Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 26

CRANKTRAIN DYNAMICS SIMULATION
Lubomr Drpal
1
, Pavel Novotn, Vclav Ptk, Martin Beran
Summary: This paper presents advanced and efficient computational approaches to an
analysis of cranktrain dynamics. The approaches are represented by complex
computational model of a powertrain (virtual engine) which is a powerful tool for
the solution of structural, thermal and fatigue problems. The computational model
of dynamics includes flexible parts and hydrodynamic approach for slide bearing
solution. Presented virtual engine is assembled as well as numerically solved as a
Multi-Body System and it is focused on cranktrain vibrations identification.
Virtual engine results are validated by measurements on a compression-ignition
in-line four-cylinder engine.
Key words: cranktrain, dynamics, vibrations, virtual engine.

INTRODUCTION
Contemporary powertrains are complex thermomechanical systems improved by large
and gradual development. Vehicle producers are still increasing the engine parameters such as
the engine performance, together with significantly decreasing the fuel consumption. In
addition, low levels of noise, vibrations and engine emissions are required by national
legislature of the European, US or Japanese markets.
However, an increase in the engine performance often leads to an increase in the
powertrain noise and vibrations. These problems can be solved by experimental or
computational approaches. Experimental methods are time-consuming and quite expensive
therefore computational methods are in use increasingly. In the sphere of powertrain
dynamics, experimental methods are often applied for complex computational model
validations because computational models can provide very exact results but only on the
condition that exact inputs are included.
All computational and measurement methods reported here are applied to four-stroke
in-line four-cylinder compression-ignition engine of these parameters:
Bore: 105 mm,
Stroke: 120 mm,
Displacement: 4.15 l,
Compression ratio: 17.8 : 1,
Rated output: 96 kW @ 2200 rpm.

1
Ing. Lubomr Drpal, Brno University of Technology, Faculty of Mechanical Engineering, Institute of
Automotive Engineering, Technick 2896/2 , 616 69 Brno, Tel.: +420 541 142 264, Fax: +420 541 143 354,
E-mail: [email protected]
Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 27
1. COMPUTATIONAL METHODS
1.1 Virtual Engine
A complex computational model of the engine (in other words a virtual engine) is
solved in the time domain. This enables different physical problems including various non-
linearities to be incorporated. The virtual engine is assembled as well as numerically solved in
MBS ADAMS. ADAMS is a general code and enables integration of user-defined models to
be made directly using ADAMS commands or user-written FORTRAN or C++
subroutines (2).
In general, the virtual engine includes all significant components necessary for noise,
vibrations and harshness (NVH) or fatigue analyses. The included modules are represented in
particular by a cranktrain, a valvetrain, a gear timing drive with an fuel injection pump and a
rubber damper. Following analyses just deal with a cranktrain as the main module of the
virtual engine.
A cranktrain module consists of solid model bodies, linearly elastic model bodies and
constraints between them.

Source: Author
Fig. 1 Virtual engine cranktrain with dynamometer model

The solid model bodies are:
Piston assembly,
Connecting rod assembly,
Crankshaft pulley,
Additional flywheel mass,
Dynamometer rotor.

The linearly elastic model bodies are modally reduced Finite Element (FE) models
suitable for dynamic simulation. These are:
Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 28
Crankshaft,
Major part of used flywheel.

A dumb-bell shaft connecting a flywheel with a dynamometer rotor is represented by a
body with defined torsional stiffness and damping. These characteristics are adjusted on
account of torsional vibrations measurement.
Some subsystems having indispensable influence on cranktrain vibrations are modelled
by their inertia characteristics or power requirement with corresponding speed ratio
(valvetrain, balancing shafts, and an injection unit).
The interaction between the crankshaft and the engine block is ensured via a non-linear
hydrodynamic journal bearing model, where pre-calculated force databases obtained when
solving separate hydrodynamic problem are used.

Source: Author
Fig. 2 Indicated pressure for virtual engine excitation

The virtual engine is excited by means of cylinder pressure defined by a high-pressure
measurement and via inertial forces from moving parts. Simulation starts from 1100 rpm and
is carried out to 2400 rpm (pressures between 22002400 rpm are extrapolated).
1.2 Validation of Computational Model
The determination of cranktrain torsional vibrations represents a fundamental step in the
powertrain development. These vibrations can significantly influence fatigue of crankshaft.
Load of each subsystem coupled with the cranktrain is also affected.
An experimental determination of torsional vibrations using, for example, laser
vibration tools is an advance and it can help to validate computational models.
A summary of a cranktrain torsional behaviour can provide a harmonic analysis of
torsional vibrations determined from crankshaft pulley angular velocity.
Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 29

Source: Author
Fig. 3 Harmonic analysis of crankshaft pulley torsional vibrations without torsional damper

Figure 3 shows harmonic analysis results of computed and measured cranktrain
torsional vibrations originated in the powertrain without a torsional damper. Brel & Kjr
Rotational Laser Vibrometer Type 2523 and POLYTEC 4000 Series Laser Vibrometer are
used for model verifications. These experimental tools enable measurements of angular
velocities of arbitrary rotating parts. For in-line four-cylinder four-stroke internal combustion
engine, there are defined the main harmonic orders (which have generally the biggest
amplitudes) as integral multiples of two. This fact is confirmed by computation as well as
measurement whereas relatively very good conformity of the computation and the
measurement is found. Possible variations are caused by not simultaneous performance of
torsional vibrations measurement and cylinder pressure measurement due to the substantial
volume of data captured per time unit.
2. CRANKSHAFT DYNAMICS SOLUTION RESULTS
The virtual engine can provide a large number of results which are necessary for
different stages of the new powertrain development. The results presented in this paper do not
provide complete information about the target powertrain because the aim of the presented
results is to give insight into some cranktrain influences, mainly those which can be validated
by measurements.
Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 30
2.1 Crankshaft Torsional Vibrations
Vibrations are common to internal-combustion engine crankshafts nevertheless
torsional vibrations in fully embedded shafts (each crank pin is next to two main pins) which
is caused by shaft flexibility is the most dangerous one. The crankshaft mechanism generates
alternating torque due to time varied combustion pressure in conjunction with the alternating
effect of reciprocating parts inertia. This torque brings the elastic crankshaft in vibrations
about the axis of rotation which is known as forced torsional vibrations. Such vibrations are
superimposed on the crankshaft oscillation that appears due to the engine run irregularity and
the static torsion from tangential forces. This type of vibrations can cause cracking and
crankshaft failure, and therefore it is very dangerous crankshaft effort (3).

Source: Author
Fig. 4 Harmonic analysis of crankshaft torsional vibrations without torsional damper

Figure 4 presents harmonic analysis results of computed crankshaft torsional vibrations
of the powertrain without any torsional damper. In this case, by crankshaft torsional
vibrations is meant angular deformation between a crankshaft pulley and a flywheel flange.
Peak-to-peak value represents a difference between maximal and minimal values for one
engine cycle. The eighth harmonic order is the most dominant and resonance of this order
occurs at engine speed 2250 rpm. The nominal engine speed is very close to this critical
engine speed, therefore, the torsional damper application seems to be appropriate to decrease
torsional vibrations and increase cranktrain component fatigue.
Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 31
2.2 Crankshaft Axial Vibrations
Crankshaft axial vibrations are a mode of vibrations resulting from the crankshaft
expanding and compressing along its axis of rotation due to cranktrain elasticity, alternating
acting forces and journal damping. This mode of vibrations affects the crankshaft web, its
pins and the other related components.

Source: Author
Fig. 5 Harmonic analysis of crankshaft axial vibrations without torsional damper

Figure 5 shows harmonic analysis results of crankshaft axial vibrations of the
powertrain obtained by dynamic simulations. The dominant orders are the second one and the
fourth one. The level of peak-to-peak half value is not very critical for whole operating speed
nevertheless it can be useful for design and location of the thrust bearing especially with
reference to a timing-drive mechanism.
2.3 Main Bearing Load
The virtual engine is able to give various bearing load data. These can be represented by
bearing forces and moments, dimensionless pin tilting angle in the plane of narrowest oil film
gap, dimensionless tilting angle in the plane perpendicular to the plane of the narrowest oil
film gap, independent bearing clearance and by relative eccentricity.
Relative eccentricity is defined as

e
D d
(1)
Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 32
where e [mm] is an absolute eccentricity of the pin centre against the centre of the
bearing shell, D [mm] is the bearing shell diameter and d [mm] stands for the pin diameter.
All is based on the assumption of an absolutely stiff pin and shell. If the centres of the pin and
shell are identical, = 0. If the pin and shell surfaces hit, the relative eccentricity reaches = 1
(3).

Source: Author
Fig. 6 Journal bearing eccentricity

Regarding the bearing life, the highest relative eccentricity in the course of the working
cycle seems to be the most important. This is also, among others, due to the fact that the
bearing response (represented by decrease in the oil film thickness) to the drive pulse is not
immediate (3).
Figure 7 presents relative eccentricity maxima of all main bearings through the entire
engine speed range. Even in the worst case, relative eccentricity amounts to approximately
= 0.92, and that is why each main bearing can be regarded as acceptable in respect to
bearing load. However, obtained bearing load data can be processed by specialized
computational software to get detail information such as bearing pressure distribution. Those
data may be then used for design of bearings, bearing clearances and shape and dimensions of
oil distribution grooves.
Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 33

Source: Author
Fig. 7 Relative eccentricity of main bearings

CONCLUSION
The introduced virtual engine presents a powerful tool for powertrain vibrations
analyses. It is able to incorporate not only linear but also non-linear behaviour of powertrain
components and subsystems.
The complex computational model should be verified by some suitable experimental
methods. The results of vibrations analyses are shown at crankshaft torsional and axial
vibrations and main bearing load that can be further used for vibrations eliminating steps such
as application and design of a torsional damper and main and thrust bearing layout. The
influence of all the modifications and new designs can be simulated and proved also by this
virtual engine. The results of simulations are also very valuable sources for more detail
analyses based on specialized software.
ACKNOWLEDGEMENT
Published results have been achieved with the financial help of the Ministry of Education,
Youth and Sports, project reg. no. CZ.1.05/2.1.00/01.0002 NETME Centre New
Technologies for Mechanical Engineering and the project FSI-S-11-8 granted by specific
university research of Brno University of Technology.
REFERENCES
(1) NOVOTN, P. Virtual Engine A Tool for Powertrain Development, Inaugural
Dissertation, Brno: Brno University of Technology, Czech republic, 2009.
Number 4, Volume VII, December 2012

Drpal, Novotn, Pitk, Beran: Cranktrain dynamics Simulation 34
(2) NOVOTN P., PTK V. New Efficient Methods for Powertrain Vibration Analysis,
Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile
Engineering, May/2010, pp. 611629. ISSN 0954-4070.
(3) DRPAL, L., NOVOTN P., PTK V., AMBRZ, R. Crankshaft Development of
Two-stroke Compression-ignition Engine With Contra-running Pistons, MECCA
Journal of Middle European Construction and Design of Cars, Volume IX, Number 01,
esk vysok uen technick v Praze MECCA, 2011, ISSN 1214-0821.
(4) KOOUEK, J. Vpoet a konstrukce spalovacch motor II, SNTL Nakladatelstv
technick literatury, Prvn vydn, Praha 1983.
(5) MILEK, S. Modln analza mechanickch kmit, Vydavatelstv VUT, Dotisk
prvnho vydn, Praha, 1996. ISBN 80-01-00872-X
(6) MSC.ADAMS/Solver C++. Users Manual 2005r2, MSC.Software Corporation,
California, 2006.