Accepted Manuscript

Fatigue life analysis of crawler chain link of excavator

Huanyu Zhao, Guoqiang Wang, Haotong Wang, Qiushi Bi, Xuefei

Li

PII: S1350-6307(16)31252-3
DOI: doi: 10.1016/j.engfailanal.2017.04.034
Reference: EFA 3114
To appear in: Engineering Failure Analysis
Received date: 29 December 2016
Revised date: 26 April 2017
Accepted date: 28 April 2017

Please cite this article as: Huanyu Zhao, Guoqiang Wang, Haotong Wang, Qiushi Bi,
Xuefei Li , Fatigue life analysis of crawler chain link of excavator, Engineering Failure
Analysis (2017), doi: 10.1016/j.engfailanal.2017.04.034

This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
 ACCEPTED MANUSCRIPT

Fatigue life analysis of crawler chain link of excavator

Huanyu Zhao Guoqiang Wang Haotong Wang Qiushi Bi Xuefei Li*

College of Mechanical Science and Engineering, Jilin University, Changchun, 130025, P.R. China
ABSTRACT: In this study, the tension of crawler chain link of excavator is measured by experimenting in
horizontal straight, pivot steering, and differential steering conditions. A virtual prototype model of excavator is
established and its accuracy is verified on the basis of comparing the simulated and experimental tension of chain

PT
link in the abovementioned three conditions and adjusting the parameters of the model. The fracture surface
morphology of chain link crack is analyzed by scanning electron microscopy. The fracture surface morphology
conforms to the fatigue failure. Finite element analysis result indicates that the maximum stress of chain link is

RI
lower than the yield strength of material. The area of the maximum stress of chain link presents a crack, which
indicates the fatigue failure of chain link. A rigid-flexible coupling virtual prototype model for simulation is

SC
established, which consists of a group of chain links of flexible body and other components of rigid bodies. The
chain link operates in an entire cycle with the crawler in horizontal straight, climbing, and pivot steering conditions.
The stress of key points in the cracked area and the fatigue life of chain link are obtained from the simulation. The
NU
comparison between the simulation and experimental data confirms the accuracy of the established virtual
prototype model of excavator. Accordingly, the fatigue life of chain link is analyzed, which is usually difficult to
achieve by experiment. The proposed analytical method for the fatigue life of components is easy and efficient and
MA

can serve as a reference for the fatigue life analysis of other key components of crawler.
Keywords: Excavator; Virtual prototype model; Experiment; Chain link; Fatigue life
E D
PT
CE
AC
 ACCEPTED MANUSCRIPT

1. Introduction

Crawler is one of the common traveling gears in construction machinery and is extensively used in
construction machinery, excavators, rollers, and dozers. The crawler supports the entire weight of machines and
works in harsh environment; hence, its components, such as track wheel [1], two-wheel bogie [2], and drive shaft
[3], usually wear and fracture. Chain link is the key component of a crawler [4, 5] and its fracture occurs when
stress is concentrated in an area caused by designing-in defects and when the mechanical properties of material are
lower than the specified values caused by manufacturing-in defects [69].
During the crawler operation, the chain link is subjected to multiple loads, including the entire weight of the
machine with track wheels and the tension generated by the meshing between the teeth of the sprocket and the pin

PT
of chain links. These cyclical loads can cause the fatigue failure of chain link, as shown in Fig. 1. The fatigue
failure of chain link leads to the incapability of tracked vehicles to travel and the significant economic losses caused

RI
by the downtime. Thus, the fatigue life of chain link must be investigated.

SC
NU
MA
D

Fig. 1. Fatigue crack of chain links

E

The fatigue life prediction of components, such as the cable shovel boom of heavy mining equipment [10, 11],
PT

the mining dragline cluster [12], and the high-speed railway brake disc [13], has been examined. The failure
fracture of crawler chain link is due to the mechanical properties of material that are lower than the standard values
and the existence of macro- and microcracks in the material structure [7]. Although the mechanical properties of
CE

material of chain links meet the requirements, cracks are still generated and fatigue damage may occur. Thus, the
fatigue life of chain link needs to be analyzed. Predicting the fatigue life of components using the experimental
approach requires much time and money. Such prediction can be conducted using the theoretical method if the load
AC

spectra are obtained on the basis of the actual load on the experimental prototype. However, this task is difficult for
components with complex structure or on harsh working environment. Therefore, an accurate virtual prototype
model must be established, the load spectrum of key components must be obtained, and the fatigue life of
components must be analyzed to improve design efficiency and reduce costs. Ying [14] established the virtual
prototype, electrical, and flexible models of the dragline front-end assembly. The author also obtained the major
motions, machine behavior, machine operating loads, and structural member stresses from the co-simulation.
Furthermore, the author used the results of stresstime history for the structural fatigue life evaluation by
PalmgrenMiners rule. Son [15] evaluated the fatigue life of hollow railway axles with tapered bore surfaces and
analyzed the fatigue life of the wheelrail contact force using the dynamic simulation results of multi-body
dynamics model. Fewer studies have attempted to predict the fatigue life of the crawler chain link. Moreover, the
chain links bear various loads that are difficult to measure through experiments. In addition, the accuracy of the
 ACCEPTED MANUSCRIPT

simulation result, which is obtained by extracting the load of key components using virtual prototype simulation
analysis, has not been verified. This study focuses on establishing the accuracy of the virtual excavator prototype
by conducting experiments on the entire machine to predict the fatigue life of chain link.
This paper is organized as follows. In Section 2, the experimental excavator prototype is described, and the
fracture surface morphology of the chain link crack is observed via scanning electron microscopy. The virtual
excavator prototype is presented in Section 3. Experimental and simulated results of the chain link tension in three
representative conditions are compared, and the accuracy of the virtual prototype model is verified in Section 4.
Section 5 presents the finite element analysis and the investigation of the fatigue life of chain links by developing a
rigid-flexible coupling virtual prototype. Finally, conclusions are presented in Section 6.

PT
2. Experiment

Verifying the accuracy and credibility of the virtual prototype model of excavator is difficult using the

RI
theoretical method. Therefore, the important performance indicators of the crawler of excavator need to be
measured through experimentation to confirm the accuracy of the virtual prototype model. Data basis for obtaining

SC
an accurate virtual prototype model must be provided, and observing the fracture surface morphology of chain link
crack by SEM is necessary. NU
2.1. Calibration of chain links

Three groups of chain links are calibrated. Strain gauges are installed on the upper inner surface of the chain
links, and the temperature compensating plates are connected. The chain links are marked as Nos. 16, as shown in
MA

Fig. 2. The chain links that connect the strain gauges are installed on the calibration test platform, as shown in Fig.
3. The micro-strain values of the pressure sensor and strain gauge are shown on the static strainmeter
simultaneously. The relationship between the measured pressure sensor and micro-strain is 1 kN=60 .
E D
PT
CE
AC

Fig. 2. Chain links connecting the strain gauges

 ACCEPTED MANUSCRIPT

PT
RI
Fig. 3. Calibration test platform of chain links
The relationship between the micro-strain and the tension of the three groups of chain links are calibrated, as
shown in Fig. 4. A linear relationship is observed, that is, 1 kN=5 .

SC
NU
MA
E D

Fig. 4. Calibration curves of chain links

PT

2.2 Measurement scheme

Fig. 5 shows the experimental prototype of excavator, which consists of the crawler, upper body, and working
CE

device. The three groups of chain links are installed on the left track link of the crawler of excavator. Label Nos. 1
and 2, Nos. 3 and 4, and Nos. 5 and 6 are paired, as shown in Fig. 6.
AC
 ACCEPTED MANUSCRIPT

Fig. 5. Experimental prototype of excavator

PT
Fig. 6. Tension measurement of chain links

2.3 SEM analysis

RI
Fig. 7 shows the SEM analysis of fracture surface of chain link crack. The crack source morphology is shown

SC
in Fig. 7(a). The fatigue crack initiates on the surface, and the slip band develops on the fracture surface because of
the plastic deformation of the surface microstructure. The enlarged image of the slip band is shown in Fig. 7(b). Fig.
7(c) shows the extending zone morphology of fatigue crack. The SEM analysis verifies that the chain link crack is
NU
caused by the fatigue failure.
MA
E D
PT

Fig. 7. Fracture surface morphology of crack

3. Virtual prototype model of excavator

CE

3.1. Establishment of the model

The three-dimensional models of the sprocket, idler, track wheel, support roller, and chain link of crawler can
AC

be imported from the Track (LM) module of the multi-body dynamics RecurDyn software. These models are
defined as the corresponding components that the system can recognize in the module. The track links are then
assembled. The integrated virtual prototype model of crawler can be established fast. Fig. 8 shows the virtual
prototype model of excavator.
 ACCEPTED MANUSCRIPT

PT
Fig. 8. Virtual prototype model of excavator
The investigated main parameters of excavator are shown in Table 1. The center of gravity position and the

RI
moment of inertia are all relative to the global coordinate system o-xyz, which is located at the middle of the
revolving platform of the upper body, as shown in Fig. 8, and x axis coincides with the forward direction, y axis is

SC
parallel to the inverse direction of gravity and z axis is determined by the right-hand rule. The moment of inertia of
x, y, and z is measured in the posture shown in Fig. 8.
Table 1
NU
The main parameters of excavator
Name of parameters Value
Weight (kg) 5960
MA

Position of gravity center (m) (0.05, 0.23, -0.03)

Moment of inertia of x axis (kgm2) 22878.62
2
Moment of inertia of y axis (kgm ) 25661.82
2
Moment of inertia of z axis (kgm ) 29278.77
D

Ground contact length (m) 2.01

Ground contact pressure (kPa) 32.5
E

Pitch radius of sprocket (m) 0.229

PT

Tooth number of sprocket 21

Track gauge (m) 1.5
Width of link pad (m) 0.4
CE

Pitch of chain link (m) 0.13

Number of chain links 40
AC

Number of track wheels 5

Pre-tension of tensioner (kN) 24
Ground friction coefficient 0.7

3.2. Main parameters setting

In the virtual prototype model of crawler, the sprocket and the pin bushings of chain links are made of steel,
and the type of friction between them is sliding friction without lubrication. Thus, the friction coefficient is 0.10.
The rotation of the idler, track wheels, and support rollers relies on the force of rolling friction with the crawler
chain links. These components are made of steel. Thus, the friction coefficient is 0.01. The axles and the bearing are
made of steel and brass, respectively, and the type of friction between them is sliding friction with lubrication; thus,
the friction coefficient is 0.03 [16]. The friction coefficients on the contact surfaces and revolution joints must be
set. These coefficients are related to the material and the type of friction and can influence the simulation results.
 ACCEPTED MANUSCRIPT

4. Comparison of simulation and experimental results

The simulation results of horizontal straight, pivot steering, and differential steering conditions on the firm
ground are conducted with the virtual prototype model of excavator. The simulated and experimental tension of
chain links in each condition are compared and the relevant parameters are adjusted. As a result, the accuracy of the
established virtual prototype model is verified.

4.1. Horizontal straight condition

During excavator operations, the crawler usually runs at a low speed of 2.3 km/h in a horizontal straight
condition. The longitudinal ground contact length of a link pad is 0.128 m, and the number of chain links is 40.

PT
Therefore, in the horizontal straight condition, the amount of time in a track link running around the track frame in
a cycle is approximately 8 s. The simulated and experimental tension of chain link with the crawler running a cycle

RI
are compared, as shown in Fig. 9. The initial position of the measured chain links is under the track wheel. The
tension increases suddenly when the chain links start meshing with the sprocket. As the crawler moves forward, the

SC
chain links move gradually to the loose side of the track link. The tension of chain links also decreases gradually.
When the chain links completely exit the meshing with the sprocket, the tension of chain links tends to be stable.
When the chain links are under the track wheel passing the idler, the fluctuation of the tension increases. The
NU
changing trend of the simulated and experimental tensions of chain links is identical, as shown in Fig. 9. However,
the curve fluctuation of the experimental results is more significant than that observed in the simulation because the
smoothness of ground surface is lower than that observed in the simulation. Thus, the shock occurs and the tension
MA

of chain links is unstable.

E D
PT
CE
AC

Fig. 9. Comparison of the simulated and experimental tension of chain links in horizontal straight condition

4.2. Pivot steering condition

The left side of the crawler of excavator goes forward, and the right side of the crawler retreats. The excavator
turns right in pivot steering condition at 1.8 km/h. The crawler runs a cycle for about 10 s. The comparison of the
simulated and experimental tension of chain links in pivot steering condition is shown in Fig. 10. The figure shows
that the tension of chain links are large in the proceeding chain links meshing with the sprocket, and the fluctuation
of the curve is significant. The changing trend of the simulated and experimental tensions of chain links is identical.
 ACCEPTED MANUSCRIPT

PT
RI
Fig. 10. Comparison of the simulated and experimental tension of chain links in pivot steering condition

4.3. Differential steering condition

SC
The left side of the crawler of excavator moves forward at 2.3 km/h, and the right side of the crawler brakes.
The excavator turns right in differential steering condition. The crawler runs a cycle for about 8 s. The comparison
NU
of the simulated and experimental tension of chain links in differential steering condition is shown in Fig. 11. The
figure shows that the changing trend of the simulated and experimental tensions of chain links is identical.
MA
E D
PT
CE

Fig. 11. Comparison of the simulated and experimental tension of chain links in differential steering condition
The simulated and experimental tension of chain links are compared in the abovementioned three conditions.
AC

The random smoothness of the ground surface and the operation of the driver in the proceeding moving vehicle
lead to the error between the simulation and experiment results. However, the error is within the acceptable limit. In
general, the identity of the simulated and experimental tension curves of chain links is appropriate. Therefore, the
established virtual prototype model of crawler is feasible and can be used to analyze the fatigue life of crawler
chain links.

5. Fatigue life analysis

The failure mechanism of materials or components shows that the fatigue process experiences three steps:
crack initiation, crack expanding, and crack fracture. The behavior starts from no crack to minimal cracks. Then,
the cracks expand and reach the critical crack size or finally fracture. The fatigue of crawler chain link belongs to a
low stress and long life problem, that is, the high cycle fatigue. Thus, the cumulative damage theory of linear
 ACCEPTED MANUSCRIPT

fatigue (PalmgrenMiner cumulative damage criterion) is used to calculate the high cycle fatigue life in RecurDyn.
This method is widely used in the engineering practice for calculating fatigue life, and the actual results can be
obtained in high accuracy.

5.1. FEA

The finite element model of chain links is established in ANASYS, and FEA in the pivot steering condition
with complex force is conducted. On the basis of the longitudinal tension, the vertical and lateral forces are
generated between the track wheels and chain links of the virtual prototype model simulation. Three kinds of loads
are the basis of the applied load in the FEA. As shown in Fig. 12, full constraint is applied on the beam elements
between the two eyes of one side of chain links, and the longitudinal tension of 21.8 kN is applied on the beam

PT
elements between the two eyes of the other side of chain links. Pre-tightening force is applied on the four groups of
bolts. The other degree of freedom constraints, except the longitudinal translation degree of freedom, are applied on

RI
the undersurface of the grouser and the two groups of chain links at the front and end. The vertical force of 6.6 kN
and the lateral force of 8.4 kN are applied on the inner node of the track wheel. Considering that the occurrence of

SC
the rigid translation in the track wheel is prevented because of the significantly large lateral force, the other degree
of freedom constraints, except the vertical and lateral degree of freedom, are applied on the inner node of the track
wheel. The mechanical properties of the material of chain links are shown in Table 2.
NU
MA
E D
PT
CE

Fig. 12. Load and constraint

Table 2
AC

The mechanical properties of material of chain links

Elastic modulus Poissons ratio Density Yield strength Ultimate strength
3)
(E, GPa) () (, kg/m (s, MPa) (b, MPa)
210 0.3 7850 735 930
The von Mises stress contours of FEA of chain links is shown in Fig. 13. The maximum stress value of 183
MPa of chain links appears at the corner of the bolt holes location of the chain link that is on the side of the bearing
of the lateral force. The position of this stress coincides with the position of the crack that appeared in Fig. 1.
However, the maximum stress value is lower than the yield strength (s=735 MPa). This result indicates that the
crack is not attributed to the significantly high stress that leads to the plastic deformation but is due to the cycle
loading. Therefore, the fatigue life of this area must be examined.
 ACCEPTED MANUSCRIPT

PT
Fig. 13. The von Mises stress contours of the FEA

5.2. SN curve

RI
The fatigue life analysis of components is based on the SN curve of materials or components. The SN curve

SC
is generally acquired by experiment, and the simple curve can be drawn by estimation. The SN curve shows the
relationship between the level of applied stress and the fatigue life of materials or components, and the curve
presents a linear relationship at a wide range of stress in the double logarithmic graph [17]. Two lines are drawn in
NU
which the left branch of the curve (high cycle fatigue) is a skew line, and the right branch (fatigue limit) is a
horizontal line [18]. The SN curve of the Basquin equation [18] is expressed as follows:
= (1)
MA

where m and C are the material constant; S is the applied stress amplitude; N is the fatigue life cycle. Take the
logarithm for both side of the Eq. (1), as follows:
lg = + lg (2)
where the material constant a=lgC/m, b=-1/m. Eq. (2) indicates that the relationship is logarithmic linear between
D

lgS and lgN in the region of high cycle fatigue.

Considering that the fatigue properties described by the S-N curve is in the high cycle fatigue condition, the
E

stress at N<103 is not applying appropriately, so assume generally when N=103 as follows:
PT

103 = 0.9 (3)

where b is the ultimate strength of material. For the metal materials, assume the fatigue limit -1 corresponding to
the number of cycles N=107 generally. Considering that the errors of estimation of the fatigue limit -1, make
CE

conservative assumption when N=107 as follows:

107 = 1 = (4)
where k is the coefficient that reflecting the different loading forms, and k=0.35 in tension.
AC

The ultimate strength of material of chain links in the study is b=930 MPa, so the S-N curve of material of
chain links is shown in Fig. 14.
 ACCEPTED MANUSCRIPT

PT
RI
Fig. 14. The S-N curve

5.3 Fatigue life calculation

SC
The rigid-flexible coupling virtual prototype model with chain link of flexible body is established in RecurDyn,
as shown in Fig. 15. The mechanical properties of the material of chain links are shown in Table 2. The simulation
NU
analysis in horizontal straight, climbing, and pivot steering conditions is performed. In the horizontal straight
condition, the virtual prototype model of excavator runs at 4.2 km/h, and the crawler runs a cycle for about 4.4 s. In
the climbing condition, the virtual prototype model of excavator runs at 4.2 km/h on the slope of 15, and the
MA

crawler runs a cycle for about 4.4 s. In the pivot steering condition, the two side track links of the virtual prototype
model of excavator run at 2.2 km/h but in opposite directions, and the crawler runs a cycle for about 8.5 s.
E D
PT
CE
AC

Fig. 15. Rigid-flexible coupling virtual prototype model

Figs. 16(a), (b) and (c) show the stress contours of chain links as the crawler runs in the horizontal straight
climbing, and pivot steering conditions, respectively. The position of the large load of chain link bearing coincides
with the position of the crack in Fig. 1. Given that the structure of this position is thin and the radius of circular
bead is small, the stress concentration may occur. The key point is selected in the position of the crack that
appeared. The curves of stress changing with time are shown in Fig. 17. The results indicate that the stress of the
key point is the amplitude change. The stress of the chain links significantly changes when passing the idler and
sprocket. In the pivot steering condition, the vertical and lateral loads from the track wheels and the tension
generated by the meshing with the sprocket are applied on the chain links simultaneously. Thus, the stress exhibits a
more significant change than those observed in the horizontal straight and climbing conditions.
 ACCEPTED MANUSCRIPT

PT
RI
Fig. 16. Stress contours of the chain links in three conditions

SC
NU
MA
E D
PT
CE
AC

Fig. 17. Stress curves of key points in three conditions

The stress spectra of the key positions of chain links are processed with Rainflow [19] for cycle-counting the
stress range of variable amplitude. Through PalmgrenMiners cumulative damage criterion [20], a variable
amplitude stress spectrum is replaced by an equivalent constant amplitude stress range using the following
equation:
1/
= [=1( )/ ] (5)

where i is the ith stress range in the spectrum which occurs Ni time. Nv is the equivalent number of stress cycles.
 ACCEPTED MANUSCRIPT

mc is the crack growth rate constant.

The results of the fatigue life analysis of key points selected on the position of the crack on the chain links in
the three conditions are shown in Table 3. The fatigue cycle of the horizontal straight condition is longer than that
of the climbing condition at the same speed. Thus, the fatigue life of the horizontal straight condition is long.
Although the fatigue lives of the climbing and pivot steering conditions are close, the speed of the climbing
condition is high. Therefore, the fatigue life of the climbing condition is long. The load of the chain links in the
pivot steering condition is more complex than those observed in the two other conditions. Thus, the fatigue life of
this condition is shorter than those of the two other conditions.
Table 3
The fatigue life analysis of chain links

PT
Condition Speed (km/h) Cycle Life (h)
6
Horizontal straight 4.2 1.1410 1393.33

RI
5
Climbing 4.2 2.7310 333.67
5
Pivot steering 2.2 1.3910 328.19

SC
6. Conclusions

In this study, the fatigue life analysis of crawler chain link of excavator is conducted. The fracture surface
NU
morphology of chain link crack is analyzed by SEM, and the results show that the fatigue crack initiates on the
surface and that an extending zone of fatigue crack is observed. Furthermore, FEA indicates that the load of chain
links is lower than the yield strength in the ultimate pivot steering condition. Thus, the generation of crack is
MA

attributed to the fatigue damage. The SN curve of material of chain links is estimated. The fatigue life analysis of
structural components is based on PalmgrenMiners cumulative damage criterion. The results indicate that the
fatigue life of chain links in the horizontal straight condition is the longest at 1393.33 h, followed by the fatigue life
of chain links in the climbing condition at 333.67 h, and the fatigue life of chain links in the pivot steering
D

condition at 328.19 h as the shortest.

E

The method for the fatigue life analysis of chain link, which is the key component of crawler, is proposed in
PT

this study. The proposed method avoids the fatigue experiment, which is cumbersome or difficult to achieve. The
method also improves efficiency and serves as a reference for the fatigue life analysis of other key components.
According to the scale of the working time of the different conditions and roads, future simulation analysis can be
CE

performed in various conditions, such as different speeds in horizontal straight condition, different angles of slope
in climbing condition, different steering radii in differential steering condition. A variety of soft roads, such as
sandy loam and clay, can also be used. The stress loading of chain links can be conducted in different conditions to
AC

obtain a highly accurate fatigue life.

Acknowledgments

The authors acknowledge the National Natural Science Foundation of China (Grant No. 51505177),
Technology Development Program of Jilin Province (Grant No. 20160520073JH) and the financial support from
the Collaborative Innovation Center of Major Machine Manufacturing in Liaoning.
 ACCEPTED MANUSCRIPT

References

[1] Arsi, M. A., Bonjak, S. M., Odanovi, Z. D., Dunji, M. M., Simonovi, A. M., Analysis of the spreader track
wheels premature damages, Engineering Failure Analysis 20 (2012) 118-136.
[2] Bonjak, S., Petkovi, Z., Zrni, N., Panteli, M., Obradovi, A., Failure analysis and redesign of the bucket wheel
excavator two-wheel bogie, Engineering Failure Analysis 17(2) (2012) 473-485.
[3] Savkovi, M., Gai, M., Petrovi, D., Zdravkovi, N., Pljaki, R., Analysis of the drive shaft fracture of the bucket

PT
wheel excavator, Engineering Failure Analysis 20 (2012) 105-117.
[4] Ryu, H. S., Bae, D. S., Choi, J. H., Shabana, A. A., A compliant track link model for high-speed, high-mobility
tracked vehicles, International Journal for Numerical Methods in Engineering 48(10) (2000) 1481-1502.

RI
[5] Rubinstein, D., Coppock, J. L., A detailed single-link track model for multi-body dynamic simulation of crawlers,
Journal of Terramechanics 44(5) (2007) 355-364.

SC
[6] Yu, Z. W., Xu, X. L., Mu, X., Failure investigation on the cracked crawler pad link, Engineering Failure Analysis
17(5) (2010) 1102-1109.
NU
[7] Bonjak, S. M., Arsic, M. A., Zrnic, N. D., Odanovic, Z. D., Dordevic, M. D., Failure analysis of the stacker crawler
chain link, Procedia Engineering 10 (2011) 2244-2249.
[8] Dudek, D., Frydman, S., Huss, W., Pogo, G., The L35GSM cast steelpossibilities of structure and properties
MA

shaping at the example of crawler links, Archives of Civil and Mechanical Engineering 11(1) (2011) 19-32.
[9] Bonjak, S. M., Momilovi, D. B., Petkovi, Z. D., Panteli, M. P., Gnjatovi, N. B., Failure investigation of the
bucket wheel excavator crawler chain link, Engineering Failure Analysis 35 (2013) 462-469.
D

[10] Yin, Y., Grondin, G. Y., Obaia, K. H., Elwi, A. E., Fatigue life prediction of heavy mining equipment. Part 1:
Fatigue load assessment and crack growth rate tests, Journal of Constructional Steel Research 63(11) (2007)
E

1494-1505.
PT

[11] Yin, Y., Grondin, G. Y., Obaia, K. H., Elwi, A. E., Fatigue life prediction of heavy mining equipment. Part 2:
Behaviour of corner crack in steel welded box section and remaining fatigue life determination, Journal of
CE

Constructional Steel Research 64(1) (2008) 62-71.

[12] Joshi, S., Price, J. W., A comparative study on application of design codes for prediction of fatigue life of a mining
dragline cluster, Engineering Failure Analysis 16(5) (2009) 1562-1569.
AC

[13] Wu, S. C., Zhang, S. Q., Xu, Z. W., Thermal crack growth-based fatigue life prediction due to braking for a
high-speed railway brake disc, International Journal of Fatigue 87 (2016) 359-369.
[14] Li, Y., Liu, W., Dynamic dragline modeling for operation performance simulation and fatigue life prediction,
Engineering Failure Analysis 34 (2013) 93-101.
[15] Son, S. W., Jung, H. S., Kwon, T. S., Kim, J. S., Fatigue life prediction of a railway hollow axle with a tapered bore
surface, Engineering Failure Analysis 58 (2015) 44-55.
[16] H. Dubbel, W. Beitz, K. H. Kttner, Handbook of Mechanical Engineering, Springer-Verlag, 1994.
[17] Lindorf, A., Curbach, M., SN curves for fatigue of bond in reinforced concrete structures under transverse tension,
Engineering Structures 32(10) (2010) 3068-3074.
[18] Kohout, J., Vechet, S., A new function for fatigue curves characterization and its multiple merits, International
Journal of Fatigue 23(2) (2001) 175-183.
 ACCEPTED MANUSCRIPT

[19] Downing, S. D., Socie, D. F., Simple rainflow counting algorithms, International journal of fatigue 4(1) (1982)
31-40.
[20] Fisher, J. W., Kulak, G. L., Smith, I. F., A fatigue primer for structural engineers. National Steel Bridge Alliance,
American Institute of Steel Construction, 1998.

PT
RI
SC
NU
MA
E D
PT
CE
AC
 ACCEPTED MANUSCRIPT

Research highlights

Fatigue life analysis of crawler chain links of excavator.

The experiment of the tensions of crawler chain links was carried out.
The fracture surface morphology of chain link crack is analyzed by scanning electron microscopy
The accurate virtual prototype model of excavator was established by comparing the simulation
results with experimental data.
The fatigue life of crawler chain links under three specific conditions was received.

PT
RI
SC
NU
MA
E D
PT
CE
AC