OPEN ACCESS

http://www.sciforum.net/conference/wsf-4
World Sustainability Forum 2014 Conference Proceedings Paper

Vibration Analysis for Environmental Sustainability

David Almorza1, Mara Carmen Carnero2,*, Carlos Lpez-Escobar3, Rafael Gonzlez-Palma1
and Pedro Mayorga4
1

Universidad de Cdiz / Escuela Superior de Ingeniera, Calle Chile 1, 11002 Cdiz, Spain
Universidad de Castilla-La Mancha / ETSII, Avda. Camilo Jos Cela s/n, 13071 Ciudad Real, Spain
3
Alcoa Power and Propulsion / Plaza Obispo 3, Madrid, Spain
4
EnerOcean S.L. / Bulevar Louis Pasteur 5, Of.321, 29010 Mlaga, Spain
2

E-Mails: david.almorza@uca.es; carmen.carnero@uclm.es; Carlos.Lopez-Escobar@alcoa.com;

rafael.gonzalez@uca.es; pedro.mayorga@enerocean.com
* Mara Carmen Carnero; Tel.: +34-926-295-300; Fax: +34-926-295-361
Received: 20 August 2014 / Accepted: 9 October 2014 / Published: 1 November 2014

Abstract: Predictive maintenance has a significant impact on the environmental

sustainability of an organization, by means of the increase in quality of the product, which
allows the consumption of raw materials to be reduced. Vibration analysis is a predictive
technique especially suitable for reciprocating and rotary machines; it consists of
periodically recording the vibration level of the machine, as it increases when there are
anomalies such as misalignment, unbalance, etc. In the manufacture of bearings for vehicles,
the final vibration of the assembled bearing is one of the quality parameters analysed to
ensure that the quality of the bearings are as desired. In external grinding processes
vibrations appear which are generated by the process itself and not by flaws in the machine
tool. These vibrations can cause defects which affect the quality of the workpieces
produced. Nevertheless, analysis of process-induced vibrations is little studied in the
literature. Vibration spectra are applied to distinguish problematic vibration frequencies
from those that are not. Intercomparison of spectrums between similar machines and studies
of the change over time of spectra or waterfall plots are used. This method considers the
possible of establishing separate vibration limits in the machining of components so as to
guarantee the quality of the final assembled product. The methodology exposed can
contribute to increase the environmental sustainability of an industrial organization due that
contribute to a more efficient use of resources.
Keywords: quality; self-induced vibrations; sustainability; bearing; spectral vibration
analysis

2
1. Introduction
Organizations are increasingly concerned about contributing to environmental sustainability, and
they may have specific sections or departments dedicated to environmental protection. Nevertheless, to
date the effect that other areas, such as maintenance, can have on the matter has not been considered.
However, the application of optimal maintenance has important repercussions for the lifetime of
assets and facilities, lengthening their use and contributing to the optimal use of resources. In
maintenance policies applied by organizations we find:
Corrective maintenance, or run to failure policy, is carried out after failure detection with the
aim of returning the failed asset to an operational condition.
Preventive maintenance is carried out at set intervals throughout the working life of the
machine, with the aim of reducing the probability of problems or wear in the machine [1].
Predictive maintenance is a maintenance policy based on measuring and recording
intermittently or continuously (on-line) certain physical parameters associated with a working
machine to obtain a diagnosis of anomalies and faults present in a machine and a prediction of
the remaining life of the machine. There exist a number of predictive diagnostic techniques
such as [2]: vibration analysis, process parameters, thermography, visual inspection, tribology,
current spectrum, ultrasound, x-rays, penetrating liquids, etc. Other nondestructive testing
techniques can be seen in NASA [3], or the American Society for Nondestructive Testing
(ANST).
Predictive maintenance, unlike other maintenance policies, contributes to guaranteeing product
quality, because the machines are operating without anomalies (such as excessive vibrations), that have
a negative effect on quality. Guaranteeing product quality also contributes to a decrease in the
consumption of raw materials and components, and the waste generated by faulty batches.
Among predictive techniques, vibration analysis is the most widely used in organizations due to its
versatility of application to a variety of machines, and the number of faults that it can detect and
diagnose.
In the manufacture of bearings for vehicles, the vibration of the assembled bearing is one of the
quality parameters analysed to ensure that the lifetime of the bearing and the noise levels during
operation are as expected. However, although vibration in machines is widely analysed in the
literature, the vibrations produced by the processes themselves are not researched. This is because the
process vibration and frequently mixed in and confused with the vibrations of the machines.
The problem analysed in this study is the appearance of chattered grinding surfaces [4]. Detection
of chatter has been analysed in [5-11] while its removal is researched in [12-17]. Profile and surface
roughness are applied as indicators of the quality of the bearing, as they affect its functional behaviour
throughout their lifetime [18].
Vibration spectra are used to distinguish problematic vibration frequencies from those that are not;
the techniques used are intercomparison of spectrums between similar machines and waterfall plots.
This research considers the possibility of establishing vibration band limits in the machining of
components so as to guarantee the quality of the final assembled product.
Firstly, the characteristics of the experimental system used are shown; next the results obtained for
automotive bearings are described.

3
2. Experimental System
The measurement systems used in this study are made up of: vibration analyzer IRD 890
Mechanalysis, accelerometer 970 from Mechanalysis, and the databases for the storage and
manipulation of vibration data are the 7090 system of IRD Mechanalysis and ODYSSEY system from
Entek-IRD. To evaluate the quality of the bearing, measurement devices model Talyrond 31 and an
analogue Federal device, calibrated by certified bodies; this equipment is available in the normal
manufacturing areas of the company.
1. Level 0. Null chattering.
2. Level 1. Intermediate chattering.
3. Level 2. Severe chattering with significant widths of displacement.
Standard ANSI B89.3.1-1972 was applied to carry out the tests.
The design of the experiment followed a classical approach, considering all the interactions, and
applying variance analysis systematically. As a result of the experimental design, 25 tests were carried
out with two different machines dedicated to the same manufacturing process.
3. Results and Discussion
Vibrations can occur in displacement, velocity and acceleration. These magnitudes can be shown
against time and against frequency, in the latter case they are spectra. The Fourier Theorem states that
any periodic signal can be decomposed as a combination of pure sinusoidal waves, whose frequencies
are multiples of a fundamental frequency, giving a spectrum in frequency [19], using the Fast Fourier
Transform (FFT) algorithm. FFT allows a time vibration signal to be transformed into components
characterized by an amplitude and a frequency.
Vibration spectra were gathered in order to identify the vibration frequencies that cause poor
quality. Since there are no studies in the literature about the maximum permissible values and their
frequencies, the concept of spectral intercomparison between similar machines, and a waterfall
analysis, have been used.
The spectral intercomparison consists in a comparison of spectra from different machines
performing the same process and with similar operating variables, while the waterfall is a graph with
the successive spectra shown as a function of the variation of the machine velocity.
Therefore, a set of experiments was performed on two similar machines to determine whether the
process-induced vibrations allow spectral comparisons to be made to show which are the operating
conditions that affect the appearance of loss of quality.
The variables included in the experimental design are: the two grinding machine-tools, the grinding
wheel diameter, the rotational speed of the grinding wheel and, as output variables, high frequency
displacement and the quality of the process assessed according to the presence of high frequency
chattering and low frequency chattering. The choice of these variables has been made after a series of
prior experiments which can be seen in [20].
The waterfall plots of Figure 1 represent the spectral characteristics in displacement, in the chosen
range of frequencies for the grinding process in the outer raceway. It can be seen that the frequency of
appearance of vibration peaks is maintained, showing that the processes retain spectral identity if the

4
process variables are not modified. From the spectral identity obtained it is possible to define
vibration bands and so quantify the vibration content in each of them.
Figure 1. Waterfalls plots.
Displacement
(micron)

50

100

150

200

250 Frequency (Hz)

Next we see whether two identical machines have different spectral identities for different
processes. The comparison of spectra measured in velocity of the two different grinding processes
obtained from two different machines is shown in Figure 2, where the differences in amplitude and
frequency for each type of process can be seen.
Figure 2. Spectra from two identical machines in different grinding processes.
Velocity
(mm/s)

Velocity
(mm/s)

100

200

300

400

500 (Hz)

100

200

300

400

500 (Hz)

At the experimental stage, spectra are obtained throughout the life of the grinding wheel, as well as
the development of chattering in the manufactured components. It is observed that low-frequency
vibration is heavily influenced by the diameter of the grinding wheel. Random high-frequency spectral
components are also observed. Therefore, two vibration bands will be taken:

From 1Hz to 44 Hz.

From 44 Hz to 250 Hz.

The experimental results for high frequency displacement lead to the hypothesis that this variable is
related to problems with quality in the processes. A variance analysis was carried out to verify this
hypothesis [21] which gave as a result the possibility of establishing a monitoring band for high
frequency displacement, entre 44 Hz and 250 Hz, which allows the stability of the process with respect
to the resulting quality to be guaranteed.
Once the maximum admissible high-frequency vibration displacement values are known, limit
spectra can be defined, controllable cycle by cycle, and even connected to continuous vibration data

5
acquisition devices so that the machines can adjust their parameters and avoid the frequency band that
is sensitive to problems of quality.
4. Conclusions
Grinding processes produce induced vibrations that depend on the process variables. These selfexcited vibrations are independent of the input variables, the amplitudes and frequencies of the
vibrations, and the chattering assessed in the workpieces.
Spectral intercomparison is a technique used to detect problems of quality in grinding processes;
this is done by taking high-frequency displacement spectra. The high-frequency vibration is of the
same order even in different machines working on the same piece, which suggests that this variable is
linked to self-excited vibrations during tool-workpiece contact. For each process it will be possible to
define a given high-frequency band where the displacements should be controlled to guarantee
operation with stable quality.
Spectral identity and intercomparison allow monitoring bands to be defined, in which the spectral
activity indicates the imminent appearance of quality problems in the parts being ground. In the
process analysed this band lies between 44 Hz and 250 Hz.
The methodology described favours the environment by preventing quality problems in the parts
being ground; this leads to more efficient use of resources as it avoids batches of defective parts. These
defective batches have used energy, raw materials, components, etc., which in the end have been
wasted. This also causes a problem of disposal of the defective parts, as in many cases they cannot be
remanufactured.
Conflict of Interest
The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.

7.

Kelly, A. Strategic Maintenance Planning, Butterwoth-Heinemann: Oxford, UK, 2006.

Carnero, M. C. Predictive maintenance in small and medium enterprises. DYNA Management
2013, 1(1), doi:10.6036/MN5790.
NASA Reliability Centered Maintenance Guide for Facilities and Collateral Equipment, National
Aeronautics and Space Administration: Washington, USA, 2000.
Harris, T. A. Rolling bearing analysis. Journal of Synthetic Lubrication 1985, 1(4), 314315.
Cho D. W.; Eman K. F. Pattern recognition for on-line chatter detection. Mechanical Systems and
Signal Processing 1988, 2(3), 279-290.
Ruxu, D.; Elbestawi, M. A.; Ullagaddi, B. C. Chatter detection in milling based on the probability
distribution of cutting force signal. Mechanical Systems and Si gnal Processing 1992, 6(4), 345362.
Gradiek, J.; Baus, A; Govekar, E.; Klocke, F.; Grabec, I. Automatic chatter detection in grinding.
International Journal of Machine Tools and Manufacture 2003, 43(14), 1397-1403.

6
8.

9.
10.
11.
12.

14.
15.
16.
17.
18.

19.
20.

21.

Cardi, A. A.; Firpi, H. A.; Bement, M. T.; Liang, S. Y. Workpiece dynamic analysis and
prediction during chatter of turning process. Mechanical Systems and Si gnal Processing 2008,
22(6), 1481-1494.
Kuljanic, E.; Totis, G.; Sortino, M. Development of an intelligent multisensor chatter detection
system in milling. Mechanical Systems and Signal Processing 2009, 23(5), 1704-1718.
Wang, L., Liang, M. Chatter detection based on probability distribution of wavelet modulus
maxima. Robotics and Computer-Integrated Manufacturing 2009, 25(6), 989-998.
Yao, Z.; Mei, D.; Chen, Z. On-line chatter detection and identification based on wavelet and
support vector machine. Journal of Materials Processing Technology 2010, 210(5), 713-719.
Shiraishi M.; Kume E.; Hoshi T. Suppression of Machine-Tool Chatter by State Feedback Control,
CIRP Annals - Manufacturing Technology 1988, 37(1), 369-372.Tarng, Y. S., Kao, J. Y., Lee, E.
C. Chatter suppression in turning operations with a tuned vibration absorber. Journal of Materials
Processing Technology 2000, 105(1-2), 55-60.
Inasaki, I.; Karpuschewski, B.; Lee, H.S. Grinding Chatter Origin and Suppression. CIRP
Annals - Manufacturing Technology 2001, 50(2), 515-534.
Govekar E.; Baus A., Gradiek J.; Klocke F.; Grabec I. A New Method for Chatter Detection in
Grinding. CIRP Annals - Manufacturing Technology 2002, 51(1):267-270.
Ema, S.; Marui, E. Theoretical analysis on chatter vibration in drilling and its suppression, Journal
of Materials Processing Technology 2003, 138(1-3), 572-578.
Zhongqun, L.; Qiang, L. Solution and Analysis of Chatter Stability for End Milling in the Timedomain. Chinese Journal of Aeronautics 2008, 21(2), 169-178.
Salgado, D. R.; Alonso, F. J.; Cambero, I.; Marcelo, A. In-process surface roughness prediction
system using cutting vibrations in turning. International Journal of Advanced Manufacturing
Technology 2009, 43(1-2), 40-51.
Muszynska, A. Vibrational Diagnostics of Rotating Machinery Malfunctions. Course Rotor
Dynamics and Vibration in Turbomachinery, Belgium, 21-25 September, 1992.
Carnero, M. C.; Gonzlez-Palma, R.; Almorza, D.; Mayorga, P.; Lpez-Escobar, C. Statistical
Quality control through overall vibration analysis. Mechanical Systems and Si gnal Processing
2010, 24, 1138-1160.
Lpez-Escobar, C.; Gonzlez-Palma, R.; Almorza, D.; Mayorga, P.; Carnero, M. C. Statistical
quality control through process self-induced vibration spectrum analysis. International Journal of
Advanced Manufacturing Technology 2012, 58, 1243-1259.

2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license.