Open Access
Add to BibSonomy
Abstract
Large strain measurement under high-temperature environment has been a hot but difficult research issue in the fields of measurement and metrology. However, conventional resistive strain gauges are susceptible to electromagnetic interference at high temperature, and typical fiber sensors will be invalid under high-temperature environment or fall off under large strain conditions. In this paper, aiming to achieve effective and precision measurement of large strain under high-temperature environment, a systematic scheme combining a well-designed encapsulation of a fiber Bragg grating (FBG) sensor and a special surface treatment method using plasma is presented. The encapsulation protects the sensor from damage while achieving partial thermal isolation and avoiding shear stress and creep, resulting in higher accuracy. And the plasma surface treatment provides a new bonding solution which can greatly improve the bonding strength and coupling efficiency without damaging the surface structure of the object under test. Suitable adhesive and temperature compensation method are also carefully analyzed. Consequently, large strain measurement up to 1500 µɛ under high-temperature (1000°C) environment is experimentally achieved in a cost-effective way.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Strain measurement is one of the key aspects of structural health monitoring, which is of great significance to assess the state of structures and to secure minimum safety conditions [1–5]. For major equipment and systems, such as aerospace engine, turbine motor and internal combustion engine (ICE), dynamic strain measurement is not only an important support in the evaluation of reliability, but also a visual basis for structural optimization and refinement design [6,7]. However, most of the major equipment work in extreme environments of high pressure and temperature, some are above 1000°C, where the mechanical properties of materials and structures are significantly different, and fatigue, loosening and wear of components, foreign bodies and machining defects can lead to abnormal strain and high-frequency vibration, resulting in equipment failure, damage and even cause major safety accidents. Therefore, the detection of core parameters such as temperature, strain and vibration of large equipment is an urgent problem to be solved [8,9].
In terms of temperature, it is generally referred to as a high temperature strain test when the test temperature exceeds 350°C, which is the maximum operating temperature limit for general sensors. In terms of strain measurement under high temperature environments, it is generally accepted that measurements exceeding 1000 µɛ can be considered as large strain measurements. Strain measurement under room-temperature or high temperature below 350°C is easy to realize using conventional welding resistive strain gauges or fiber sensors bonded with epoxy resin. Welding metal sensors such as strain gauges has been an effective and commonly used traditional solution, but it will damage the metal material properties and the structural mechanical properties of the equipment, and metal sensors are easily oxidized under high-temperature environment, so it is not allowed or suitable in many occasions.
Lightweight fiber optic sensors such as fiber Bragg grating (FBG) and Fabry-Perot (FP) sensor cavities are more suitable for compact space, narrow, complex structure of the working environment, and cannot destroy the original mechanical structure, hence are therefore popular recently [10–13]. As a revolutionary sensing technology, FBG sensors perceive the signal by analyzing the transmitted or reflected light signal in the optical fiber [14–17]. Compared with traditional electrical sensors such as resistance strain gauges, FBG has the characteristics of small size, light weight, low loss, high sensitivity, and resistance to electromagnetic interference, corrosion resistance and multiplexing [18–20]. However, considering the above requirements, fiber optic sensors generally face an industry difficulty, that is, their effective installation in extreme environments. The core is to maintain effective mechanical transfer efficiency under high temperature and strong vibration conditions. FBG sensor not only needs to be able to measure large strain in high temperature environment, but also needs to be closely integrated with the structure to be tested and maintain high mechanical transfer efficiency. Besides, the collected strain data usually contains two parts, one part is the real strain caused by mechanical load or thermal stress and the other is the heat output caused by high temperature environment. The combination of the two causes a large change in sensing value, which further increases the difficulty of measurement.
For non-interventional measurements, conventional robust and reliable installation method, such as welding and screws, cannot be used, because they would cause destructive damage to the surface of the structure, making it unable to maintain its original appearance, and the cost of testing such damage is high. Inorganic adhesive composed of ceramic material can bear high temperature above 1000°C, but it can only provide small binding force and has a small thermal expansion coefficient, which eventually results in fall off of the sensor and failure of the strain measurement [14–17]. National Aeronautics and Space Administration (NASA) has used the metallic coating prepared by plasma spraying process on the surface of workpiece to improve the binding efficiency of fiber sensors. However, the device is expensive, bulky and energy-consuming, so it is inconvenient to use and has the risk of damaging the surface due to the ultra-high temperature at which it operates [15,21]. Therefore, improving the binding strength and coupling efficiency of fiber optic sensors in high temperature environment is now becoming the key to solve the problem of high temperature strain testing [22–26].
In this paper, aim to achieve non-injury, on-line and fast strain measurement under extreme high-temperature environment, we carry out a series of studies and come up with a complete set of solutions from the encapsulation of FBG sensors to the installation process and the selection of adhesives. We explore and modified the encapsulation method of FBG sensors which can achieve partial thermal isolation and avoid shear stress and creep while ensuring effective measurement. We also propose a bonding method of FBG sensor using plasma surface treatment, where by injecting ions into the surface of the alloy, new functional groups are added, and the surface infiltration property as well as the activity of the material are changed. Consequently, the adhesion and infiltration properties are improved, which strengthens the bonding strength and coupling efficiency without damaging the surface structure. Eventually, benefited from the aforementioned technological improvements and structural innovations, dynamic large strain up to 1500 µɛ measurement at ultra-high temperature over 1000°C is experimentally achieved without complex steps and high costs.
2. Method
2.1 Principle of FBG sensing and the existing problems
As shown in Fig. 1(a), FBG is a periodic modulator of optical fiber refractive index, which generates a resonant structure and reflects a specific wavelength, the so-called Bragg wavelength:
where ${n_{eff}}$ is the efficient refractive index of fiber and $\mathrm{\Lambda }$ is the modulation period, i.e., the grating constant [15]. According to Erdogan's coupling mode theory [27,28], the transmitted light and reflected light modes of FBG are solved, and the analytical expression formula of reflected light of uniform grating can be obtained with:
Fig. 1. Principle of FBG strain sensing and existing problem when applied to high-temperature conditions. (a) Basic principle of FBG sensors that in-fiber grating reflects light with specific wavelength. (b) Typical structure of FBG-based sensing system. (c) Adhesive force. ${\sigma _1}$ is the force between the object and adhesive, and ${\sigma _2}$ is the force between the FBG sensor and adhesive. (d) Radial strain. (e) Measurement failure due to creep. BBLS: broadband light source; CIR: optical circulator.
The principle of a FBG sensing system using spectrometric interrogation method based on a broadband light source (BBLS) is shown in Fig. 1(b). The FBG sensor is installed on the object to be measured with adhesive. The emitted light of the BBLS is reflected by FBG sensors through a circulator. The reflected FBG spectrum is sampled and then analyzed with follow-up algorithms, which is usually called FBG signal interrogation. FBG technology has been developed for decades and is approaching maturity, but it still faces several difficulties when the system is applied to high temperature and large strain environment. First, because the greater the vibration frequency and strain, the greater the stress, the adhesive should be able to maintain a large enough adhesive force at high temperature to transfer the large stress steadily (shown in Fig. 1(c)). Second, lateral (i.e., radial) strain will make the measurement inaccurate or destroy the FBG grating area, as shown in Fig. 1(d). Besides, quartz material becomes soft at high temperature, and creep occurs at high frequency strain, which leads to measurement failure, as shown in Fig. 1(e). In order to achieve large strain measurement in high temperature environment, the above three problems have to be solved.
2.2 Encapsulation of FBG sensor
Since FBG is thin-diameter, fragile and easily broken, it is difficult to use directly especially in extreme environments. Therefore, FBG sensors are generally protected by encapsulation. However, the conventional packaging encapsulation using metal strain gauge is no longer applicable in high temperature environment. Here, we first introduce the common encapsulations can be applied to certain occasions, and then introduce a well-designed, special encapsulation which can not only protect the sensor from high temperature but also ensure good mechanical transmission, as shown in Fig. 2.
Fig. 2. Three types of common encapsulations of FBG sensors and our designed encapsulation. (a) Apply the polyimide coated FBG directly to the surface with adhesive. (b) The FBG is protected by a small section of steel tube. (c) Fix the tube to the strain gauge with a thin groove by adhesive, and then pass the FBG through the tube. (d) The designed encapsulation of FBG sensor which can not only protect the sensor from high temperature but also ensure good mechanical transmission.
The first is to apply the polyimide coated FBG directly to the surface with adhesive, as shown in Fig. 2(a). The second is that a small section of steel tube (slightly longer than the length of the grating area) with high-temperature adhesive is glued to the surface of the to-be-tested sample, then the FBG is passed through so that the grating is suspended inside the tube, and the two ends of the FBG outside the grating are fixed to the surface with adhesive, as shown in Fig. 2(b). The third is to fix the tube to the strain gauge with a thin groove by adhesive, and then pass the FBG through the tube so that the grating is suspended inside the tube, while the fiber is fixed to the strain gauge by adhesive on both sides of the tube, as shown in Fig. 2(c).
The three encapsulation methods have different stress transfer processes. In the first encapsulation method, the strain of the measured material is transferred directly to the sensor, so the transfer efficiency is high, but shear forces can also be transferred in, which can damage the grating area and lead to inaccurate measurement. In the latter two methods, the tube plays the role of protecting the grating area and can retain only the axial tension, so they have better accuracy. The difference is that the stress transfer of the third method needs to go through the strain gauge first, which increases the stability of stress transfer, but also increases the need for adhesive adhesion, because the larger the mass of the sensing unit, the greater the need for adhesion. The multilayer strain transfer link increases the possibility of strain loss along the strain transfer in the case of large strain.
To address potential risks and problems that still exist, we designed a precise encapsulation structure of FBG as shown in Fig. 2(d). The FBG is passed through the protective casing, and the sealing glue at both ends makes the FBG and the casing become a whole. And the grating area is protected by a ceramic casing. When working, the base of stainless steel is installed on the surface of the object to be measured. The space below the casing and the substrate can be filled with thermal insulation asbestos. In addition to the strain measurement grating, a suspended temperature measurement grating is arranged in the ceramic casing for temperature compensation. The main view and the top view of the designed structure are shown in Fig. 3(a). The surface strain of the object will drag the two bases to slide, as the ${f_L}$ and ${f_L}$ shown in Fig. 3(a), and further drag the support beam and the protective casing integrated with it to move. And the schematic diagram of force transmission is shown in Fig. 3(b). The mechanical transmission loss of the structure is caused by three aspects: sliding friction between the mounting base and the surface of the structure, sliding friction between the support beam and the lower mounting hole, and sliding friction between the protection casing and the upper mounting hole. The equivalent plastic strain of the frictional force of the double rough surface in sliding contact is:
Fig. 3. Stress and strain transfer analysis. (a) The main view and the top view of the structure. (b) Schematic diagram of force transmission. (c) and (d) are the distribution of the displacement of the support and the whole structure.
The compensation principle of this loss can be vividly reflected in finite element simulation. As an example, a model of the structure is established and finite element analysis is carried out using the finite element analysis software Ansys. Based on the material properties and dimensions of the structure in actual application scenarios, displacement excitation is applied to the base of the structure, taking into account the actual working conditions, and a displacement along the FBG axis is applied to the structure with the magnitude increasing linearly with the step size while maintaining the load constraint direction unchanged. In a schematic simulation, the distribution of the displacement of the support and the whole structure are shown in Figs. 3(c) and 3(d), where the base is set as 10 mm long and a displacement of ${10^{ - 3}}$ mm is applied to both bases, resulting in a total strain of 200 µɛ. It can be seen that the displacement of the upper end of the protective sleeve is $9.9305 \times {10^{ - 7}}\; m$, the effective contact length between the protective sleeve and the FBG is 18 mm, and the average strain on the protective sleeve is 55.16 µɛ. Furthermore, since the protective sleeve is connected to the FBG sensor as a whole and the length of the sensing grating region of the FBG is 7 mm, the average strain reflected on the FBG is 141.84 µɛ. Therefore, the strain transmission loss at this time on the FBG sensor is 58.16 µɛ, and the transmission efficiency is 70.92%. Because the FBG is connected with the protective casing through the sealing glue, it is transformed into the stretch of the grating area, and the strain value can be obtained after compensating the loss caused by the mechanical transmission.
The advantages of the designed structure are as follows. 1) The FBG sensor is separated from the high temperature in the environment for the superior thermal insulation performance of ceramic and asbestos, which can prevent the damage of high temperature to the FBG. 2) The two mounting bases are flexible so it can be used in more complex applications and has a larger area to provide higher adhesion. Additionally, protection is also required for non-grating segments, we use nylon casing and armored hoses to provide double protection. Nylon casing provides flexible protection for the fiber and prevents the fiber from being damaged by irregular edge cutting when passing through the armored hose. The armored hose further provides rigid protection, which can make the fiber work normally and stably under the harsh environment of high temperature and large strain. The comparison of the performance of the sensors of each encapsulation method is shown in Table 1. The choice of which encapsulation method should be based on the actual application situation.
Table 1. Comparison of performance of the sensors of the four encapsulations.
2.3 Surface treatment using plasma
The contaminants on the surface of the object to be tested, such as superalloy, will reduce the quality of the bonding, such as uneven, bubbles, etc. And these random bonding problems are often fatal in extreme environmental measurements. In order to optimize the effective coupling efficiency between the fiber grating sensor and the surface of the sample to be tested, targeted surface treatment using plasma is performed on the surface in our work. The plasma surface treatment system is depicted in Fig. 4(a). Plasma surface treatment is an excellent surface modification method, which is widely used in surface modification of various materials. Since all substances will become plasma state at an extreme high temperature, so the plasma state is called the “fourth state” of substance. Plasma contains a large number of active particles, such as high energy electrons, ions, free radicals, excited gas atoms, molecules and photons. Plasma surface treatment causes collision, scattering, excitation, heterogeneous, defects, crystallization and amorphization by injecting particles or gases into the material surface. These particles interact with the material surface such as etching and cleaning, oxidation, activation and polymerization, so as to achieve the treatment effect of changing the surface properties of the material.
Fig. 4. Schematic diagram of surface treatment using plasma. (a) Plasma surface treatment system mainly including an air compressor, a plasma generator and a plasma nozzle. (b, c) Plasma or plasma-activated chemically active substances are used to react chemically with the material surface dirt, such as the oxidation reaction between the reactive oxygen species and the organic matter on the surface of the. (d) Highly active chemical bonds react with other substances and form stable chemical bonds, thus improving the adhesion paste effect. (e) Contact angle of the liquid on the solid surface used to reflect the degree of immersion. Solid - gas surface tension for ${\gamma _{SV}}$, liquid and gas surface tension for ${\; }{\gamma _{LV}}$, solid - liquid surface tension for ${\; }{\gamma _{SL}}$. (f) and (g) are the contact angles of the untreated and treated sample surfaces.
On the one hand, plasma can remove organic pollutants and impurities on the surface, remove dust particles and static electricity. When the plasma interacts with the surface, as shown in Fig. 4(b), plasma or plasma-activated chemically active substances are used to react chemically with the material surface dirt, such as the oxidation reaction between the reactive oxygen species in the plasma and the organic matter on the surface of the material. The organic pollutants on the surface of the material are decomposed into carbon dioxide and water. Besides, the physical effects such as the bombardment of pollutants by high-energy particles of plasma are used. For example, the surface pollutants of the object are cleaned by active argon plasma (as shown in Fig. 4(c)), and the volatile pollutants are discharged by vacuum pump due to the bombardment. On the other hand, plasma can introduce new functional groups on the material surface and form highly active chemical bonds with the treated surface. As shown in Fig. 4(d), the highly active chemical bonds are more likely to react with other substances and form stable chemical bonds, thus achieving the purpose of improving the adhesion paste effect by the increase of surface tension after plasma treatment.
To better evaluate the process parameters for implementing surface modification using plasma technology, wettability refers to the extent to which the liquid spreads over the solid surface. An effective contact between the adhesive and the object surface is achieved and conditions are created for their physicochemical bonding only if the adhesive and the object surface have excellent wettability. The degree of immersion is usually measured by the contact angle of the liquid on the solid surface, as shown in Fig. 4(e). Solid - gas surface tension for ${\gamma _{SV}}$, liquid and gas surface tension for ${\; }{\gamma _{LV}}$, solid - liquid surface tension for ${\; }{\gamma _{SL}}$, contact angle is the tangent of the solid surface and the surface of the liquid droplet, the size of its value can be expressed in Young's formula of: ${\gamma _{SV}} - {\gamma _{SL}} = {\gamma _{LV}}cos\theta $. In order to investigate the effect of plasma treatment on the surface, the contact angles of the untreated and treated sample surfaces were compared using a contact angle measuring instrument, and the results are shown in Figs. 4 f and 4 g. It can be seen that after the plasma treatment, there is a significant reduction in the contact angle and the surface wettability of the sample is significantly improved.
After the plasma treatment process, the impregnation properties and adhesion of the surface are improved. The next step of coating adhesive should be carried out immediately after the completion of surface treatment, so that the surface and the adhesive were closely integrated and the mechanical transmission were maintained, which paves the way of precision measurement of large strain under extreme high temperature environment. Compared with other dry processes such as radiation, laser, electron beam and corona treatment, the unique feature of plasma is that the depth of plasma surface treatment only involves a very thin layer of the substrate surface. According to the observation results of electron spectroscopy for chemical analysis (ESCA) and scanning electron microscope (SEM) for chemical analysis, it can be inferred that it is generally within the range of tens to thousands of angstroms (nanoscale) from the surface, so that the interface physical properties can be significantly improved without any influence on the material phase.
2.4 High-temperature adhesive and temperature compensation
The ideal high-temperature adhesive is characterized by good fluidity before curing and no cracks and other defects in the high-temperature environment after curing. The thermal expansion should match that of the sample to be tested, and the compressive strength and flexural strength should be large enough, and the curing method should be simple. In our work, three types of high-temperature are used whose parameters are compared in Table 2. Silicate, aluminosilicate and zirconia are three common high-temperature adhesives.
Table 2. Parameters and properties of different adhesives.
The bonding method should follow the following criteria. First, the adhesive, as the intermediate between the sample and the FBG, should be in good contact with the sample to ensure the bond strength and avoid falling off. Second, the adhesive needs to be applied evenly to avoid damage to the fiber grating under high temperature. And last but not least, the bonding process should be simple to minimize potential damage. In the process of strain sensing, there is a multi-stage strain transfer process: the sample to be tested → substrate of sensor → adhesive layer → FBG, and each stage causes strain loss, which reduces the strain transfer efficiency of FBG and makes the strain felt by FBG inaccurate with the real strain of the measured structure. Therefore, it is necessary to compare the strain efficiency of the three adhesives.
Since different adhesives have different modulus of elasticity, the relationship between the strain transfer efficiency and the modulus of elasticity of the adhesive layer can be calculated as shown in the Fig. 5. It can be seen from the Fig. 5 that as the elastic modulus of the adhesive increases, the strain transfer efficiency also increases. After the elastic modulus is more than 30 GPa, the growth of strain transfer efficiency tends to slow down and gradually approaches the limit. The strain transfer efficiency of the zirconia is the highest because the modulus of elasticity of the silicate is about 30 GPa, that of the aluminosilicate is about 45 GPa, and that of the zirconia is 210 GPa. Besides, the main component zirconium oxide has a fracture toughness of 5∼6 MPa. Considering all the parameters, we choose the zirconia as our adhesive. Additionally, the greater the fracture toughness, the stronger the resistance to brittle fracture. The geometrical parameters of the hand-applied production adhesive are not easily controlled, resulting in human interference to the stability and reliability of the test data, so the adhesive shape should be standardized as much as possible.
Because FBG is both temperature and strain sensitive, temperature compensation is an indispensable step when measuring strain in high temperature environment. We calibrated the temperature and strain responses of FBG sensors in detail in a laboratory environment. As shown in the Fig. 6(a), the FBG is connected to a FBG interrogator, and the grating area is placed in a high-temperature oven. The fibers at each end of the FBG are glued to the platform, which can be moved to provide tension strain. Put the grating of the FBG into the capillary tube and seal it with the probe, and fix it with high temperature adhesive. Using high temperature tape to fix the thermocouple with the capillary steel tube to ensure that the probe and the grating area are attached together, and put it into the high temperature oven for heating. After the oven stops heating, read the maximum value of the center wavelength, and then read the maximum value of the thermocouple temperature, and take five consecutive readings, and use the average value as the result.
Fig. 6. Temperature compensation setup and results. (a) Temperature compensation experimental setup. (b) Experimental results of FBG strain response and the linear fitting curve and the prediction interval. (c) Fitting errors of strain response fitting. (d) Experimental results of FBG strain response and the quadratic polynomial fitting curve and the prediction interval. (e) Fitting errors of temperature response fitting. w means the Bragg wavelength. dw means the shift of Bragg wavelength.
According to (2) and (4), the FBG strain response is close to linear relationship, so a linear fit is used, and the fitting results and fitting errors are shown in Figs. 6(b) and 6(c). It can be seen that the error of strain measurement is about 12 µɛ. And within a small range, the wavelength shift can be considered to have a linear relationship with temperature. However, at high temperatures, the relationship between wavelength and temperature is approximately quadratic. Here, using the typical 1555 nm as the base point, the quadratic polynomial fitting results and fitting errors are shown in Figs. 6(d) and 6(e). The error of measurement is less than 5°C in the range of 0–800°C, and is less than 10°C below 1000°C. The above parameters can be used according to actual needs in practical applications.
3. Experiments and results
In order to verify our high temperature measurement scheme, a series of experiments were carried out including high temperature test and large strain measurement experiments of both strong impact strain and vibration of a high-power engine. In the experiments, quartz FBG sensors that wrote directly using femtosecond laser and protected with polyimide coating layer were used. And the FBG sensors were interrogated with a 5 kHz rate commercial FBG interrogator (shown in Fig. 7(d)).
Fig. 7. Experimental setup and results of continuous heating and transient shock excitation. (a) A superalloy was continuously heated using a hand-held flamethrower. (b) A strong and quantitative transient shock excitation was applied to the surface where the sensors are mounted (c) plasma surface treatment of on the surface of a high-power fuel engine. (d) The employed commercial FBG interrogator. (e) The thermocouple used for real-time temperature monitoring. (f) A close-up of the plasma nozzle. (g) Recorded temperature curve of FBG #1 under continuous heating. (h) and (i) are the response curves of the FBGs #1 and #2 to the transient shock excitation. Experimental results of an engine test. (j) and (k) Recorded the wavelength shifts of FBGs #4 and #5 that mounted on the surface of a high-power engine. FBG #4 is encapsulated with the designed encapsulation structure.
First, three encapsulated FBG sensors, #1, #2 and #3, were mounted close together on a piece of superalloy taken from the combustion chamber of a large engine. The first two, FBGs #1 and #2, were mounted with the same zirconia high-temperature adhesive, and the difference is the surface with or without plasma treatment, and the third FBG #3 was mounted with the usual epoxy adhesive. After mounting the sensors, the superalloy was continuously heated using a hand-held flamethrower whose flame can reach a maximum temperature of 1200°C, as shown in Fig. 7(a). A thermocouple (Fig. 7(e)) is used for real-time monitoring of the temperature. Soon the FBG #3 falls off because the epoxy vaporizes at high temperatures. FBGs #1 and #2 behave similarly under heating and Fig. 7 g records the heating curve of FBG #1. The maximum error between the experimentally measured temperature and the temperature measured by the thermocouple was within 5°C in the whole process. It can be seen that the sensor maintains good performance at temperatures close to 1000°C. At the temperature, the polyimide coating will fail, but the grating will retain its function. Note that because of the hand-held heating, the heating process is not particularly linear. The experimental results show that the chose zirconia adhesive can withstand extremely high temperatures up to 1000°C.
And then, as shown in Fig. 7(b), a strong and quantitative transient shock excitation was applied to the surface where the sensors are mounted. The hammer is dropped naturally from the same height every time to ensure that the impact force given in the experiment is roughly the same. And the response curves after compensating temperature of the FBGs #1 and #2 to the excitation are shown in Figs. 7 h and 7i respectively. The instantaneous strain can be up to 1500 $\mu \varepsilon $. Due to the use of high temperature adhesives, both of them can remain intact at high temperatures and have a certain response to impact excitation. Observably, the FBG #1 with surface treating using plasma behaves more like those in the normal temperature condition. From the perspective of background noise, it can be seen that the sensor #2 without plasma surface treatment has a higher background noise, because the uneven adhesion will lead to uneven force and thus noise. In general, the experiments show that our plasma surface treatment technology can greatly improve the bonding effectiveness between the adhesive and the surface to be tested, and thus improve the measurement performance.
Also, we applied our scheme to a practical application that an encapsulated FBG sensor #4 was mounted on the surface of a high-power fuel engine with our zirconia high-temperature adhesive after plasma surface treatment (Fig. 7(c)), and the other end of the FBG sensor is connected to the interrogator through a fixed platform and finally to the PC. At the same time, another non-encapsulated FBG sensor #5 was also mounted nearby the above FBG sensor. During the experiments, once the engine is started, the strain signals as well as the thermal signals from its surface is transmitted to the FBG sensors which transform the signals into the shifts of the central wavelength and transmits them to the interrogator for demodulation and processing. The engine has a warm-up time of about 1 s and starts to vibrate violently and gradually heats up at 2 s, and the engine’s surface temperature can reach about 500°C. The recorded wavelength shifts are shown in Figs. 7(j) and 7 k respectively. It can be seen that the sensor #4 shown in Fig. 7(j) performs much better than the sensor #5 shown in Fig. 7 k. The former maintains good stability and measures high frequency strain at high temperatures, while the latter is likely to have fallen off or the sensor is broken. Due to the large strain measurement under extremely high temperature conditions, there are currently no effective non-invasive measurement instruments available. Therefore, the tests conducted so far have been exploratory and preliminary research. Besides, the theoretical model of the engine can also provide effective reference for the test data. The test data and simulation model of the engine achieves a very high degree of match, which is in line with the actual engine conditions. We believe that this is sufficient to confirm the effectiveness of our scheme.
In conclusion, the experimental results indicate that the proposed measurement scheme combining our special encapsulation and plasma surface treatment based on FBG sensors makes it possible to measure large strains at extremely high temperatures.
4. Conclusion and future work
In this work, aiming at the issue of large strain measurement under extreme high-temperature environment, a systematic fiber sensing scheme comprising of encapsulation protection, plasma surface treatment, high-temperature adhesive and temperature compensation scheme is proposed. The designed encapsulation has the advantages of protecting the grating area, partial thermal isolation and avoiding shear stress and creep. The plasma surface treatment technology is able to improve the impregnation properties and adhesion of the surface, and thus achieves a strong and effective coupling between the surface of the sample and the adhesive. It can also effectively reduce the residual stress during the temperature change process, which is one of the important factors leading to noise and grating damage. Consequently, precision measurement of large strain up to 1500 $\mu \varepsilon $ under high temperature up to 1000°C is experimentally achieved in a cost-effective way. It can be believed that our high temperature large strain measurement scheme has broad application potential in many fields, especially in aerospace, large engine, turbine and other major equipment health monitoring. Worthy to note that, limited by the properties of quartz itself, there are still improvements to be studied. For example, new FBG materials, such as sapphire, can enhance the robustness of the grating itself. In addition, sensors with hybrid structure and special encapsulation are also orientations worth further researched.
Funding
National Natural Science Foundation of China (61825501, 62205036); The Graduate Research and Innovation Foundation of Chongqing, China (CYB22061); Postdoctoral Research Foundation of China (2021M700614); Chongqing Postdoctoral Science Foundation (cstc2021jcyj-bshX0083).
Disclosures
The authors declare no conflicts of interest.
Data availability
The data within this paper and other findings of this study are available from the corresponding author upon reasonable request.
References
1. H. Zhang, Z. Du, Z. Fang, S. Wang, and X. Chen, “Sparse Decomposition Based Aero-engine's Bearing Fault Diagnosis,” J. Mech. Eng. 51(1), 97 (2015). [CrossRef]
2. X. Yang, “Data Acquisition and Process System for Temperature Field of Some aeroengine,” Computer Automated Measurement & Control (2003).
3. Y. S. Wang, X. M. Dong, W. Liu, M. G. Huang, L. I. Xin, and H. X. Cui, “Research on Developments of High Temperature Testing Technology for Aero-Engine,” Measurement & Control Technology (2017).
4. X. J. Shi and L. Q. Wang, “TEHL Analysis of Aero-engine Mainshaft Roller Bearing Based on Quasi-dynamics,” J. Mech. Eng. 52(3), 86 (2016). [CrossRef]
5. J. B. Du and Z. Y. He, “Sensitivity enhanced strain and temperature measurements based on FBG and frequency chirp magnification,” Opt. Express 21(22), 27111–27118 (2013). [CrossRef]
6. K. Guo, J. He, L. Shao, G. Xu, and Y. Wang, “Simultaneous Measurement of Strain and Temperature by a Sawtooth Stressor-Assisted Highly Birefringent Fiber Bragg Grating,” J. Lightwave Technol. 38(7), 2060–2066 (2020). [CrossRef]
7. Q. Bian, C. Bauer, A. Stadler, M. Lindner, M. Jakobi, W. Volk, A. W. Koch, and J. Roths, “In-Situ High Temperature and Large Strain Monitoring During a Copper Casting Process Based on Regenerated Fiber Bragg Grating Sensors,” J. Lightwave Technol. 39(20), 6660–6669 (2021). [CrossRef]
8. F. H. Zeng, Z. Y. Li, X. Gui, X. L. Fu, and H. H. Wang, “Wear measurement based on the length variation of a sacrificial FBG,” Opt. Express 28(16), 23189–23197 (2020). [CrossRef]
9. C. E. Campanella, A. Giorgini, S. Avino, P. Malara, R. Zullo, G. Gagliardi, and P. De Natale, “Localized strain sensing with fiber Bragg-grating ring cavities,” Opt. Express 21(24), 29435–29441 (2013). [CrossRef]
10. H. T. Zhao and Y. Qiu, “Strain transfer of surface-bonded fiber Bragg grating sensors for airship envelope structural health monitoring,” J. Zhejiang Univ. Sci. A 13(7), 538–545 (2012). [CrossRef]
11. B. A. Tahir, J. A. Saktioto, M. Fadhali, R. A. Rahman, and A. Ahmed, “A study of FBG sensor and electrical strain gauge for strain measurements,” J. Optoelectron. Adv. Mater. 10(10), 2564–2568 (2008).
12. H. E. R. Shiuh-Chuan and T. S. A. I. Chang-Yu, “Strain measurement of fiber optic sensor surface bonding on host material,” Trans. Nonferrous Metals Soc. China 19(S1), s143–s149 (2009). [CrossRef]
13. I. Laarossi, P. Roldán-Varona, M. A. Quintela-Incera, L. Rodríguez-Cobo, and J. M. López-Higuera, “Ultrahigh temperature and strain hybrid integrated sensor system based on Raman and femtosecond FBG inscription in a multimode gold-coated fiber,” Opt. Express 27(26), 37122–37130 (2019). [CrossRef]
14. Y. J. Rao, “In-fibre Bragg grating sensors,” Meas. Sci. Technol. 8(4), 355–375 (1997). [CrossRef]
15. A. D. Kersey and M. A. Davis, “Fiber grating sensors,” J. Lightwave Technol. 15(8), 1442–1463 (1997). [CrossRef]
16. R. Kashyap, “Theory of Fiber Bragg Gratings - ScienceDirect,” Fiber Bragg Gratings 58(4), 119–193 (1999). [CrossRef]
17. J. Wang, T. Huang, F. Duan, Q. Cheng, F. Zhang, and X. Qu, “Fast peak-tracking method for FBG reflection spectrum and nonlinear error compensation,” Opt. Lett. 45(2), 451–454 (2020). [CrossRef]
18. M. S. Ferreira, P. Roriz, J. Bierlich, J. Kobelke, K. Wondraczek, C. Aichele, K. Schuster, J. L. Santos, and O. Frazão, “Fabry-Perot cavity based on silica tube for strain sensing at high temperatures,” Opt. Express 23(12), 16063–16070 (2015). [CrossRef]
19. T. Habisreuther, T. Elsmann, A. Graf, and M. A. Schmidt, “High-temperature strain sensing using sapphire fibers with inscribedfirst-order Bragg gratings,” IEEE Photonics J. 8(3), 1–8 (2016). [CrossRef]
20. C. M. Petrie, N. Sridharan, A. Hehr, M. Norfolk, and J. Sheridan, “High-temperature strain monitoring of stainless steel using fiber optics embedded in ultrasonically consolidated nickel layers,” Smart Mater. Struct. 28(8), 085041 (2019). [CrossRef]
21. Y. A. Abdel-Aziz, A. El-Hameed, M. I. Ismail, L. A. Hashim, and A. Gregorio, “Effects of Space Plasma on an Oxide Coating of Spacecraft's Surface Materials,” Adv. Space Res. 68(3), 1601–1612 (2021). [CrossRef]
22. J. Marsden, I. Herszberg, and H. C. Li, “Durability of Surface Mounted Optical Fibres for Use in Structural Health Monitoring,” in Engineers Australia (2007).
23. A. Piazza, L. W. Richards, and L. D. Hudson, “High-Temperature Strain Sensing for Aerospace Applications,” in WRSGC Summer Test and Measurements Conference (2008).
24. S. H. Meng, C. Du, W. H. Xie, S. Y. Huo, and L. Y. Song, “Application of high-temperature optical fiber sensor in temperature and strain testing of hot structure,” J. Solid Rocket Technology (2013).
25. P. Zeng, Y. Wang, F. Chen, Y. Sun, Q. Zhang, X. Huang, and Z. You, “Analysis of the Effect of Adhesives on Strain Transfer for Surface Bonded Polyimide Fiber Bragg Grating,” Chin. J. Sens. Actuators 32(1), 43–49 (2019). [CrossRef]
26. Z. Wei, W. Chen, Y. Shu, J. Wu, and X. Lei, “Degradation of sensing properties of fiber Bragg grating strain sensors in fatigue process of bonding layers,” Opt. Eng. 53(4), 046102 (2014). [CrossRef]
27. J. Skaar and M. K. Risvik, “A genetic algorithm for the inverse problem in synthesis of fiber gratings,” J. Lightwave Technol. 16(10), 1928–1932 (1998). [CrossRef]
28. J. Skaar, L. Wang, and T. Erdogan, “On the synthesis of fiber Bragg gratings by layer peeling,” IEEE J. Quantum Electron. 37(2), 165–173 (2001). [CrossRef]
