Acoustic intensity technique applied to monitor planetary gears
Introduction
Gear transmission systems are critical components in many industrial applications (power generation, automobiles, agricultural and construction machinery, helicopters, etc.) [1], [2], [3], [4]. In consequence, the reliability requirements for this type of elements are increasing every day, both due to the economic costs and to the human life risk, to which a system failure can lead [5], [6], [7], [8], [9], [10].
In the last decades, the use of planetary gear trains has increased due to their compactness and lightness, especially when high-torque levels and/or high transmission ratios are involved. The increase in their use has been accompanied by more demanding technical requirements and has led to the need for improving their design, and more specifically, their reliability. Although the vibration sources for planetary gears are the same as in fixed-axe trains, planetary gears have unique features which add complexity to the problem. Regarding this aspect, many researchers have developed models, both theoretical and experimental, in order to predict and optimize their dynamic behaviour under different operating conditions [11], [12], [13]. Likewise, several experimental techniques have been proposed, mainly focused on both the theoretical model validation and the development of predictive maintenance techniques for this mechanical system [14], [15], [16].
Its compactness implies important difficulties in monitoring this kind of gears by means of vibration and/or strain gauges measurements. Specifically, the sensors installation inside the train and their wiring are hampered by the scarce free volume in its interior and by the complex relative movement among its components. Even the use of telemetry systems instead of wires and slip rings is not always a valid solution. Among other drawbacks, its use implies the modification of the mass of the systems and the possible introduction of unbalances. The installation of external sensors is simpler but needs a contact with the device and requires space, which sometimes is unavailable.
The present work aims to offer an alternative to the abovementioned methods, by using acoustic measurements for condition monitoring [17], [18], [19], [20] and its application to predictive maintenance. For the sake of simplicity and in order to offer a competitive alternative, acoustic characterization proposed in this paper is developed by measuring sound intensity [21], [22] and pressure. With this approach, the sensors can be placed in an easier and quicker way and, more importantly, they do not need to be in contact with the components. To guarantee the success of the procedure, it is vital that the acoustic surrounding of the system remains approximately stable, which makes it suitable for a wide range of applications, from monitoring a wind turbine gearbox inside its nacelle to control quality in a gearbox assembly line. In order to achieve this goal, a set of measurements were developed in a planetary test bench [23].
In order to validate the proposed acoustic technique, it has been used to characterise the behaviour in operation of a specific planetary test bench, which were previously developed by the authors, extensively presented in the literature [23], [24], [25] and summarised in Section 2 for the sake of simplicity. The proposed acoustic technique is detailed in the third section. In Section 4, the characterisation results are shown when this acoustic technique is used in a perfectly lubricated gear set, and then compared with those obtained in presence of defects (no lubrication and a notch on the tooth root), which are presented in Section 5. Finally, conclusions of the study are detailed in Section 6.
Section snippets
Test bench set-up
The test bench used in this work was previously designed and built by the research team [23], [24], [25] in order to experimentally validate an advanced computational model of planetary gear transmissions. It is a back-to-back configuration (Fig. 1) with a mechanical power recirculation composed by two identic planetary stages with a sun, three planets, a ring and a carrier each. The teeth number of the sun, planets and ring is respectively 16, 24 and 65. One of the stages is the test specimen,
Measurement tool developed to perform the acoustic characterisation
One of the main tasks of this work was the development of a unified computer tool which allows for the measurement control, the treatment of the registered data and the representation of the resultant parameters. The Matlab® environment was chosen as programming platform for this tool due to its built-in powerful routines of application in each of the aforementioned aspects. The measuring chain that controls the computer tool consists of the following elements: Intensity probe (type-2
Test bench acoustic characterisation
In order to perform the acoustic characterization of the test bench without defects, both intensity and sound pressure measurements were carried out for different operating conditions. For this purpose, 9 load levels were established from 200 Nm to 1800 Nm (besides the level 0 which represents idle conditions) and 3 motor speeds: 300, 600 and 900 rpm. These values were considered to be representative of the potential operating conditions to which the gear train will be subjected in its actual
Maintenance application
Once the acoustic characterization of the test bench was carried out, the application of the tool in the predictive maintenance of gears were assessed. In order to check the validity of these techniques in maintenance applications, two different tests were performed. In the first, a run-out of lubrication was reproduced in the test bench, whilst, in the second, a defective planet was incorporated.
Conclusions
In this study, the use of acoustic measurements of pressure and intensity in the characterization and condition monitoring of planetary gear boxes was assessed. The measurements were carried out on a planetary gearbox installed in a specific test bench. The gearbox was run under normal operating conditions and with two different defects separately. The acoustic results were presented through frequency spectra and intensity maps.
Mesh frequencies are better identified in acoustic pressure spectra
CRediT authorship contribution statement
P. Garcia Fernandez: Conceptualization, Writing - original draft, Writing - review & editing, Supervision, Investigation. Ana de-Juan: Writing - original draft, Writing - review & editing, Investigation. A. Diez-Ibarbia: Writing - original draft, Writing - review & editing, Investigation. J. Sanchez-Espiga: Writing - original draft, Investigation. A. Fernandez del Rincon: Writing - original draft, Supervision, Investigation.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgment
This work has been supported by project DPI2017-85390-P funded by the Spanish Ministry of Economy, Industry and Competitiveness.
References (25)
On holographic reconstruction without reference: application to machinery noise and comparison with multi-reference holography approach
Appl Acoust
(2017)- et al.
Prediction of airborne radiated noise from lightly loaded lubricated meshing gear teeth
Appl Acoust
(2015) - et al.
Review of condition monitoring and fault diagnosis technologies for wind turbine gearbox
Procedia CIRP
(2013) - et al.
Detection of gear failures via vibration and acoustic signals using wavelet transform
Mech Syst Sig Process
(2003) - et al.
Multiple-teeth defect localization in geared systems using filtered acoustic spectrogram
Appl Acoust
(2013) - et al.
On the identification of the angular position of gears for the diagnostics of planetary gearboxes
Mech Syst Sig Process
(2017) - et al.
Vibro-acoustic propagation of gear dynamics in a gear-bearing-housing system
J Sound Vib
(2014) A revised model for the extraction of periodic waveforms by time-domain averaging
Mech Syst Sig Process
(1987)- et al.
Comparison of experimental and operational modal analysis on a back to back planetary gear
Mech Mach Theory
(2018) - et al.
Effects of variable loading conditions on the dynamic behaviour of planetary gear with power recirculation
Measurement
(2016)
Acoustic monitoring of engine fuel injection based on adaptive filtering techniques
Appl Acoust
The application of the sound-intensity technique to defect detection in rolling-element bearings
Appl Acoust
Cited by (6)
Numerical evaluation of the accuracy in the load sharing calculation using strain gauges: Sun and ring gear tooth root
2022, Mechanism and Machine TheoryCitation Excerpt :In the last decades, numerous experimental works have been developed to characterize the planetary transmissions behaviour. In order to study the dynamical behaviour of a planetary transmission diverse techniques are appropriate, accelerometry [23–26], acoustic emission [26,27], and acoustic intensity [28] among others. Another technology that is employed for experimental studies is extensometry, which allows to calculate the stains in the root tooth and the amount of load in such tooth [29–31].
A method of in-situ monitoring multiple parameters and blade condition of turbomachinery by using a single acoustic pressure sensor
2022, Mechanical Systems and Signal ProcessingCitation Excerpt :Mongia et al. [25] pointed out in the research that vibration, temperature, and acoustic signals are the most widely used parameters of machinery health monitoring, and summarized the acoustic signals-based studies conducted for monitoring the health of the rotating machinery. Fernandez et al. [26] studied the possibility of using acoustic measurements for the planetary gear-set characterization, as well as for predictive maintenance of this mechanical system. Generally, the acoustic signals of health monitoring are weak and acoustic analysis is sensitive to environmental noise.
Multi-component fault classification of a wind turbine gearbox using integrated condition monitoring and hybrid ensemble method approach
2022, Applied AcousticsCitation Excerpt :CM uses complex signal processing techniques along with decision making algorithms to identify and quantify the defects. Over the past decade, different CM techniques based on acoustic, vibration, lubrication oil quality, motor current, and temperature were proposed by researchers [6–15]. Elasha et al. [11] compared the bearing fault diagnosing performances of the acoustic and vibration based CM techniques.
Optimal sensor placement for identifying multi-component failures in a wind turbine gearbox using integrated condition monitoring scheme
2022, Applied AcousticsCitation Excerpt :Defect identification and quantification in machinery components are achieved with the help of sophisticated signal processing and decision support algorithms in a CM system. Various authors have implemented different CM techniques based on vibration, quality of lubrication oil, temperature, acoustic signals, and motor current [4–13]. Individual CM techniques showed certain limitations; for example, individual vibration analysis and wear debris based CM analysis can only diagnose about 30–40% of faults [14], shifting focus toward integrated condition monitoring (ICM).