(2016) Crack Detection For A Jeffcott Rotor With A Transverse Crack - An Experimental Investigation

Mechanical Systems and Signal Processing ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Mechanical Systems and Signal Processing

journal homepage: www.elsevier.com/locate/ymssp
Crack detection for a Jeffcott rotor with a transverse crack: An

experimental investigation
Chaozhong Guo, Jihong Yan n, Weicheng Yang
School of Mechatronics Engineering, Harbin Institute of Technology, China
a r t i c l e i n f o abstract
Article history: In this paper, an experimental investigation is carried out to verify the theoretical results
Received 18 March 2016 of the dynamic behavior and the EMD based crack detection method for the cracked rotor
Received in revised form proposed in our former research. The breathing crack in the rotor is simulated by a real
5 June 2016
fatigue crack. The whirl orbits during passage through the 1/2, 1/3 and 1/4 subcritical
Accepted 10 June 2016
speeds are investigated. The dynamic responses in these subcritical speed zones are de-
composed into several subcomponents by the EMD method, and the variation of the high-
Keywords: frequency component are studied. As a comparison, the fast Fourier transform method is
Jeffcott rotor used to derive the amplitude variation of the high order frequencies from the frequency
Breathing crack
spectra of the experimental vibration signal. The experimental results are well concordant
Crack detection
with the theoretical analysis, which indicates that the EMD based crack detection method
Empirical mode decomposition
is practicable.
& 2016 Published by Elsevier Ltd.
1. Introduction
In large rotating machinery, such as turbine, generator and aero-engine, the rotor is one of the most important parts. Fatigue
cracks have great potential to cause catastrophic failures in rotors which can lead to huge economic loss, or even serious human
injury. Crack detection has attracted the attention of researchers all over the world based on the idea that changes of a rotor's
dynamic behavior could be used for general fault detection since 1970s [1]. During the past 40 years, a great deal of papers have
been published on the theoretical and experimental study of dynamic analysis of the cracked rotor [2–4]. It has been shown that
accurate modeling of breathing crack in a damaged rotor system is a critical step in well understanding the dynamic effects of the
crack and in identifying its vibration features. There are several crack models proposed by former researchers to calculate the
stiffness of the cracked rotor [5–14]. In 2011, Al-Shudeifat and Butcher [15] developed new breathing functions for the cracked
rotor by using Fourier series based on their former research [16] which used the integration method to calculate the true
stiffness. These functions were in a close and independent mathematical form and were proved to be more accurate than the
Mayes' model. However, in Ref. [15], the function of cross-coupling stiffness was not provided. In our previous paper [17], a new
function of the cross-coupling stiffness was developed to complete the breathing model in Ref. [15] so that the breathing
mechanism of the crack could be fully represented. The proposed function was applied to a Jeffcott rotor, the typical whirl orbits
[17] and stability region [18] of which were investigated, respectively.
Based on the accurate development of time-varying stiffness functions, the governing equations of the cracked rotor
system can be built and the dynamic behaviors are studied for crack identification by numerous researchers. Haji et al. [19]
n
Corresponding author.
E-mail address: jyan@hit.edu.cn (J. Yan).
http://dx.doi.org/10.1016/j.ymssp.2016.06.011
0888-3270/& 2016 Published by Elsevier Ltd.
Please cite this article as: C. Guo, et al., Crack detection for a Jeffcott rotor with a transverse crack: An experimental
investigation, Mech. Syst. Signal Process. (2016), http://dx.doi.org/10.1016/j.ymssp.2016.06.011i
2 C. Guo et al. / Mechanical Systems and Signal Processing ∎ (∎∎∎∎) ∎∎∎–∎∎∎
used orthogonal natural frequencies (ONFs) to identify the crack for a non-rotating cracked rotor. The simulated results
showed that the unique characteristics was the appearance of the sharp, notched peaks at the crack location of the nor-
malized ONF curve but round peaks at non-cracked locations. Silani et al. [20] modeled the crack breathing behavior with a
finite element method, and the dynamic response of the cracked rotor was examined with the short time Fourier transform
for small crack detection based on the presence of higher harmonics. Sekhar [21] used the continuous wavelet transform
(CWT) to detect transverse cracks during a rotor system passing through its critical speed. The subcritical response peaks
appeared in the CWT coefficient plots but not in the time-history plots, which was used as the indicator of the presence of
cracks. Sinou [22] detected the breathing crack in a nonlinear rotor system by numerically using the 2X and 3X super-
harmonic frequency components at the associated subcritical resonant peaks. Babu et al. [23] applied the Hilbert–Huang
transform (HHT) to transient response of a cracked rotor. In the HHT spectrum, the instantaneous frequency fluctuations
showed up as a result of the sub-harmonics which is a typical characteristic of a crack. In the former paper [17], we used
empirical mode decomposition (EMD) to extract the fault feature of a transverse crack in a Jeffcott rotor by investigating the
variations of super-harmonic components. Other researches [24–28] also focused on this topic and several methods have
been proposed to detect the crack in rotor systems based on the typical dynamic behavior including the whirl orbits,
spectrums of FFT, WT and HHT during passing through the subcritical speed zones. However, most of these researches are
based on the simulation without any experimental verification.
Only few papers provided experiment results for the cracked rotor. In Ref. [29], Darpe et al. set up experiments on both
slotted and cracked shafts to verify the analytical findings. A slot of width 1 mm was machined on the rotor near the disc
using a slitter to simulate the open crack. In terms of the real breathing crack, a three-point-fatigue machine was used to
generate a fatigue crack by adding cyclic loading to an initiate slit at a desired location. The orbits and FFT spectra during the
rotor passing through 1/3 and 1/2 critical speed were consistently matching with theoretical findings. Similar method was
adopted to make fatigue crack in Refs. [30,31]. However, in [30] the typical orbits in the subcritical speed zones were not as
regular as the analytical results, and in [31] the frequency analysis was performed by the wavelet transform while none
typical orbit was provided. Dong et al. [32] and Lin et al. [33] used the wire-cut electrical discharge machine to make a slit on
the rotor which was treated as an open crack. Therefore, it is still necessary to experimentally verify the dynamic behavior
and the crack detecting method for breathing cracks.
In this paper, based on the theoretical research of the dynamic analysis and the crack diagnostic method provided in Ref.
[17], an experiment was set up on a rotor testbed. A real fatigue was made near the middle location of the shaft. The
experiment results are compared with the analytical findings to investigate whether the proposed crack detecting method is
feasible in practice.
2. Equations of motion of a cracked Jeffcott rotor
A Jeffcott rotor with a transverse crack considered in Ref. [17] is adopted in the present study which is shown as Fig. 1. A
brief description of the modeling process for the cracked rotor is given below.
The coordinates built on the cross-section where the crack is located are shown in Fig. 2.
The equations of motion of the cracked rotor system can be given as
mu¨ + cu̇ + k1( t )u + k12( t )v = med Ω2 sin(Ωt + β )

mv¨ + cv ̇ + k21( t )u + k2( t )v = med Ω2 cos(Ωt + β ) − mg (1)
where u is the shaft displacement in the x1 direction, v is the shaft displacement in the y1 direction, m is the mass of the rigid
disk, c is the external damping, med is the mass unbalance, β is the angle between the mass unbalance and the weak
direction of the crack, Ω is the rotating speed, and g is the gravitational acceleration.
Considering the neutral axis shift, the time-varying moment of inertia of the cross-section at the crack location is
approximated by the Fourier series expansion. The independent functions of the moment of inertia and production
of inertia (cross-moment of inertia) were given as
I X¯ ≅ I − f1(t )I11 (2)
Rigid disk
Massless shaft
Crack
Support
Fig. 1. Cracked Jeffcott rotor system [17].
C. Guo et al. / Mechanical Systems and Signal Processing ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3
Fig. 2. Coordinate systems of the cracked rotor on the cross-section [15,17].
IY¯ (t ) ≅ I + f1(t )I11 + f2 (t )I22 (3)
p
A1 2θ¯2 sin(π − kθ¯2)
¯ ¯ ( t ) ≅ I XY
I XY ¯¯ ∑ sin(kΩt )
k=1 π 2 − k 2θ¯22 (4)
where
⎡⎛ n ⎞ (n /2)− 1
⎛ n⎞ ⎤
f1(t ) = cosn( Ωt /2) = 2−n ⎢ ⎜ ⎟ + 2 ∑ ⎜ ⎟cos( n − 2j )Ωt /2⎥
⎢⎣ ⎝ n/2⎠ ⎝ j⎠ ⎥⎦
j=0 (5)
1 ⎡ θ + θ2 ⎤
p
2 cos(iθ2) − cos(iθ1)
f2 (t ) = ⎢ 1 − ∑ cos(iΩt )⎥
π ⎢⎣ 2 ( 2 1) i = 1
θ − θ i2 ⎥⎦ (6)
The detailed calculation process can be found in Ref. [15]. For the non-dimensional crack depth (crack depth h/radius of
the shaft R) μ = 0.1, the time-varying moments of inertia which represent the breathing mechanism of the transverse crack
during the rotating process are shown in Fig. 3. It can be seen that compared with the moments of inertia in X and Y
direction, the product of inertia is much smaller.
The complete stiffness matric of the Jeffcott rotor can be calculated by
⎡ k (t ) k (t )⎤ 48E ⎡ I ¯ (t ) I ¯ ¯ (t )⎤
1 12
K(t ) = ⎢ ⎥= ⎢ X XY
⎥
⎢⎣ k21(t ) k2(t ) ⎥⎦ l3 ⎣ IYX
¯ ¯ (t ) IY¯ (t ) ⎦ (7)
where E is the modulus of elasticity, l is the length of rotor.
Fig. 3. Breathing process in one periodic for μ = 0.1.

Fig. 4. Rotating speed vs. amplitude of the cracked rotor for normalized crack depth: (a) μ = 0.01, (b) μ = 0.05, (c) μ = 0.1 [17].
3. Dynamic analysis and crack detection based on EMD method
It is interesting to investigate the dynamic behavior of the cracked rotor during passage through the subcritical speeds
which can support the crack detection. Here, the simulation results obtained by integrating the equations of motion with the
fourth order Runge–Kutta method in the Ref. [17] are cited. Fig. 4 shows the transverse amplitude of the cracked rotor versus
the rotating speed for different crack depths. We can see that the sub-resonances appear when a crack exists in the rotor.
Moreover, as the crack depth gets larger, the amplitude of that sub-resonance gets stronger as well, which well indicates
that the existence of a crack does influence the vibration of a rotor system. Consequently, the whirl orbits when the rotor
passes through these subcritical speed zones are examined. In Fig. 5, the upper four subplots show the orbit evolution
around the 1/3 subcritical speed zone. It can be seen that two inner loops appear in Fig. 5(a) and get larger as the rotating
Fig. 5. Whirl orbits of the cracked rotor during the passage through 1/3 and 1/2 of the first critical rotating speed for μ = 0.2 [17].
-3
x 10
3.5
0.1 3X 1X
Amplitude (mm) 3
Amplitude (mm)
2.5
0.08
2
0.06 1.5
1
0.04
0.5
0.02 0
820 840 860 880 900 820 840 860 880 900
Rotating speed (rpm) Rotating speed (rpm)
Fig. 6. Average amplitudes of the IMFs in the neighborhood of 1/3 of the critical speed for μ = 0.1: (a) 3X-component, (b) 1X-component [17].
speed raises more close to the 1/3 subcritical speed as shown in Fig. 5(b), after which the inner loops become smaller and
finally disappear in Fig. 5(d). During this process, the orbit changes its direction by nearly π/2 rad. Similar behaviors have
been observed during the rotor passing through the 1/2 subcritical speed as shown in Fig. 5(e)–(h) with the difference that
only one inner loop appears during the evolution and its direction change is about π rad. These typical orbits agree with the
results in Refs. [15,29,30,34]. In Ref. [15] where a finite element model was used to model the cracked rotor, the orbit
changes its direction by nearly π/2 rad as well during passing through the 1/3 subcritical speed, while the orbit direction
changes by about π rad during passing through the 1/2 subcritical speed. The simulation results of Refs. [29,34] about the
orbit changing direction are almost the same with Fig. 5. In Ref. [30], the orbit orientation changes by about 3π /4 rad in the
region of 1/2 subcritical speed, while the typical orbits in the region of 1/3 subcritical speed were not provided. It is noted
that the orbits of a rotor with an open crack is quite different which were discussed in [35] where no such inner loops
appeared in the open cracked rotor system. Hence, the unique change of the whirl orbits is considered as one of the typical
features for identification of breathing cracks.
Based on the dynamic analysis in these subcritical speed zones, an EMD based crack detection method was proposed in
Ref. [17], the basic idea of which was to analyse the variation of super-harmonic components. The original vibration signals
were decomposed by EMD method into several sub-signals from high to low frequencies, named intrinsic mode functions
(IMFs). The average amplitudes of these IMFs for μ = 0.1 in the neighborhoods of the 1/3 and 1/2 subcritical speeds are
shown in Figs. 6 and 7, respectively. It can be seen that the amplitude variation curves of 3X and 2X components around the
1/3 and 1/2 subcritical speeds have apparent peaks as the rotating speed increases while that of the 1X components does
not have. This finding can be employed in developing a methodology for identifying the vibration signature for a breathing
crack in rotary systems based on the EMD analysis.
In order to make a comparison, a further simulation by adopting the system parameters of the experiment testbed listed
in Table 1 are performed following the same procedure as above, where the relative crack depth used in the experiment is
-3
x 10
0.045 3.5
2X 1X
0.04 3
Amplitude (mm)
Amplitude (mm)
0.035
2.5
0.03
2
0.025
1.5
0.02
0.015 1
0.01 0.5
1260 1280 1300 1320 1260 1280 1300 1320
Fig. 7. Average amplitudes of the IMFs in the neighborhood of 1/2 of the critical speed for μ = 0.1: (a) 2X-component, (b) 1X-component [17].
Table 1
System parameters of the rotor testbed.
Description Value Description Value
Length of rotor, L 0.4 m Gravitational acceleration, g 9.8 m/s2

Radius of rotor, R 5 × 10−3 m Mass unbalance, med 4 × 10−4 kg⋅m
Density of rotor, ρ 7.85 × 103 kg/m3 Mass unbalance angle, β 0 rad
Modulus of elasticity, E 2.1 × 1011 N/m2 Disk mass, m 0.5 kg
about μ = 0.64 . The whirl orbits during the passage through 1/3 and 1/2 subcritical speeds are shown in Fig. 8, and the
corresponding variation curves of the super-harmonic components are plotted in Figs. 9 and 10, respectively. It can be seen
that the dynamic behavior is quite similar with the previous simulation results shown in Figs. 5–7, which indicates that the
crack system modeling and the EMD based crack detection method proposed in [17] is feasible for even larger crack depth in
simulation.
4. Experiment investigations
In order to verify the numerical findings, an experiment is set up on a rotor testbed as shown in Fig. 11 which mainly
consists of a cracked shaft with a central disk, a drive motor, two eddy current sensors, a PXI data acquisition device. The
main system parameters are listed in Table 1. The diameter of the steel shaft is 10 mm which is supported by two identical
ball bearings with a span of 400 mm. The disk mass is 0.5 kg with a certain unbalance which generates the exciting force
Fig. 8. Simulation whirl orbits of the cracked rotor using the experiment parameters for μ = 0.64 .
0.07 0.06
3X 1X
0.06
Amplitude (mm) 0.05
Amplitude (mm)
0.05
0.04
0.04
0.03
0.03
0.02 0.02
0.01 0.01
1050 1100 1150 1200 1250 1050 1100 1150 1200 1250
Fig. 9. Average amplitudes of the IMFs in the neighborhood of 1/3 of the critical speed for μ = 0.64 : (a) 3X-component, (b) 1X-component.
0.14 0.2
2X 1X
0.12
0.15
Amplitude (mm)
Amplitude (mm)
0.1
0.08 0.1
0.06
0.05
0.04
0.02 0
1600 1700 1800 1600 1700 1800
Fig. 10. Average amplitudes of the IMFs in the neighborhood of 1/2 of the critical speed for μ = 0.64 : (a) 2X-component, (b) 1X-component.
Eddy current Disk Cracked shaft DC motor
Bearing 1 Bearing 2
Fig. 11. Experimental setup of the rotor testbed.
Fig. 12. The precutting crack and cross section of the fatigue crack.
Fig. 13. Experimental whirl orbits of the uncracked rotor during the passage through 1/2 critical rotating speed.
Fig. 14. Experimental whirl orbits of the cracked rotor during the passage through 1/2 critical rotating speed (the blue dot curve is the orbit of the cracked
rotor, and the grey dash curve is the orbit of the uncracked rotor). (For interpretation of the references to color in this figure legend, the reader is referred to
the web version of this article.)
during the rotation. The motor can provide a maximum rotating speed of 10,000 rpm. Two eddy current sensors are fixed
near the crack location to measure the displacements of the rotor in vertical and horizontal directions separately. A PXI data
acquisition box produced by the National Instruments Company is used to collect the vibration data of the cracked rotor
system in company with a LabVIEW virtual instrument.
An important preparation is to make the artificial transverse crack in the shaft. For this purpose, a very fine slit with a
0.2 mm width is precut using a wire-electrode cutting machine. Then the shaft is placed in a three-point-bending machine
and a real fatigue crack is generated by adding cyclic loading to the precut shaft at the slit location. The precutting crack and
the typical cross section of the fatigue crack rotor obtained after the experiment is shown in Fig. 12. In Fig. 12(b), a clear
boundary of the fatigue and non-fatigue area can be seen. The fatigue crack depth is about 3.2 mm. It is noted that compared
with the pre-loading force, the alternation stress is much smaller, and the lasting time of the experiment is very short.
Therefore, the crack depths before and after the rotating experiment can be treated as the same.
In the experiment, a coast-up and rundown operation is first performed on a shaft without any crack. Based on the
investigation of vibration intensity, it can be seen that the first critical speed of the rotor system is about 3300 rpm where
the system has a very apparent resonance. In order to make comparison with the orbits of the cracked rotor, the whirl orbits
around the 1/2 subcritical speed are shown in Fig. 13. It is noted that in the experiment, all the orbits are of a similar shape.
Based on the theoretical analysis, it is well know that the orbits for an unbalanced Jeffcott rotor are in a circular pattern. In
Fig. 13, the orbits are nearly circles, which agrees well with the theoretical results and proves that the testbed works in a
good condition. When the orbits of the cracked rotor are plotted in Figs. 14 and 15, apparent changes show up as the rotor
passes through the 1/2 and 1/3 subcritical speed.
In Fig. 14, one inner loop appears and the direction changes by about π rad when the cracked rotor passes through the 1/2
critical speed zone as shown form Figs. 14(a)–(e). This experimental observation agrees well with the theoretical results in
Figs. 5(e)–(h) and 8(e)–(h) which were firstly provided in Ref. [15] in the aspect of the inner loop number and the angle of
orientation change. As mentioned previously, the same breathing functions were used in both Ref. [15] and this paper. The
difference is that in the former paper the FEM was employed to model the rotor system while in this paper the Jeffcott
model was used. The agreement indicates that the breathing functions and system modeling methods in both papers are
correct, and also proves that the former theoretical research well reveals the inherent characteristics of the real system.
When the rotor passes through the 1/3 critical speed, two loops appear and the angle of direction change is about π/2 rad
as shown in Fig. 15. In addition, the orbit evaluation during passage through 1/4 critical speed is shown in Fig. 16 where
three inner loops appear, the reason of which is that the 4X frequency component dominates the orbits. In Ref. [15], the
simulation result of the rotor passing through the 1/4 subcritical speed was provided and agreed well with the experimental
result here. However, there was no experimental observation provided in [15] as well as other previous literature.
In Section 3, the simulation results have shown that the EMD method is an effective tool to detect the breathing crack on a
Fig. 17. Average amplitudes of the IMFs in the neighborhood of 1/3 of the critical speed for experimental data.
rotor by investigating the variation of super-harmonic components as shown in Figs. 6 and 7, as well as Figs. 9 and 10. Following
the same procedure, the experimental signals during passage through the 1/3 and 1/2 subcritical speed are decomposed into
several nearly monochromatic IMFs corresponding to the basic and super-harmonic components. Then the amplitude variations
of these components around 1/3 and 1/2 subcritical speeds are plotted in Figs. 17 and 18 with the solid ‘*’ curves, respectively.
From Fig. 17(a), it can be seen that similar with the simulation results in Fig. 9, the variation curve of 3X-component
shows a single apparent peak during passage through the 1/3 subcritical speed. However, the curve shape of this experi-
mental data is not as symmetric as the simulation result shown in Fig. 9(a). The reason is that in the experiment there are
Fig. 18. Average amplitudes of the IMFs in the neighborhood of 1/2 of the critical speed for experimental data.
several uncertainties involved in the rotor testbed and the data acquisition system, even the manufacturing process of the
fatigue crack. Likewise, in Fig. 18(a), the 2X-component shown with the solid ‘*’ curve also has the same property. As for the
basic harmonic component 1X, the variation curve does not show any regularly changing pattern as shown in Figs. 17(b) and
18(b). Additionally, the FFT method is adopted to extract the high frequency information during the passage through 1/3 and
1/2 subcritical speeds as a comparison of the EMD method. The variations of the high frequency amplitudes are shown in
Figs. 17 and 18 by using the dash ‘o’ curves. It can be seen that there are small differences between the results obtained by
these two methods in Fig. 17, while in Fig. 18, the differences are relatively larger. However, the patterns of these two kinds
of curves are very similar. In both Figs. 17(a) and 18(a), the single peak in the 1/3 and 1/2 subcritical speed zones appears.
The above observations proves the simulation findings very well, which indicates that the variation of the super-harmonic
components in the subcritical speed zones can be used as one feature for crack diagnosis and the EMD based crack detecting
method is feasible in practice.
5. Conclusions
This paper experimentally investigates the dynamic behavior of a Jeffcott rotor and the EMD based crack detection
method as an extension of our previous research. A real fatigue crack was made on a steel shaft by a three-point-bending
machine. The cracked shaft was installed into a rotor testbed and the dynamic vibration signals were collected by acqui-
sition devices. The experimental whirl orbits during passage through the 1/3 and 1/2 subcritical speeds are in good
agreement with the theoretical results in terms of both the number of inner loops and the orientation change when the
loops appear and disappear in this process. In the 1/3 subcritical speed region, two inner loops show up which indicates the
3X-component is in domination; while in the 1/2 subcritical speed region, only one inner loop shows up. In the experiment,
the whirl orbits with three inner loops are also observed during passage through the 1/4 subcritical speed region. The EMD
method is used to extract the fault features for crack detection following the procedure proposed in the former paper [17].
The results show that the variations of the particular high-frequency components have unique signature during the passage
through the 1/3 and 1/2 subcritical speeds, while that of the fundamental frequency component is in a fluctuating form. This
experimental finding is quite concordant with the previous simulation result. As a comparison, the FFT method is also
employed to investigate the amplitude variation of high frequency obtained from the spectra of the experimental signal. The
result shows that although small differences exist, the general patterns of the corresponding variation curves are agreed
well, which proves that the variation of the super-harmonic component in the 1/3 and 1/2 subcritical zones is a stable
feature for crack detection, and further verifies that the EMD based crack detection method proposed in the former lit-
erature is reliable and practical.
Due to the limitation of the crack making devices and the accuracy of the testbed, the crack depth used in this research is
not very small. In the future work, more experiments for smaller cracks will be performed to find the minimum crack depth
that can be detected experimentally by the proposed method so that the performance of this method for the early detection
of cracks could be tested. And how the initial crack width influences the dynamic behavior of the cracked rotor system is
also an interesting and important topic which has not been discussed currently.
Acknowledgment
This work was funded by the National Natural Science Foundation of China (#51505099).
References
[1] J.T. Sawicki, M. Friswell, Z. Kulesza, A. Wroblewski, J.D. Lekki, Detecting cracked rotors using auxiliary harmonic excitation, J. Sound Vib. 330 (2011)
1365–1381.
[2] C. Kumar, V. Rastogi, A brief review on dynamics of a cracked rotor, Int. J. Rotating Mach. 2009 (2009) 1–6.
[3] A.S. Sekhar, Some recent studies on cracked rotors, in: Proceedings of the IUTAM Symposium on Emerging Trends in Rotor Dynamics, IUTAM Con-
ference Proceeding, 2011, pp. 491–503.
[4] A. Bovsunovsky, C. Surace, Non-linearities in the vibrations of elastic structures with a closing crack: a state of the art review, Mech. Syst. Signal
Process. 62–63 (2015) 129–148.
[5] A.D. Dimarogonas, S.A. Paipetis, Analytical Methods in Rotor Dynamics, Applied Science Publishiers, London and New York, 1983.
[6] A. Ebrahimi, M. Heydari, M. Behzad, A continuous vibration theory for rotors with an open edge crack, J. Sound Vib. 333 (2014) 3522–3535.
[7] N. Bachschmid, P. Pennacchi, E. Tanzi, Cracked Rotors – A Survey on Static and Dynamic Behavior Including Modelling and Diagnosis, Scientific
Publishing Services Pvt, Ltd., Chennai, India, 2010.
[8] R. Gasch, Dynamic behavior of a simple rotor with a cross sectional crack, in: Proceedings of IMechE Conference on Vibrations in Rotating Machinery,
London, 1976, pp.123–128.
[9] I.W. Mayes, W.G.R. Davies, Analysis of the response of a multi-rotor-bearing system containing a transverse crack in a rotor, J. Vib. Acoust. Stress Reliab.
Des. 106 (1984) 139–145.
[10] J. Sawicki, M. Friswell, Z. Kulesza, A. Wroblewski, J.D. Lekki, Detecting cracked rotors using auxiliary harmonic excitation, J. Sound Vib. 330 (2011)
1365–1381.
[11] J.-J. Sinou, Detection of cracks in rotor based on the 2x and 3x super-harmonic frequency components and the crack–unbalance interactions,
Commun. Nonlinear Sci. Numer. Simul. 13 (2008) 2024–2040.
[12] M.A. Al-Shudeifat, Stability analysis and backward whirl investigation of cracked rotors with time-varying stiffness, J. Sound Vib. 348 (2015) 365–380.
[13] O.S. Jun, H.J. Eun, Y.Y. Earmme, C.-W. Lee, Modelling and vibration analysis of a simple rotor with a breathing crack, J. Sound Vib. 155 (1992) 273–290.
[14] R.T. Liong, C. Proppe, Application of the cohesive zone model for the evaluation of stiffness losses in a rotor with a transverse breathing crack, J. Sound
Vib. 332 (2013) 2098–2110.
[15] M.A. Al-Shudeifat, E.A. Butcher, New breathing functions for the transverse breathing crack of the cracked rotor system: approach for critical and
subcritical harmonic analysis, J. Sound Vib. 330 (2011) 526–544.
[16] M.A. AL-Shudeifat, E.A. Butcher, C.R. Stern, General harmonic balance solution of a cracked rotor-bearing-disk system for harmonic and sub-harmonic
analysis: analytical and experimental approach, Int. J. Eng. Sci. 48 (2010) 921–935.
[17] C. Guo, M.A. AL-Shudeifat, J. Yan, L.A. Bergman, D.M. McFarland, E.A. Butcher, Application of empirical mode decomposition to a Jeffcott rotor with a
breathing crack, J. Sound Vib. 332 (2013) 3881–3892.
[18] C. Guo, M.A. AL-Shudeifat, J. Yan, L.A. Bergman, D.M. McFarland, E.A. Butcher, Stability analysis for transverse breathing cracks in rotor systems, Eur. J.
Mech. A/Solids 42 (2013) 27–34.
[19] Z.N. Haji, S.O. Oyadiji, The use of roving discs and orthogonal natural frequencies for crack identification and location in rotors, J. Sound Vib. 333 (2014)
6237–6257.
[20] M. Silani, S. Ziaei-Rad, H. Talebi, Vibration analysis of rotating systems with open and breathing cracks, Appl. Math. Model. 37 (2013) 9907–9921.
[21] A.S. Sekhar, Crack detection through wavelet transform for a run-up rotor, J. Sound Vib. 259 (2003) 461–472.
[22] J.-J. Sinou, Detection of cracks in rotor based on the 2x and 3x super-harmonic frequency components and the crack-unbalance interactions, Commun.
Nonlinear Sci. Numer. Simul. 13 (2008) 2024–2040.
[23] T.R. Babu, S. Srikanth, A.S. Sekhar, Hilbert–Huang transform for detection and monitoring of crack in a transient rotor, Mech. Syst. Signal Process. 22
(2008) 905–914.
[24] D. Guo, Z.K. Peng, Vibration analysis of a cracked rotor using Hilbert–Huang transform, Mech. Syst. Signal Process. 21 (2007) 3030–3041.
[25] Y.L. Lin, F.L. Chu, The dynamic behavior of a rotor system with a slant crack on the shaft, Mech. Syst. Signal Process. 24 (2010) 522–545.
[26] A.K. Darpe, A novel way to detect transverse surface crack in a rotating shaft, J. Sound Vib. 305 (2007) 151–171.
[27] M. Silani, S. Ziaei-Rad, H. Talebi, Vibration analysis of rotating systems with open and breathing cracks, Appl. Math. Model. 37 (2013) 9907–9921.
[28] G.I. Giannopoulos, S.K. Georgantzinos, N.K. Anifantis, Coupled vibration response of a shaft with a breathing crack, J. Sound Vib. 336 (2015) 191–206.
[29] A.K. Darpe, K. Gupta, A. Chawla, Transient response and breathing behaviour of a cracked Jeffcott rotor, J. Sound Vib. 272 (2004) 207–243.
[30] T. Zhou, Z. Sun, J. Xu, W. Han, Experimental analysis of cracked rotor, J. Dyn. Syst. Meas. Control 127 (2005) 313–320.
[31] Z. Ren, S. Zhou, C. E, M. Gong, B. Li, B. Wen, Crack fault diagnosis of rotor systems using wavelet transforms, Comput. Electr. Eng. 45 (2015) 33–41.
[32] H.B. Dong, X.F. Chen, B. Li, K.Y. Qi, Z.J. He, Rotor crack detection based on high-precision modal parameter identification method and wavelet finite
element model, Mech. Syst. Signal Process. 23 (2009) 869–883.
[33] Y. Lin, F. Chu, Numerical and experimental investigations of flexural vibrations of a rotor system with transverse or slant crack, J. Sound Vib. 324 (2009)
107–125.
[34] T.H. Patel, A.K. Darpe, Influence of crack breathing model on nonlinear dynamics of a cracked rotor, J. Sound Vib. 311 (2008) 953–972.
[35] M.A. AL-Shudeifat, On the finite element modeling of the asymmetric cracked rotor, J. Sound Vib. 332 (2013) 2795–2807.

(2016) Crack Detection For A Jeffcott Rotor With A Transverse Crack - An Experimental Investigation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(2016) Crack Detection For A Jeffcott Rotor With A Transverse Crack - An Experimental Investigation

Uploaded by

Copyright:

Available Formats

Mechanical Systems and Signal Processing ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Mechanical Systems and Signal Processing

Crack detection for a Jeffcott rotor with a transverse crack: An

2. Equations of motion of a cracked Jeffcott rotor

mu¨ + cu̇ + k1( t )u + k12( t )v = med Ω2 sin(Ωt + β )

Fig. 1. Cracked Jeffcott rotor system [17].

Fig. 2. Coordinate systems of the cracked rotor on the cross-section [15,17].

IY¯ (t ) ≅ I + f1(t )I11 + f2 (t )I22 (3)

where E is the modulus of elasticity, l is the length of rotor.

Fig. 3. Breathing process in one periodic for μ = 0.1.

3. Dynamic analysis and crack detection based on EMD method

Description Value Description Value

Length of rotor, L 0.4 m Gravitational acceleration, g 9.8 m/s2

Eddy current Disk Cracked shaft DC motor

Fig. 11. Experimental setup of the rotor testbed.

You might also like