. 9. Fourier spectrum signal for testes

bearing

For validating the laboratory test

performed on the bearing there were used
analytical methods. The analytical method
used is based on the classical theory for
identifying the bearing natural frequencies
of the following form [5]:
• Rotating frequency of the inner ring:

Fig. 7. Transducer and impact point for the

f i = n / 60 ; (1)
turbocharger rotor

The configuration and placement for the • Rotating frequency of the cage to the
transducers and the impact point for the outer ring:
hammer were maintained in the same
position as used in the software simulation. f ce = n(1 − DW cos α / d m ) / 120 ; (2)
The used configuration is presented in
Figure 8. • Rotating frequency of the cage to the
The natural frequencies obtained for the inner ring:
turbocharger bearing are presented in
Figure 9. f ci = n(1 + DW cos α / d m ) / 120 ; (3)
Boricean, C.C., et al.: Diesel Engine Turbochargers: Analysis and Testing 5

• Rotating frequency of the rolling element maintaining, in this case the same
to the outer ring: specifications as the factory ones. It were
performed several tests in which the speed
[1 − ( DW cos α / d m ) 2 ] of the turbocharger rotor achieved a
fb = n ⋅ dm , (4)
(120 ⋅ DW ) maximum speed value of 86000 rpm. In
Figure 10, it could be observed the
where: n - rotational speed; Dw - ball turbocharger mounted on the test rig.
diameter; dm - medium diameter of the The rotor precession for 50000 rpm is
bearing; α - contact angle. highlighted in Figure 11.

5. Dynamical Testing of the Considered

Turbocharger

In order to validate the data sets obtained

by simulations it were performed several
test rig dynamical tests on a turbocharger
specialized and approved test rig. The test
rig used in performing the tests was
supplied by Schenk. For a correct
measurement of the dynamical behaviour
of the turbocharger rotor it was considered
the simulation of different parameters on Fig. 11. Rotor precession at 50777 rpm
the test that are related to: oil temperature;
rotor acceleration time, as in real The rotor precession for 86000 rpm is
functioning conditions of the tested highlighted in Figure 12.
turbocharger [8]. It is to be mentioned that
the tested turbocharger is a new generation
of diesel engine turbocharger supplied by
Garrett for Mercedes-Benz for a 3.0 litre
V6 diesel engine.
In order to establish a reference unbalance,
the first test was accomplished by using
the turbocharger in undismantled condition,

Fig. 12. Rotor precession at 86320 rpm

6. Results and Conclusions

Consulting also the technical specifications

supplied for different types of turbochargers
we can mention the followings:
• the unbalance at 86000 rpm is 586 um/s,
a relatively high value fir this category of
Fig. 10. Turbocharger mounted on test rig turbochargers;
6 Bulletin of the Transilvania University of Braşov • Series I • Vol. 5 (54) No. 1 - 2012

• mathematical models can’t highlight References

the complex functioning behaviour of a
turbocharger, due to the complexity of 1. Bathe, K.J., Wilson, E.L.: Numerical
mathematical equations necessary in this Methods in Finite Element Analysis.
cases; New Jersey. Pretince-Hall Inc., 1976.
• the vibrations produced by turbochargers 2. Deutsch, I.: Rezistenţa materialelor
are generally generated by rotor precessions; (Strenght of Materials). Bucureşti.
• vibrational phenomena occurring at the Editura Tehnică, 1976.
level of the turbocharger are complex and 3. Ehric, F.F.: Handbook of
also are strongly influenced by vibrations Rotordynamics. New York. McGraw-
generated by the functioning of internal Hill Inc., 1992.
combustion engines; 4. Esfandiari, R., Lu, B.: Modeling and
• the rotor stabilizes the movement Analysis of Dynamic Systems. CRC
amplitude over an axis that doesn’t concur Press, 2010.
with the central geometric axis OX of the 5. Gafiţanu, M.: Rulmenţi. Vol. I, II
considered model. The bearings stiffness (Bearings. Vol. I, II). Bucureşti.
plays an important role over the rotor Editura Tehnică, 1985.
precession. The vibrational phenomena of 6. Harris, M.: Shock and Vibration.
the turbocharger are not only related to the Handbook Fifth Edition. New York.
actual vibrations of its component parts, McGraw-Hill, 2002.
but also to the vibrations that are induced 7. Munteanu, M.: Metoda elementelor
by the functioning of the internal finite (Finit Element Method). Braşov.
combustion engine, which is the main Editura Universităţii Transilvania,
source of vibrations at the level of 1997.
vehicles. Unbalanced masses and 8. Radeş, M.: Dynamics of Turbo-
eccentricity lead to vibration phenomena machinery. Vol. I, II, III. Bucureşti.
increasing. Editura Printech, 2007.
9. Roşca, I.C.: Vibraţii mecanice
Acknowledgements (Mechanical Vibrations). Braşov.
Editura Infomarket, 2002.
Author: Boricean Cosmin C-tin. This 10. Wu, J.J.: Prediction of Lateral
paper is supported by the Sectoral Vibration Characteristics of a Full
Operational Programme Human Resources Size Rotor-Bearing System by Using
Development (SOP HRD), financed from Those Scale Models. In: Journal of
the European Social Fund and by the Finite Elements in Analysis and
Romanian Government under the contract Design 43 (2007) Issue 10, p. 803-816.
number POSDRU/88/1.5/S/59321. 11. *** Ansys 13.0: Help.