.:Maintenance Online:.

	The Role of Vibration Monitoring In Predictive Maintenance	Issue : Jan-10
Company : Schaeffler (UK) Ltd

The role of vibration monitoring in predictive maintenance Part 1: Principles and Practice r. S. J. Lacey, Dr S J Lacey, Engineering Manager Schaeffler UK (INA FAG) ABSTRACT Unexpected equipment failures can be expensive and potentially catastrophic, resulting in unplanned production downtime, costly replacement of parts and safety and environmental concerns. Predictive Maintenance (PdM) is a process of monitoring equipment as it operates in order to identify any deterioration, enabling maintenance to be planned and operational costs reduced. Rolling bearings are critical too and used extensively in rottating equipment, and when they fail unexpectedly can result in a catastrophic failure with associated high repair and replacement costs. Vibration based condition monitoring (CM) can be used to detect and diagnose machine faults and form the basis of a PdM strategy. In this, the first part of a comprehensive two-part review which builds on an earlier contribution on this topic, the author discusses the role of CM in maintenance, the importance of identifying asset criticality and justifying the economics when implementing a CM programme, and the basic principles and techniques of monitoring and analysing the vibration of bearings in rotating machinery. In the second part, to be published in the next issue, he will present examples of the successful application of these techniques to a wide range of industrial plant. INTRODUCTION As greater demands are being placed on existing assets, either for higher output or increased efficiency, the need to understand when things are starting to go wrong is becoming more important. Also, as plant and equipment becomes more complex and automated the need to have a properly structured and funded maintenance strategy is becoming more important. There is also a need to understand the operation of your own equipment so that improvements in plant output and efficiency can be realised. In today's increasingly competitive world all of these issues are of key importance and can only be achieved through a properly structured and financed maintenance strategy which meets the business needs. Maintenance can often be a casualty as businesses downsize to save costs. How often have we heard the words 'we have had no problems since the equipment was installed so we don't need condition monitoring' This is often born out of ignorance and the failure to undertake a proper risk assessment to identify the criticality of existing assets so that the potential return on investment (ROI) of a properly funded maintenance strategy can be determined. The need to run a plant at a higher efficiency, yet often with fewer people, puts increasing pressure on all concerned when equipment fails prematurely. When equipment does fail it is often at the most inconvenient time: either in the middle of a key process, at a weekend or in the middle of the night, when obtaining replacement parts may be difficult and labour costs high due to overtime. While there is never a good time for equipment to fail, with the technology available today there is simply no excuse for not taking the necessary steps to protect key assets. This can be achieved by minimising the risk of early and unexpected failures through a properly structured and funded maintenance strategy which will ultimately reduce overall operational costs. The cost of not having a robust maintenance strategy should not be underestimated. It should not be looked at simply as an up front cost, but viewed as an investment to safeguard and protect key assets, reducing the need for costly repairs and protecting the output from key processes. In some industries maintenance is now the second highest or even the highest element of operating costs. As a result, in the last two or three decades it has moved from almost nowhere to the top of the league as a cost control priority. The need to contain costs and run plant for longer and more reliably means that there is a growing awareness of the need to prevent unnecessary equipment failures. Central to this is having a maintenance strategy which is based on monitoring key assets to detect when things are starting to go wrong, enabling plant outage to be better planned, both in terms of resource availability, spare components, repairs etc. As a result the risk of missing important contract deadlines is reduced and customer confidence is improved. Until recently many industries have taken, and still take, the reactive approach to maintenance, because this has no up front costs, but it can result in many hours or days of plant downtime or lost production. While this may have been acceptable in the past, the increasing complexity and automation of equipment has meant this is now not a cost effective option. Having a clear and robust maintenance strategy that is fully supported by senior management is becoming more important, particularly in industries where it not only has a major impact on costs but also on the health and safety of employees, particularly in situations where secondary damage and a catastrophic failure may result. MAINTENANCE APPROACH Maintenance is traditionally performed either at fixed time intervals – so called preventive maintenance, or by corrective maintenance when a breakdown or fault actually occurs. In the latter it is often necessary to perform the maintenance actions immediately, but in some cases it may be deferred, depending on the criticality of the equipment. With predictive maintenance an advanced warning is given of an impending problem and repairs are only carried out when necessary and can be planned to avoid major disruption. These three approaches are summarised in Figure 1 and briefly discussed in the next section. Reactive maintenance Often referred to as the 'run till failure' approach, this involves fixing problems only after they occur. Of course, this is the simplest and cheapest approach in terms of up front maintenance costs, but often results in costly secondary damage along with high costs as a result of unplanned downtime and increased labour and parts costs. Because there are no up front costs it is often seen as an easy solution to many maintenance problems– or as having no strategy at all. In rotating equipment, rolling element bearings are one of the most critical components, both as regards their initial selection and, just as importantly, in terms of how they are maintained. Bearing manufacturers give detailed guidelines as to what maintenance is required and when – which is often overlooked (and which can have disastrous consequences in terms of poor quality output, reduced plant efficiency or equipment failure). Monitoring the condition of rolling bearings is therefore essential, and vibration based monitoring has great potential for detecting the early onset of a fault. Preventive maintenance With Preventive Maintenance (PM), machinery is overhauled on a regular basis regardless of the condition of the parts. This normally involves the scheduling of regular machine/plant shutdowns, whether or not they are required. The process may reduce the incidence of unwanted failures but it also leads to increased maintenance costs because parts are replaced when this is not necessarily required. There is also the risk of infant mortality due to human error during the time the asset is taken out of service for repair, adjustment, or installation of replacement parts. Other risks include installing a defective part, incorrectly installing or damaging a replacement part, or incorrectly reassembling parts. With some predictive maintenance programmes a large proportion of the plant will not experience failure, which can lead to a significant cost savings from unnecessary repair work. Often a direct result of preventive maintenance is that much of the maintenance is carried out when there is nothing wrong in the first place. With predictive maintenance the only costs incurred are those that become necessary as a result of an advanced warning of a problem. Predictive maintenance Predictive maintenance (PdM) is the process of monitoring the condition of machinery as it operates in order to predict what might fail. In this way maintenance can be planned and it gives an opportunity to change only those parts that are showing signs of deterioration or damage. The basic principle of predictive maintenance is to take measurements that expedite the prediction of when parts will breakdown and what will go wrong. These can be measurements of such things as machine vibration or of plant operating data such as flow, temperature, or pressure. Continuous monitoring detects the onset of component problems in advance so that maintenance is performed only when needed. With this type of approach unplanned downtime is reduced or eliminated and the possibility of catastrophic failure is minimised. It allows parts to be ordered more effectively (reducing parts inventory) and manpower can be scheduled (increasing efficiency and reducing the costs of overtime). The main benefits of PdM are – * Improvement of machine reliability through the effective prediction of equipment failures. * Reduction of maintenance costs by minimising downtime through the scheduling of repairs. * Increase of production through greater machine availability. * Reduction of energy consumption. * Extension of bearing service life. * Improvement of product quality Rolling bearings are often a key element in many different types of plant and equipment, spanning all market sectors. On one hand they can be of a standard design, readily available and low cost commodity items costing only a few pounds (e.g. those on electric motors, fans, gearboxes) while on the other hand they can be of bespoke design with long lead times and costing hundreds of thousands of pounds (e.g. those found on Wind Turbines, Steel Plants etc. ) However, they have one thing in common: if they fail unexpectedly they can result in plant and equipment outage resulting in lost production costing from a few thousands to many millions of pounds. With a predictive maintenance strategy such large costs can be avoided by giving advanced warning of a potential problem enabling remedial action to be planned and taken at a convenient time. Replacing a bearing in a gearbox is preferable to replacing the whole gearbox, or replacing a motor bearing better than having to send the motor to a rewinder to make expensive repairs and replace parts. At the heart of many predictive maintenance strategies is condition monitoring, which detects potential defects in critical components (e.g. bearings, gears etc.) at an early stage, thereby enabling the maintenance activity to be planned, saving both time and money and preventing secondary damage to equipment, which can often be catastrophic. IDENTIFYING ASSET CRITICALITY Rolling bearings are used extensively in almost every type of rotating equipment whose successful and reliable operation is very dependent on the type of bearing selected, the bearing fits, the installation and the maintenance requirements (e.g. re-lubrication). The deterioration of rolling bearings can result in expensive equipment failures with high associated costs. Unplanned downtime, the costly replacement of equipment, health and safety issues and environmental concerns are all potential consequences of a maintenance strategy that fails to monitor and predict equipment problems before they escalate into something bigger. Assessing the criticality of an asset to the overall operation of the plant is therefore essential in terms of determining the type of condition monitoring required and whether it is necessary at all. In some cases where a plant has a large number of low cost assets where replacements are readily available and/or are not deemed critical a reactive or preventive approach may well be appropriate. Even if an asset does warrant condition monitoring then a decision must be made, not only on the technology but also whether the asset warrants continuous (on-line system) or non-continuous (patrol) monitoring. To help with this decision, assets are often categorised as falling into one of three categories depending on their criticality (see Figure 2)1. Assets falling in Category A are deemed to be critical and their failure can result in one or more of the following – * Total or major interruption of the process. * A significant safety risk, such as a fire, toxic leak, or explosion. * Long lead times and/or significant repair costs. A good example of this would be the main turbine-generator trains in a large power plant. For such assets, it is the cost of failure that is of primary concern. Other examples would be the main rotor bearing or gearbox bearings in a wind turbine. Because of the generally remote location, failure of these (which may lead to secondary damage) makes costly the replacement of parts and the hire of equipment and labour. Category C, or non essential assets, are at the opposite end of the spectrum and the reasons for monitoring these, if indeed they are monitored at all, might be to prevent failure by eliminating root cause and having more effective maintenance planning. Category B on the other hand covers essential assets and might, for instance, be pumps or compressors where a standby unit is available in the event of a failure and which then may become Category A. RETURN ON INVESTMENT (ROI) As already discussed, equipment failure can be expensive and potentially catastrophic, resulting in unplanned downtime, missed customer schedules, costly machine replacements or repairs as well as safety and environmental concerns. By initiating a predictive maintenance strategy unforeseen failures are minimised, which can yield an impressive ROI. Another major benefit of introducing CM as part of a predictive maintenance programme is that it generates a greater understanding of the equipment critical to the process and also allows more time to be spent on improving the overall condition of the assets and improving the efficiency of key processes. When justifying PdM the following should be taken into consideration – (1) Direct costs Labour * Normal and overtime labour for – – planned repair activities – unplanned repairs Materials * Parts replaced * Machinery replaced (2) Indirect costs * Lost production (£/hr) * Outside services * I nsurance * Parts inventory Total annual potential cost reduction (1 + 2) (3) PdM programme costs * Site survey * Capital equipment * Any additional labour * Training * Initial set-up and baseline Total annual PdM costs (3) By contracting out your PdM there will be no capital equipment and training costs and the benefits tend to be more immediate because of the use of highly trained staff. However, it is often more beneficial to keep the activities in house so that you become more familiar with your own plant, equipment and processes, which gives benefits of not only preventing unplanned downtime but enables more time to be spent on mitigating potential failure modes and improving process efficiency. The cement industry provides a good example of the implementation of CM reducing repair costs and lost production costs. The failure of a large gearbox could cause a three-week shutdown and extensive repair costs may be typically €50,000 to €100,000. To prevent such damages F'IS (FAG Industrial Services, Schaeffler Group) installed an eight-channel FAG DTECT X1 system and trained the customer's staff, who received three months support – total cost €18,000. Detecting deterioration of the gearbox early resulted in a repair cost of €5000, saving the customer at least €27,000. More importantly, the company avoided lost production amounting to around €6000/hour. CONDITION MONITORING Condition monitoring is a process where the condition of equipment is monitored for early signs of deterioration so that the maintenance activity can be better planned, reducing down time and costs. This is particularly important for continuous process plants where failure and downtime can be extremely costly. The monitoring of vibration, temperature, voltage or current, and oil analysis are probably the most commonly used techniques in this area. Vibration is the most widely used and not only has the ability to detect and diagnose problems but has the potential to provide a prognosis, i.e. an indication of the remaining useful life and possible failure mode of the machine. However, prognosis is much more difficult and often relies on the continued monitoring of the fault to determine a suitable time when the equipment can be taken out of service (or relies on known experience with similar problems). Vibration monitoring Probably the most widely used PdM technique, and with few exceptions can be applied to a wide variety of rotating equipment. Because the mass of the rolling elements is generally small compared to the machine, the velocities generated are usually small and result in even smaller movements of the bearing housing, difficult for the vibration sensor to detect. Machine vibration comes from many sources, e.g. bearings, gears, imbalance etc., and even small amplitudes can have a severe effect on the overall machine vibration, depending on the transfer function, damping and resonances (see Figure 3). Each source of vibration will have its own characteristic frequencies which can be discrete frequencies or sum and/or difference frequencies. At low speeds it is still possible to use vibration, but a greater degree of care and experience is required and other techniques, such as measuring shaft displacement or Acoustic Emission (AE), may yield more meaningful results – although the former is not always easy to apply. Also, while AE may detect a change in condition it has limited diagnostic capability. Vibration is used successfully on wind turbines, where the main rotor speed is typically between 5 and 30 RPM. In a wind turbine there are two main groups of vibration frequencies generated – gear mesh and bearing defect frequencies. This can result in complex vibration signals which can make frequency analysis a formidable task. However, techniques such as enveloping (see later), which has a high sensitivity to faults that cause impacting, can help reduce the complexity of the analysis. Bearing defects can excite higher frequencies which can be used for the basis of detecting incipient damage. Vibration measurement can be generally characterised as falling into one of three categories, viz. detection, diagnosis and prognosis. Detection generally uses the most basic form of vibration measurement, where the overall vibration level is measured on a broadband basis in a range, for example, of 10 to 1,000 Hz or 10 to 10,000 Hz. In machines where there is little vibration other than from the bearings the spikiness of the vibration signal, indicated by the Crest Factor (peak/RMS), may imply incipient defects, whereas the high energy level given by the RMS level may indicate severe defects. Generally speaking, other than to the experienced operator this type of measurement gives limited information, but can be useful when used for trending, where an increasing vibration level is an indicator of a deteriorating machine condition. Trend analysis involves plotting the vibration level as a function of time and using this to predict when the machine must be taken out of service for repair, or at least a more in-depth survey must be performed. Another way of using the measurement is to compare the levels with published vibration criteria for different types of equipment. Although broadband vibration measurements may provide a good starting point for fault detection it has limited diagnostic capability, and although a fault may be identified it may not give a reliable indication of where the fault is, i.e. bearing deterioration or damage, unbalance, misalignment etc. Where an improved diagnostic capability is required frequency analysis is normally employed, which usually gives a much earlier indication of the development of a fault and, secondly, the source of the fault. Having detected and diagnosed a fault the prognosis – what is the remaining useful life and possible failure mode of the machine or equipment? – is much more difficult and often relies on the continued monitoring of the fault to determine a suitable time when the equipment can be taken out of service, or relies on known experience with similar problems. Generally, rolling bearings produce very little vibration when they are fault free and have distinctive characteristic frequencies when faults develop. A fault that begins as a single defect, e.g. a spall on the raceway, is normally dominated by impulsive events at the raceway pass frequency, resulting in a narrow band frequency spectrum. As the damage worsens there is likely to be an increase in the characteristic defect frequencies and sidebands, followed by a drop in these amplitudes and an increase in the broadband noise with considerable vibration at shaft rotational frequency. Where machine speeds are very low the bearings generate low energy signals which again may be difficult to detect. Also, bearings located within a gearbox can be difficult to monitor because of the high energy at the gear meshing frequencies, which can mask the bearing defect frequencies. Overall vibration level This is the simplest way of measuring vibration and usually consists of measuring the Root Mean Square (RMS) vibration of the bearing housing or some other point on the machine with the transducer located as close to the bearing as possible. This technique involves measuring the vibration over a wide frequency range, e.g. 10 to 1,000 Hz or 10 to 10,000 Hz. The measurements can be trended over time and compared with known levels of vibration, or pre-alarm and alarm levels can be set to indicate a change in the machine condition. Alternatively measurements can be compared with general standards. Although this method represents a quick and low-cost method of vibration monitoring, it is less sensitive to incipient defects, i.e. defects in the advanced condition, and has a limited diagnostic capability. Also, it is easily influenced by other sources of vibration, e.g. unbalance, misalignment, looseness, electromagnetic vibration etc. In some situations, the Crest Factor (Peak-to-RMS ratio) of the vibration is capable of giving an earlier warning of bearing defects. As a local fault develops this produces short bursts of high energy which increase the peak level of the vibration signal, but have little influence on the overall RMS level. As the fault progresses, more peaks will be generated, until finally the Crest Factor will reduce but the RMS vibration will increase. The main disadvantage of this method is that in the early stages of a bearing defect the vibration is normally low compared with other sources of vibration present and is therefore easily influenced, so any changes in bearing condition can be difficult to detect. Frequency spectrum Frequency analysis plays an important part in the detection and diagnosis of machine faults. In the time domain the individual contributions, e.g. unbalance, bearing faults, gear faults, etc., to the overall machine vibration are difficult to identify. In the frequency domain they become much easier to identify and can therefore be much more easily related to individual sources of vibration. It is not always possible to rely on the amplitude of bearing discrete frequencies to provide information on defect severity because each machine will have different mass, stiffness and damping properties. Even identical machines could have different system properties and this affects the amplitudes of similar sized bearing defects. Often of real significance is the pattern of the bearing defect frequencies in determining the defect severity. Study of the number of bearing-related harmonic frequencies, frequency sidebands and characteristic features within the time waveform data can be a much more reliable method of determining when action needs to be taken than relying on monitoring amplitude alone. As already discussed, a fault developing in a bearing will show up as increasing vibration at frequencies related to the bearing characteristic frequencies, making detection possible at a much earlier stage than with overall vibration. Envelope spectrum When a bearing starts to deteriorate the resulting time signal often exhibits characteristic features which can be used to detect a fault. Also, bearing condition can rapidly progress from a very small defect to complete failure in a relatively short period of time, so early detection requires sensitivity to very small changes in the vibration signature. As already discussed, the vibration signal from the early stage of a defective bearing may be masked by machine noise making it difficult to detect the fault by spectrum analysis alone. The main advantage of envelope analysis is its ability to extract the periodic impacts from the modulated random noise of a deteriorating rolling bearing. This is even possible when the signal from the rolling bearing is relatively low in energy and 'buried' within other vibration from the machine. Like any other structures with mass and stiffness the bearing inner and outer rings have their own natural frequencies which are often in the kilohertz range. However, it is more likely that the natural frequency of the outer ring will be detected due to the small interference or clearance fit in the housing. If we consider a fault on the outer ring, as the rolling element hits the fault the natural frequency of the ring will be excited and will result in a high frequency burst of energy which decays and then is excited again as the next rolling element hits the defect. In other words the resulting time signal will contain a high frequency component amplitude modulated at the ball pass frequency of the outer ring. In practice, this vibration will be very small and almost impossible to detect in a raw spectrum, so a method to enhance the signal is required. By removing the low frequency components through a suitable high pass filter, rectifying and then using a low pass filter, the envelope of the signal is left, the frequency of which corresponds to the repetition rate of the defect. This technique is often used to detect early damage in rolling element bearings and is also often referred to as the High Frequency Resonance Technique (HFRT) or Envelope Spectrum. Cepstrum analysis Vibration spectra from rotating machines are often very complex, containing several sets of harmonics and also sidebands as a result of various modulations. When trying to identify and diagnose possible machine faults a number of characteristics of the vibration signal are considered, including harmonic relationships and the presence of sidebands. Cepstrum analysis can simplify this, because single discrete peaks in the cepstrum represent spacing of harmonics and sidebands in the spectrum, i.e. the cepstrum identifies periodicity within the spectrum. Cepstrum analysis converts the spectrum back into the time domain i.e. plots amplitude versus time ('quefrency') and harmonics ('rhamonics'). Vibration monitoring can also be used to gain valuable information about the condition of machining processes. In the manufacture of rolling bearings grinding of the raceways is a critical process in the achievement of a high surface finish and roundness, essential to achieving the required service life. Figure 4 shows cepstra of shoe force obtained during the shoe centreless grinding of bearing outer ring raceways2. In this case, as the severity of the dressing process increases, i.e. increasing diamond infeed, the amplitude of the first peak, 2.38 ms, increases along with the number of rhamonics. The quefrency of 2.38 ms corresponds with the wheel rotational frequency, 420 Hz., because as the severity of the dressing operation increases it has a significant effect on wheel form, and hence work piece quality, and the vibration signal becomes more highly modulated at wheel rotational speed. ROLLING ELEMENT BEARINGS Rolling contact bearings are used in almost every type of rotating machinery, whose successful and reliable operation is very dependant on the type of bearing selected as well as the precision of all associated components, i.e. shaft, housing, spacers, nuts etc. Bearing engineers generally use fatigue as the normal failure mode, on the assumption that the bearings are properly installed, operated and maintained. Today, because of improvements in manufacturing technology and materials, bearing fatigue life (which is related to sub-surface stresses) is generally not the limiting factor and probably accounts for less than 3% of failures in service. Unfortunately, however, many bearings fail prematurely in service because of contamination, poor lubrication, mis-alignment, temperature extremes, poor fitting/fits or unbalance. All these factors lead to an increase in bearing vibration, and condition monitoring has been used for many years to detect degrading bearings before they catastrophically fail and incur the associated costs of downtime or of significant damage to other parts of the machine. Rolling element bearings are often used in noise sensitive applications, e.g. household appliances or electric motors, which often use small to medium size bearings. The elimination of bearing vibration is therefore becoming increasingly important from both an environmental consideration and because it is synonymous with the achievement of quality. Vibration monitoring has now become a well accepted part of many PdM regimes and relies on the well known characteristic vibration signatures which rolling bearings exhibit as the rolling surfaces degrade. However, in most situations bearing vibration cannot be measured directly, and so the bearing vibration signature is modified by the machine structure, and this situation is further complicated by vibration from other equipment on the machine, e.g. electric motors, gears, belts, hydraulics, structural resonances etc. (see Figure 3). This often makes the interpretation of vibration data difficult other than by a trained specialist and can in some situations lead to a mis-diagnosis resulting in unnecessary machine downtime and costs. Bearing characteristic frequencies Although the fundamental frequencies generated by rolling bearings can be expressed in relatively simple formulae they cover a wide range and can interact to give very complex signals. This is often further complicated by the presence of other sources of mechanical, structural or electromechanical vibration on the equipment. For a stationary outer ring and rotating inner ring the fundamental frequencies are derived from the bearing geometry as follows: f_c/o = f_r/2 [1 – d/D Cos Y ] f_c/i = f_r/2 [1 + d/D Cos Y ] f_b/o = Z f_c/o f_b/i = Z f_c/i f_b = D/2d f_r [1 – (d/D Cos Y)2 ] where f_r = Inner ring rotational frequency, f_c/o = Fundamental train (cage) frequency relative to outer ring, f_c/i = Fundamental train frequency relative to inner ring, f_b/o = Ball pass frequency of outer ring (BPFO), f_b/i = Ball pass frequency of inner ring (BPFI), f_b = Rolling element spin frequency, D = Pitch circle diameter, d = Diameter of roller elements, Z = Number of rolling elements, Y = Contact angle. The bearing equations assume that there is no sliding and that the rolling elements roll over the raceway surfaces. However, in practice this is rarely the case and due to a number of factors the rolling elements undergo a combination of rolling and sliding. In addition, the operating contact angle á may be different from the nominal value. As a consequence, the actual characteristic defect frequencies may differ slightly from those predicted, but this is very dependent on the type of bearing, operating conditions and fits. Generally the bearing characteristic frequencies will not be integer multiples of the inner ring rotational frequency, which helps to distinguish them from other sources of vibration. Since most vibration frequencies are proportional to speed, it is important when comparing vibration signatures that data is obtained at identical speeds. Speed changes will cause shifts in the frequency spectrum, causing inaccuracies in both the amplitude and frequency measurement. In variable speed equipment spectral orders may sometimes be used, where all the frequencies are normalised relative to the fundamental rotational speed. This is generally called 'order normalisation', where the fundamental frequency of rotation is called the first order frequency. Because of the elastic properties of the raceway materials ball pass frequencies can be generated due to variable compliance or as the rolling elements pass over a defect on the raceways. The frequency generated at the outer and inner ring raceway can be roughly estimated as 40 % (0.4) and 60% (0.6) of the inner ring speed times the number of rolling elements respectively. Unfortunately, bearing vibration signals are rarely straight forward and are further complicated by the interaction of the various component parts, but this can be often used to our advantage in order to detect a deterioration or damage to the rolling surfaces. Analysis of bearing vibration signals is usually complex, and the frequencies generated will add and subtract and are almost always present in bearing vibration spectra. This is particularly true where multiple defects are present. However, depending upon the dynamic range of the equipment, background noise levels and other sources of vibration bearing frequencies can be difficult to detect in the early stages of a defect. Nevertheless, over the years a number of diagnostic algorithms have been developed to detect bearing faults by measuring the vibration signatures on the bearing housing. Usually these methods take advantage of both the characteristic frequencies and the 'ringing' frequencies (i.e. the natural frequencies) of the bearing. By measuring the frequencies generated by a bearing it is often possible not only to identify a problem but also to identify the cause. While it may be only be necessary to identify that a bearing is starting to deteriorate, and to then plan the timing of its change, a more detailed analysis of the vibration can often give some vital clues as to what caused the problem in the first place. This can be further enhanced by inspecting the bearing after removal from the equipment, especially when the fault has been identified early. Bearing defects A rolling contact bearing is a complex vibration system in which the components – i.e. rolling elements, inner raceway, outer raceway and cage – interact to generate complex vibration signatures3. Although rolling bearings are manufactured using high precision machine tools and under strict cleanliness and quality controls, like any other manufactured part they will have degrees of imperfection and will generate vibration as the surfaces interact through a combination of rolling and sliding. Nowadays, although the amplitudes of surface imperfections are in the order of nanometres, vibrations can still be produced in the entire audible frequency range (20 Hz to 20 kHz). Whereas surface roughness and waviness result directly from the bearing component manufacturing processes, discrete defects refer to damage of the rolling surfaces due to assembly, contamination, operation, mounting, poor maintenance etc. These defects can be extremely small and difficult to detect and yet can have a significant impact on vibration critical equipment, or can result in reduced bearing life. This type of defect can take a variety of forms: indentations, scratches along and across the rolling surfaces, pits, debris and particles in the lubricant. During the early development of the fault the vibration tends to be impulsive but this changes as the defect progresses and becomes larger. The type of vibration signal generated depends on many factors, including the loads, internal clearance, lubrication, installation and type of bearing. Because defects on the inner ring raceway have to travel across a number of interfaces – e.g. lubricant film between both inner ring raceway and rolling elements, and rolling elements and outer ring raceway and outer ring-housing interface – they tend to be more attenuated than outer ring defects and therefore can sometimes be more difficult to detect. When a defect first starts a single spectral line can be generated at the ball pass frequency and then, as the defect becomes larger, it allows movement of the rotating shaft and the ball pass frequency becomes modulated at shaft rotational speed. This modulation generates a sideband at shaft speed. As the defect increases in size more sidebands may be generated, until at some point the ball pass frequency may no longer be generated, but a series of spectral lines spaced at shaft rotational speed. A defective rolling element may generate vibration at twice rotational speed as the defect strikes the inner and outer raceways. The vibration produced by a defective ball may not be very high, or may not be generated at all, as it is not always in the load zone when the defect strikes the raceway. Also, as the defect contacts the cage it can often modulate other frequencies (i.e. ball defect frequency, ball pass frequency or shaft rotational frequency) and show up as a sideband. The cage rotational frequency can be generated in a badly worn or damaged cage. In a ball bearing the rolling elements may never generate ball rotational frequency, or twice ball rotational frequency, because of the combination of rolling and sliding and the constant changing of the ball rotational axis. In cylindrical roller bearings the damage often occurs all the way around the majority of the rolling element surface, so the rolling element rotational frequency may never be generated. Variable compliance This occurs under radial or misaligning loads; it is an inherent feature of rolling bearings and it is completely independent of quality. Radial or misaligned loads are supported by a few rolling elements confined to a narrow region and the radial position of the inner ring with respect to the outer ring depends on the elastic deflections at the rolling element-raceway contacts (see Figure 5). The outer ring of the bearing is usually supported by a flexible housing which generally has asymmetric stiffness properties described by the linear springs with different stiffnesses. As the bearing rotates individual ball loads, and hence elastic deflections, change to produce a relative movement between the inner and outer rings. The movement takes the form of a locus which under radial load is two dimensional and contained in a radial plane, while under mis-alignment it is three dimensional. The movement is also periodic with base frequency equal to the rate at which the rolling elements pass through the load zone. Frequency analysis of the movement yields the base frequency and a series of harmonics. So even a geometrically perfect bearing will produce vibration because of the relative periodic movement between the inner and outer rings due to raceway elastic deflections. Variable compliance vibration is heavily dependent on the number of rolling elements supporting the externally applied loads; the greater the number of loaded rolling elements, the less the vibration. For radially loaded or misaligned bearings 'running clearance' determines the extent of the load region, and hence, in general, variable compliance increases with radial internal clearance. A distinction is made between running clearance and radial internal clearance (RIC). When fitted to a machine the former is normally less than the RIC owing to differential thermal expansion and interference fit of the rings. In high speed applications the effect of centrifugal force should also be considered. Variable compliance vibration levels can exceed those produced by roughness and waviness of the rolling surfaces. However, in applications where vibration is critical it can be reduced to a negligible level by using ball bearings with the correct level of axial pre-load. Bearing speed ratio The bearing speed ratio (ball pass frequency divided by the shaft rotational frequency) is a function of the bearing loads and clearances and can therefore give some indication of the bearing operating performance. When abnormal or unsatisfactory lubrication conditions are encountered, or when skidding occurs, the bearing speed ratio will deviate from the normal or predicted values. If the bearing speed ratio is below predicted values it may indicate insufficient loading, excessive lubrication or insufficient bearing radial internal clearance, which could result in higher operating temperatures and premature failure. Likewise, a higher than predicted bearing speed ratio may indicate excessive loading, excessive bearing radial internal clearance or insufficient lubrication. An experienced analyst cannot only use vibration monitoring to detect deterioration in bearing condition but also to make an assessment of whether the equipment is operating satisfactorily at initial start-up. In electrical machines two deep-groove radial ball bearings are commonly used to support the shaft, one bearing fully located while the other is non-locating and free to slide in the housing to compensate for axial thermal expansion of the shaft. It is not unusual for bearings to fail catastrophically due to thermal pre-loading or cross-location as a result of insufficient clearance between the bearing outer ring and housing, resulting in the non-locating or 'floating' bearing failing to slide in the housing, i.e. the bearings become axially loaded. The effect of this axial load is to increase the operating contact angle, which in turn increases the ball pass frequency of the outer ring. For a ball bearing the contact angle can be estimated from the following expression – Y = Cos-1 [1 – RIC / [(2 (r_o+r_i-D)] ], where Y = Contact angle, RIC = Radial internal clearance, r_o = Raceway groove radius of outer ring, r_i = Raceway groove radius of inner ring, D = Ball diameter. Because a radial ball bearing is designed to have a radial internal clearance in the no-load condition, it can also experience axial play. Under an axial load this results in the ball-raceway contact having an angle other than zero. As the bearing radial internal clearance, hence axial play, increases so does the contact angle. For a correctly assembled motor under pure radial load the contact angle will be zero and the BPFO will be given by the expression – f_b/o = Zf_r/2 [1 – d/D ] On the other hand, if cross location occurs (outer ring not free to move in housing) then the bearing radial internal clearance will be lost by the relative axial movement between the inner and outer rings, the bearings become axially loaded and the BPFO will increase due to the increase in contact angle. The amplitude of BPFO is likely to be small until the bearing becomes distressed and it may not always be possible to detect the BPFO, particularly if using a linear amplitude scale. A log or dB amplitude scale may be better, but again care should be exercised because there may be other frequencies which may be close to the BPFO. A good example of how the bearing speed ratio can be used to identify a potential problem is shown in Figure 6, which shows a vibration acceleration spectrum measured axially at the drive end (DE) on the end cap of a 250 kW electric motor. The measurements were obtained during a 'run-up' test prior to installing in the plant. For a nominal shaft speed of 3000 RPM the calculated BPFO was 228.8 Hz, giving a bearing speed ratio of 4.576. The measured BPFO was 233.5 Hz (see Figure 6), giving a bearing speed ratio of 4.67, an increase of 2%. The BPFO of 233.5 Hz corresponds to a contact angle of 25°, which strongly suggested that the Type 6217 bearing was experiencing a high axial load. The most probable cause was that the bearing had been installed too tightly in the housing so that it could not move freely in the housing as the shaft of the motor expanded and contracted. Soon after installation the motor failed catastrophically. Examination of the inner ring revealed that the ball running path was offset from the centre of the raceway towards the shoulder. After a thorough investigation of all the bearing fits, it was confirmed that there was insufficient clearance between the outer ring and housing of the non- locating bearing, resulting in cross location (thermal loading) which was consistent with the vibration measurements taken prior to installation. A number of harmonics and sum and difference frequencies – relating to the ball pass frequency of the outer ring (233.5 Hz), the cage rotational frequency (21 Hz) and the inner ring rotational frequency – are also evident in the spectrum of Figure 6. When the motor was rebuilt with new bearings and the correct bearing fits, the 'run-up' test was repeated prior to installation (see Figure 7). The raw spectrum shows no characteristic bearing frequencies, but when both the amplitude and frequency scales are expanded a discrete peak at 229 Hz becomes evident [see Figure 7(b)], which matches very closely with the predicted ball pass frequency, f_b/o, of the outer ring, of 228.8 Hz. This motor went on to operate successfully. REFERENCES 1. Bently Nevada, Application Note, Asset Categorisation. 2. Lacey S J, Vibration monitoring of the internal centreless grinding process, Part 2: experimental results, Proc. Inst. Mech. Eng., Vol 24, 1990 3. Lacey S J, An overview of bearing vibration analysis, Schaeffler (UK) Technical Publication. steve.lacey@schaeffler.com www.schaeffler.co.uk IN THE NEXT ISSUE – PART II: APPLICATIONS This second and concluding part of the paper will present examples of the use of vibration monitoring to detect and diagnose problems on rotating equipment ranging from electric motors to large crushing machines used for mining and processing. Also included will be examples taken from the FAG WiPro Condition Monitoring System used for monitoring the condition of wind turbine drive trains. Click to Download

Click here to enquire