Thread Content
I. Overview: Mechanical fault diagnosis is a new discipline that has developed over the past twenty years. It is an important part of modern equipment maintenance techniques, and it is increasingly becoming a key indicator of the modernization of equipment maintenance management. The application of this technology plays a crucial role in ensuring the safety of mechanical equipment, improving product quality, reducing maintenance costs, and preventing environmental pollution. There are various methods that can be used in the condition monitoring and fault diagnosis of mechanically faulty equipment. Such as vibration monitoring technology, oil analysis technology, infrared temperature measurement technology, acoustic emission technology, non-destructive testing technology, etc. Among them, vibration monitoring technology is a commonly used basic method, as the theories and measurement techniques related to vibrations are well-developed and simple to implement. Furthermore, statistics show that 90% of mechanical failures can be detected through vibration measurement. Vibration measurement and signal analysis have always been key methods for predictive maintenance, and equipment departments in various industries typically start carrying out this work by focusing on these two aspects. Vibration monitoring technology is “a technique for detecting, analyzing, and processing vibration signals of equipment, as well as for fault identification and prediction.” II. Simple Diagnosis and Precision Diagnosis: Equipment condition monitoring technology refers to the process of testing certain characteristic parameters of equipment (components, parts), and comparing the obtained values with the specified normal values in order to determine whether the equipment is operating normally or abnormally (i.e., whether there is a fault); this is also known as simple diagnosis. Equipment fault diagnosis technology not only needs to determine whether the condition of mechanical equipment is normal, but more importantly, it must estimate the cause, location, and severity of machine failures. Thus, it is called precision diagnosis. At present, simple diagnosis (condition monitoring) is more widely used, while precise diagnosis is actually applied in production only in limited cases, and mainly in high-precision equipment. This situation is the same in Europe, America, and Japan, and it is widespread. This indicates that simple diagnosis is relatively mature and easy to implement, whereas precise diagnosis remains a developing technology and is not yet fully mature. Additionally, the cost of precision diagnosis is relatively high; it requires sophisticated instruments and must be carried out by engineers with specialized training, which is why it is only applied to important equipment. This is something worth noting when developing and promoting diagnostic technologies in our country. The focus should currently be on promoting simple diagnostics or condition monitoring. At the same time, efforts are being made to develop precision diagnostic technologies so that they can be put into use as soon as possible. According to relevant statistics, simple diagnostic instruments can resolve 50% of the faults that occur during equipment operation. It is evident from this the important role of simple diagnostics in equipment management and maintenance. Nippon Steel believes that in large steel complexes, to ensure the proper operation of all system equipment, two technologies must be implemented. The first is to effectively monitor the condition of machines, that is, “equipment monitoring technology”. The second is accurate diagnostic methods. That is: “Precision diagnostic technology”. 1. “Equipment monitoring technology”. Such technologies should meet the following requirements: ● They must be able to measure various types of machinery quickly; ● They should enable on-site conclusions to be drawn; ● They need to be cost-effective and easy to use. At present, Nippon Steel has developed a range of hardware specifically designed for use with various equipment in steel plants. Nippon Steel classifies the hardware used for equipment monitoring into three categories: (1) The first category consists of small instruments that can be carried by routine inspectors and that provide quantitative results. For example: machine detectors, oil film detectors, small vibration meters, motor detectors, acoustic emission crack detectors, etc. (2) The second category consists of monitoring instruments, which are usually installed on-site and capable of performing automatic data sampling, automatic data recording, comparison with trend management charts, display of anomaly detection charts, etc. Examples include rotary machinery monitors, compressor monitors, hot rolling mill monitors, etc. (3) The third category is data control systems, which require data obtained from the measurements made by the first category of instruments as input in order to function. It can automatically process large amounts of measurement data, record it, store it, compare it with control charts, and issue control signals, among other functions. 2. “Precision diagnosis technology”: When measurements are taken using monitoring equipment and compared with the charts or values used for trend management, it becomes necessary to conduct precision diagnosis when a fault is detected or when a potential fault is predicted. Through precision diagnosis, it is not only possible to determine whether a fault indeed exists, but also, if a fault is present, its location, cause, and severity can be identified. Nippon Steel believes that such technology should be carried out by specialized technicians known as precision diagnosis experts. In this regard, Nippon Steel has developed two types of instruments in recent years: (1) Specialized instruments primarily used for rotating machinery; these instruments usually require the use of various software technologies (such as diagnostic manuals and criteria for assessing vibration severity), and they are used in conjunction with analyses of the equipment’s condition. Examples include automatic diagnostic systems for rotating machinery, automatic balancers, and acoustic emission crack analyzers. (2) The other category consists of general-purpose instruments, which are used for measurements and analysis aimed at condition diagnosis. For example: signal recorders, signal analyzers, transfer function analyzers, etc. III. Applications of vibration measurement in machine condition monitoring 1. Advances in measurement methods A well-designed machine also has a low natural vibration level. But when the machine wears out, the foundation sinks, and the components deform, various subtle changes begin to occur in its dynamic performance: the shafts become misaligned, the components start to wear, the rotor becomes unbalanced, and the gaps increase. All these factors are reflected in the increase in vibrational energy. Therefore, increased vibration is often a sign that a machine is about to fail, and vibration can be measured from the outside surface of the machine. In the past, equipment engineers relied on experience, using touch and listening to determine whether a machine was functioning properly or if its faults were worsening. But today the machine’s rotation speed is very high, and many warning vibrations occur in the high-frequency range; therefore, they can only be detected using instruments. The method is: when the machine is operating properly, it exhibits typical vibration level and spectral characteristics. As the machine failure progresses, the dynamic behavior of the machine as well as the forces acting on its components also change, thereby affecting the vibration energy levels and the shape of the spectrum of the machine. Through the measurement and analysis of such vibrations, we can determine changes in the machine’s operating condition as well as whether maintenance is required. It is worth noting when using vibration measurement for machine condition monitoring that success does not necessarily depend on substantial initial investment in analytical instruments equipped with advanced computers. Many successful cases began with the use of less expensive analog vibration meters and analyzers in representative units. When experience is gained and expansion is planned, it naturally evolves to the use of faster and more powerful instruments. It is wise to purchase high-quality instruments from the start. If many new solutions are undertaken by engineers with limited experience in vibration measurement, coupled with inaccurate and inconsistent measurement results, this will **limit the reliability of condition-based maintenance**. Shanghai Xingsheng Testing Instruments Co., Ltd. developed a series of portable measuring instruments following this approach. Products such as VIB-5, VIB-10, VIB-15, VIB-20, BT2000, BA2010, etc. 2. Measurement instruments and systems The instrument systems used for periodic monitoring of vibrations can be classified into three categories based on their complexity: (1) Simple systems The simplest systems consist of a direct-reading pocket vibration meter used to measure vibration levels within a certain frequency range. Compare the measurement results with general standards or the specific reference values for each machine. For those who gain experience in condition monitoring using vibration measurements, a high-quality handheld vibrator is necessary. This type of vibrometer can measure the root mean square (RMS) value or the peak value of single vibration acceleration or velocity within a frequency range of 10–1000 Hz or 10–10000 Hz. The RMS value of the velocity can be directly compared with standard vibration severity ratings to determine the extent of repair required. Although broadband vibrometers have limited utility in early fault detection, diagnosis, and damage prediction, they are still necessary. (2) Vibration condition monitoring system (a) Basic frequency analysis system. There are several configuration options for the frequency analysis systems used for machine condition monitoring. If you want to start with a limited investment and have only dozens of measurement points (rather than hundreds or thousands), battery-powered vibration analyzers and vibration level recorders are superior. This supporting device can sequentially generate the narrowband spectrum diagrams for each monitoring point. Each measurement and analysis takes about a few minutes, and the data for each measurement point are manually recorded on a log sheet. Record the reference spectrum for each measurement point and copy it onto a transparent card. The spectra obtained later are placed below the reference spectrum, and by comparing the two cards, the differences can be identified immediately. If the amplitude at a certain frequency increases, its amplitude-over-time curve is plotted, allowing for the prediction of its trend. Some users only perform regular, simple broadband monitoring; frequency analysis is used only when there are significant changes in the amplitude level. This approach is useful for diagnosing the progression of faults, but it cannot be used for early warning or for analyzing patterns in trend development. (b) Advanced machine condition monitoring systems: When a large number of machine sensors need to be monitored, it is reasonable to use more sophisticated systems. The operator can use an accelerometer and a data collector (or data recorder) to directly record the vibration signals at each test point. After returning to the equipment department, it is sent to a computer where software is used for analysis. The advantages of this approach are faster speed, enhanced functionality of the instruments, improved detection capabilities, and reduced measurement costs. (3) Permanent machine vibration monitoring: The machine vibration monitoring systems mentioned above are all based on periodic measurements and calibrations. Permanent monitoring is first applied to those individual machines that play a crucial role in the production process; when sudden changes occur in these machines, it is possible to issue warnings to the control room immediately or within a few minutes, allowing effective actions to be taken before catastrophic failures occur. Such systems are widely used in power generation and the petrochemical industry to monitor equipment such as turbines, feed pumps, and compressors. The primary requirements for a permanent condition monitoring system are high reliability, long-term stability, as well as the ability to withstand harsh environments and false alarms. A robust machine design, the capability to operate in humid and dusty conditions, along with regular environmental testing, are necessary to meet these stringent requirements. Front-line equipment that is particularly robust, such as accelerometers, cables, and junction boxes, also needs to be able to operate in high temperatures. This system also includes an automatic testing system, so that in the event of an alarm, operators can immediately check which instruments are functioning properly.