John Alexander Steel

Detection of incipient cavitation in pumps using acoustic emission

Abstract This work concerns the detection of incipient cavitation in pumps using acoustic emission (AE). Three activities have been pursued in this context: (a) the construction of a small-scale rig for the investigation of cavitation detection using AE sensors; (b) the acquisition of data on a 75 kW single-stage centrifugal pump in an industrial test loop under normal running and cavitation conditions; (c) the determination of parameters that could be used for the early diagnosis of cavitation within pumps. In the laboratory-scale apparatus water was pumped around a short loop by a 3 kW centrifugal pump. The flow loop contained a section specifically designed to induce cavitation by means of reducing the pressure level to that of the vapour pressure of the fluid. This apparatus was used to produce a variety of well-controlled cavitation conditions which were useful in determining the suitability of AE for the detection of cavitation. The industrial-scale tests consisted of progressively reducing the net positive suction head in a 75 kW pump while recording the AE signals at various points on the test loop and pump. Results are presented from both laboratory and full-scale tests which demonstrate the feasibility of detecting incipient cavitation using AE in the face of background noise from normal running of the pump. The features of AE which are indicative of cavitation are also seen to change continuously as NPSH is decreased. Thus early detection of cavitation is possible, certainly before any indication is seen on the dynamic head.

STATISTICAL ENERGY ANALYSIS OF STRUCTURE-BORNE SOUND TRANSMISSION AT LOW FREQUENCIES

Abstract If statistical energy analysis is to be used for the study of structure-borne sound transmission then it is necessary for the response of the structure to be controlled by resonant modes. At low frequencies there are few modes and this places a limit on the frequency range over which statistical energy analysis can be used. It is shown that for transmission between plates it is the modes in the receiving subsystem that affect the power flow.

Recent developments in monitoring of engines using acoustic emission

The application of acoustic emission (AE) for monitoring internal combustion (IC) engines is reviewed in this paper. Recent developments in monitoring mechanical events and processes using AE are discussed. The high spatial and temporal fidelity of the AE signals acquired from engines in service make it possible to focus monitoring strategies on individual events and processes. This gives AE the advantages of earlier fault diagnosis and source location whereas other techniques generally monitor symptoms of faults. Monitoring of engine speed, event timing, and reconstitution of processes such as injection and combustion are also possible. These capabilities of AE monitoring are discussed with reference to the development of AE data handling and analysis approaches for engines.

The effect of workmanship on sound transmission through buildings: Part 1—Airborne sound

Abstract A study of airborne sound transmission through a building has shown that parts of the building which appear to be identical do not have the same acoustic performance. This difference cannot be explained by differences in the dimensions or material properties, nor by variations in flanking transmission. It is therefore concluded that the variation, which is approximately 2 dB, is due to workmanship.

Acoustic emission source location for steel pipe and pipeline applications : the role of arrival time estimation

Abstract Acoustic emission (AE) can be generated by a number of different fault conditions in pipes, including localized fluid-mechanical disturbances, such as local impingement or erosion, growing fatigue cracks or crack face rubbing, external impacts and leaks, each of which may have its own temporal and frequency characteristics. In this paper, a linear array of sensors is applied with the ultimate aim of locating and reconstituting the time-domain and frequency-domain signatures of AE sources in pipes. Experiments have been carried out with a simulated source on sections of line pipe and process pipe using a linear array. The acquired signals have two identifiable components and methods are demonstrated for separating these components automatically and determining their group velocities. A range of techniques, including a wavelet transform technique, a cross-correlation technique, and a filtering and thresholding technique are applied to obtain arrival times for various modes of the signal. Finally, methods are proposed for giving automatic source location in large length-diameter ratios with an accuracy of better than 5 per cent.

Sound transmission through a double leaf partition with edge flanking

Lightweight double leaf partitions are widely used and with proper design give good sound isolation. However, when these walls are used as party walls between dwellings, then precautions are necessary to prevent the transmission of fire and smoke. This is usually carried out by placing a firestop in the cavity. This firestop introduces flanking transmission paths reducing the airborne transmission loss of the wall. A simple model is developed which can predict vibration transmission across this type of structural connection. The structural vibration transmission loss can then be used with a more general statistical energy analysis model to give the sound transmission through the entire system. Predicted airborne transmission loss results for a variety of different materials are compared with measured results and good agreement is obtained.

On-line power estimation of large diesel engines using acoustic emission and instantaneous crankshaft angular velocity

Abstract It is of interest to be able to gauge the power output of an engine from sensors that do not interfere with engine performance and are non-invasive. Acoustic emission (AE) analysis is one such technique, whereby information can be obtained from sensors placed on the external surface of the engine, requiring very little preparation. This paper shows that there is a strong relationship between AE energy and the variation of the instantaneous crankshaft angular velocity (ICAV) waveform over the engine cycle for a range of engine loads and that both types of measurement give a good indication of engine power on a ‘per cylinder’ basis. AE and shaft encoder signals were acquired from two large (>7 MW), low-speed, two-stroke, diesel engines over a range of power outputs. The ICAV waveform (inferred from the shaft encoder signals) for one engine cycle was found to give a resolution that showed compression and expansion strokes of each cylinder. The variance of this waveform was found to correlate well with the power output of the engine, a finding that is in accord with those of a number of other authors. Additionally, the AE energy associated with fuel injection/combustion during each cycle was also related to engine power output, a finding that has not explicitly been reported elsewhere. It is shown that, by combining both the AE and the ICAV data, and accounting for differences in engine specification such as number of cylinders, the observations of the two engines can be generalized, in that both the standard deviation in the ICAV waveform and the AE energy generated during the injection/combustion phase increase with increasing engine power output. This relationship can therefore be used to indicate engine power per cylinder and hence indicate whether cylinders are performing poorly. Provided that sufficiently robust sensors are available, such measurements could be made on-line and included in a preventative maintenance plan.

Monitoring of Rail-Wheel Interaction Using Acoustic Emission (AE)

This work presents the results of field measurements and laboratory studies carried out with a view to developing ways to monitor rail-wheel interaction using Acoustic Emission. It is known that impact, wear and cracking generate AE and it is therefore expected that axle loads, wheel out-of-roundness, speed and traction will influence the AE generated by an interaction. It is hoped that the extent of the effect might be sufficient to permit a measure of “interaction intensity” that could be used to quantify cumulative damage by wear and contact fatigue. In the field measurements, AE was acquired as a train with 20 moving sources of AE (20 wheels) passed a single sensor position and a laboratory rig has been devised which uses a single wheel whose condition, speed and loading can conveniently be modified. Simulated source tests have indicated that the AE wave characteristics on real rails are similar to those in the laboratory rig. A simplified analytical model, devised for AE waves propagating from a moving source(s), based on a ‘vehicle’ speed and wave damping coefficients, has been compared to measured results. As a wheel rolls towards a sensor and then away from the sensor the measured AE generally rises and falls in a predictable way. The effects of wheel and rail surface features appear to complicate the results by introducing sharp spikes in the signals. The numerical model for AE wave propagation from the moving sources (wheels) shows good agreement with the more slowly changing envelope of the signals.

A SMART software package for maintenance optimisation of offshore wind turbines.

Offshore Wind Turbine (OWT) maintenance costs in between 20 – 35% of the lifetime power generation cost. Many techniques and tools that are being developed to curtail this cost are challenged by the stochastic climatic conditions of offshore location and the wind energy market. A generic and OWT centric software packages that can smartly adapt to the requirement of any offshore wind farm and optimise its maintenance, logistics and spares-holding while giving due consideration to offshore climate and market conditions will enable OWT operators to centralise their operation and maintenance planning and make significant cost reductions. This work aims to introduce the idea of a comprehensive tool that can meet the above objectives, and give examples of data and functions required. The package uses wind turbine condition monitoring data to anticipate component failure and proposes a time and maintenance implementation strategies that is developed as per the requirements of HSE and government regulations for working in the offshore locations and at heights. The software database contains key failure analysis data that will be an invaluable asset for future researchers, turbine manufacturers and operators, that will optimise OWT power generation cost and better understand OWT working. The work also lists some prevalent tools and techniques developed by industries and researchers for the wind industry.

Rail–wheel contact stress assessment using acoustic emission: a laboratory study of the effects of wheel flats

This article presents part of a wider laboratory study of the acoustic emission (AE) generated during rail/wheel interaction. The particular experiments reported here use simulated wheel defects with a view to developing methods of in situ wheel flat diagnosis using rail-mounted sensors. The analytical approach recognizes the similarity between rolling bearing defects [11] and wheel defects and exploits the principle of demodulated frequency resonance [23] to recognize specific defect frequencies associated with the number, intensity, and disposition of the wheel flats. A set of experiments were carried out on a scaled test rig consisting of a single wheel running on a circular track, during which AE was recorded at a fixed point on the track. The wheel had a set of three asymmetrically distributed flats machined onto it and the rolling speed and axle load were varied. A previously developed time-based analytical model [17] validated against control experiments with an undamaged wheel, was used to reconstruct the normal pattern for a wheel running round the track and to provide a means of removing the effect of attenuation as the defect pattern moves away from and back towards the stationary sensor. Because the wheel flats contact the track at a known frequency, associated with wheel rotational speed, they introduce a pulsatile aspect to the signal, although at a frequency much lower than the frequency of the carrier (AE) wave. Accordingly, once the signal had been corrected for attenuation, it was enveloped to a suitable frequency range, and a frequency-based pulse-train model was used to devise an approach to matching the measured spectra to the expected pulse train spectra in a way similar to the concept of defect frequencies in bearings [24]. For all the enveloped time-series over the full range of conditions, the pattern of flats was clearly visible and reasonably reproducible between consecutive rotations, but with sufficient variation to warrant a spectral approach. The pattern of harmonics was, as expected, dependent on the number, intensity (size), and circumferential disposition of the wheel flats, as well as the axle load, while the rotational speed of the wheel controlled the fundamental frequency. Some frequencies lower than the fundamental were found and these were associated with track rotational frequencies and unwanted frequencies due to normal wheel rolling. These lower frequencies were removed using a high-pass filter [22] to focus on wheel flat diagnosis. It is concluded that, with appropriate calibration and modification, this approach could be used for the diagnosis of wheel defects (as distinct from track defects) in real railway wheels using either track-mounted or wheel-mounted sensors.