See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/274703035

Condition monitoring of rotating machinery through vibration analysis

Article  in  Journal of Scientific and Industrial Research · April 2014

CITATIONS READS
3 3,231

2 authors:

Senthilkumar Mouleeswaran Sendhilkumar S.

PSG College of Technology Akshaya College of engineering and technology, Coimbatore, India
179 PUBLICATIONS   1,327 CITATIONS    19 PUBLICATIONS   26 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Solar Collector View project

Rotor-bearing system View project

All content following this page was uploaded by Sendhilkumar S. on 23 August 2018.

The user has requested enhancement of the downloaded file.

Journal of Scientific & Industrial Research
Vol. 73, April 2014, pp. 258-261

Condition Monitoring of rotating machinery through Vibration Analysis

S Sendhil Kumar1* and M Senthil Kumar2
1*
Anna University, Chennai
2
Department of Production Engineering, PSG College of Technology, Coimbatore

Received 31 August 2012; revised 21 July 2013; accepted 17 December 2013

Condition monitoring, commonly referred to as predictive maintenance, a proven approach to improve the reliability and
productivity in any industries. Its underlying philosophy is that technology can be used to measure and evaluate the
condition of plant assets and equipment; enabling intelligent decisions in regard to maintenance activities. Maintenance
processes are adopted by almost all the industries and are going well with preventive maintenance, but could not totally
implement the predictive maintenance to practice, especially in the small and medium scale industries. Predictive
maintenance by condition monitoring techniques will increase the efficiency, industrial safety and will reduce the
maintenance cost. This paper gives the brief idea of condition monitoring in a nutshell referring to vibration analysis.

Key words: Condition monitoring, Maintenance, Vibration

Introduction analyses to perform and to interpret the results. In this

Condition monitoring is the process of monitoring paper authors discusses the effort performed by
a parameter of condition in machinery, such that a various researchers in the area of the condition
significant change is indicative of a developing monitoring and also developed a test rig capable of
failure. The application of condition monitoring simulating machine faults, namely unbalance, shaft
allows maintenance to be scheduled, or other actions misalignment and its effects are investigated.
to be taken to avoid the consequences of failure,
before the failure occurs which is typically much Overview of vibration monitoring
more cost effective than allowing the machinery to Condition of any machineries can be monitored
fail. Out of the various condition monitoring either online or offline mode through a proper data
techniques, vibration signal analysis is used in a acquisition and efficient signal processing unit.
general way for fault diagnosis of mechanical Vibration monitoring comprises of several emerging
equipment. Any change in a system condition will and promising hardware, software components that
result in a change in dynamic response, so vibration offer a flexible and yet reliable approach for
signals carry a great deal of information about the diagnosing a new systems. A model based technique
equipment condition. Many techniques have been for fault diagnosis of rotor–bearing system is
reported for fault diagnosis based on vibration signals, described1.The fault condition and locations of faults
and they aim at finding some efficient fault features are successfully detected by this model based
from the vibration signals. However, it is still a technique. The effect of the coupling misalignment on
challenge to look for effective techniques that can the bearing vibration, adapting arithmetically the
isolate specific features from the vibration signals. exciting forces or moments due to the misalignment is
The developments of vibration measurement and evaluated2.
analysis for monitoring the condition of rotating
Diagnostic systems
machinery while in operation are most important one
The basic principles of machinery diagnostics and
in the vibration monitoring. These have been in all
several specific malfunction symptoms supported by
three areas of interest, namely fault detection,
simple mathematical models are presented3. The
diagnosis and prognosis. Of these areas, diagnosis and
various malfunctions included are unbalance,
prognosis still require an expert to determine what
—————— excessive radial load, rotor-to-stator rubbing, fluid
*
Author for correspondence induced vibrations, and loose stationary and rotating
Email: [email protected] parts, coupled torsional-lateral vibration excitation,
 KUMAR & KUMAR: CONDITION MONITORING OF ROTATING MACHINERY 259

and rotor cracking. The diagnostic systems4 using the introduced in the model which increases the
Spectra Quest Machinery Fault Simulator, Triaxial harmonics with misalignment can easily be modelled
vibration measurements were taken at each end of the by using finite element analysis. The location of the
coupling on the motor and rotor bearing housings. coupling with respect to the bending mode shape has
Data was collected at several other locations of the a strong influence on the vibrations. They numerically
Simulator. The results indicate that the speed and the evaluated the effects of coupling misalignment on the
coupling stiffness have a strong effect on the vibration 2X vibration response of a rotor-coupling bearing
spectra. The level and type of misalignment had a system. The problems connected with dynamics of
significant effect on the vibration signature. mechanical system at certain radial misalignment of
shafts are investigated in the offered work11. As an
Data acquisition system objective of research, the system of two shafts
The various test5 on a machinery fault simulator connected by elastic centrifugal ring coupling. In the
under various operating conditions is performed. coupling, connecting radially misaligned shafts, an
Operating data is simultaneously acquired using a internal moment of resistance to rotation is arising.
multi-channel data acquisition system. Since
misalignment produces dominant motion at the rotor Response of the system
running speed and its harmonics, this data is used to A clear correlation between misalignment &
construct an operational deflection shape (ODS). The vibratory symptoms that it produces to aid machinery
emphasis is on correlating the ODS of the machine diagnostic engineers to distinguish between shaft
when properly aligned with its ODS following misalignment and other problems that exhibit similar
induced shaft misalignment. The results of this work vibration symptoms using vibration analysis is
will provide a new perspective of machinery fault established12.The dynamic response of two Jeffcott
detection. The aim is to develop a more reliable tool rotors13 connected by rigid- type couplings with
for determining shaft misalignment and other machine parallel misalignment is studied. The study
faults from operating data. investigated the steady state and transient response of
the system and concluded that the presence of lateral
Severity of defects and torsional coupling is coupled. The general
The rise of vibration problems associated with
equations of motion are derived and given in
structures, which are more delicate and intricate
dimensionless form to represent the general case. The
machines, which are faster and more complex, and
expected vibration frequency for a misaligned
production processes, which are automated and
metallic-disc flexible coupling based on the analysis
interlinked is explained6.Information7 on the practical
of the structural vibrations produced by misalignment
treatment of alignment methods and preparations is
is determined14. Further, they showed experimentally,
provided. A general guideline for alignment
through real-time spectrum analysis, that all the
tolerances was illustrated. A theoretical model of a
theoretically predicted vibration frequencies are
complete motor rotor flexible coupling, discusses the
produced by a misaligned metallic-disc coupling, but
system response undergoing angular misalignment is
the 2× and 4× running speed components show the
presented8. An experiment on a rotor dynamic test
largest changes as misalignment increases.
apparatus to predict the vibration spectrum for shaft
misalignment is performed9.
Prognosis
Analysis of defects The steady- state vibration response at integer
Various researchers have analysed the presence and fraction of the first bending natural frequency is
effects of the defects in a rotating system. The discussed15. Full spectra and orbit plots are efficiently
improper aligning of shafts through couplings often used to reveal the unique nature of misalignment
leads to severe vibration problems in many rotating coupling characteristics under misalignment condition
machines is explained clearly10. The rotor bearing Full spectrum analysis of vibration signal. Several
system is modelled using higher order finite elements investigators have provided vibration identification
by considering deflection, slope, shear force, bending charts which indicate that the coupling misalignment,
moment with eight degrees of freedom per node. The generally, produces a frequency which is twice the
reaction forces and moments developed due to shaft speed frequency. An experiment to study the
flexible coupling misalignment are derived and vibration parameters of the system such as amplitude,
260 J SCI IND RES VOL 73 APRIL 2014

velocity and acceleration in vertical direction were mounted on the mid-way of the bearing supports. The
measured at the bearing points is performed16. A bearing pedestals and motor support are firmly
proposed back propagation neural network has been mounted on steel base. Using the radial screws the
used for vibration parameters prediction of a rotating disk is fastened to the rotor shaft, there are four
mechanical system at the bearing points using real threaded holes symmetrically at a radius of 35 mm to
data. Fixed system measurements such as working hold any desired amount of unbalance mass. By
speeds were used as inputs to the network. This kind varying the bearing pedestals various misalignments
of neural network predictors would be designed as a can be induced. Using the tachometer the speed is
controller of such systems. Finally, vibration measured and it is varied by adjusting the power
parameters of rotating systems could be predicted for supply to the motor. Using the sound and vibration
all directions. tool kit available in the LABVIEW software a
condition monitoring system is developed interfacing
Rotor-Bearing Test Rig: Identification of with a non-contact accelerometer of make NI 9233
Unbalance and Misalignment with a sensitivity of 0.9 g. An effective continuous
The rotor-bearing test rig is fabricated which signal extraction system is developed for monitoring
consists of an AC motor, a self designed coupling, a of misaligned rotor and unbalanced rotor signatures
single-disk rotor. The rotor shaft of length 900mm is and the extracted signals are diagnosed in time
supported by three identical pillow radial bearings of domain analysis. The experiment is conducted for the
bearing span of 500 mm. The diameter of the rotor speeds ranging from 600 to 2000 rpm. The signals are
shaft is 12 mm, a disk of outer diameter 50 mm, captured by placing the accelerometer in axial
weighing 1000 grams with the provision for position on the top of bearing housing. After
unbalance of 40 grams at a radius of 30 mm is achieving perfect alignment in offline conditions, then
a base line data is captured this can be used for further
comparison. The responses are compared across the
three bearing ends. For various modifications,
amplitude of the vibrations is measured and the
spectrum is plotted with horizontal scale of speed in
orders of RPM and vertical scale is velocity in
mm/sec. The response illustrates the level of the
velocity is within the limit in comparison with the
vibration severity chart. The velocity spectrum for the
coupling end of the rotor-bearing setup is shown in
the Fig 1. An experimental based investigation has
been made into the dynamics and control of rotor
Fig. 1-Velocity spectrum at CE
vibration due to unbalance and shaft misalignment.
The vibration along the drive side is lesser in
comparison to coupling end of the rotor. The level of
vibration due to unbalance is greater than the normal
condition. The vibration value at non drive end of the
test rig is illustrated in Fig 2. Values of the velocity
significantly vary with respect to the types of faults.
For parallel misalignment the velocity is lesser in
comparison with the angular misalignment, the
velocity for the unbalance significantly increases with
respect to the speed.
Conclusion
The various aspects of condition monitoring,
vibration analysis, predictive maintenance, and related
sensors are reviewed and presented in this paper. The
Fig. 2-Velocity spectrum at NDE procedure for detection, diagnosis & prognosis is
 KUMAR & KUMAR: CONDITION MONITORING OF ROTATING MACHINERY 261

discussed. The vibration monitoring test rig is 5 Surendra N.Ganeriwala (Suri) & Zhuang Li & Mark H.
developed with various modes of machine failure is in Richardson, Using operating deflection shapes to detect
misalignment in rotating equipment, IMAC XXVI, February
deed capable of generating common machine faults. 4-7 (2008).
The base signal was found to be within the safe limit 6 Al-Hussain K M, Dynamic stability of two rigid rotors
as per the vibration severity chart. Then as per the connected by a flexible coupling with angular misalignment,
design of experiments the various faults are J Sou Vib, 266 (2003) 217–34.
7 Pitorwoski, Shaft alignment handbook, MarcelDekker,
stimulated in the test rig, correspondingly, the NewYork, 1986.
vibration signals are extracted. From all the 8 Xu M & Marangoni R D, Vibration analysis of a motor-
characteristic curves it is found that the velocity value flexible coupling-rotor system subject to misalignment and
increases after reaching the speed of 800 rpm. In unbalance-Part I and II, J Sou Vib, 176 (1994) 663-679.
comparing the spectrum of all the defects, unbalance 9 Hariharan V & Srinivasan P S S, Vibration analysis of shaft-
ball bearing system, Indian J Sci Tech, 2 9 (2009) 45-50.
has produced higher vibrations at all the end 10 Sekhar S & Prabhu B S, Effects of Coupling misalignment
conditions. The values of displacement and velocity on vibrations of rotating machinery, J Sou Vib, 185 4 (1985)
for a speed of 1000 rpm is higher for a faulty 655-671.
condition in comparing with healthy condition 11 Saavedra P N & Ramirez D E, Vibration analysis of rotors
for the identification of shaft misalignment Part- I:
irrespective of the drive,coupling or non drive ends. Theoretical analysis, Proc. Instn Mech Engrs.Part-C. J.
The values of velocity significantly increases as the Mech Engg Sci, 218 (2004) 971-985.
misalignment increases from 0 mm to 2 mm. 12 Bachschmid N, Pennacchi P & Vania A, Accuracy of
modeling and identification of malfunctions in rotor systems:
References Experimental results, J Brazil Soci Mech Engrs, 22 3 (2000)
1 Arun Kr. Jalan & Mohanty A R, Model based fault 22-42.
diagnosis of a rotor–bearing system for misalignment 13 Al-Hussain K M & Redmond I, Dynamic response of two
and unbalance under steady-state condition, J Sou Vib, 17 rotors connected by rigid mechanical coupling with parallel
(2009) 20-30. misalignment, J Sou Vib, 249 3 (2002) 483-498.
2 Simon G, Prediction of vibration of large turbo-machinery on 14 Dewell D L & Mithchell L D, Detection of a misaligned disk
elastic foundation due to unbalance and coupling coupling using spectrum analysis, J Vib Aco, Str, Rel
misalignment, Proc of the Instn of Mech Engrs, 206 (1992) Des,106 (1984) 9-16.
29-39. 15 Tejas H Patel & Ashish K Darpe, Experimental
3 Agnes Muszynska, Vibrational diagnostics of rotating investigations on vibration response of misaligned rotors,
machinery malfunctions, Intl J of Rot Mach, 1 3-4 (1995) Mech Sys Sig Processing, 23 (2009) 2236–52.
237-266. 16 Menderes Kalkat, Sahin Yildirim & Ibrahim Uzmay, Design
4 Ganeriwala S, Patel S & Hartung H A, Observations of artificial neural networks for rotor dynamics analysis of
concerning misalignment vibration signature, Spectra Quest rotating machine system, J Brazil Soci Mech Engrs,15 (2004)
Inc (1998). 573-88.

View publication stats