Optical Fibre, New Developments Part 2 docx

Opticalbresinaeronautics,roboticsandcivilengineering 29 4. Application in robotics The number of application domains of robotic systems is rapidly growing and in particular the service robotics is becoming the most popular. In such application field a high degree of autonomy is required for the robot and thus a large number of exteroceptive sensors appear necessary. When multifingered robotic hands are considered, the requirement of minimally invasiveness for the sensory system is of major importance due to the limited space available in a mechanical structure with several degrees of freedom. In a robotic hand, different exteroceptive sensors are required to ensure stable grasping and manipulation of objects. Among these, sensing of both contact point and contact force appears mandatory for any control algorithm which intends to achieve such goals. Even though many different technologies have been explored and tested to build tactile sensors, like piezo-resistive (Liu et al., 1993), capacitive (Morimura et al., 2000), piezoelectric (Krishna & Rajanna, 2002), magneto-resistive (Tanie, 1986), optoelectronic approaches demonstrated their potential since the beginning of tactile sensors development (Maekawa et al., 1993). Also, on the market optoelectronic tactile sensors can be found that measure distributed tactile information, but such tactile information is generally limited to pressure force, and spatial resolution is coarse, a few millimetres order. Generally, a commercial sensor accurately responds to a load of 0.25 N or more up to 2 N, but such a range can be too narrow for manipulation tasks. More recently, a number of different optical approaches have been pursued, among which the solution based on an LEDs matrix has been presented in (Rossiter & Mukai, 2005) and the solution based on a CCD camera is reported in (Ohka et al., 2006). Among optical approaches, those based on the use of optical fibres appear particularly suitable for pressure sensing, thanks to the low size and minimum invasivity of fibres themselves. Since the advent of fibre optics, it has been recognized that optical fibres can be used as effective pressure (and tactile) sensors. One of the earliest demonstrations of such a capability relied on the pressure-induced displacement of a diaphragm placed close to the tip of an optical fibre (Cook & Hamm, 1979). The fibre was operated in reflection mode, so that changes in reflected intensity can be used as a measure of the pressure applied on the diaphragm. In case of tactile sensing, such an approach presents the disadvantage of requiring a complex micromachining at the tip of the fibre. Another possible approach is based on the intensity loss resulting from pressure-induced bending of the fibre (Fields et al., 1980). However, in this case the response of the sensor is highly nonlinear due to the exponential dependence of the bending loss on the radius of curvature of the fibre. More complex examples can be found based on interferometric approaches, where the changes in the optical phase are used as transducer mechanism to sense the pressure (Saran et al., 2006, Wang et al., 2001, Yuan et al., 2005). Interferometric sensors exhibit high sensitivity, but also present some disadvantages, such as low tolerance to external disturbances, and periodicity in their response. Fig. 13. Sketch of the single sensing element (taxel). Recently, we proposed a solution based on the scattering of the light illuminating the surface of urethane foam (De Maria et al., 2008). The configuration makes use of a couple of emitter/receiver fibres placed at the edge of a micromachined well covered by the foam. The distance between the two fibres can be chosen in order to ensure a desired sensitivity of the sensing element. As a demonstration of the effectiveness of the proposed configuration, we present the results of two sensors, in which the relative distance between the two fibres was properly selected in order to fit the range of pressures to be detected. Top and lateral schematic views of a single taxel are shown in Fig. 13. The sensor works as follows: the light emitted by the illuminating fibre is scattered by the internal surface of the urethane foam and a fraction of its power is collected by the receiving fibre, depending on the applied pressure. In particular, when one applies a pressure on the external surface of the urethane foam, the distance between the tip of the collecting fibre and the internal surface of the foam is reduced, and this will result in an increased fraction of power collected by the receiving fibre. The use of a scattering surface, such as that of the urethane foam employed for the realization of the prototypes, is justified by the fact the multiple scattering permits to smooth (average-out) local variation of light intensity within the cavity, and thus reduce the sensitivity of the collected power on micro-displacements of the illuminating and/or receiving fibre. As the power collected by the receiving fibre is a function of the pressure applied on the foam surface, it can be used as a measure of the applied force. Obviously, the collected light is also a function of the relative distance between the illuminating and the receiving fibres. In our experiments, such a distance was kept constant and was not a function of the applied force. However, we can exploit such dependence, by choosing an opportune distance giving rise to a desired sensitivity of the sensor on the applied pressure. Generally speaking, a smaller distance will result in a higher sensitivity, so that smaller pressures can be measured. On the other hand, a higher sensitivity implies a reduced dynamic range, i.e. the sensor response will saturate at lower pressure levels. Hence, a trade-off must be found between sensitivity and dynamic range. One advantage of the proposed technique is that it can be easily extended to a number of taxels, so as to acquire a pressure distribution. Figure 14 shows a possible configuration of a matrix of taxels to realize a complete tactile sensor able to detect both contact point and contact force applied on a finite area. Two different taxels have been produced with the same well and two different distances between the emitting/receiving fibres, i.e. 10 m and 200 m. The micromachined well size is 5x5 mm 2 . The optical source was a superluminescent LED operating at a central wavelength of 1550nm, and having an output optical power of 3mW. The output pigtail of Illuminatin g Receivin g Scatterin g Urethane A pplied force Scatterin g OpticalFibre,NewDevelopments30 the source was connected to the illuminating fibre, whereas the receiving fibre was connected to an InGaAs photodiode, whose output signal was fed to an oscilloscope having an input impedance of 1 M. Both illuminating and receiving fibres were SMF-28, single- mode optical fibres. The two prototypes have been calibrated with a load cell mounted as shown in Fig. 15. The results corresponding to the calibration of the first prototype are reported in Fig. 16 (left), where the output voltage, proportional to the optical power collected by the receiving fibre, is plotted against the load applied to the sensor. As expected, the sensitivity is very high but with a limited dynamic range. Moreover, to test the repeatability of the measurements, different sets of measurements have been collected and two of them are reported in the figure. Fig. 14. Schematic diagram of a 10-taxel tactile sensor. Fig. 15. Experimental set-up for the fibre-optics based taxel. The second prototype, as expected, had a lower sensitivity but wider dynamic range, as shown by the calibration curve of Fig. 16 (right). In both cases, the sensitivity is certainly better than the typical values of commercial optical tactile sensors. Load cell for calibration Illuminating fibre Receiving fibre Tactile sensor Illuminating fibre Receiving fibres 5. Conclusions In this chapter, a number of experimental demonstrations on the use of the optical fibre sensor technology have been reported. It has been shown that different application fields can take advantage of the peculiar characteristics of optical fibre sensors. In particular, distributed fibre sensors have great potentiality in the field of structural health monitoring, as they permit to perform continuous measurements of the quantity of interest. On the other hand, fibre Bragg grating technology offers high sensitivity and accuracy, and in general it benefits from the immunity to electromagnetic interference, in common with other fibre- optic sensors. Finally, the small size and minimally invasiveness of optical fibres have been demonstrated to be useful in robotic applications, where the use of fibre-optics may lead to efficient exteroceptive sensing systems. Fig. 16. Calibration curves of the first prototype with 10 m distance between the fibres (left) and of the second prototype with 200 m distance between the fibres (right). 6. References Agrawal, G.P. (2001). Nonlinear Fibre Optics. Academic Press, San Diego. Barnoski, J.K. & Jensen, S.M. (1976). Fibre waveguides: A novel technique for investigating attenuation characteristics. Appl. Opt. Vol. 15, No. 9, 2112-2115. Bernini, R.; Crocco, L.; Minardo, A.; Soldovieri, F. & Zeni, L. (2002). Frequency-domain approach to distributed fibre-optic Brillouin sensing. Opt. Lett. Vol. 27, No. 5, 288- 290. Bernini, R.; Fraldi, M.; Minardo, A.; Minatolo, V.; Carannante, F.; Nunziante, L. & Zeni, L. (2006a). Identification of defects and strain error estimation for bending steel beams using time domain Brillouin distributed fibre sensors. Smart Materials and Structures Vol. 2, 612-622. Bernini, R.; Minardo, A. & Zeni, L. (2006b). An accurate high resolution technique for distributed sensing based on frequency domain Brillouin scattering. IEEE Photonics Technology Letters Vol. 18, No. 1, 280-282. 140 160 180 200 220 240 260 280 300 0 50 100 150 200 250 300 350 Weight [g] Output Voltage [mV] Measured Cubic fitting Opticalbresinaeronautics,roboticsandcivilengineering 31 the source was connected to the illuminating fibre, whereas the receiving fibre was connected to an InGaAs photodiode, whose output signal was fed to an oscilloscope having an input impedance of 1 M. Both illuminating and receiving fibres were SMF-28, single- mode optical fibres. The two prototypes have been calibrated with a load cell mounted as shown in Fig. 15. The results corresponding to the calibration of the first prototype are reported in Fig. 16 (left), where the output voltage, proportional to the optical power collected by the receiving fibre, is plotted against the load applied to the sensor. As expected, the sensitivity is very high but with a limited dynamic range. Moreover, to test the repeatability of the measurements, different sets of measurements have been collected and two of them are reported in the figure. Fig. 14. Schematic diagram of a 10-taxel tactile sensor. Fig. 15. Experimental set-up for the fibre-optics based taxel. The second prototype, as expected, had a lower sensitivity but wider dynamic range, as shown by the calibration curve of Fig. 16 (right). In both cases, the sensitivity is certainly better than the typical values of commercial optical tactile sensors. Load cell for calibration Illuminating fibre Receiving fibre Tactile sensor Illuminating fibre Receiving fibres 5. Conclusions In this chapter, a number of experimental demonstrations on the use of the optical fibre sensor technology have been reported. It has been shown that different application fields can take advantage of the peculiar characteristics of optical fibre sensors. In particular, distributed fibre sensors have great potentiality in the field of structural health monitoring, as they permit to perform continuous measurements of the quantity of interest. On the other hand, fibre Bragg grating technology offers high sensitivity and accuracy, and in general it benefits from the immunity to electromagnetic interference, in common with other fibre- optic sensors. Finally, the small size and minimally invasiveness of optical fibres have been demonstrated to be useful in robotic applications, where the use of fibre-optics may lead to efficient exteroceptive sensing systems. Fig. 16. Calibration curves of the first prototype with 10 m distance between the fibres (left) and of the second prototype with 200 m distance between the fibres (right). 6. References Agrawal, G.P. (2001). Nonlinear Fibre Optics. Academic Press, San Diego. Barnoski, J.K. & Jensen, S.M. (1976). Fibre waveguides: A novel technique for investigating attenuation characteristics. Appl. Opt. Vol. 15, No. 9, 2112-2115. Bernini, R.; Crocco, L.; Minardo, A.; Soldovieri, F. & Zeni, L. (2002). Frequency-domain approach to distributed fibre-optic Brillouin sensing. Opt. Lett. Vol. 27, No. 5, 288- 290. Bernini, R.; Fraldi, M.; Minardo, A.; Minatolo, V.; Carannante, F.; Nunziante, L. & Zeni, L. (2006a). Identification of defects and strain error estimation for bending steel beams using time domain Brillouin distributed fibre sensors. Smart Materials and Structures Vol. 2, 612-622. Bernini, R.; Minardo, A. & Zeni, L. (2006b). An accurate high resolution technique for distributed sensing based on frequency domain Brillouin scattering. IEEE Photonics Technology Letters Vol. 18, No. 1, 280-282. 140 160 180 200 220 240 260 280 300 0 50 100 150 200 250 300 350 Weight [g] Output Voltage [mV] Measured Cubic fitting OpticalFibre,NewDevelopments32 Bernini, R.; Minardo, A.& Zeni, L. (2008). Vectorial dislocation monitoring of pipelines by use of Brillouin-based fibre-optics sensors. Smart Materials and. Structures. Vol. 17, 015006. Cavallo, A.; May, C.; Minardo, A.; Natale, C.; Pagliarulo, P. & Pirozzi, P. (2009). Modelling and control of a smart auxiliary mass damper equipped with a Bragg grating for active vibration control, Sensors and Actuators A, in press. Cook, R.O. & Hamm, C.W. (1979). Fibre optic lever displacement transducer. Appl. Opt. Vol. 18, No. 19, 3230-3241. Culshaw, B. & Dakin, J. 1997. Optical Fibre sensors Vol. 4. Artech House Publishers, 0890069409. De Maria, G.; Minardo, A.; Natale, C.; Pirozzi, S. & Zeni, L. (2008). Optoelectronic Tactile Sensor Based on Micromachined Scattering Wells. FIRST MEDITERRANEAN PHOTONICS CONFERENCE, European Optical Society Topical Meeting, 25–28 June 2008, Ischia, Italy. Fields, J.N.; Asawa, C.K.; Ramer, O.G. & Barnowski, M.K. (1980). Fibre Optic Pressure Sensor. J. Acoust. Soc. Am., Vol. 67, 816-818. Garus, D.; Krebber, K.; Schliep, F. & Gogolla, T. (1996). Distributed sensing technique based on Brillouin optical-fibre frequency-domain analysis. Opt. Lett., Vol. 21, No. 17, 1402-1404. Kersey, A. D.; Davis, M. A.; Patrick, H. J.; LeBlanc, M.; Poo, K. P.; Askins, A.G.; Putnam, M. A. & Friebele, E. J. (1997). Fibre grating sensors, Journal of Lightw. Technol., vol. 15, no. 8, pp. 1442-1462. Krishna, G.M. & Rajanna, K. (2002). Tactile sensor based on piezoelectric resonance. Proc. of 2002 IEEE Conference on Sensor, pp. 1643- 1647. Liu, L.; Zheng, X. & Li, Z. (1993). An array tactile sensor with piezoresistive with single crystal silicon diaphragm. Sensors and Actuators-A32, 193-196. Maekawa, H.; Tanie, K. & Komoriya, K. (1993). A finger-shaped tactile sensor using an optical waveguide. Proc. of 1993 IEEE International Conference on Systems, Man and Cybernetics, pp. 403-408. Measures, R.M. (2002). Structural monitoring with fibre optic technology. Academic press, San Diego. Morimura, H.; Shigematsu, S. & Machinda, K. (2000). A novel sensor cell architecture and sensing circuit scheme for capacitive fingerprint sensors. IEEE Journal of Solid State Circuits, Vol, 35, 724-731. May, C.; Pagliarulo, P. & Janocha, H. (2006). Optimisation of a magnetostrictive auxiliary mass damper. Proc. 10th International Conference on New Actuators ACTUATOR2006, Bremen, Germany, pp. 344–348. Nikles, M.; Thevenaz, L. & Robert, P.A. 1997. Brillouin gain spectrum characterization in single-mode optical fibres. J. Lightw. Technol., Vol. 15, No. 10, 1842 – 1851. Ohka, M.; Kobayashi, H.; Takata, J. & Mitsuya, Y. (2006). Sensing Precision of an Optical Three-axis Tactile Sensor for a Robotic Finger. Proc. Of the 15th IEEE International Symposium on Robot and Human Interactive Communication, pp. 214-219. Rossiter, J. & Mukai, T. (2005). A novel tactile sensor using a matrix of LEDs operating in both photoemitter and photodetector modes. Proc. of 2005 IEEE Conference on Sensor, pp. 994-997. Saran, A.; Abeysinghe, D.C. & Boyd, J.T. (2006. Microelectromechanical system pressure sensor integrated onto optical fibre by anodic bonding. Appl. Opt., Vol. 45, 1737- 1742. Tanie, K. (1986). Advances in tactile sensors for robotics. Proc. of the 6th Sensor Symposium Japan, pp. 63-68. Udd. E. (2002). Overview of fibre optic sensors, In: Fibre Optic Sensors. Francis T. S. Yu; Shizhuo Yin, pp. 1-40, Routledge, 978-0-203-90946-1, USA. Wang, A.; Xiao, H.; Wang, J.; Wang, Z.; Zhao, W. & May, R.G. (2001). Self-calibrated interferometric-based-optical fibre sensors. J. Lightw. Technol., Vol. 19, No. 10, 1495- 1501. Yuan, S.; Ansari, F.; Liu, X. & Zhao, Y. (2005). Optical fibre based dynamic pressure sensor for WIM sensor. Sens. and Actuat. A, Vol. 120, No. 1, 53-58. Zhao, Y. & Liao, Y. (2004) “Discrimination methods and demodulation techniques for fibre Bragg grating sensors”, Opt. Lasers Eng., vol. 41, pp. 1-18. Opticalbresinaeronautics,roboticsandcivilengineering 33 Bernini, R.; Minardo, A.& Zeni, L. (2008). Vectorial dislocation monitoring of pipelines by use of Brillouin-based fibre-optics sensors. Smart Materials and. Structures. Vol. 17, 015006. Cavallo, A.; May, C.; Minardo, A.; Natale, C.; Pagliarulo, P. & Pirozzi, P. (2009). Modelling and control of a smart auxiliary mass damper equipped with a Bragg grating for active vibration control, Sensors and Actuators A, in press. Cook, R.O. & Hamm, C.W. (1979). Fibre optic lever displacement transducer. Appl. Opt. Vol. 18, No. 19, 3230-3241. Culshaw, B. & Dakin, J. 1997. Optical Fibre sensors Vol. 4. Artech House Publishers, 0890069409. De Maria, G.; Minardo, A.; Natale, C.; Pirozzi, S. & Zeni, L. (2008). Optoelectronic Tactile Sensor Based on Micromachined Scattering Wells. FIRST MEDITERRANEAN PHOTONICS CONFERENCE, European Optical Society Topical Meeting, 25–28 June 2008, Ischia, Italy. Fields, J.N.; Asawa, C.K.; Ramer, O.G. & Barnowski, M.K. (1980). Fibre Optic Pressure Sensor. J. Acoust. Soc. Am., Vol. 67, 816-818. Garus, D.; Krebber, K.; Schliep, F. & Gogolla, T. (1996). Distributed sensing technique based on Brillouin optical-fibre frequency-domain analysis. Opt. Lett., Vol. 21, No. 17, 1402-1404. Kersey, A. D.; Davis, M. A.; Patrick, H. J.; LeBlanc, M.; Poo, K. P.; Askins, A.G.; Putnam, M. A. & Friebele, E. J. (1997). Fibre grating sensors, Journal of Lightw. Technol., vol. 15, no. 8, pp. 1442-1462. Krishna, G.M. & Rajanna, K. (2002). Tactile sensor based on piezoelectric resonance. Proc. of 2002 IEEE Conference on Sensor, pp. 1643- 1647. Liu, L.; Zheng, X. & Li, Z. (1993). An array tactile sensor with piezoresistive with single crystal silicon diaphragm. Sensors and Actuators-A32, 193-196. Maekawa, H.; Tanie, K. & Komoriya, K. (1993). A finger-shaped tactile sensor using an optical waveguide. Proc. of 1993 IEEE International Conference on Systems, Man and Cybernetics, pp. 403-408. Measures, R.M. (2002). Structural monitoring with fibre optic technology. Academic press, San Diego. Morimura, H.; Shigematsu, S. & Machinda, K. (2000). A novel sensor cell architecture and sensing circuit scheme for capacitive fingerprint sensors. IEEE Journal of Solid State Circuits, Vol, 35, 724-731. May, C.; Pagliarulo, P. & Janocha, H. (2006). Optimisation of a magnetostrictive auxiliary mass damper. Proc. 10th International Conference on New Actuators ACTUATOR2006, Bremen, Germany, pp. 344–348. Nikles, M.; Thevenaz, L. & Robert, P.A. 1997. Brillouin gain spectrum characterization in single-mode optical fibres. J. Lightw. Technol., Vol. 15, No. 10, 1842 – 1851. Ohka, M.; Kobayashi, H.; Takata, J. & Mitsuya, Y. (2006). Sensing Precision of an Optical Three-axis Tactile Sensor for a Robotic Finger. Proc. Of the 15th IEEE International Symposium on Robot and Human Interactive Communication, pp. 214-219. Rossiter, J. & Mukai, T. (2005). A novel tactile sensor using a matrix of LEDs operating in both photoemitter and photodetector modes. Proc. of 2005 IEEE Conference on Sensor, pp. 994-997. Saran, A.; Abeysinghe, D.C. & Boyd, J.T. (2006. Microelectromechanical system pressure sensor integrated onto optical fibre by anodic bonding. Appl. Opt., Vol. 45, 1737- 1742. Tanie, K. (1986). Advances in tactile sensors for robotics. Proc. of the 6th Sensor Symposium Japan, pp. 63-68. Udd. E. (2002). Overview of fibre optic sensors, In: Fibre Optic Sensors. Francis T. S. Yu; Shizhuo Yin, pp. 1-40, Routledge, 978-0-203-90946-1, USA. Wang, A.; Xiao, H.; Wang, J.; Wang, Z.; Zhao, W. & May, R.G. (2001). Self-calibrated interferometric-based-optical fibre sensors. J. Lightw. Technol., Vol. 19, No. 10, 1495- 1501. Yuan, S.; Ansari, F.; Liu, X. & Zhao, Y. (2005). Optical fibre based dynamic pressure sensor for WIM sensor. Sens. and Actuat. A, Vol. 120, No. 1, 53-58. Zhao, Y. & Liao, Y. (2004) “Discrimination methods and demodulation techniques for fibre Bragg grating sensors”, Opt. Lasers Eng., vol. 41, pp. 1-18. OpticalFibre,NewDevelopments34 OpticalFibreSensorSystemforMultipointCorrosionDetection 35 OpticalFibreSensorSystemforMultipointCorrosionDetection JoaquimF.Martins-FilhoandEduardoFontana X Optical Fibre Sensor System for Multipoint Corrosion Detection Joaquim F. Martins-Filho and Eduardo Fontana Department of Electronics and Systems, Federal University of Pernambuco Brazil 1. Introduction Over the past thirty years there has been intense research and development on optical fibre sensors for many applications, basically because of their advantages over other technologies, such as immunity to electromagnetic interference, lightweight, small size, high sensitivity, large bandwidth, and ease in signal light transmission. The applications include sensing temperature, strain, pressure, current/voltage, chemical/gas, displacement, and biological processes among others. To accomplish those, different optical technologies have been employed such as fibre grating, interferometry, light scattering and reflectometry, Faraday rotation, luminescence and others. A review on fibre sensors can be found in (Lee, 2003). Corrosion and its effects have a profound impact on the infrastructure and equipment of countries worldwide. This impact is manifested in significant maintenance, repair, and replacement efforts; reduced access, availability and production; poor performance; high environmental risks; and unsafe conditions associated with facilities and equipment. There have been some efforts from different countries to estimate the cost of corrosion and the results indicate that it can reach 2 to 5% of the gross national product. For example, corrosion damage represented an estimated cost of US$ 276 billions in the United States of America in 2002 (Thompson et al., 2005). Therefore, corrosion monitoring is an important aspect of modern infrastructure in industry sectors such as mining, aircraft, shipping, oilfields, as well as in military and civil facilities. Optical fibre-based corrosion sensors have been investigated in recent years mainly because of the advantages obtained by the use of optical fibres, as already pointed out. A short review of the technologies employed in the fibre-based corrosion sensors can be found in (Wade et al., 2008). The reported applications include corrosion monitoring in aircrafts (Benounis & Jaffrezic-Renault, 2004), in the concrete of roadways and bridges (Fuhr & Huston, 1998) and in oilfields. 2. Corrosion Monitoring in Deepwater Oilfield Pipelines In the oil industry, to which we focus the sensing approach described in this chapter, a very challenging problem is that related to surveillance and maintenance of deepwater oilfield pipelines, given the harsh environment to be monitored and the long distances involved. 3 OpticalFibre,NewDevelopments36 These structures are subject to corrosion and sand-induced erosion in a high pressure, high temperature environment. Moreover, the long distances (kilometres) between the corrosion points and the monitoring location make the commercially available instruments not appropriate for monitoring these pipelines. Costly, regularly scheduled, preventive maintenance is then required (Staveley, 2004; Yin et al., 2000). Electronic and electromagnetic-based corrosion sensors (Yin et al., 2000; Vaskivsky et al., 2001; Andrade Lima et al., 2001) are also not suitable in these conditions. Fibre optic based corrosion sensors are ideal for this application. However, the sensing approaches reported in the literature are either single point (Qiao et al., 2006; Wade et al., 2008) or use a stripped cladding fibre structure that requires a high precision mechanical positioning system with moving parts for light detection, which compromises the robustness of the sensor system (Benounis et al., 2003; Benounis & Jaffrezic-Renault, 2004; Saying et al., 2006; Cardenas- Valencia et al., 2007). An optical fibre PH sensor has been recently developed for the indirect evaluation of the corrosion process in petroleum wells (Da Silva Jr. et al., 2007). It employs a fibre Bragg grating mechanically coupled to a PH-sensitive hydrogel, which changes its volume according to the PH of the medium. Thus, the change in PH is translated into a mechanical strain on the Bragg grating, which can be interrogated by standard optical methods. Although it can easily be multiplexed for multipoint measurements, this technique is limited to the evaluation of the chemical corrosion due to acid attack inside the well, disregarding the combined effects of other important sources of corrosion, such as mechanical (erosion), chemical, thermic and biological (microorganisms). The oil industry can also make use of the time domain reflectometry (TDR) technique to evaluate the corrosion process inside pipelines and oil wells (Kohl, 2000). The proposed scheme involves the deployment of a metallic cable inside and along the pipeline or well. The conductor is exposed to the fluid at selected locations such that it should be susceptible to the same corrosive processes as the pipeline. A signal generator launches a pulsed electrical signal to the conductor cable and an electronic receiver measures the reflected pulses intensity and delay. The reflections come from the locations where the exposed cable was affected by the corrosion process, which changes its original impedance. This TDR technique has also been applied to the monitoring of corrosion in steel cables of bridges (Liu et al., 2002). Although this technique has the advantage of being multipoint or even distributed, it is limited in reach. For practical purposes the maximum distance covered by the sensor is about 2 km. This is suitable for standard wells, but not for deep oilfields, especially those from the recently discovered presalt regions in Brazil, which are over 6 km deep. 3. A Multipoint Fibre Optic Corrosion Sensor We have recently presented for the first time the concept and first experimental results of a fibre-optic-based corrosion sensor using the optical time domain reflectometry (OTDR) technique as the interrogation method (Martins-Filho et al., 2007; Martins-Filho et al., 2008). Our proposed sensor system is multipoint, self-referenced, has no moving parts and can detect the corrosion rate several kilometres away from the OTDR equipment. These features make it very suitable to the problem of corrosion monitoring of deepwater pipelines in the oil industry. It should be pointed out, however, that the approach is not limited to this specific application and can be employed to address a number of single or multipoint corrosion detection problems in other industrial sectors. In this chapter we present a detailed description of the sensor system, further experimental results and theoretical calculations for the measurement of the corrosion rate of aluminium films in controlled laboratory conditions and also for the evaluation of the maximum number of sensor heads the system supports. 3.1 Sensor Setup Our proposed sensor system consists of several sensor heads connected to a commercial OTDR equipment by a single-mode optical fibre and fibre couplers. Figure 1 shows the corrosion sensor setup. The OTDR is connected to a 2 km long single mode optical fibre. Directional couplers can split the optical signal such that a small fraction (3 to 9%) is directed to the sensing heads. The OTDR operates at 1.55 m, with a pulsewidth of 10 ns, which corresponds to a spatial resolution of 2 m. The OTDR is set to measure 50000 points for the total distance of 5 km (one point every 10 cm). The optical fibres and couplers are standard telecommunication devices. The sensor heads have 100 nm of aluminium deposited on cleaved fibre facets by a standard thermal evaporation process and they are numbered from 1 to 11 in Fig. 1. Fig. 1. Schematic diagram of the corrosion sensor. Sensor heads are numbered. Fibre lengths and split ratios are shown. 3.2 Results For laboratory measurements the corrosion action was simulated by controlled etching of the Aluminium film on the sensor head. We used 25 H 3 PO 4 : 1 HNO 3 : 5 CH 3 COOH as the Al-etcher. The expected corrosion rate of Al from this etcher is 50 nm/min. Figure 2-a shows the OTDR trace where each peak, numbered from 1 to 11, indicates the reflection from the corresponding sensing head. The head number 6 is immersed in the Al-etcher. As the aluminium is being removed from the fibre facet the reflected light measured in the OTDR decreases, as shown in Fig. 2-b. In Fig. 3 we plot the ratio of peak (point A) to valley (point B) of the reflected light shown in Fig. 2-b as a function of the aluminium corrosion time. Figure 3 shows that up to 60 seconds of corrosion there is no significant change in the OTDR measured reflected light, since the aluminium is still too thick. Further up from this point the reflection drops to a minimum and then stabilizes at a constant level. The constant level means that the corrosion process on the fibre facet has ended. We obtain the corrosion rate by taking the deposited metal thickness and the time taken to reach the constant level, as show in Fig. 3. 40m40m40m40m40m40m 40m40m40m40m 10m10m10m10m10m10m 10m10m10m10m 4 96 5 95 7 93 7 93 6 94 6 94 5 95 9 91 8 92 3 97 1 2 3 4 5 6 7 8 9 10 OTDR 11 Optical Fiber 2km OpticalFibreSensorSystemforMultipointCorrosionDetection 37 These structures are subject to corrosion and sand-induced erosion in a high pressure, high temperature environment. Moreover, the long distances (kilometres) between the corrosion points and the monitoring location make the commercially available instruments not appropriate for monitoring these pipelines. Costly, regularly scheduled, preventive maintenance is then required (Staveley, 2004; Yin et al., 2000). Electronic and electromagnetic-based corrosion sensors (Yin et al., 2000; Vaskivsky et al., 2001; Andrade Lima et al., 2001) are also not suitable in these conditions. Fibre optic based corrosion sensors are ideal for this application. However, the sensing approaches reported in the literature are either single point (Qiao et al., 2006; Wade et al., 2008) or use a stripped cladding fibre structure that requires a high precision mechanical positioning system with moving parts for light detection, which compromises the robustness of the sensor system (Benounis et al., 2003; Benounis & Jaffrezic-Renault, 2004; Saying et al., 2006; Cardenas- Valencia et al., 2007). An optical fibre PH sensor has been recently developed for the indirect evaluation of the corrosion process in petroleum wells (Da Silva Jr. et al., 2007). It employs a fibre Bragg grating mechanically coupled to a PH-sensitive hydrogel, which changes its volume according to the PH of the medium. Thus, the change in PH is translated into a mechanical strain on the Bragg grating, which can be interrogated by standard optical methods. Although it can easily be multiplexed for multipoint measurements, this technique is limited to the evaluation of the chemical corrosion due to acid attack inside the well, disregarding the combined effects of other important sources of corrosion, such as mechanical (erosion), chemical, thermic and biological (microorganisms). The oil industry can also make use of the time domain reflectometry (TDR) technique to evaluate the corrosion process inside pipelines and oil wells (Kohl, 2000). The proposed scheme involves the deployment of a metallic cable inside and along the pipeline or well. The conductor is exposed to the fluid at selected locations such that it should be susceptible to the same corrosive processes as the pipeline. A signal generator launches a pulsed electrical signal to the conductor cable and an electronic receiver measures the reflected pulses intensity and delay. The reflections come from the locations where the exposed cable was affected by the corrosion process, which changes its original impedance. This TDR technique has also been applied to the monitoring of corrosion in steel cables of bridges (Liu et al., 2002). Although this technique has the advantage of being multipoint or even distributed, it is limited in reach. For practical purposes the maximum distance covered by the sensor is about 2 km. This is suitable for standard wells, but not for deep oilfields, especially those from the recently discovered presalt regions in Brazil, which are over 6 km deep. 3. A Multipoint Fibre Optic Corrosion Sensor We have recently presented for the first time the concept and first experimental results of a fibre-optic-based corrosion sensor using the optical time domain reflectometry (OTDR) technique as the interrogation method (Martins-Filho et al., 2007; Martins-Filho et al., 2008). Our proposed sensor system is multipoint, self-referenced, has no moving parts and can detect the corrosion rate several kilometres away from the OTDR equipment. These features make it very suitable to the problem of corrosion monitoring of deepwater pipelines in the oil industry. It should be pointed out, however, that the approach is not limited to this specific application and can be employed to address a number of single or multipoint corrosion detection problems in other industrial sectors. In this chapter we present a detailed description of the sensor system, further experimental results and theoretical calculations for the measurement of the corrosion rate of aluminium films in controlled laboratory conditions and also for the evaluation of the maximum number of sensor heads the system supports. 3.1 Sensor Setup Our proposed sensor system consists of several sensor heads connected to a commercial OTDR equipment by a single-mode optical fibre and fibre couplers. Figure 1 shows the corrosion sensor setup. The OTDR is connected to a 2 km long single mode optical fibre. Directional couplers can split the optical signal such that a small fraction (3 to 9%) is directed to the sensing heads. The OTDR operates at 1.55 m, with a pulsewidth of 10 ns, which corresponds to a spatial resolution of 2 m. The OTDR is set to measure 50000 points for the total distance of 5 km (one point every 10 cm). The optical fibres and couplers are standard telecommunication devices. The sensor heads have 100 nm of aluminium deposited on cleaved fibre facets by a standard thermal evaporation process and they are numbered from 1 to 11 in Fig. 1. Fig. 1. Schematic diagram of the corrosion sensor. Sensor heads are numbered. Fibre lengths and split ratios are shown. 3.2 Results For laboratory measurements the corrosion action was simulated by controlled etching of the Aluminium film on the sensor head. We used 25 H 3 PO 4 : 1 HNO 3 : 5 CH 3 COOH as the Al-etcher. The expected corrosion rate of Al from this etcher is 50 nm/min. Figure 2-a shows the OTDR trace where each peak, numbered from 1 to 11, indicates the reflection from the corresponding sensing head. The head number 6 is immersed in the Al-etcher. As the aluminium is being removed from the fibre facet the reflected light measured in the OTDR decreases, as shown in Fig. 2-b. In Fig. 3 we plot the ratio of peak (point A) to valley (point B) of the reflected light shown in Fig. 2-b as a function of the aluminium corrosion time. Figure 3 shows that up to 60 seconds of corrosion there is no significant change in the OTDR measured reflected light, since the aluminium is still too thick. Further up from this point the reflection drops to a minimum and then stabilizes at a constant level. The constant level means that the corrosion process on the fibre facet has ended. We obtain the corrosion rate by taking the deposited metal thickness and the time taken to reach the constant level, as show in Fig. 3. 40m40m40m40m40m40m 40m40m40m40m 10m10m10m10m10m10m 10m10m10m10m 4 96 5 95 7 93 7 93 6 94 6 94 5 95 9 91 8 92 3 97 1 2 3 4 5 6 7 8 9 10 OTDR 11 Optical Fiber 2km [...]...38 Optical Fibre, New Developments   50 (a) 11 1 40 2 3 6 5 7 8 9 10 Intensity (dB) 4 30 20 10 0 2. 0 2. 1 2. 2 2. 3 2. 4 2. 5 2. 6 Distance (Km) 45 (b) A Corrosion Time (s) Intensity (dB) 40 68.5 76.9 82. 0 84.4 87.6 91.4 94.8 35 30 25 B 20 15 2, 338 2, 340 2, 3 42 2,344 2, 346 2, 348 2, 350 2, 3 52 2,354 2, 356 Distance (Km) Fig 2 (a) OTDR trace, corresponding to the intensity... mode, the reflectance is given by  R 1  r 12 r23 exp j 2k 0 2 2 d  , r 12  r23 exp  j 2k 0  2 d where ri , i 1  (1)  i 1   i  i 1   i (2) is the normal incidence reflectivity at the interface between media i and i+1 (i = 1, 2) , k0 = 2 /λ, εi is the relative electrical permittivity of medium i (i = 1, 2, 3) and d is the metal film thickness Optical Fibre Sensor System for Multipoint... structures under impact waves in real time 0 .25 0 .2 0.15 Sensor Output (V) 0.1 0.05 0 -0.05 -0.1 -0.15 -0 .2 -0 .25 0 5 10 15 20 Time (s) 25 30 35 -20 2Hz 12. 5Hz 18Hz -40 Amplitude (dB) -60 -80 -100 - 120 -140 0 5 10 15 20 25 30 Frequency (Hz) 35 40 45 50 Fig 13 Field test of traffic impact monitoring with 5 cars or trucks passing: (a) Time trace of the signal around 2 Hz, after 1 Hz-5 Hz bandpass filter; (b)... Intrusion, Submarine and Optical Ground Wire Fibers 61 0 ,20 0,15 0,10 0,05 Amplitude (V) 0,00 -0,05 -0,10 0,10 0,08 0,06 0,04 0, 02 0,00 -0, 02 -0,04 -0,06 -0,08 -0,10 3,6 3,8 4,0 4 ,2 4,4 Fiber O ptic A ccelerom eter 3,6 3,8 4,0 4 ,2 4,4 T im e (s) Fiber Optic Accelerometer (b) 113,75 23 5 3 62, 5 Amplitude (dB) -60 535 7 42, 5 -80 -100 - 120 -140 -160 -180 0 20 0 400 600 Frequency (Hz) 800 1000 Fig 12 Impact detection... September 20 00, available in http://www.freepatentsonline.com/6114857.html Lee, B (20 03) Review of the present status of optical fiber sensors Optical Fiber Technology, Vol 9, 20 03, pp 57–79, ISSN 1068- 520 0 Lide, D R (20 04) Handbook of Chemistry and Physics, 85th Edition, CRC Press, ISBN 0-84930485-7, USA, 20 04, pp 12- 133 – 12- 156 Liu, W.; Hunsperger, R G.; Chajes, M J.; Folliard, K J & Kunz, E (20 02) Corrosion... surface-plasmon spectroscopy Physical Review B, Vol 37, No 7, 1988, 3164-31 82, ISSN 0163-1 829 Fuhr, P L & Huston, D R (1998) Corrosion detection in reinforced concrete roadways and bridges via embedded fiber optic sensors, Smart Materials and Structures, Vol 7, 1998, pp 21 7 22 8, ISSN 0964-1 726 44 Optical Fibre, New Developments Kohl, K T., (20 00) System and method for monitoring corrosion in oilfield wells... the electrostrictive constant  e , is defined (Boyd, 20 03) as     e  o       T ( 12) o is the average density of the material If we assume that the sound wave propagates along the fibre optical axis, then the density fluctuation propagates according to (Boyd, 20 03)  2   2     2  2  ' 0 (13)    VA t 2 z 2  t  z 2 where ' is the damping parameter and VA the velocity... May/June 20 02, pp 21 7 -22 3, ISSN 0899-1561 Malitson, I H (1965) Interspecimen comparison of the refractive index of fused silica Journal of the Optical Society of America, Vol 55, No 10, 1965, 120 5- 120 9 Martins-Filho, J F.; Fontana, E.; Guimaraes, J.; Pizzato, D F & Souza Coelho, I J (20 07) Fiber-optic-based Corrosion Sensor using OTDR, Proceedings of the 6th Annual IEEE Conference on Sensors, pp 11 72- 1174,... immune, to a certain extent, to small optical power fluctuations that may occur due to changes in the OTDR signal power, optical fibre and fibre coupler loss variations along the sensor system 24 100nm 22 Relative Intensity (dB) 20 Corrosion Rate 18 16 Metal Thickness 14 12 10 0 8 6 4 0 20 40 60 80 100 120 140 Corrosion Time (s) Fig 3 Relative intensity obtained from Fig 2- b, as a function of the corrosion... and Optical Ground Wire Fibers 51 Fig 2 (a) Small-scale view of experimental ACF of SOP, DGD, PSP, and CCF of SOP and PSP (b) Comparison of theoretical and experimental ACF of SOP and DGD (c) Comparison of experimental ACF of SOP during the day (11:00 A.M to 18:00) and the night (23 :00 to 6:00 A.M the next morning) with corresponding curve fittings From Zhang et al (20 06)  52 Optical Fibre, New Developments . 10m10m10m10m 4 96 5 95 7 93 7 93 6 94 6 94 5 95 9 91 8 92 3 97 1 2 3 4 5 6 7 8 9 10 OTDR 11 Optical Fiber 2 km Optical Fibre, New Developments3 8  2. 0 2. 1 2. 2 2. 3 2. 4 2. 5 2. 6 0 10 20 30 40 50 Intensity (dB) Distance (Km) (a) 1 2 3 4 5 6 7 8 9 10 11 . Vol. 18, No. 1, 28 0 -28 2. 140 160 180 20 0 22 0 24 0 26 0 28 0 300 0 50 100 150 20 0 25 0 300 350 Weight [g] Output Voltage [mV] Measured Cubic fitting Optical Fibre, New Developments3 2 Bernini,. (Km) (a) 1 2 3 4 5 6 7 8 9 10 11 2, 338 2, 340 2, 3 42 2,344 2, 346 2, 348 2, 350 2, 3 52 2,354 2, 356 15 20 25 30 35 40 45 Intensity (dB) Distance (Km) Corrosion Time (s) 68.5 76.9 82. 0 84.4 87.6 91.4 94.8 B A (b) Fig. 2.