Through health monitoring, the development of intelligent bearing solutions will help improve the reliability of aeroengines. The purpose of this project is to develop an intelligent bearing system for the ground test verification machine with ultra-high propulsion efficiency (UHPE), including a fully integrated self powered wireless sensor system for the next generation aircraft. This paper provides an overview of the existing technologies of intelligent bearings, showing the structure of the integrated sensing system, focusing on the parameters to be monitored by aeroengines and the selection methods of sensing technologies. At present, most of the intelligent bearings that have been developed can be used in automobile, railway and other industries. Their availability is limited and they are not suitable for the harsh environment (such as high temperature and high vibration level) that jet engines experience. The primary monitoring contents include vibration, temperature, load, spindle movement, speed and wear debris. Select appropriate sensing technology based on the classification method of durability under extreme environment inside the engine, as well as size, weight, sensitivity, working frequency band, installation method, data processing requirements and energy consumption of energy collection and wireless transmission.
Introduction
Rolling bearing is one of the most important parts in jet engine. The condition monitoring of jet engine bearings is helpful to detect bearing faults and predict bearing life. The intelligent integrated sensing bearing developed can realize online condition monitoring. The bearing is called intelligent bearing, which is composed of small low-power sensors and has self energy supply capability for wireless communication and data transmission. Intelligent bearing field of vision will promote online condition monitoring to a new level. However, at present, most of the existing intelligent bearing technologies can be used in automobiles, railways, wind power equipment, etc. Due to the complex and challenging environment and operating conditions of jet engines, including extremely high spindle speed, high vibration frequency and high temperature, the development of intelligent bearings that can be used in jet engines is very limited. The main shaft of a jet engine and the bearings for a turbine are exposed to temperatures of about 200 ℃ and 300 ℃ respectively. High temperature lubricating oil also presents a harsh environment for the sensor. Other challenges include limited input power, limited space and availability of wired channels, and unavailability of high temperature resistant electronic components on the market. In addition, the use of magnetic sensors or materials around jet engine bearings is strictly restricted, as they may adsorb metal debris and cause blockage. In jet engines, bearings are installed in sealed metal cases, which severely limits wireless data transmission. Therefore, although the technology has made significant progress in recent years, the research and development of intelligent bearings for jet engines is still a challenge.
The first step of this work is to determine the sensor components used for bearing condition monitoring in the harsh environment where the jet engine is located, as well as the sensor integration in the intelligent bearing that can measure a series of parameters indicating the bearing status. At the same time, we develop an energy collection technology that can collect and transmit data wirelessly, which is a key part of intelligent bearings.
Main Purpose of the Project
*Identify the sensors on the market that are suitable for monitoring the condition of jet engine bearings, especially those that can operate in high temperature and corrosive lubricating oil environments of jet engines, and use them in jet engines;
*Identify low-power sensors to reduce energy consumption;
*Identify and develop energy collection technologies suitable for jet engine environments;
*Optimize the energy consumption of the sensor system and develop an energy distribution strategy;
*Develop wireless communication system for data transmission through metal chassis of jet engine.
In order to verify the selected technology and intelligent sensor system, a series of tests on parts and small bearing levels will be carried out in the laboratory. A test head is designed for the test bed of small bearings to simulate the real environment of jet engines. This paper focuses on the development of intelligent bearing sensor components. First of all, this paper summarizes the existing intelligent bearing technology, and discusses the challenges faced by the sensor system in the jet engine environment. After that, this paper describes the method of sensor selection and the structure of intelligent bearing, and finally gives the conclusion.
Summary of intelligent bearing technology
In the past three decades, a lot of work has been done in the development of sensing bearings. Initially, the research focused on installing multiple sensors on the bearing to measure the parameters that can indicate the bearing status. Vibration, speed and temperature are considered to be the most important parameters for the on-line monitoring of bearing status. It is then expanded to include load and lubrication monitoring.
The installation of intelligent units is an important aspect in the development of intelligent bearings. The sensor unit was first installed on the bearing pedestal, which was developed into embedding the sensor into the bearing ring. The sensors of most bearings on the market are connected through a wired refurbished ring system. Most of these bearings can be used in the automotive and railway industries, such as the sleeve to be installed in the axle box bearing unit with integrated sensors in the railway industry. In a word, great progress has been made in the development of sensing bearing technology. However, until now, the number of commodities available, such as axle box bearings, NSK motion and control, active sensor bearings, and integrated rotary sensor bearings, is still limited. The focus of research has shifted from sensing bearing (wired sensor unit) to intelligent bearing (self powered wireless sensor system). In order to remove the power supply for online monitoring of intelligent bearings, wireless sensor systems and self powered sensor units for energy collection are very popular. However, the intelligent bearing with self energy supply and wireless sensor system is still in the research and development stage, and there is no product on the market yet? Similarly, the development of thin film sensors and MEMS has shifted the focus of research to embedding sensors into the inner and outer rings of bearings. Most of the development of sensing and intelligent bearing technologies have been applied to the railway and automobile industries, but less attention has been paid to jet engine bearings. Traditionally, jet engine bearings are monitored by measuring vibration and oil debris monitoring. The purpose of this research is to develop an integrated intelligent bearing system for a new generation of jet engine based on the existing knowledge of intelligent bearing technology and the operating conditions of jet engine bearings.
Challenges in Developing Intelligent Bearings for Jet Engines
As mentioned above, although smart bearings used in other fields have been developed, there are still no smart bearings available for jet engines due to some major challenges. In the initial phase of this study, these challenges became clearer, helping to identify sensor technologies for jet engines.
The jet engine bearings operate at high speed (3,000rpm - 10,000rpm), high temperature (＞ 200 ℃) and high vibration (＞ 100g). In addition, the jet engine stays in the so-called hot soak state to store the heat, which can not dissipate heat even after the engine stops working, thus raising the bearing temperature to 250 ℃.
In order to simulate the environment of jet engines, planned tests of bearings will be carried out at 150 ℃ to 250 ℃. This is a major challenge for most existing electronic devices, as they can only work in environments up to 80 ℃. Finding sensors and related technologies suitable for high temperature environment is the main obstacle in the development of intelligent bearings for jet engines. More than 90% of accelerometers are designed and manufactured for use in environments below 80 ℃.
The second challenge is the high spindle speed (3,000rpm - 10,000rpm), which creates a high vibration environment with high amplitude. This not only makes it more difficult to improve the durability of the sensor, but also poses a major challenge for measuring vibration, cage speed and other values (see below for details). In addition, in order to simulate the performance of jet engines, smaller bearings are used on the test bench, so it will run to a higher speed (between 25,000 rpm and 30,000 rpm) to achieve a pitch diameter similar to that of jet engines.
In addition to temperature limits, smart bearings for jet engines require low energy consumption to enable wireless power and data transmission using appropriate energy collection technologies. There are further restrictions in the jet engine environment, such as low energy consumption requirements (resulting in limited airborne data processing and storage), small sensor installation space, inflexible engine design after adding customer requirements, inability to use magnetic sensors due to metal debris blockage, and inability to use optical sensors (the use of oil will hinder optical performance).
For sensors that meet the high temperature requirements, they should also be tested to ensure that they can be exposed to the high temperature (such as 180 ℃) lubricating oil of jet engines. In general, jet engines use gas engine oils and/or high thermal stability (HTS) oils. These oils are aggressive and can cause chemical damage to the sensor in a high temperature environment for a long time. Lubricants can also damage the connectors and cables of the sensors in the engine.
With regard to high temperatures, if it is necessary to stick the sensor to the bearing/housing, you must select the appropriate adhesive or epoxy resin, as most adhesives cannot be used in high temperatures. Before use, the impact of corrosive oil environment on the adhesive shall also be inspected. In order to verify the selected sensor and its connector and cable, this study conducted a pre-test in a high-temperature lubricating oil environment before it was integrated into a small bearing test bench.
Selection of sensors
One of the most important tasks in the development of integrated intelligent bearings is to carefully select the commercial off the shelf (COTS) sensors suitable for the operating conditions of jet engine bearings. Initially, prior to the development of fully integrated smart bearings in mounted/embedded bearings, the sensors were mounted in bearing pedestals on the test bench. As mentioned above, it is considered to use sensors capable of measuring vibration, temperature, cage speed, spindle displacement and load during the development of intelligent bearings.
To ensure that the most suitable sensor is selected for the smart bearing, the COTS sensor is selected using the method shown in Figure 1. In the aerospace industry, in order to reduce costs, COTS sensor based solutions have received more and more attention. Any COTS sensors to be deployed in jet engines need to meet the high performance standards mentioned by the aerospace industry. The selection of sensors is based on the information and knowledge of bearing monitoring obtained from literature and standards, bearing design, bearing environment and operating conditions, and other requirements. The selection process can be divided into two parts: a) identification of correct methods and technologies; b) Identification of the most suitable sensor for the prior art. The first part identifies the techniques applicable to the measurement of specific parameters. For example, there are many available methods to measure bearing temperature, such as thermocouple, MEMS technology, etc. For this application, the reason why the thermocouple is selected instead of MEMS technology is that the thermocouple does not need to input electrical energy and can measure temperature in a wide range. Based on the techniques identified in the first part, the second part focuses on the selection (modeling and manufacturing) of specific sensors.
Select sensors capable of measuring vibration, cage speed and load for smart bearings of jet engines. The following subsections provide details of selection.
01 Vibration
Vibration monitoring is one of the most important and commonly used methods for bearing condition monitoring, because vibration monitoring can provide diagnostic information according to the specific characteristic frequency of the bearing to identify faulty components. Even small defects on the bearing mating surface, if not detected in time, can lead to bearing failure. According to the geometry, the number of rolling elements and the spindle speed, the defects of rolling bearings will produce a specific frequency. The expected defect frequency can be calculated using the formula given in Reference 16. The detection of these frequencies is helpful to predict the life of jet engine bearings. For the planned bearing tests, the expected defect frequency has been calculated based on the bearing design and spindle speed. These calculations provide information for selecting the appropriate sensor for the test bearing.
In order to effectively measure the vibration, the sensor shall be used to measure the vibration? It is installed on the bearing next to the contact area (near the load area), where the rolling element of the bearing directly contacts the raceway. The area near the load where the sensor is installed is also the high-temperature area of the jet engine bearing, and the temperature can be as high as 250 ℃. The rapid speed of jet engine leads to high defect frequency. Therefore, the charging mode accelerometer technology meets the requirements, while displacement and vibration based technologies are not appropriate.
In addition to the stringent requirements for the accelerometer temperature and frequency range, the sensor resonant frequency is also important. For the required frequency range (>25kHz), the resonant frequency must be at least twice the accelerometer operating frequency? triple. This means that the accelerometer has a resonant frequency of at least 50 kHz. The resonance and operating frequency of the accelerometer are inversely proportional to the sensitivity, that is, the higher the resonance frequency, the lower the sensitivity, and vice versa. In this case, the higher resonance frequency is preferred, because the sensitivity can be controlled by the amplifier.
The installation method is another factor to be considered when selecting sensors. To ensure that the accelerometer is securely mounted on the bearing in high vibration and high temperature environments, only bolt and screw mounted sensors are available. It is not feasible to stick the accelerometer on the bearing by adhesive installation, because it will not only reduce the operation and resonance frequency, but also act as a vibration attenuator. In addition, in the high temperature environment, the adhesive ability will decline with time, which cannot meet the requirements of long-term operation.
According to the criteria defined in the selection method, hundreds of COTS accelerometers provided by different manufacturers were screened, and only eight sensors met the requirements of operating frequency, resonance frequency and other characteristics. The spindle speed is very fast (25,000 rpm - 30,000 rpm); Therefore, it is expected that the defect frequency will also develop towards the higher end of the spectrum. At harmonic frequencies of 5 and 10, the expected flaw frequencies are 28 kHz and 56 kHz, respectively. The operating and resonant frequencies of these accelerometers are greater than 15kHz and 45kHz, respectively. The two accelerometers with the highest resonant frequency have been selected, with frequencies of 90kHz and 100kHz respectively. The operating frequency of both accelerometers is 20kHz. In addition, there are sensors with operating frequency up to 30kHz. However, the given operating frequency is higher than other accelerometers, but the resonant frequency falls within the harmonic frequency generated by the bearing defect frequency. Therefore, the use of this accelerometer is impractical and will not be used in the test.
02 Cage speed
In jet engines, the rotating speed of bearing components is very fast, and the sliding between raceways and rolling elements will cause early failure. The relative sliding between mating surfaces will produce a large amount of surface shear stress. For bearings in high-speed rotation, sliding will cause the actual speed of the rolling element to be lower than the theoretical value. The slip effect cannot be monitored by vibration, but it can be monitored by measuring the speed of the cage.
The rotational speed of the cage can be measured by non-contact methods such as eddy current, capacitive sensor, magnetic sensor and optical sensor. However, due to a series of reasons, the harsh environment of the jet engine limits the use of magnetic, capacitive and optical sensors. For example, it is not allowed to put magnetic components into the air oil tank, because the magnetic sensors will gather wear debris, causing danger. Optical sensors cannot make accurate measurements because light diffracts and scatters in the bearing lubricating oil environment. The measuring range of capacitive sensor is limited, and the lubricating oil has a significant impact on the measuring accuracy.
The eddy current sensor meets all requirements for measuring the rotational speed of the bearing cage of the jet engine, including high temperature, high rotational speed and the available space around the engine bearing. The cage speed is measured by calculating the time for each ball to pass through the eddy current detection probe. As shown in Figure 2, each time the ball passes through the probe, a distorted square wave will be generated due to magnetic field interference. When a pulse is generated when a certain rate is reached, this rate is called switching frequency and can be calculated by multiplying the number of balls by the cage speed. For the bearing on the undersized test bench, the theoretical value of the cage speed is about half of the spindle speed. Between 12500 rpm and 15000 rpm, the number of balls is 20. The resulting switching frequency is between 250000 and 300000. Measuring such a high switching frequency value is a challenge for most COTS sensors. Combining other factors that need to be considered, such as temperature, probe range and reaction time, ball surface area and oil immersion, selecting an eddy current sensor suitable for jet engine bearings becomes challenging.
Fig. 2 Measuring the rotational speed of cage with switching frequency
The temperature around the bearing in the air oil groove can be as high as 200 ℃. A typical eddy current sensor consists of a sensing unit connected with electronic components, which can withstand a high temperature of 80 ℃. The solution to this problem is to use cables to separate the sensing unit from the electronic components. Unfortunately, this will greatly slow down the response time of the sensor and reduce the switching frequency. The switching frequency of most eddy current sensors that can be used at high temperatures can only reach a few hundred hertz, while the switching frequency of jet engine holder speed measurement is about several thousand hertz. One possible solution is to apply a squeeze in the cage that can be measured by a turbine sensor for each cycle. However, this is based on the feasibility of replacing existing jet engine bearing designs.
The choice of eddy current probe is often based on its measurement range, probe area and measurement target size. Similarly, the measurement range is directly related to the probe size, that is, when the probe size increases, the measurement range will also increase, and vice versa. However, for a given target, it is recommended that the probe size be less than or equal to the target size (see Figure 2). In order to maximize the detection, the shape of the measuring object (e.g. cage) should preferably be rectangular (see Figure 2). If it is a ball bearing, the visible surface area of the eddy current probe is very small, so it is better to choose a probe that is smaller. However, this in turn will narrow the measuring range of the probe. This can be adjusted if the sensor is mounted next to the bearing. In addition, the cage rotating at high speed may have a small amount of axial displacement, which requires the sensor to be installed at a safe distance to avoid contact with the bearing during operation.
Combined with all the challenges in the selection process, it was found that only two eddy current probes met the conditions, and they were selected for the development of intelligent bearings. These two probes will be tested on the small bearing test bench to measure the bearing retention? The capability of rotating speed shall be evaluated. Later in the project, the feasibility of specially designed cages will also be explored.
03 Load
The jet engine bearing bears the load both axially and radially. The real-time monitoring of the load on the bearing can help to understand the dynamic state of the engine under complex operating conditions. Load cells are usually used to measure the load, but because of their heavy weight and size, they are not practical, so they are not suitable for jet engine bearings. Therefore, in this application, an alternative method is selected to evaluate the load by measuring the elastic deformation of the fixed bearing ring with a strain gauge. There are many methods to measure strain, and three of them may be applicable to the harsh environment of jet engines, including resistance strain gauges, light gratings and surface acoustic wave devices. The light grating measurement system is very large and needs a lot of energy to support operation. Similarly, SAW sensors need to be further developed in order to be able to measure strain in the harsh environment of jet engines. Therefore, resistance strain gauge is selected in this project to measure the strain of jet engine bearing.
In order to measure the elastic deformation of the outer ring, it is recommended to install the strain gauge directly on the bearing (fixed) outer ring. The strain gauge shall be installed outside the bearing, and the radial and axial strains shall be measured along the secondary side. On the outer ring, the strain gauge is exposed to high temperatures up to 250 ° C. As mentioned above, a suitable adhesive (or adhesive) should be selected to complete the long time sensing. Similarly, after a period of time, the chemical aggressiveness of the lubricant will also weaken the bonding effect. Therefore, the strain gauge must be protected from aggressive lubricating oil. When oil seepage occurs between the strain gauge connectors, it will immediately cause sensor failure.
In addition, during operation, the outer ring of the jet engine bearing undergoes severe temperature changes, and the strain measurement largely depends on the temperature of its environment. In order to obtain accurate strain measurement results, temperature compensation must be applied. It can be realized by T-type strain gauge (see Figure 3), and differential strain can be measured by establishing a plate bridge circuit. However, due to the limited space available on the bearing rings, especially on the raceway side, this presents another challenge. Figure 3 shows that in order to measure the radial strain, the strain gauge should be installed outside the ferrule. However, the total width of the outer ring to be tested is 5.5 mm. Considering all the limitations and requirements, it is recognized that only two types of T-strain gauges are applicable to the test bearing. The dimensions of these two strain gauges are 5.6 mm × 5.6 mm (rectangular) × 5.4mm (round).
Conclusion
Through preliminary research, it is clear that the harsh environment of jet engines poses a great challenge to the development of intelligent bearings. Two of the main challenges - high temperature and high speed, as well as many other challenges also limit the choice of sensors for jet engine bearings. Based on literature description and industry experience, the most important parameters selected for the monitoring of jet engine bearings are vibration, temperature, cage speed, spindle displacement and load. A methodology is used to select the sensing technology suitable for aviation bearings. After comprehensive screening of COTS sensors, it is found that only a few sensors meet the requirements. The future work will focus on the pre test of the selected sensor in the high temperature and oil immersion environment before the test on the small bearing test bed.
More about WBM Tapered Roller:
WBM produces high-precision tapered rollers: Capacity: 10 million pcs per month ;Residual capacity: 4 million pcs per month; 5 roller production lines; 7 cold heading machines (3 Nationals, 4 Domestic); Heat treatment capacity: 9 tons per day.
Welcome to contact WBM for more information!