INDUSTRIAL GAS TURBINE CONDITION MONITORING SYSTEM: AN OVERVIEW

39th Engine Systems Symposium, Cranfield University ESS39/2013/27 Volume 1 INDUSTRIAL GAS TURBINE CONDITION MONITORING SYSTEM: AN OVERVIEW S. INABUAYE-OMIMI [email protected] Department of Power and Propulsion, Cranfield University, Cranfield Bedfordshire MK43 0AL, UK Abstract Condition monitoring has become a cornerstone of predictive/condition based maintenance. With knowledge of the operating condition of a machinery ( in this case the gas turbine), Operators and maintenance personnel alike can be warned in advance of an impending fault and potential failure Thus enabling maintenance planning in an optimized manner with the ultimate goal of reducing maintenance cost since a significant portion of all operating cost results from maintenance. The purpose of this paper is to examine the basic practices in condition monitoring of gas turbine. It looks at the various condition monitoring techniques with more details on vibration analysis since it is the most useful and arguable the best condition monitoring technique. It also examined the two major condition monitoring systems (i.e. online and offline systems), the merit and challenges of this maintenance management tool. The importance and benefits of condition monitoring was established. However major challenge now inherent in modern condition monitoring practices is the extent or scope of condition monitoring to embark on. This can only be ascertained only when a proper detailed evaluation of the profitability of the proposed system is carried out. Keywords: Condition Monitoring; Vibration; Defects; Trending Apparently, condition monitoring is arguably the most improperly used and misunderstood plant optimization programs [1]. Many industrialists consider it as just a maintenance scheduling tool. Nevertheless, to get the full benefits of condition monitoring, this scope must be expanded to that where it is capable of effectively managing plant machinery, logistics and labor requirements. In truth, condition monitoring is a management technique which observes and analyses machine condition, identify changes in its operating condition and give warning of potential failure for the purpose of optimizing the all round operation of machinery. It is therefore a major component of predictive maintenance. Since gas turbines are always critical to plant operations, condition monitoring becomes a necessity in ensuring that its operation and the entire operation of the plant is optimized. There exist different methods and techniques of carrying out condition monitoring of gas turbines. The proper selection and combination of these techniques determines the productivity of the condition based maintenance employed. 1. Introduction Maintenance ideologies of gas turbine have evolved over the past decades. From run-to-breakdown to periodic scheduling (preventive) and presently, condition based maintenance (predictive). During the era of run-to-breakdown, maintenance actions on gas turbine engines were only carried out when they fail. Output level was high since there were no regular interruptions in service. However since a stitch in time would always save nine, this was not desirable as maintenance costs (which would have been lower if the problem was attended to early) were very high. The ideology therefore changed to one in which inspections and maintenance activities were carried out at planned intervals (preventive) as recommended by OEMs. This meant that the engine was stopped when maintenance was to be carried out and some parts replaced even if they were still in good operating condition. Clearly, this ideology was not the solution if engine operation and maintenance was to be optimized. The best solution was to move from preventive to condition based maintenance since the Health of the engine is ascertained by monitoring its operating condition and therefore appropriate maintenance plan and decision can be taken. 1 ESS39/2013/27 39th Engine Systems Symposium, Cranfield University Volume 1 2. Condition monitoring techniques 2.1. Vibration analysis 2.1.1 Principles Vibration analysis is the most used and versatile c ondition monitoring technique. Since the gas turbin e is rotating machinery, vibrations occurrence durin g its operation becomes inevitable. More so, these vibration contains crucial information on the cond ition and health of the gas turbine at a given time interval. Different processes such as bearing fault s, imbalance, looseness, etc produces energy at dif ferent frequencies. If these frequencies are distingu ished from one another through spectrum analysis, then faults and potential failure can be identified [2]. The vibration signatures produced by the gas turbi ne are complex containing a variety of sinusoidal waveforms and frequencies which are all connecte d fundamentally to the RPM of the engine. Analy zing this becomes cumbersome hence the necessity arises to transform these signatures from time do main to frequency domain. This is accomplished with the aid of the Fast Fourier transforms (FFT). Some basic fundamentals of vibration include the following: Period: this is the time taken for a signal to comp lete one cycle and it is best shown with a single frequency [3]. Figure 1: phase difference between displacement (x), velocity (v) and acceleration (a) [4]. Time domain: when vibration data is displayed and analyzed as a function of time it is said to be in time domain [5].There is minimal loss of data before inspection thus allowing for more detailed analysis to be carried out. Conversely, fault diagnostics is less efficient as a result of too much data available. Frequency domain: in the frequency domain, vibration data is displayed and analyzed as a function of frequency. Typically, data acquisition is done in the time domain and converted to the frequency domain through the FFT method. It is also called spectrum. Faults are easily detected here and trended over time as the condition of the gas turbine deteriorates. A major disadvantage is that a large portion of the transient, non repetitive components may be lost during the transformation (from time domain to frequency domain) process. Amplitude: it is the peak displacement of a vibrati ng body about its equilibrium position. Displacement: simply distance from a reference po sition in a specified direction. Velocity: is the rate at which displacement changes with time Acceleration: is the rate at which velocity changes with time Harmonics: basically harmonics refer to vibration signals which have frequencies that are precise multiple of a reference fundamental frequency. Figure 2.2: A Typical spectrum [7] 2.1.2 Vibration instrumentation Transducers Phase relation: vibration can either be viewed as displacement, velocity or acceleration and these are sinusoidal in nature. They will complete one cycle in the same time. Conversely the do not move at the same time hence a phase difference exist between them. In principle there is a phase difference of 180⁰ between acceleration and velocity and velocity lags acceleration by 90⁰ and leads displacement by 90⁰. This is shown in the diagram below. Vibration transducers are sensors capable of measuring vibration and converting to an electrical signal. There are different types of transducers used in vibration monitoring of industrial gas turbines. Proximity probes: proximity probes are transducers used to detect and measure the displacement of devices (i.e. gas turbine shaft relative to its bearing). In industrial gas turbine (and gas turbine at large), 2 ESS39/2013/27 39th Engine Systems Symposium, Cranfield University high temperature proximity probes are used. Typically, two proximity probes are mounted in the radial and axial (X-Y) positions to measure the relative motion of bearing and shaft since the fluid-film (journal) bearings are used extensively in industrial gas turbines. Volume 1 are option- al. They are used in conjunction with the proximity probes for checking malfunctions and also as a shut- down protection when vibration exceeds the set limits. Transducer cables: provides connection between the transducers mounted on the gas turbine and the monitoring system. Routing these cables through areas with high temperature and direct contact with oil is usually challenging when dealing with gas turbines. However this can be overcome by routing cables through oil return lines thus ensuring the cable remains well below their upper temperature rating [6]. Conversely this can also present oil sealing problems. A solution provided by Bentley Nevada is using High pressure feed through cable assembly [6]. Figure 2.3: Bentley Nevada proximity probes Accelerometers: most gas turbine use high temperature piezoelectric accelerometers for sensing vibration with acceleration units as output. These transducers are often used by aero-derivative engines because of the rolling element bearing mostly employed. Since this type of bearing couples shaft and bearing vibrations to the bearing support, accelerometers are mounted on the bearing casing. The specific mounting positions are carefully chosen to coincide with the point of maximum bearing-to- structure load path [6]. In industrial gas turbine applications, they are used lube oil and seal oil pumps and motors. 2.1.3 Vibration Measurement application Broad band pass (overall vibration): the broad band measurement also known as Root Mean Square (RMS) gives a total value of vibration measured at clearly defined points of measurement. Usually broad band frequencies are unfiltered. Broad band trending basically serves as a notification to show that there is a change in the condition of the gas turbine. Narrow Band trending: Here overall vibration frequencies are filtered or narrowed through a narrow band pass filter to contain only the basic frequencies that are needed to identify a particular defect of fault hence it provides a medium to quickly monitor conditions of specific components (i.e. like bearings and gears) without manually analyzing vibration signatures. Key phasor probe: the key phasor probe is basically used as a once-per-turn reference signal [6]. It produces a voltage pulse each time a notch (created in the shaft) goes past the probe. It is a very useful diagnostics tool. By integrating the signal from the two proximity probes and the key phasor probe, a wave shape elliptical in nature is created representing the shaft orbit [7]. Thus it is possible to identify abnormalities and defects by simple observation of changes in ideal shaft orbit pattern. 2.1.4 Vibration limits Vibration limits are very useful in vibration analysis. Although it often takes a lot more to ascertain the overall condition of the gas turbine, however, these limits give an overall initial indication of the condition of the gas turbine and prompt the gas turbine user on the next line of action to take depending on the vibration limit zone at that particular time. The following tables show the different vibration limits zones as given in ISO 10816 (part 4) for industrial gas turbines. Figure 2.1: vibration limits zones Figure 2.4: key phasor diagram showing shaft orbit [10] Seismic vibration transducers: typically seismic transducers are mounted on the gas turbine casing or specifically on the bearing housing in the radial direction. They are mostly used in large industrial gas turbines where there is notable motion between bearing housing and free space, otherwise these transducers ZONE ZONE DESCRIPTION VALUE (mm/s)(r.m.s) A B C Vibration of new machines Considered acceptable Unsatisfactory and closely monitored Severe enough to cause damage < 4.5 4.5 - 9.3 9.3 - 14.7 D 3 > 14.7 ESS39/2013/27 39th Engine Systems Symposium, Cranfield University Volume 1 2.1.5 Common faults 2.2. Tribology Monitoring The following are some of the faults detected by vibration analysis techniques which results in very high vibrations. Imbalance: this occurs when a part or component is out of balance (i.e. excess mass on one side of a rotor). Typical abnormal vibration caused by unbalance will at 1 � RPM Misalignment: misalignment can be angular, offset or a combination of both. Typically in industrial gas turbines, it will occur between stub shafts connected by a coupling, bearings and between the gears of the auxiliary drive system. Typical vibration is produced at 1 RPM (since it causes unbalance) and 2 RPM harmonics [8]. [9] describes tribology as “the science and engineering of surfaces in relative motion which includes the study and application of the principles of friction, lubrication and wear”. Tribology analysis simply entails the monitoring of the overall condition of the gas turbine lubrication and seal oil. The two major Tribology monitoring techniques are Oil analysis and wear debris analysis. There industrial gas turbine there is need to also monitor the operation of the entire oil systems. The two major reasons for conducting tribology monitoring are as follows;  to ascertain the condition of the oil so that scheduled oil change can be planned in an optimized manner.  to analyze the content of debris in the oil so as to identify component wear rate and there- fore plan maintenance action optimally. Mechanical looseness: looseness occurs when there is a slack in components fitted together. Looseness faults may be caused by loose foundation support, loose bearing in its housing, loose bearing on its shaft, among others. Vibration is typically produced at a frequency of 2 RPM. Other common faults detected by vibration include the following: Bearing defects Resonance Gear problems Hydraulic and Aerodynamic problems Oil whirl 2.2.1 Oil analysis Oil monitoring is a technique used for monitoring and evaluating the physical condition of the lubrication oil. The following are some of the test carried out in this analysis. Viscosity test: the important of viscosity cannot be overemphasized. Here the actual viscosity of the test sample is compared to a reference unused oil. Too low a viscosity reduces the oil wedge thickness of the journal bearing. Conversely, very high viscosity may resist the flow of oil in the lubrication system and there reduce its lubricity. Table 2.2: common faults detected from Known Fault Main frequency imbalance 1 � RPM misalignment Looseness Bearing defect Gear mesh defect Oil whirl Resonance Comment Leads to severe vibration 2nd harmonics and severe vibration. Problem can be 2 � RPM solved using laser alignment. Typically high harmonic content and random 1 � RPM phase High frequency Shock pulse technique Vibration can be used to deter(N � RPM) mine the intensity of damage. Gear freq.=no Customarily regulated by the running speed of teeth (i.e. gearbox input & rotational output speed) freq. of the defect gear Characteristically Sub-RPM 0.43-0.48 of RPM with phase usually unstable Natural freOccurs when there is a quency of coincidence between component component natural frequency and excitable frequency Figure 2.5: Adhesion meter- viscosity test instrument [10] Contamination test: the most common contamination of lube oil is water. Other contaminants include oxidants, microbes, etc. these contaminants reduce the overall quality of the lubrication oil. Capacitance analysis can be used to ascertain the amount of contaminants in the oil [11]. Solid content test: traces of solids are often contained in lubricating oils. Wear rate of sleeves and shaft can be significantly increased by the presence of these sol-id traces. Techniques like particle count, blotter spot analysis, among others can be used to determine the magnitude of these traces. 4 ESS39/2013/27 39th Engine Systems Symposium, Cranfield University Volume 1 radation is as a result of normal wear and aging of the gas turbine. Recoverable degradation are artificial and are caused by external factors like compressor fouling, FOD, etc. compressor fouling is the most common source of gas turbine compressor degradation. After proper performance monitoring and trending, the optimum intervals for compressor washing can be planned. The figure below shows the effect of offline compressor washing on flow capacity at different time intervals. Degradation test: As lubrication oils age in use, so does their overall properties degrade. Properties such as dispersancy, emulsibility, additive content can be tested and monitored. If the degradation increases above a particular set standard then an oil change is carried out. Therefore oil change is done only when it is necessary. Thus preventing premature oil change out and helps provide optimum stock of inventory. 2.2.1 Wear debris analysis Wear debris analysis gives detailed and specific information about the wearing condition of the gas turbine components involved (i.e. shaft, bearings). The quantity, size and composition of the particles present in the lube oil give an insight into the condition of the lubricated components. All components are bound to wear over time. Therefore it is impossible to get a wear-free scenario during the analysis. Nevertheless it is the particle size and the rate of wear that is important and trended over a period of time. Typical wear type encountered in gas turbine bearing is the rolling fatigue wear. There are two basic methods of wear debris analysis, they are; Spectrograpy: in this technique, oil samples collected from drain ports on the gas turbine lube system is burnt leaving only the traces of debris. These particles are then analyzed and trended. One major disadvantage in this technique is the limitation in the size of the particles detectable. Particles below 10 μm are difficult to detect [12] Figure 2.6: effect of compressor washing on flow deviation [14] By careful study and trending of the component characteristics, fuel flow, component efficiencies, power output, EGT, pressure ratio and thermal efficiency, information of the performance condition can be obtained. Also by monitoring the pressure drop across the inlet air inlet filters, useful information about the condition of the filters can be obtained which prompts the operators to plan for proper and optimal filter change out. Performance monitoring like efficiency map, wash advisor, online optimizers, etc can be used to carry out online or offline performance monitoring. Ferrography: ferography is similar to spectrography; the difference being that magnetic field is used to separate larger sizes of particles up to 100μm. Nevertheless only ferrous particles can be separated. [12] 2.4. Acoustic emission monitoring Materials emit elastic waves when there is an abrupt change in the stress of the material. This phenomenon is known as acoustic emission (AE). This sudden redistribution of stress is brought about by external stimulating factors like pressure and variation of load. The varying nature of performance parameters coupled with the fact that the gas turbine operates under severe conditions (high stress levels), make the occurrence of acoustic emissions inevitable. Figure 2.6: Ferrography wear particle analysis [13]. 2.3. Performance Monitoring Performance monitoring involves the measurement and analysis of thermodynamic parameters such as temperature and pressures and then comparing them with a set of expected value. Like every other turbo machine, the gas turbine is bound to degrade with time. Degradation can either be recoverable or non recoverable. Non recoverable deg- Typically AE transducers (of about 6-10) are place on the surface of the casing to monitor the acoustic emissions dissipated by the stator vanes which are always in direct contact with the turbine casing [15]. Early detection of these cracks are important so that the components affected is replaced before it lead to a significant damage of the turbine. 5 ESS39/2013/27 39th Engine Systems Symposium, Cranfield University Volume 1 none of them can absolutely guarantee an all round condition monitoring of the gas turbine. In order to achieve the desired goal of optimizing the gas turbine maintenance strategy, a combination of these all or some of these techniques becomes necessary. This combination makes up a condition monitoring system. Defects such as fracture in vanes, cracking of discs and casing, etc can be detected from information gotten from the AE monitoring record. The acoustic emissions are then trended with trend lines established. Deviations from these bench marks would therefore suggest that there is degradation in the component measured. There are two systems of condition monitoring. They include online condition monitoring and offline condition monitoring systems. 2.1. Online (continuous) Monitoring system Here the health and operating condition of the eng ine is monitored continuously and permanently. Mo nitoring devices like transducers and sensors are in stalled permanently on the engine with data collect ion and analysis done automatically i.e. no physica l presence needed on site. Most gas turbines use online condition monitoring systems since their roles are always critical to the operation of the plant. Abrupt changes in the ope rating condition of the gas turbine prompts meticul ous investigation or a possible automatic shut dow n of the gas turbine. Basically and online condition monitoring system will consist of the following:  Permanently mounted vibration sensors  Permanently mounted speed sensors  Permanently installed dynamic pressure sen sors  High temperature sensors  Field connection cables  Junction boxes  Diagnostic module for data collection  Network cabling  Online management host computer  Client computers  Online vibration software.  Interface to the GT plant DCS (data contr ol system [3] Figure 2.7: Acoustic emission monitoring [16] 2.5. Thermography Thermography is a condition monitoring technique that uses temperature i.e. infrared energy to determ ine the condition of the engine. The gas turbine operates under very high temperat ure conditions. Consequently this condition is subj ect to the material integrity of the blade. Therefor e it becomes very important to monitor these temp eratures. By monitoring and trending the temperature profile of components like combustion chamber, exhaust ducts and NGVs, local hot spots and unusually hi gh temperatures can be detected. Because of the e xcessively high temperatures involved in the gas t urbine, Non contact infrared sensors are used for monitoring. The data collection diagnostic module typically use s 32 channels for each with the choice of collecti ng a wide range of static and dynamic inputs fro m permanently installed sensors. The online vibrati on software is used for monitoring, carrying out a nalysis and reporting. The dynamic sensors monitor pulsations in the co mbustor and consequently combustion instability (h umming) can be detected. Figure 2.7: Non- contact infrared temperature sensor [17] 3. Monitoring system Different types of online condition monitoring syst ems are commercially available. The basic gas turbine condition monitoring techniques have been discussed in the previous section. However it should be noted that while some of these techniques are considered fundamental, 6 ESS39/2013/27 39th Engine Systems Symposium, Cranfield University Volume 1 plant operations. Since maintenance cost contributes to a significant amount of operating cost, it then becomes necessary to ensure that this cost is reduced by carrying out maintenance optimally thereby maximizing gas turbine utilization and preventing catastrophic failures. Condition monitoring has played a successful role in realizing this objective. An effective condition monitoring system should be able to derive the following benefits:  Provide early warning signs of potential failure and hence improve maintenance planning strategy.  Increased gas turbine availability and reliability.  Reduced inventory of spare parts.  Reduce gas turbine downtime.  Reduced overall maintenance cost. 2.2. Offline (periodic) Monitoring system In the offline monitoring strategy, Data acquisition and analysis is done intermittently with the help of hand held devices like data collector. Here a trained personnel is required to carry out this operation. It is mostly applied to uncritical equipments. In some cases, permanent monitoring hardware are used for this type of monitoring but data is only collected at specific times (Mechefske, C., K. 2005). This monitoring strategy is applied to the gas turbine auxiliary system i.e. the lubrication and seal oil pumps, motors and bearings especially in plants where the gas turbine have a spare auxiliary system hence making them not so critical as compared to the gas turbine. Data acquisition here can be done using portable hand-held vibration meter or data collectors. Data collectors are mostly used because of the capabilities of performing on-the-spot diagnostics and or balancing. A portable vibration data collection system typically consists of the following;  Data collector  Portable vibration sensor with probe or magnet  Switching boxes  Host computer  Monitoring and management software Some commercially available data collection instruments are shown below. On the other hand, setting up an effective condition monitoring system does not come free of charge. A lot has to be put in place. Sometimes, these CM systems can be very complex depending on the number of techniques integrated in the system. It is a well known fact that as complexity of a system increases, its reliability decreases. The following are some demerits of condition monitoring.  Cost of acquiring CM equipments and setting up the CM system are usually significant.  Very skilled personnel are needed thereby incurring extra cost. (i.e. cost of training personnel  Reliability of the system can be an issue if it is too complex. Figure 3.1: Portable data collectors/analyzers [3]. Figure 3.2: Return on investment (ROI) curve [11]. 4. Pros and Cons of condition monitoring Considering the curve above, it can be seen that as the quality (i.e. complexity) of the condition monitoring system increases, the net savings increases. Nevertheless at some point, the procurement and maintenance costs of the CM system exceeds the net savings realized by using the Condition monitoring system. The importance and significance of condition monitoring in today’s maintenance strategy (i.e. predictive/condition based maintenance) cannot be overemphasized, most especially for very expensive equipments like gas turbine which are always critical to An important question then arises, “How detailed and thorough should an engine CM system be?” 7 ESS39/2013/27 39th Engine Systems Symposium, Cranfield University Volume 1 References 4. Conclusion The significance and relevance of condition monitoring in gas turbine application have been established. The success of condition monitoring as an optimal and integral maintenance management tool have prompted wide acceptance among gas turbine operators and the industry at large. Consequently improving the quality of these CM systems have been a major interest to operators and OEMs alike. [1] [2] However, it has also been established that these improvements does not come without a cost. Some setbacks have also been identified. In conclusion, there have to be a proper evaluation and scaling of the merits and demerits involved in setting up a particular CM system. The quality, extent and comprehensiveness of the CM system should be driven by the profitability and conformance to the ultimate goal of condition based maintenance. Also the reliability of the system should be considered to a great extent. [3] [4] [5] [6] [7] Recommendations [8] From indications and trends in condition based maintenance, it is evident that there will be a continual increase in the growth, development, quality and hence complexity of these condition monitoring systems as a result of the huge losses associated with machine downtime. It is recommended that manufacturers of these systems should ensure that the CM systems are very reliable since failure of the system affects the operation of the engine adversely. It is also recommended that operators should ensure that condition monitoring personnel are trained adequately in order to maximize the effectiveness since these CM systems are becoming more complex. [9] [10] [11] [12] Acknowledgements [13] Special thanks to the Almighty God who has given me the grace and strength in the duration of this work. I am also grateful to my Parents for their all round support and encouragements throughout the duration of the preparation of this paper. [14] [15] [16] 8 M. Liebler, "Rockwell Automation," September 2005. [Online]. Available: www.ferret.com/au/c/Rockwell-Automation/conditionmonitoring-misused-and-misunderstood-n678999. [Accessed August 2013]. B. Rao, Handbook of Condition Monitoring, Oxford: Elsevier Advanced Technology, 1996. S. R. W. Mills, Vibration Monitoring & Analysis Handbook, Northampton: British Institute of Non Destructive Testing, 2010. CBMApps, "CBMApps," 2010. [Online]. Available: http://www.cbmapps.com/docs/28. [Accessed August 2013]. C. K. Mechefske, "Machine Condition Monitoring and Fault Diagnostics," in Vibration and Shock Handbook, Florida, CRC press, Taylor and Francis Group, 2005, pp. 25-35. M. Maalouf, "Gas Turbine Vibration Monitoring: An Overview," Orbit, vol. I, pp. 48-62, 2005. Novo Pignone S.p.A, "Vibration and Axial Displacement Monitoring," Florence, 2002. PRUFTECHNIK LTD., An Engineers Guide to Shaft Alignment,Vibration Analysis, Dynamic Balancing and Wear Debris Analysis, 8th ed., Straffordshire: PRUFTECHNIK LTD, 2002. IET , "Tribology," 2013. [Online]. Available: http://www.theiet.org/communities/tribology/overview/ what-is.cfm. [Accessed August 2013]. DHgate.com, 2004. [Online]. Available: http://www.dhgate.com/store/product/viscosity-test-inst rument-digital-adhesion/13999861-142297266.html. [Accessed August 2013]. L. Yi-Guang, Gas Turbine Diagnostics, 1.5 ed., Cranfield: Cranfield University, 2013. B. K. N. Rao, "Condition monitoring and the integrity of industrial systems," in Handbook of Condition Monitoring: Techniques and Methodology, 1st ed., A. Davies, Ed., London, Chapman and Hall, 1998, pp. 1-50. B. Jawa, "ferrography wear particle analysis," June 2008. [Online]. Available: http://budak-jawa.blogspot.co.uk/2008/06/ferrographywear-particle-analysis.html. [Accessed August 2013]. Power Engineering International, "PEI," December 2012. [Online]. Available: http://www.powerengineeringint.com/articles/print/volu me-20/issue-11/features/safeguarding-performance-gasturbine.html. [Accessed August 2013]. M. G. B. Z. V. G. Jaya P. Koduru, "Online acoustic emission monitoring of combustion turbines for compressor vane crack detection," in Industrial and Commercial Applications of Smart Structures Technologies , San diego, 2013. NDT , "NDT resource center," 2002. [Online]. Available: http://www.ndt-ed.org/EducationResources/Community College/Other%20Methods/AE/AE_Intro.htm. ESS39/2013/27 [17] [18] [19] [20] 39th Engine Systems Symposium, Cranfield University [Accessed August 2013]. Omega engineering inc., 2013. [Online]. Available: http://www.omega.com/ppt/pptsc_lg.asp?ref=os137&N av=. [Accessed August 2013]. G. Hands, "NDT.net," Resonant Inspection A 'New' NDT technique, vol. II, p. 7, 1997. Fajar, 2011. [Online]. Available: http://freevibrationanalysis.blogspot.co.uk/. [Accessed August 2013]. General Electric Company, "Bently Nevada Sensors and Transducers," 2013. [Online]. Available: http://www.ge-mcs.com/en/bently-nevada-sensors-and-t ransducers/proximity-probes.html. [Accessed August 2013]. 9 Volume 1