Engineering Failure Analysis 79 (2017) 845–851

Contents lists available at ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier.com/locate/engfailanal

Optimization of noisiness of mechanical system by using a MARK

pneumatic tuner during a failure of piston machine
Lucia Žuľováa, Robert Gregaa, Jozef Krajňáka, Gabriel Fedorkob,⁎, Vieroslav Molnárb
a
Faculty of Mechanical Engineering, Technical University of Kosice, Letna 9, 042 00, Kosice, Slovak Republic
b
Technical University of Kosice, Park Komenskeho 14, 042 00, Kosice, Slovak Republic

AR TI CLE I NF O AB S T R A CT

Keywords: The stage of construction and operation of the machines requires monitoring of the steps causing
Mechanical system negative effects on humans and environment in the form of noise during their use. Current
Pneumatic tuner technologies allow for appropriate optimization of mechanical systems as part of these machines.
Noisiness One possible optimization of the mechanical systems is considered to be the use of pneumatic
tuners. The aim of this article is to demonstrate the effect of the pneumatic tuner on noisiness of
mechanical system even in case of failure of mechanical drive part. Performed experimental
measurement presents the change of noise in mechanical system by changing a pressure of
gaseous medium in pneumatic tuner's compression space. The solution is an issue of appropriate
mechanical tuning system when there is a change of dynamic parameters of the system and thus
to change the entire mechanical system noise. Subsequent obtained results set out suggestions for
the use of pneumatic tuners in mechanical systems in order to achieve the lowest possible noise
during their operation mode.

1. Introduction

In technical practice, machinery and their equipment are included into process of energy conversion. When converting energy, a
change in movement, or more precisely a reduction of some of physical quantities characterizing mechanical parameters of energy
transfer occur. Such changing characteristic parameter is considered to be an operating speed of mechanical system. During this
energy conversion, there a loss - idle energy also occurs. The part of this energy changes into a vibrating motion of individual parts of
the mechanical system and consequently the noise level increased. The question of noise of machineries is nowadays a very sub-
stantial issue which effects the sphere of machinery dynamics, their maintenance and technical diagnostics [1–5].
In terms of dynamics, every machine or machinery performing a mechanical movement can be considered as a mechanical
system. The sources of dynamic forces induce mechanical motion of machine's surfaces. Subsequently, the vibrations of surfaces are
transmitted into the air, which in technical practice are perceived as noise. The noise generated by mechanical systems has a negative
impact on durability, reliability and safety of mechanical systems [6–9].
During an operating work of these mechanical systems, either transient or steady running, torsional shaking or vibrations may
arise and their intensity depends on the dynamic properties of the system, e.g. natural frequency and time of excitation. The men-
tioned torsional vibrations and shaking become a noise source in the mechanical systems [10–13].
The signs of failures of individual parts of machines or machineries can be considered as results of increased vibration and
consequential noise. Typical signs of predictable failure, manifested by increased noise are:

⁎
Corresponding author.
E-mail address: gabriel.fedorko@tuke.sk (G. Fedorko).

http://dx.doi.org/10.1016/j.engfailanal.2017.05.044
Received 1 March 2017; Received in revised form 12 May 2017; Accepted 29 May 2017
Available online 30 May 2017
1350-6307/ © 2017 Elsevier Ltd. All rights reserved.
L. Žuľová et al. Engineering Failure Analysis 79 (2017) 845–851

- release of rotors and hub,

- fatigue failures of shaft,
- breaks of rotating machine parts,
- failures of gears and flexible couplings [14–18].

In case of our examination, a failure of piston machine, which is part of the mechanical drive, is monitored. The failure is
represented by a decommissioning of one of the piston cylinder from running.
Mentioned examination includes proper tuning of the mechanical system besides in failure-free mode even in failure mode. The
carried tuning affects dynamic properties of the mechanical system and thus changes noisiness of the entire mechanical system. One
of the possible solutions of the optimizing of mechanical system is the use of pneumatic tuners in mechanical drive. By changing the
pressure of the gaseous medium in the pneumatic compression space of tuners, it is possible to change a torsional stiffness, and hence
the natural frequency of the mechanical system. According to several authors, the most proper pneumatic tuner in mechanical
systems is pneumatic flexible coupling [19,20].
The basic characteristic of the pneumatic flexible coupling is a compression space, located between the drive and driven parts of
the coupling. A compression space is formed by one, optionally more pneumatic flexible elements that are, or may not be inter-
connected. Pneumatic flexible element is filled with a gaseous medium (usually air). Construction and operating principles of the
pneumatic flexible couplings allows for a wide variability of their strength and operational characteristics that belong to their
exceptional properties.
For propagation of the properties of pneumatic flexible coupling, it has become an integral part of the observed mechanical
system, which was created for experimental measurements to fulfill the purpose of this article. By demonstration of the effect of the
pneumatic flexible couplings, the noisiness of mechanical system was monitored [21,22,23]. The main aim of the article is to present
the influence of the pneumatic tuner on the noise of the mechanical system, which we examined in a failure-free mode and failure-
mode of the mechanical system.

2. Preparation for laboratory experiment

To provide the noise experimental measurements, the mechanical system was assembled in laboratory conditions. The realized
mechanical system (Fig. 1) consists of a three-cylinder air compressor (4), which is driven via pneumatic flexible coupling (3) by a
three-phase asynchronous electric motor (1). Another part of the mechanical system located after electromotor is a two-speed
gearbox (2). Specific parameters of devices are in Table 1. The mechanical system operates in the range of operating speed. A work
mode of the mechanical system can be continuously varied by a changing the operating speed of the electric motor. Into the me-
chanical system, a three-cylinder compressor is included and it is used as a producer of torsional vibration. The failure of piston
machine is simulated by decommissioning of one piston cylinder which causes a significant change in torsional vibration.
Long-lasting operation of mechanical systems, which causes fatigue and aging of particularly flexible shaft couplings, as well as
randomly generated phenomena induced by change of characteristics of the drive and driven equipment, it affects the steady running
of the mechanical system particularly in terms of dangerous torsional vibration and noise. As a result of the mentioned phenomena, a
tuned mechanical system becomes out of tune. In this case, the flexible shaft coupling is not able to reduce and in some cases even
eliminate the increasing torsional vibration, causing the noise of the whole system. To ensure proper tuning of vibrating mechanical
systems, the application of the pneumatic flexible coupling is suitable solution for the mechanical system's tuning.
Changing the pressure of the gaseous medium in the pneumatic flexible coupling affects its stiffness and damping properties, thus
it can optimize the whole system. Fig. 2 shows a pneumatic flexible shaft coupling designated 4-1/70-T-C used in measurements of
mechanical system noise.
Torsional stiffness k is allowed to be a dynamic property that significantly affects the natural frequency of the mechanical drive.
Changing the natural frequency of the mechanical drive, it is possible to avoid the resonance area [19,24,25]. In Fig. 3, courses of
torsional stiffness k at various value of air pressure in the pneumatic flexible coupling depending on the load torque Mk are showed.

Fig. 1. Experimental mechanical system. 1 - three-phase asynchronous electric motor, 2 - two-speed gearbox, 3 - pneumatic flexible coupling, 4 - three-cylinder air
compressor.

846
L. Žuľová et al. Engineering Failure Analysis 79 (2017) 845–851

Table 1
Specific parameters of devices included in measured mechanical system.

1. Three-phase asynchronous electric motor

• Power: 11 kW
• Operating speed: 0–3000 min− 1
2. Two-speed gearbox
• Helical gearing
• Gear ratio: 1:1
3. Pneumatic flexible coupling 4-1/70-T-C
• 4 – number of pneumatic flexible elements (PFE)
• 1 – number of bellow on PFE
• 70 – outer diameter of PFE in mm
• T – type of coupling in terms of construction; T - tangential
• C – full interconnection of PFE
4. Three-cylinder air compressor
• Maximum overpressure: 600 kPa
• Type: ORLIK 3JSK-75

From the courses of the torsion stiffness k is clear that the course of the torsional stiffness has linear character at constant pressure in
the pneumatic flexible coupling. Only in the case of the lowest pressure, a course of torsional stiffness k becomes non-linear. In-
creasing course of torsional stiffness is monitored with increasing load of the mechanical system. The courses of torsional stiffness k
create an area which can limit the minimum and maximum torsional stiffness designated as kmin - kmax. Then according to needed
tuning, by changing air pressure in the pneumatic flexible coupling, any torsional stiffness can be achieved in defined area.

3. Measurement method

In order to obtain the results of noisiness of the mechanical system, several experiments in the laboratory of measurement and
tuning of torsional vibration were conducted. Thereafter we were able to evaluate recommendations, it was necessary to determine
the quantity of the measurement. Our aim was not to analyze the noise source, but to assess the possibility of applying a pneumatic
tuner to ensure the overall noise level changes in mechanical propulsion, even in case of failure of the piston machine. After careful
consideration, as a measuring quantity was chosen A-weighted equivalent sound pressure level LAeq, that values were obtained from
sound-level meter during the measurement. During measurement, a sound-level meter was placed at a height of 1.5 m from the floor
on a tripod. Distance of sound-level meter from the mechanical system was 1 m. Measuring range and working range of the me-
chanical system during the measurement were as follows:
- range of the gas pressure 200 kPa ÷ 700 kPa.
- operating speed of electric motor: 200 rpm− 1 ÷ 1000 rpm− 1.
- constant load of the mechanical system.

Fig. 2. Pneumatic flexible coupling used in the mechanical system.

847
L. Žuľová et al. Engineering Failure Analysis 79 (2017) 845–851

Fig. 3. Courses of the torsional stiffness of pneumatic flexible coupling.

Scheme of measurement apparatus with a sound-level meter placement is shown in Fig. 4. During noise measurements, following
diagnostic conditions have been observed:
- required microphone distance from the noise source,
- Microphone placed in the path of sound waves from the measured object,
- Appropriately adjusted measuring range of sound-level meter.
- Keeping documentation of measuring run.
To measure noise, these devices were used:
- Hand-held sound-level meter Brüel & Kjær Type 2250,
- Microphone Type 4189, nominal sensitivity: 50 mV·Pa − 1,
- Acoustic calibrator Brüel & Kjær.

4. Results and discussion of experiment

4.1. Results of the experiment in failure-free mode

The primary aim of the measurement was to identify the impact of pneumatic tuner on the noisiness of the mechanical system
which will work in failure-free mode. Mechanical system worked in range of operating speed. During each speed mode, dynamic
characteristics of pneumatic tuner has been changed by changing a pressure in pneumatic flexible coupling varied in the range of
200 kPa ÷ 700 kPa. For each pressure change, repeated noise measurement was realized. In Fig. 5 can be seen a noise courses LAeq of
the mechanical system for failure-free mode. From the course of noise level LAeq it is evident that pneumatic tuner markedly in-
fluences noisiness of the mechanical system. Obviously, the range of the impact is more than 20 dB. An important finding is the fact
that the pneumatic tuner is required to be filled with different pressure during various operating speed to achieve the minimum value
of noise level LAeq. This state can be called as the optimum tuning of mechanical systems in terms of noise level LAeq.
Another important finding that can be analysed from Fig. 5 is that the mechanical system passes through resonance. The re-
sonance occurs in operating speed from 550 rpm− 1–750 rpm− 1, depending on the air pressure in the pneumatic tuners. As the
analysis shows, for each operating speed mode it is suitable a different pressure in the pneumatic tuners, thus confirming the

Fig. 4. Scheme of measurement apparatus with a sound-level meter placement.

848
L. Žuľová et al. Engineering Failure Analysis 79 (2017) 845–851

Fig. 5. Courses of noise level LAeq for the operating range of mechanical systems in failure-free mode.

assumption that the change of torsional stiffness k of pneumatic tuner affects the noise level LAeq of the mechanical system.

4.2. Results of the experiment in failure mode

In other experiment, we simulate a failure of piston machine of the mechanical drive. Piston machine was represented by
compressor in which one of the cylinders did not work. Following this simulated failures, we were again carried out measurements of
noise level LAeq. Fig. 6 shows courses of noise level LAeq for the operating range of mechanical systems with various tuned pneumatic
coupling.
For running of the machine with a failure, the pressure change in pneumatic tuner also affects the noise level LAeq of the me-
chanical system. The range of noise level LAeq varies in 20 dB. However, a resonance, which occurs in failure-free mode, is not
monitored. Also in this case, the air pressure change alters the torsional stiffness, which affects the natural frequency of the me-
chanical system and consequently the noise level LAeq.

4.3. Discussion of the experiment in both mods

The courses of minimum noise level LAeq by optimizing the pressure in the pneumatic tuners for failure-free mode and failure
mode is shown in Fig. 7. Seen courses present trend both mode which is increasing, but in any mode, different values are obtained.
For failure-free mode was noise level LAeq of mechanical system in under-resonance area (operating speed range up to 550 rpm− 1)
comparable lower than noise level LAeq in failure mode.
Minimal noise level LAeq is achieved by optimizing the air pressure in the pneumatic tuners. While optimizing an air pressure, the
torsional stiffness k of the pneumatic tuner is also optimized, whereby we change the natural frequency of the mechanical system. In
Fig. 8 is shown a course of torsional stiffness k for achievement of the minimum noise level LAeq in failure-free mode and failure mode
of piston machine. By the pneumatic tuner, a natural frequency of the mechanical system was varied in order to achieve a minimum

Fig. 6. Courses of noise level LAeq for the operating range of mechanical systems in failure mode.

849
L. Žuľová et al. Engineering Failure Analysis 79 (2017) 845–851

Fig. 7. The courses of minimum noise level LAeq by optimizing the pressure in the pneumatic tuners for failure-free mode and failure mode.

noise level LAeq. The courses of torsional stiffness k in failure-free mode and failure mode monitored in operating speed range from
500 rpm− 1 ÷ 600 rpm− 1 are identical. However, at identical torsional stiffness k in both modes, the minimum noise level LAeq
(Fig. 7) at that operating speed is significantly different. On the contrary, at the same value of the minimum noise level LAeq at an
operating speed of 650 rpm− 1 and 850 rpm− 1 is the value achieved at significantly different torsional stiffness k.
This finding confirms that the achievement of a minimum value of noise level LAeq is influenced by the change of torsional stiffness
of the pneumatic tuners. In the case of mechanical drives proportionality is not applicable, because the minimum value of the
torsional stiffness k of the pneumatic tuner will not ensure the minimum values of noise level LAeq.
Experiment was preceded by careful preparation and several partial experimental measurements. Subsequently, the complete
noise measurement of the mechanical system proceeded without unpredictable problems.

5. Conclusions

From the measured values and analysis during the observed noise levels LAeq it can be stated that by changing the pressure of the
gaseous medium in the compression space of the pneumatic tuner, noise level LAeq varies across the mechanical system, even in case
of failure of the piston machine. From the measured noise levels LAeq, the impact of pneumatic tuners is visible within their tuning.
This ensures during a running state, it is possible to avoid unwanted resonance and consequently there is not a dangerous torsional
oscillation monitored.
Experimental measurements of the impact on pneumatic tuner on noisiness of the mechanical systems confirmed their influence
on noisiness in range of 20.4 dB during particular operating speed. The obtained results confirm the change in noise level LAeq of
mechanical system where difference of 13.5 dB was monitored at the lowest and highest pressure in the compression at operating
speed 750 rpm− 1.
We can definitely state that even in case of failure of any members of the drive; just by applying the pneumatic tuners can
effectively reduce the noise level LAeq. The basis for the creation of such a reduction is tuning of dynamic properties of the mechanical
system by optimizing the torsional stiffness k of the pneumatic tuner where noise level LAeq of the mechanical system is the lowest.
The presented results confirm the fact that the appropriate ways of tuning contribute to improving the reliability and technical

Fig. 8. The courses of torsional stiffness k for achievement of the minimum noise level LAeq in failure-free mode and failure mode.

850
L. Žuľová et al. Engineering Failure Analysis 79 (2017) 845–851

level of all torsional vibrating mechanical systems. Our future research will be focused mainly on the field of torsion vibration tuning
in mechanical drives.

Acknowledgements

This work is a part of these projects VEGA 1/0258/14, VEGA 1/0619/15, VEGA 1/0063/16, VEGA 1/0327/17, KEGA 018TUKE-
4/2016.

References

[1] G. D'Elia, E. Mucchi, M. Cocconcelli, On the identification of the angular position of gears for the diagnostics of planetary gearboxes, Mech. Syst. Signal Process.
83 (2017) 305–320, http://dx.doi.org/10.1016/j.ymssp.2016.06.016.
[2] R.B. Randall, J. Antoni, Rolling element bearing diagnostics—a tutorial, Mech. Syst. Signal Process. 25 (2011) 485–520, http://dx.doi.org/10.1016/j.ymssp.
2010.07.017.
[3] L. Jakubovičová, P. Zavadinka, J. Jakubovič, Transport duty cycle measurement of hybrid drive unit for mixing drum, Adv. Intell. Syst. Comput. 393 (2016)
219–224.
[4] L. Charroyer, O. Chiello, J.-J. Sinou, Parametric study of the mode coupling instability for a simple system with planar or rectilinear friction, J. Sound Vib. 384
(2016) 94–112, http://dx.doi.org/10.1016/j.jsv.2016.07.037.
[5] F.J. Jiménez Espadafor, J.A. Becerra Villanueva, D. Palomo Guerrero, M. Torres García, E. Carvajal Trujillo, F. Fernández Vacas, Measurement and analysis of
instantaneous torque and angular velocity variations of a low speed two stroke diesel engine, Mech. Syst. Signal Process. 49 (2014) 135–153, http://dx.doi.org/
10.1016/j.ymssp.2014.04.016.
[6] J. Kuľka, E. Faltinová, M. Kopas, M. Mantič, Diagnostics and optimisation of crane track durability in metallurgical plant, Diagnostyka 17 (2016) 41–46.
[7] P. Folęga, G. Wojnar, R. Burdzik, Ł. Konieczny, Dynamic model of a harmonic drive in a toothed gear transmission system, J. Vibroeng. 16 (2014) 3096–3104.
[8] P. Charles, J.K. Sinha, F. Gu, L. Lidstone, A.D. Ball, Detecting the crankshaft torsional vibration of diesel engines for combustion related diagnosis, J. Sound Vib.
321 (2009) 1171–1185, http://dx.doi.org/10.1016/j.jsv.2008.10.024.
[9] E. Lumnitzer, M. Andrejiová, A. Goga Bodnárová, Verification of the impact of the used type of excitation noise in determining the acoustic properties of
separating constructions, Measurement 78 (2016) 83–89, http://dx.doi.org/10.1016/j.measurement.2015.09.030.
[10] D. Pisarski, C.I. Bajer, B. Dyniewicz, J.M. Bajkowski, Vibration control in smart coupled beams subjected to pulse excitations, J. Sound Vib. 380 (2016) 37–50,
http://dx.doi.org/10.1016/j.jsv.2016.05.050.
[11] B. Łazarz, G. Wojnar, C. Piotr, Early fault detection of toothed gear in exploitation conditions, Maint. Reliab. 1 (2011) 68–77.
[12] L. Jakubovičová, M. Sága, M. Vaško, Impact analysis of mutual rotation of roller bearing rings on the process of contact stresses in rolling elements, Manuf.
Technol. 13 (2013) 50–54.
[13] Ł. Konieczny, R. Burdzik, J. Warczek, P. Czech, G. Wojnar, J. Młyńczak, Determination of the effect of tire stiffness on wheel accelerations by the forced vibration
test method, J. Vibroeng. 17 (2005) 4469–4477.
[14] P. Czech, G. Wojnar, R. Burdzik, J. Konieczny, Łukasz Warczek, Application of the discrete wavelet transform and probabilistic neural networks in IC engine fault
diagnostics, J. Vibroeng. 16 (2014) 1619.
[15] V. Dekys, A. Sapietova, Use of MSC.ADAMS software product in modelling vibration sources, Komunikacie 18 (2016) 101–107.
[16] J. Kulka, M. Mantic, G. Fedorko, V. Molnar, Failure analysis of increased rail wear of 200 tons foundry crane track, Eng. Fail. Anal. 67 (2016) 1–14, http://dx.
doi.org/10.1016/j.engfailanal.2016.05.032.
[17] M. Puškár, P. Bigoš, M. Kelemen, R. Tonhajzer, M. Šima, Measuring method for feedback provision during development of fuel map in hexadecimal format for
high-speed racing engines, Measurement 50 (2014) 203–212, http://dx.doi.org/10.1016/j.measurement.2014.01.005.
[18] L. Jakubovičová, M. Sága, P. Kopas, M. Handrik, M. Vaško, Some notes on analysis of bending and torsion combined loading, Mach. Dyn. Res. 34 (2010) 97–105.
[19] P. Kaššay, J. Homišin, P. Čopan, M. Urbanský, Verification of torsional oscillating mechanical system dynamic calculation results, Zesz. Nauk. Politech. Ślaskiej.
1907 (2014) 29–34.
[20] J. Homišin, P. Kaššay, M. Puškár, M. Moravič, Continuous tuning of ship propulsion system by means of pneumatic tuner of torsional oscillation, Int. J. Marit.
Eng. 158 (2016) 231–238.
[21] H. Madej, P. Czech, Discrete wavelet transform and probabilistic neural network in ic engine fault diagnosis, Maint. Reliab. (2010) 11–14.
[22] J. Krajňák, J. Homišin, R. Grega, M. Urbanský, The analysis of the impact of vibrations on noisiness of the mechanical system, Pneuma 17 (2016) 21–26.
[23] R. Grega, J. Homisin, The chances for reduction of vibrations in mechanical system with low-emission ships combustion engines, Int. J. Marit. Eng. 157 (2015)
235–240.
[24] B. Lazarz, G. Wojnar, H. Madej, P. Czech, Evaluation of gear power losses from experimental test data and analytical methods, Mechanika 8 (2009) 56–63.
[25] J. Homisin, Tuning methods of mechanical systems by means of torsional oscillation tuner application, Pneuma 61 (2006) 32–35.

851