Beruflich Dokumente
Kultur Dokumente
Reprinted From: New SI Engine and Component Design and Engine Lubrication
and Bearing Systems
(SP-2093)
400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-0790 Web: www.sae.org
By mandate of the Engineering Meetings Board, this paper has been approved for SAE publication upon
completion of a peer review process by a minimum of three (3) industry experts under the supervision of
the session organizer.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise,
without the prior written permission of SAE.
SAE Permissions
400 Commonwealth Drive
Warrendale, PA 15096-0001-USA
Email: permissions@sae.org
Fax: 724-776-3036
Tel: 724-772-4028
ISSN 0148-7191
Copyright © 2007 SAE International
Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE.
The author is solely responsible for the content of the paper. A process is available by which discussions
will be printed with the paper if it is published in SAE Transactions.
Persons wishing to submit papers to be considered for presentation or publication by SAE should send the
manuscript or a 300 word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE.
Printed in USA
2007-01-0258
1
In order to estimate fatigue life of crankshafts, Prakash crankshafts were discussed. Their review also included
et al. [4] performed stress and fatigue analysis on three cost analysis and potential geometry optimizations of
example parts belonging to three different classes of crankshaft.
engines. The classical method of crankshaft stress
analysis (by representing crankshaft as a series of rigid In this paper, first dynamic load analysis of the
disks separated by stiff weightless shafts) and an FEM- crankshaft investigated in this study is presented. This
based approach using ANSYS code were employed to includes a discussion of the loading sources, as well as
obtain natural frequencies, critical modes and speeds, importance of torsion load produced relative to bending
and stress amplitudes in the critical modes. A fatigue load. FE modeling of the crankshaft is presented next,
analysis was also performed and the effect of variation of including a discussion of static versus dynamic load
fatigue properties of the material on failure of the parts analysis, as well as the boundary conditions used.
was investigated. This was achieved by increasing each Results from the FE model are then presented which
strain-life parameter (σf′, εf′, b and c) by 10% and includes identification of the critically stressed location,
estimating life. It was shown that strength and ductility variation of stresses over an entire cycle, and a
exponents have a large impact on life, e.g. a 10% discussion of the effects of engine speed as well as
increase of b leads to 93% decrease in estimated life. torsional load on stresses. A comparison of FEA
stresses with those obtained from strain gages of a
A geometrically restricted model of a light automotive crankshaft in a bench test is also presented. Finally,
crankshaft was studied by Borges et al. [5]. The conclusions are drawn based on the analysis preformed
geometry of the crankshaft was geometrically restricted and results presented.
due to limitations in the computer resources available to
the authors. The FEM analysis was performed in ANSYS LOAD ANALYSIS
software and a three dimensional model made of
Photoelastic material with the same boundary conditions The crankshaft investigated in this study is shown in
was used to verify the results. This study was based on Figure 1 and belongs to an engine with the configuration
static load analysis and investigated loading at a specific shown in Table 1 and piston pressure versus crankshaft
crank angle. The FE model results showed uniform angle shown in Figure 2. Although the pressure plot
stress distribution over the crank, and the only region changes for different engine speeds, the maximum
with high stress concentration was the fillet between the pressure which is much of our concern does not change
crank-pin bearing and the crank web. and the same graph could be used for different speeds
[9]. The geometries of the crankshaft and connecting rod
Shenoy and Fatemi [6] conducted dynamic analysis of from the same engine were measured with the accuracy
loads and stresses in the connecting rod component, of 0.0025 mm (0.0001 in) and were drawn in the I-DEAS
which is in contact with the crankshaft. Dynamic analysis software, which provided the solid properties of the
of the connecting rod is similar to dynamics of the connecting rod such as moment of inertia and center of
crankshaft, since these components form a slide-crank gravity (CG). These data were used in ADAMS software
mechanism and the connecting rod motion applies to simulate the slider-crank mechanism. The dynamic
dynamic load on the crank-pin bearing. Their analysis analysis resulted in angular velocity and angular
was compared with commonly used static FEA and acceleration of the connecting rod and forces between
considerable differences were obtained between the two the crankshaft and the connecting rod.
sets of analysis. Shenoy and Fatemi [7] optimized the
connecting rod considering dynamic service load on the
component. It was shown that dynamic analysis is the
proper basis for fatigue performance calculation and
optimization of dynamically loaded components. Since a
crankshaft experiences similar loading conditions as a Fz
connecting rod, optimization potentials of a crankshaft Fx Fy
could also be obtained by performing an analytical
dynamic analysis of the component.
2
Table 1: Configuration of the engine to which the
crankshaft belongs 100 40000
Mass of the piston assembly 0.417 kg -20 0 180 360 540 720-10000
-40
Connecting rod length 120.78 mm -60
-20000
40
Forces applied to the crankshaft cause bending and
35
torsion. Figure 1 demonstrates the positive directions
30 and local axis on the contact surface with the connecting
Cylinder Pressure (bar)
3
In many studies the torsional load is neglected for the at the fillets where the stresses are higher due to stress
load analysis of the crankshaft, and this is because concentrations. As a crankshaft is designed for very long
torsional load is less than 10 percent of the bending load life, stresses must be in the linear elastic range of the
[10]. In this specific engine with its dynamic loading, it is material. Therefore, all carried analysis are based on the
shown in the next sections that torsional load has no linear properties of the crankshaft material. The meshed
effect on the range of von Mises stress at the critical crankshaft with 122,441 elements is shown in Figure 6.
location. The main reason of torsional load not having
much effect on the stress range is that the maxima of The dynamic loading of the crankshaft is complicated
bending and torsional loading happen at different times because the magnitude and direction of the load
(see Figure 4). In addition, when the peak of the bending changes during a cycle. There are two ways to find the
load takes place the magnitude of torsional load is zero. stresses in dynamic loading. One method is running the
FE model as many times as possible with the direction
Figure 5 compares the magnitude of maximum torsional and magnitude of the dynamic force. An alternative and
and bending loads at different engine speeds. As can be simpler way of obtaining stress components is
seen in this figure, the maximum of total load magnitude, superposition of static loading. The main idea of
which is equal to the maximum of bending load superposition is finding the basic loading positions, then
decreases as the engine speed increases. The reason applying unit load on each position according to dynamic
for this situation refers to the load sources that exist in loading of the crankshaft, and scaling and combining the
the engine at 355 degree crank angle. At this crank stresses from each unit load. In this study both methods
angle these two forces act in opposite directions. The were used with 13 points over 720 degrees of crankshaft
force caused by combustion which is greater than the angle. The results from 6 different locations on the
inertia load does not change at different engine speeds crankshaft showed identical stress components from the
since the same pressure versus crankshaft angle is used two methods.
for all engine speeds. The load caused by inertia
increases in magnitude as the engine speed increases.
Therefore, as the engine speed increases, a larger
magnitude of inertia force is deducted from the
combustion load, resulting in a decrease of the total load
magnitude.
25
20
Force Magnitude (kN)
15
5
It should be noted that the analysis is based on dynamic
loading, though each finite element analysis step is done
0
in static equilibrium. The main advantage of this kind of
2000 2800 3600 analysis is more accurate estimation of the maximum
Engine Speed (RPM)
and minimum loads. Design and analyzes of the
crankshaft based on static loading can lead to very
conservative results. In addition, as was shown in this
Figure 5: Comparison of maximum and range of bending section, the minimum load could be achieved only if the
and torsional loads at different engine speeds analysis of loading is carried out during the entire cycle.
The minimum value of von Mises stress which is
obtained at the minimum load is needed for the stress
range calculation and considering it zero will lead to
smaller values for the stress range.
FE MODELING OF THE CRANKSHAFT
As the dynamic loading condition is analyzed, only two
5
The FE model of the crankshaft geometry has about 10 main loading conditions are applied to the surface of the
quadratic tetrahedral elements, with the global element crankpin bearing. These two loads are perpendicular to
length of 5.08 mm and local element length of 0.762 mm each other and their directions are shown in Figure 1 as
4
Fx and Fy. Since the contact surface between connecting 2 experiences the highest stress at this moment.
rod and crankpin bearing does not carry tension, Fx and Therefore, element number 2 was selected as the critical
Fy can also act in the opposite direction to those shown element. Figure 10 shows the maximum stress, mean
in Figure 1. Any loading condition during the service life stress, and stress range at the engine speed of 2000
of the crankshaft can be obtained by scaling and rpm at different locations. It can be seen that element
combining the magnitude and direction of these two number 2 not only has the maximum von Mises stress,
loads. but it also carries the largest stress range and mean
stress among other locations. This is important in fatigue
Boundary conditions in the FE model were based on the analysis since the range and mean stress have more
engine configuration. The mounting of this specific influence than the maximum stress. This is another
crankshaft is on two different bearings which results in reason for why having the stress history of critical
different constraints in the boundary conditions. One side elements are more useful than static analysis of the
of the crankshaft is fixed to the engine block by a ball crankshaft.
bearing and the other side is rolling over a journal
bearing. When under load, only 180 degrees of the a
4
bearing surfaces facing the load direction constraint the 5
7 c d
motion of the crankshaft. Therefore, a fixed semicircular A
surface as wide as the ball bearing width was used to 6 1 A-A b
model that section. This indicates that the surface can
not move in either direction and can not rotate. The other
side was modeled as a fixed thin semicircular ring which
A
only holds the crankshaft centerline in its original position
3
and acts as a pivot joint. In other words, the journal
2
bearing is modeled in a way that allows the crankshaft to
rotate about axis 1 as well as slide in direction 3 as
occurs in a journal bearing. These defined boundary
Figure 8: Locations on the crankshaft where the stress
conditions are shown in Figure 7. Boundary conditions variation was traced over one complete cycle of the
rotate with the direction of the load applied such that the engine, and locations where strain gages were mounted
semicircular fixed surface and ring face the direction of
the load.
150
Stress Magnitude (MPa)
Fixed surface
in all degrees 100
2 Fixed ring in of freedom
directions 1 & 2 over 180o
3 1
over 180o
50
5
Maximum Minimum Range Mean stress at location number 2 shown in Figure 9 remains
250
the same with and without considering torsional load.
This is due to the location of the critical point which is not
influenced by torsion since it is located on the crankpin
200
bearing. Other locations such as 1, 6, and 7 in Figure 8
experience the torsional load. Figure 12 shows changes
in minimum, maximum, mean, and range of von Mises
150 stress at location 7 with considering torsion and without
Stress Magnitude (MPa)
100
-50
Location Number 80
60
-40
Figure 11 shows the effect of engine speed on minimum,
maximum, mean and range of stress. This figure -60
indicates the higher the engine speed, the lower the von -80
Mises stress. It should, however, be noted that there are Engine Speed (RPM)
150
according to the fixture of the test assembly. The fixture
constraints the motion of the shaft on the left side of the
100 crankshaft in Figure 8 and a load is applied on the right
50 side of the crankshaft with a moment arm of 44 cm.
Therefore, the crankshaft is experiencing bending as a
0
cantilever beam. Applying load in the direction of axis 2
-50 in Figure 7 will result in stresses at locations a and b, and
1500 2000 2500 3000 3500 4000 applying load in the direction of axis 1 in the same figure
Engine Speed (RPM) will result in stresses at locations c and d. Analytical
calculations based on pure bending equation, Mc/I, show
the magnitude of stresses to be the same and equal to
Figure 11: Variation of minimum stress, maximum 72 MPa at these locations, for a 890 N load. The values
stress, mean stress, and stress range at location 2 on obtained from experiments are tabulated in Table 2. FEA
the crankshaft as a function of engine speed results are also shown and compared with experimental
results in this table. As can be seen, differences between
FEA and strain gage results are less than 7 percent for
The effect of torsional load was discussed in the load different loading conditions. This is an indication of the
analysis section, and was pointed out that it has no effect accuracy of the FE model used in this study.
on the stress range of the critical location. The von Mises
6
Table 2: Comparison of stress results from FEA and
strain gages located at positions shown in Figure 8 200
2
7
modeling due to the relatively complex geometry of Using a Geometrucally Restricted Finite Element
the crankshaft. Model”, SAE Technical Paper No. 2002-01-2183,
8. Using the rainflow cycle counting method on the Society of Automotive Engineers
critical stress history plot shows that in an entire 6. Shenoy, P. S. and Fatemi, A., 2006, “Dynamic
cycle only one peak is important and can cause analysis of loads and stresses in connecting rods,”
fatigue damage in the component. IMechE, Journal of Mechanical Engineering Science,
Vol. 220, No. 5, pp. 615-624
7. Shenoy, P. S. and Fatemi, A., "Connecting Rod
REFERENCES Optimization for Weight and Cost Reduction", SAE
Paper No. 2005-01-0987, SAE 2005 Transactions:
1. Henry, J., Topolsky, J., and Abramczuk, M., 1992, Journal of Materials and Manufacturing
“Crankshaft Durability Prediction – A New 3-D 8. Zoroufi, M. and Fatemi, A., "A Literature Review on
Approach,” SAE Technical Paper No. 920087, Durability Evaluation of Crankshafts Including
Society of Automotive Engineers Comparisons of Competing Manufacturing
2. Guagliano, M., Terranova, A., and Vergani, L., 1993, Processes and Cost Analysis", 26th Forging Industry
“Theoretical and Experimental Study of the Stress Technical Conference, Chicago, IL, November 2005
Concentration Factor in Diesel Engine Crankshafts,” 9. Fergusen, C. R., 1986, “Internal Combustion
Journal of Mechanical Design, Vol. 115, pp. 47-52 Engines, Applied Thermo Science,” John Wiley and
3. Payar, E., Kainz, A., and Fiedler, G. A., 1995, Sons, New York, NY, USA
“Fatigue Analysis of Crankshafts Using Nonlinear 10. Jensen, E. J., 1970, “Crankshaft strength through
Transient Simulation Techniques,” SAE Technical laboratory testing,” SAE Technical Paper No.
Paper No. 950709, Society of Automotive Engineers 700526, Society of Automotive Engineers
4. Prakash, V., Aprameyan, K., and Shrinivasa, U., 11. Stephens, R. I., Fatemi, A., Stephens, R. R., and
1998, “An FEM Based Approach to Crankshaft Fuchs, H. O., 2001, “Metal Fatigue in Engineering,”
Dynamics and Life Estimation,” SAE Technical 2nd edition, John Wiley and Sons, New York, NY,
Paper No. 980565, Society of Automotive Engineers USA
5. Borges, A. C. C., Oliveira, L. C., and Neto, P. S.,
2002, “Stress Distribution in a Crankshaft Crank