Durability of Embedded Fibre Optic Sensors in …8910/...Application of fibre optic sensors have...

Durability of Embedded Fibre Optic Sensors

in Composites

Klas Levin

Doctoral Thesis Report 2001-8

Durability of Embedded Fibre Optic Sensors in Composites

by

Klas Levin

Department of Aeronautics Royal Institute of Technology

SE-100 44 Stockholm Sweden

Report 2001-8 ISSN 0280-4646

Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan i Stockholm framlägges med offentlig granskning för avläggande av teknisk doktorsexamen måndagen den 21 maj 2001 kl 13.15 i Kollegiesalen, Administrationsbyggnaden, Kungliga Tekniska Högskolan, Valhallavägen 79, Stockholm. Fakultetsopponent är Professor Anders Ulfvarson, Institutionen för marin teknik, Chalmers Tekniska Högskola.

Durability of embedded fibre optic sensors in composites

i

PREFACE

The work presented in this doctoral thesis was carried out at FFA, The Aeronautical Research Institute of Sweden, presently, a division of FOI, The Swedish Defence Research Agency. The research was financially supported by the FMV, Defence Materiel Administration.

There are a number of people who deserves my gratitudes as they have been important to this work. First of all I am deeply indebted to Dr Pavel Sindelar who by his visionary view directed me into the field of structural health monitoring. I also want to thank him for his support during the years at FFA. I am deeply grateful to T. Lic Rolf Jarlås and Jérôme Matrat for their contribution and helpfulness. I also wish to express my sincere gratitude to Sören Nilsson and Dr Arne Skontorp sharing their experience with me. Christophe Paget and Dr Gunnar Melin are both thanked for their enthusiasm, as well as all the others at the Structures Department are acknowledged.

I want to thank Professor Jan Bäcklund for accepting me as a student at the Department of Aeronautics at the KTH, and that he finally encouraged me to finish this thesis. Finally, but not least, I want to thank my wife Tuula for her patience and support.

Stockholm in April 2001

Klas Levin


ii


iii

ABSTRACT

This thesis concerns various aspects of the durability of fibre optic sensors embedded in composite. Since repair or replacement of embedded sensors is not generally possible, the functional reliability of embedded sensors is one of the most important prerequisites for successful use. The main research objective was to investigate the interaction between the sensor and the composite, and how this is affecting the mechanical and optical sensor response. Fibre optic sensors embedded in composite structures induce local stress concentrations when the composite is subjected to mechanical loads and environmental changes such as temperature and moisture. A complex transfer of stresses through the interfaces between the embedded sensor and the composite occurs and can result in large local stresses in the composite and a significant change in the response of the embedded sensor. These stress concentrations make the interfaces susceptible to debonding.

The sensor performance was studied experimentally and numerically. Some basic results were generated for the EFPI and Bragg grating sensors. The phase-strain response was determined during static and fatigue loading. The results showed that the sensors were more reliable in compression than in tensile static and fatigue loading. Generally, the sensor reliability during loading was significantly improved for the Bragg grating sensors over that of the EFPI sensor, as an effect of the sensor geometry. This was also demonstrated in the investigations on impacts. Impacts do not necessarily result in damage in the composite, but might cause debonding or other failure modes in the sensor area. Large, local stress concentrations occur at several positions in the EFPI sensor, which pointed out that this sensor type was not suitable for embedded applications.

The shift in focus from the sensor concept based on the EFPI sensor to that based on the Bragg grating sensor manifested itself in several studies. The calculated deformation field around an embedded optical fibre was verified in experiments using a high-resolution moiré interferometric technique. Furthermore, the improvement in the coating technology was verified. A significant higher interfacial strength was obtained with the silane-treated glass surface. The results indicated that at least a twofold improvement of the shear strength was obtained.

To simultaneously measure the in-plane strain components and the temperature change, embedded Bragg grating sensors were arranged in a rosette configuration. The relationship between the optical response from each sensor and the strains in the laminate was numerically and analytically established.

Damage lead to stress redistribution in the sensor region, which may influence the output from the embedded Bragg grating sensor. The effect was numerically evaluated for interfacial damage, and was compared to that of a sensor with undamaged interface. The results showed that debonding might have a significant influence, in particular for combined thermal and mechanical loading.

Keywords: composites, fibre optic sensor, embedded, EFPI sensor, Bragg grating sensor, durability, fatigue, impact, strain measurement, interface, stress analysis


iv


v

TABLE OF CONTENTS

PREFACE ....................................................................................................................I

ABSTRACT................................................................................................................III

DISSERTATION....................................................................................................... VII

DIVISION OF WORK BETWEEN AUTHORS .......................................................... IX

INTRODUCTION.........................................................................................................1

1. BACKGROUND......................................................................................................1

2. FIBRE SENSOR TECHNOLOGY...........................................................................3 2.1 Optical fibres and Bragg gratings ...................................................................3 2.2 Fibre optic strain sensors ...............................................................................4 2.3 Bragg grating demodulation and multiplexing ................................................6 2.4 Strain and temperature measurements..........................................................7 2.5 Temperature discrimination............................................................................8

3. COMPOSITES WITH EMBEDDED SENSORS....................................................10 3.1 Microstructural effects ..................................................................................10 3.2 Structural integrity.........................................................................................10 3.3 Sensor performance .....................................................................................11

4. SUMMARY OF APPENDED PAPERS.................................................................13

REFERENCES..........................................................................................................16

PAPER A.......................................................................................................... A1-A13 PAPER B.......................................................................................................... B1-B14 PAPER C............................................................................................................ C1-C8 PAPER D............................................................................................................ D1-D9 PAPER E.......................................................................................................... E1-E10 PAPER F ...........................................................................................................F1-F16 PAPER G .........................................................................................................G1-G10


vi


vii

DISSERTATION

This dissertation consists of the following seven appended papers:

Paper A:

K. Levin & S. Nilsson, "Examination of reliability of fibre optic sensors embedded in carbon/epoxy composites", Third International Conference on Intelligent Materials and Third European Conference on Smart Structures and Materials, ICIM-3/ECCM-3, Proc. SPIE 2779, Lyon, June 3-5 1996, pp.222-229

Paper B:

K. Levin & R. Jarlås, "Vulnerability of embedded EFPI-sensors to low-energy impacts", Smart Structures and Materials, Vol. 6, June 1997, pp. 369-382

Paper C:

R. Jarlås & K. Levin, ”Location of embedded fiber optic sensors for minimized impact vulnerability", Journal of Intelligent Material Systems and Structures, Vol. 10, March 1999, pp.187-194

Paper D:

L.G. Melin, K. Levin, S. Nilsson, S.J.P. Palmer & P. Rae, ”A study of the displacement field around embedded fibre optic sensors”, Composites: Part A, Vol. 30, 1999, pp.1267-1275

Paper E:

K. Levin, ”Interfacial strength of optical fibres”, Proc. Twelfth International Conference on Composite Materials, ICCM-12, Paris, July 5-9, 1999, Paper 640

Paper F:

J. Matrat, K. Levin & R Jarlås, ”Implementation of a Bragg grating strain rosette embedded in composites”, Smart Structures and Materials 2001: Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials, Proc. SPIE 4328, March 6-8, 2001, Paper 4328-18

Paper G:

J. Matrat, K. Levin & R. Jarlås, ”The effect of debonding on strain measurement with embedded Bragg grating sensors,” Proc. Second International Workshop on Structural Health Monitoring 2000, Stanford, CA, Sep. 8-10, 1999, pp. 651-660


viii


ix

DIVISION OF WORK BETWEEN AUTHORS

Paper A:

The work was initiated and defined by K. Levin, who performed the experiments. S. Nilsson did the numerical analysis. The paper was jointly written by the authors.

Paper B and C:

The works were initiated and defined by K. Levin. The experimental work was performed by K. Levin. The numerical analysis was carried out by R. Jarlås. The papers were jointly written by the authors.

Paper D:

The work was initiated by K. Levin. L.G. Melin did mainly the optical testing with assistance of S.J.P Palmer and P. Rae. K. Levin made the additional experimental work. S. Nilsson performed the numerical analysis. The paper was jointly written by L.G. Melin and K. Levin.

Paper F and G:

The work was initiated and defined by K. Levin. Numerical analysis was performed by J. Matrat with assistance of R. Jarlås and K. Levin. The papers were written by J. Matrat and K. Levin.


x


1

INTRODUCTION

1. BACKGROUND

A challenge to finding more cost-effective ways to reduce maintenance costs and enhancing structural safety has been created by the continual ageing of the aircraft fleets, a more cost driven design of new military aircraft, rotorcraft and the future use of large capacity aircraft. Also, in marine vessels and other vehicles where new design concepts including composite structures have been introduced, a need exist for more beneficial inspection methods. The important driving force for built-in sensors in structures is improved ways to monitor operational loads. In many applications such as in the area of civil engineering (bridges, tunnels, dams etc.) the knowledge of the loads is generally low, and significant rewards can be gained by incorporating sensors into the structures.

Historically, structural monitoring in aircraft has been accomplished with a limited number of resistive strain gauges together with manual flight logs and accelerometers [1]. The development of sensor technology and microprocessors has made it possible to incorporate more data processing on-board the aircraft. Such systems include strain sensors, parametric and other sensors together with diagnostic and prognostic capacities, as schematically indicated in Figure 1. This might allow monitoring of the structure during service with less efforts and resources and at the same time obtain more reliable and efficient data. The operation and support of the aircraft may be sufficiently improved while reducing cost for maintenance and improving the structural performance. Similar approaches also exist for other vehicles and structures.

Figure 1. Proposed architecture at Boeing of a structural health management system [2].


2

At research level, the field of structural condition and health monitoring has attracted many contributors over the last 20 years with the objective to develop systems for monitoring operational loads and even damage detection. Various sensor techniques have been proposed. Currently, these solutions are in most cases developed based on fibre optic sensor technology. The use of optical fibre sensors to measure strain and temperature, or even to detect damage in structures are finding more applications due to the sensors’ small size, sensitivity and immunity to electromagnetic interference. Other advantages of fibre optic sensors originate from the fact that they are dielectric and therefore can be used in hazardous environments where electrical sensors would not be safe, and that the losses are very small in optical fibres allowing the signal to be transmitted over a long distance. Fibre optic sensors have been used to monitor strain but are also finding applications in other areas, such as cure process monitoring for composites, pressure, current, chemical and acoustic emission monitoring, indicating the versatility of the fibre optic sensor technique.

Many activities in Europe, Asia and the US are focused on application of large fibre optic sensor arrays for various types of structures. Recently the research has led to a number of demonstration projects. Application of fibre optic sensors have been demonstrated in a military boat [3], as indicated in Figure 2, and a large yacht [4] and in civil structures such as bridges [5]. Field testing has also been conducted to demonstrate utility of fibre optic sensor systems in aircraft structures as in the flight testing of a BAe Jetstream in the Brite-EuRam project MONITOR [6]. Other applications are the full-scale testing of a fighter F-15 [7] and the testing of a wing structure from a civil aircraft [8]. The fibre optic sensors have in these applications been either surface attached or embedded within the material. The major reason to embed sensors is that they will be more protected from external damage. In the US Air Force close to half of the structural repairs is due to damage caused during maintenance [9]. On the other hand, attached sensors are easier to install, and may be refitted during service.

Figure 2a. Sensor location in the Fast patrol boat KMN Skjold [3].

Figure 2b. Strain data from a pair of fibre optic sensors attached to the hull [3].


3

2. FIBRE SENSOR TECHNOLOGY

2.1 Optical fibres and Bragg gratings

Current optical fibre sensor technology has arisen largely on the basis of optical components and especially optical fibres made available as a result of the growth in their use for telecommunication purposes. An example is the development of optical fibres where the technology to produce low loss optical fibres has been developed from the sixties to be highly sophisticated a few decades later. The same fibre technology is also applicable for manufacturing of optical fibre sensors. As these optical fibre sensor techniques have been evolving rapidly, different types of optical fibres have been developed to suit many applications. For sensing purposes, mainly one type of fibre has been considered, which is the single mode optical fibre. The main characteristic is that only one mode is allowed to propagate in the fibre core, which serves as the light guide.

The optical fibres are small, lightweight and flexible. The diameter of the fibres is typically 125 µm. In a single mode fibre, the core has a diameter which is typically less than 9 microns. The fibre is protected from moisture and from mechanical contact using a polymer coating or a hermetically closed coating made of metal or ceramics [10]. The most common polymeric coating material for sensor applications is polyimide. It has excellent high temperature properties and generally bond well with various adhesives, glass and polymers such as polyesters and epoxies. To further improve the bond between the coating and the optical fibre a sizing may be used.

4000

3000

2000

1000

0

Stre

ss (M

Pa)

10-3

10-2

10-1

100

101

102

103

104

105

Time (hours)

Day Month Year 10 Y

Acrylate-coated Fit to K-model Polyimide-coated Fit to K-model

4000

3000

2000

1000

0

Stre

ss (M

Pa)

10-3

10-2

10-1

100

101

102

103

104

105

Time (hours)

Day Month Year 10 Y

Acrylate-coated Fit Uncoated - Series 1 Uncoated - Series 2 Uncoated - Series 3

Figure 3a Applied stress as a function of time-to-failure for the acrylate and polyimide coated fibres [13].

Figure 3b Applied stress as a function of time-to-failure for the acrylate and uncoated fibres [13].

The strength and durability of optical fibres is among the major concerns in embedded fibre sensory applications. Since it is generally impossible to repair a damaged optical fibre embedded in a structure, it is of vital importance to ensure that the optical fibre system will survive and remain functional for the life of the structure. The intrinsic short-term failure strain of an optical fibre is very high, normally above 5 %, as shown in Figure 3 [11,12,13]. However, it was found that in order to achieve these good properties, the glass surface of the fibre must be protected from the environment, i.e. shielded from moisture and mechanical


4

contact. Failure mechanisms in silica glass have been found to be accelerated due to moisture. It is believed that this mechanism is mainly responsible for the ageing of optical fibres [12,13]. However in [13] it was indicated that even uncoated fibres could remain functional for years at typical composite load levels in aircraft applications.

The failure strain of uncoated fibres is often less than 1 %. However, the strength and ageing process of the optical fibre are strongly affected by the Bragg grating production method. To obtain access to the fibre core when the Bragg gratings are written, the coating must be locally removed. Most gratings are made by first stripping the optical fibre, either mechanically or chemically. Fibres with mechanically removed coating have very low strength; therefore, chemical stripping is generally preferred. For polyimide coatings this is normally performed with hot sulphur acid. However, in many applications, it is preferable to directly fabricate the gratings in the draw tower or to write gratings directly through a special fibre coating. It is preferable since the strength is only slightly affected as the exposure of the glass to the ambient environment and further handling are reduced to a minimum. It is however difficult to effectively implement such a technique, as it results in an inflexible grating fabrication.

Sensors based on Bragg gratings have received the most focus during the recent years. This is a result of the relatively simple fabrication method and the capability to multiplex Bragg grating sensors. When a germanium-doped silica core fibre is exposed to ultraviolet radiation, this leads to a permanent change in refractive index of the germanium-doped region [14]. Photosensitivity is often increased by sensitising the germanium-doped fibres with hydrogen to obtain gratings with high reflectivity. For commonly used production techniques, the degradation of the optical properties of the Bragg gratings has been shown to be small at 300ºC or below.

2.2 Fibre optic strain sensors

Two different sensor principles, the intracore Bragg grating and the interferometric Fabry-Pérot, have attracted most interest for localised strain and temperature sensing. Their multiplexibility and versatility lends itself to many sensing applications. Many other different sensor principles exist, such as the polarimetric approach [e.g. 15] or the Brillouin scattering technique [16]. However, these have not achieved general acceptance due to the fact that the optical systems based on such sensors have either low sampling rates, low spatial resolution or low strain resolution.

The intracore Bragg grating sensor is based on the principle that the Bragg grating reflects the central wavelength of the light, as indicated in Figure 4. A uniform periodic grating is used, which is same as the sensor length. Typically the sensor length is a few millimeters long to achieve gratings with large reflectivity. Strain and temperature induce a shift in the spectral location, which is detected.


5

Figure 4. Principle of the intracore Bragg grating sensor [21].

The interferometric Fabry-Pérot sensors have been used in many different configurations, although it is the extrinsic Fabry-Pérot Interferometer (EFPI) [e.g. 17] and the interferometric Bragg grating sensor [e.g. 18] that have received the most interest. Common for these sensors are that they have two (semi-) reflective mirrors which create the Fabry-Pérot cavity. The principle of the interferometric sensors is that the two reflections interfere and the intensity at the detector varies as a sinusoidal function of the length between the mirrors. The length is directly affected by the strain and temperature. The mirrors can be realised with deposition of reflective material inside the fibre or at straight cleaved fibre ends as shown in Figure 5, or by writing two semi-reflective Bragg gratings in the fibre core as seen in Figure 6.

Figure 5. An EFPI-sensor. Figure 6 An interferometric sensor based on two semi-reflective Bragg gratings.

The cavity in the EFPI sensor inherently results in mechanical weakness. These sensors have the disadvantages that they have stress concentrations due to diameter discontinuities between the lead fibres and the alignment tube or partially reflecting splice. These geometrical features resulted in static failure strain of approximately 1 % or less [17]. Sensors with Bragg gratings have a much higher strength, similar to that of optical fibres.


6

2.3 Bragg grating demodulation and multiplexing

Multiplexing is a method in which two or more sensors utilize the same light source, detector and signal processor. Multiplexing is necessary to reduce the number of fibre leads and connectors. Different concepts for multiplexing of fibre optic sensors have been reviewed in [18-21]. In theory, multiplexing can be achieved in many fashions such as in the time, wavelength, spatial, phase and frequency domain. The two most important methods are the time and wavelength multiplexing techniques for which many different configurations have been suggested.

Figure 7. Example of a wavelength multiplexing technique [21].

The concepts utilizing wavelength division multiplexing have to accommodate the wavelength response of the Bragg grating sensors associated with the change in strain and temperature. Each sensor grating is written with a specific Bragg wavelength for a unique identification of each sensor. The difference in wavelength between sensors is associated with the spectral width of the light source and the strain range. This will limit the number of sensors that a fibre can host. Typically, the maximum sensors with full strain range (1 %) are less than ten with currently available light sources. However, there exist different techniques to expand the number of sensors such as sequentially addressing each sensorised fibre or using multiple light sources, although this creates a more complex configuration. An example of a wavelength multiplexing technique is shown in Figure 7.

For interferometric sensors, time division multiplexing is the standard multiplexing technique. The principle is based on addressing each sensor in the fibre with different delays from the source and detector. For such systems each generated light pulse produces twice as many pulses representing the two reflected pulses from each sensor. The interference of the light is then made in the detection unit, as schematically shown in Figure 8. The number of sensors is limited due to that fact that each sensor reflects a part of the light resulting in a reduced light intensity at the successive sensors.


7

Figure 8. A schematic view of a time domain multiplexing system [22].

2.4 Strain and temperature measurements

In this section, the response of embedded fibre optical sensors to strain and temperature is discussed.

Once an optical fibre is embedded, the interpretation of the sensor response becomes complex due to the effects of the interaction between the sensor and the surrounding material in which the sensor is embedded. The strain state in the sensor core is not necessary the same as the strain in the surrounding material. Several studies have addressed this interaction. In the early studies of embedded fibre optic sensors, the difference between embedded and free sensors were not fully understood.

With illumination from a broadband light source, only a narrow spectral line is reflected along the core at the Bragg wavelength, λb, which is the wavelength at maximum reflectivity. The reflective light from the sensor is independent of intensity fluctuation of the light. The Bragg wavelength is dependent on the period of the grating, Λ, and the refractive index of the fibre, ne, as

Λ= nb 2λ . (1)

The period of the grating is changed from the mechanical strain and temperature. Therefore the Bragg wavelength of the reflected light is sensitive to these parameters. In the optical sensor system, the sensor response is encoded in terms of the changed Bragg wavelength [23,24]. The strain and temperature in the fibre core is linear proportional to the Bragg wavelength shift,


8

( )( )

∆+

+++−=

∆ TPPPn

b

ξεεεελ

λ332212111112

2

11 21

2 (2)

where ε11, ε22 and ε33 are the sensor strains, and ∆T and ξ are the temperature variation and the thermo-optic constant, respectively. P11 and P12 are the photo-elastic constants (the Pockel’s constants). The equation is based on the assumption that the strain is approximately constant over the gauge length and that the sensor is in a non-birefringent fibre. Equation 2 has recently been extended for non-uniform conditions such as chirped gratings (gratings with a non-uniform period) [25] and large stress gradients [26].

Similarly, the Fabry-Pérot interferometric sensor, consisting of two gratings, provides interference between the two reflected signals and exhibits the same linear relation between the phase shift, ∆φ, and the change in strain and temperature [23,24],

( )( )

∆+

+++−=∆ TPPPnnkL ξεεεεφ 332212111112

2

11 21

2 (3)

where L is the gauge length of the sensor and k is the free-space propagation constant.

For the extrinsic Fabry-Pérot interferometric sensor, Equation 3 can be further simplified in that the sensor has a cavity formed in air instead of a glass waveguide. This implies that the transverse sensitivity disappears and the thermal-optic constant, ξ, is changed to thermal expansion for the sensor gauge length, ξα, as:

( )

∆+−=∆ TPnnkL aξεεφ 1112

2

11 2. (4)

A disadvantage with the sensor based on Bragg gratings is that the sensor is significantly more temperature sensitive due to the sensor being formed in the glass rather than by an air cavity. Another point is that the transverse sensitivity is negligible in the EFPI sensor whereas the transverse strain in Bragg grating based sensors has a significant influence on the sensor output. Despite this, the ease of fabrication and multiplexing of Bragg gratings has made the EFPI sensor obsolete.

Studies have been performed to determine the full strain field in the composite with embedded fibre optic sensors. Various type solutions have been proposed and tested by using different measurement principles such as Bragg grating in polarization preserving optical fibres [27], multiple sensors with different strain sensitivities [28], and multiple sensors embedded with different orientations [29, 30, Paper F].

2.5 Temperature discrimination

One of the challenges of strain measurement is always the separation of mechanical strains from thermally induced strains. With a single measurement of the Bragg wavelength, it is not possible to differentiate between the effects of changes in strain and temperature. Therefore,


9

in these concepts the principle is to use two measurements to be able to separate the strain and temperature. For a successful determination of strain and temperature, the effect of sensor location on the measurements should be insignificant. Temperature compensation in the strain measurement can be achieved with a varity of detection schemes. Separation of mechanical strains from thermally induced strains has been tried with some success using sensor fibres with different core dopings that induced differences in the thermo-optical constant [31,32] or using two sensors with different thermal sensitivity [33,34]. In this case, the wavelengths changes of two single gratings, one in each fibre, were monitored subjected to the same temperature and strain embedded in a composite.

The change in peak Bragg wavelength of the fibre grating can be expressed as,

TKKT T ∆+∆=∆ εελ ε),( (5)

assuming that the strain and thermal response on the wavelength shift is independent. If this assumption is valid, we can solve the following equation system where 1 and 2 refer to the two wavelengths:

∆∆

=

∆∆

TKKKK

T

T ελλ

ε

ε

22

11

2

1 (6)

by simultaneously measuring the response of the Bragg gratings in terms of wavelength shift for the two wavelengths. As the photo-elastic and thermo-optic coefficients are wavelength dependent, wavelength changes of each of the two gratings will be different although each grating has the same level of strain. The technique can be extended to interferometric sensors by replacing the wavelength shift with the phase shift in Equation 5.

Other methods used for strain-temperature discrimination are based on dual wavelengths for measuring the response at two different wavelengths [35], long-period gratings [36,37], two different optical modes, or a sensor uncoupled to the strain [38]. These methods basically work using the same principle as the above methods in that they are based on that the measurands strain and temperature being separated by a two-parameter scheme. As the sensitivity at two different Bragg wavelengths is different with strain and temperature, they can be determined by a simultaneous measurement. The limitations of these techniques are that they show some degree of a non-linear thermal sensitivity and have limited strain resolution. A resolution on the order of 20 microstrain and a temperature resolution better than 1 ºC over a strain range of 800 microstrain can be achieved [39,40].

A few of these concepts are based on two fibres spatially embedded close to each other so that each sensor experiences the same condition. However, there exist a required minimum distance between the two fibres such that no interaction occurs in the local strain field around the optical sensors. In [41] it was shown that a distance larger than 0.3 mm is sufficient to avoid any interaction. An alternative way is to use dual cores in one fibre, but this concept is envisaged having large problems to connect the sensors with the optical system.


10

3. COMPOSITES WITH EMBEDDED SENSORS

3.1 Microstructural effects

The basic understanding of the changes in microstructure as a consequence of embedding optical fibres was obtained in studies reported in [42,43]. There is a significant interaction between the composite material and an embedded optical fibre, giving rise to changes in geometry on the local scale. An optical fibre embedded off-axis to the reinforcing carbon fibres creates a matrix pocket around the optical fibre, as shown in Figure 9.

The geometrical effect was studied and quantified for typical cases. Generally, sensors should be embedded parallel to the surrounding carbon fibres in order to disturb the reinforcing fibres the least and avoid matrix pockets. The effect of small local variation in fibre volume around the optical fibre found in [43] for parallel optical and reinforcing fibres was analysed in [41]. The finding was that this effect is negligible. The resulting strain field around the optical fibre is significantly changed and the interfacial stresses increase [44].

Optical fibre

Matrix pocket

Figure 9. An optical fibre embedded between a +45º- and a -45º-ply.

3.2 Structural integrity

Embedded optical fibre sensors and their associated optical fibres should not affect the structural integrity of the host composite structure. A decrease in the load-carrying ability of the structure is not acceptable in most cases especially in composite materials as the strive is


11

to increase strain allowables and thereby make them more competitive. As described above, embedded optical fibres may disrupt the composite microstructure. In addition, the difference in thermo-elastic properties may cause stress concentrations. From a structural integrity point-of-view, as well as from a fibre survivability viewpoint, the optimal position of an optical fibre in a composite is between two plies with reinforcing fibres oriented along the optical fibre. In this position, the optical fibre is most protected as the reinforcing fibres will carry axial loads, and potentially threatening matrix cracks will grow along the optical fibre, not through it.

Studies have shown that the effect of embedded optical fibres on the strength and fatigue life of composites laminates is generally unaffected by the presence of a small (125 µm) optical fibre, with the optical fibre in the direction of the surrounding co-linear plies [45-50]. In static tension or compression, there were no significant adverse effects of the embedded optical fibres, because the optical fibres did not initiate or indirectly cause fibre-dominated failures in the composite. In the case of an optical fibre loop, where the fibre changes direction and locally disturbs the microstructure considerably, no premature failures associated with optical fibres were found [48,49].

Structures introduce an additional complexity; thus, a need existed to investigate whether the introduction of embedded optical fibres into regions with internal ply-drops and bolts would affect the static load-carrying ability. Therefore, studies were conducted on both holes, ply-drops and bolted joints with optical fibres [47-49]. Similar studies were performed to evaluate if impact on plates with embedded optical fibres can result in reduction in structural integrity of the composite [51-54]. The results indicated that the embedded optical fibres did not affect the structural integrity, as they do not initiate failure.

3.3 Sensor performance

The sensor performance has to be excellent and reliable during the entire life of the structure. Since repair or replacement of embedded sensors is not generally possible, the functional reliability of embedded sensors is one of the most important prerequisites for successful use. This will be associated with a disruption of data and also induce cost for installation of new sensors attached to the surface.

Sensors and transducers embedded in composite structures induce local stress concentrations when the composite is subjected to mechanical loads and environmental changes such as temperature and moisture. This is an effect of the local geometry and difference in stiffness between the composite and the fibre optic sensor. A complex transfer of stresses through the interfaces between the embedded fibre and the composite occurs and can result in large local stresses in the composite and a significant change in the response of the embedded sensor. These stress concentrations make the interfaces susceptible to debonding [55, Paper A]. If there is debonding at the interface, this may lead to unacceptable stress redistribution in the sensor, which may lead to undesired changes in the sensor output [Paper A, Paper G].


12

The sensor performance has been studied experimentally and numerically. The first numerical investigations concerned the transfer of stresses to optical fibres in different laminate configuration, for optical fibre embedded parallel and off-axis to the reinforcing carbon fibres [44,56-61]. These studies showed that the matrix pocket created for optical fibres creates large stress concentrations, which are significantly larger than those occurring in optical fibres embedded parallel to the reinforcement. These studies were accompanied by analytical solutions [62-69], which focused on reduction of the stress concentration with a selection of coating properties. They found that a reduction in the interfacial stresses might be achieved, although only a few viable alternatives exist to polyimides, limiting the useful of these results.

In [41], a detailed numerical study showed a complex interaction between the cavity and the stresses at the interface on the EFPI sensor. This study was further extended in [Paper A] to understand the coupling between the local stresses induced from mechanical and thermal loading and the effect of the lay-up. One concern was that impacts, although they do not necessarily result in damage in the composite, might cause debonding or other failure modes in the sensor area [Paper B, Paper C]. Impacts generate high, local stress levels that can induce damage in the sensor or damage in the composite in the region of the sensor. Initially the work concerned embedded EFPI sensors. The critical impact energy level was predicted for different thickness positions of the sensor. The results were correlated with impact experiments indicating that in some positions the EFPI sensor was vulnerable to impacts. The study was shifted to addressing the impact vulnerability of embedded Bragg grating sensors. The objectives were further extended to more cases and conditions. It was found that the best position to embed sensors was close to the mid-plane, similar to that found for the EFPI sensor. Although the results showed that an impact might not initially affect the sensor performance, subsequent fatigue loading could result in premature degradation in sensor performance. This could occur if the interfacial strength between the coating and the glass was poor.

Other important findings from the numerical studies [70] were that the number of 0°-plies in which the sensor is embedded does not lead to significant change of the stress field around the sensor. No large changes in the Bragg grating signal response and the interfacial stresses occur as an effect of the coating thickness.

The calculated deformation field around the embedded optical fibre was verified in experiments using a high-resolution moiré interferometric technique [56-57, 59, Paper D]. These studies generated a deeper understanding of the mechanics of embedded sensors.

Experimental evaluation of the limit of sensor performance and the associated damage mechanisms to the degradation was tested in [Paper A] for the EFPI sensor. The phase-strain response was determined during static and fatigue loading indicating a strain sensitivity of embedded sensors which was linear up to approximately 0.20 % strain in tensile static and fatigue loading. In compression, the sensor was significantly more reliable. Similar experimental studies were also performed for embedded Bragg grating sensors. Generally, the sensor reliability during loading was significantly improved for the Bragg grating sensors over that of the EFPI sensor, probably as an effect of the sensor geometry as well as the improved coating technology. Damage lead to stress redistribution in the sensor region, which


13

may influence the output from the embedded sensor. Another study has addressed the influence of matrix cracking in the composite on the sensor output [71]. It was found that the sensor response is not significantly affected by the presence of damage in the composite because the stress redistrubution as an effect of matrix damage in the composite is small on the strain in the fibre core. The interfacial damage effect was numerically evaluated [Paper G], and was compared to that of a sensor with undamaged interface. The results showed that debonding might have a significant influence, in particular for combined thermal and mechanical loading. The study pointed out the importance of a strong bond between the sensor and the composite.

Improvement in the coating technology was evaluated. Initially, acrylate coated fibres were used similar to those common in the telecommunication industry. It was shown early that these could not be used in embedded applications since the thermal stability is significant reduced above 100 ºC. Polyimide soon became the prevalent coating material. However it was shown that in the case of highly strained embedded sensors the interface between the silica glass and the polyimide failed [Paper A]. A significant higher interfacial strength could be obtained with a silane-treated glass surface [72, 73, Paper E]. The results in Paper E indicated that at least a twofold improvement of the shear strength was obtained.

4. SUMMARY OF APPENDED PAPERS

Paper A

One of the most important tasks in achieving a reliable performance of optical fibre sensors embedded in aircraft health monitoring and usage systems is to clarify their functional reliability. In this study, the limits of sensor functionality are assessed for a number of loading conditions at room temperature.

It was observed that under static tensile loading the functional strain limit for EFPI-sensors was approximately 0.20%, whereas the functional limit for embedded Bragg grating sensors was above the strain level (0.4-0.5%) for the initiation of matrix cracking in the cross-ply laminate. In compression, the function of the EFPI-sensor was unimpaired up to a strain level that was more than four times larger than that in tension. In constant amplitude fatigue loading, the functional strain limit of the EFPI-sensor was similar to that in static tensile loading, i.e. 0.20% at 105 cycles. The silica glass/polyimide coating interface in the EFPI-sensor was identified as the sites where failure was initiated. The finite element analysis showed that the effects of curing stresses and moisture in laminates have significant influence on the functional reliability of sensors. The stress concentrations on the interface caused by the cavity in the EFPI-sensor are the main reason for the lower reliability obtained for EFPI-sensors than for embedded Bragg grating sensors.


14

Paper B

The purpose of this investigation was to evaluate the vulnerability to low-energy impacts of EFPI sensors embedded in carbon fibre composites. The task was to determine the critical contact load, which leads to deterioration in the sensor function. A bending dominated case, typically resulting from an impact far from any supporting boundary, was chosen in this investigation of impact vulnerability of a modified EFPI-sensor embedded at different depths in a composite plate.

An important finding from the experiments was that the depth position of the sensor in the plate had a significant influence on the contact load at which the sensor function deteriorated. The best position was in the middle of the plate. When the sensor was embedded closer to the plate surfaces, the sensor function became vulnerable to impact. It was shown that the function of the sensor deteriorated as an effect of debonding. In the finite element analysis the effect of embedded fibre optic sensors on the local stress distribution was a subject of special interest. The analysis supported the findings from the experiments that debonding of the sensor is the most critical failure mode for the sensor function, it is however difficult to predict the load level for initiation of damage. A new debonding criterion was proposed to improve the agreement with the experimental data.

Paper C

Vulnerability of embedded Bragg grating sensors to impact has been studied in composite laminated plates. The effects of the distance from the supporting boundary and the sensor thickness position were determined. The main objective was to develop an accurate numerical procedure to calculate the local stresses at the sensor from impact in order to determine the criticality with respect to damage initiation. A semi-analytical and a finite element based method were used to consider the contact at impact. From the analysis, it was found that a lower impact energy was required to cause damage in the sensor region when the sensor was embedded close to the boundary. At locations far from boundaries, sensors should not be embedded close to the plate surfaces. For sensors embedded close to a boundary, a thickness position close to the upper surface was found to be less suitable than other positions.

Paper D

Deformation fields around optical fibres embedded in carbon fibre/epoxy composite specimens have been measured using moiré interferometry. The inclusion of the optical fibre resulted in large strain gradients. Calculated displacements from finite element analysis were compared to the experimental results. The numerical analysis showed that the displacement field on the specimen surface is smoothed out through the moiré grating thickness, an effect which is most pronounced at the material interfaces. With this influence taken into consideration a reasonable good quantitative agreement between the experiments and the


15

finite element analysis was obtained. The finite element analysis also showed that the grating stiffness did not affect the measured displacements as long as the grating had a lower stiffness than the specimen.

Paper E

The interfacial strengths of three different optical fibres were evaluated using a push-out test method. Polyimide coated optical fibres either without or with a silane sizing on the silica glass, and optical fibres with aluminium coating were tested. A significant increase in the interfacial strength was achieved with the silane sizing. The failure mode shifted from the silica glass/polyimide interface to the polyimide/composite. For the two other fibres the interfacial strength was similar. Pre-conditioning of samples with polyimide coated optical fibres indicates that moisture can result in a significant reduction in interfacial strength for fibres without sizing whereas the sizing has a protective effects on the interfaces to moisture.

Paper F

Four embedded Bragg grating sensors were arranged in a rosette configuration to simultaneously measure the in-plane strain components and the temperature change. The sensors were embedded in neighbouring plies to realise a point-like measurement. Assuming linear thermoelastic behaviour, the relationship between the optical response from each sensor and the strains in the laminate was established. A finite element model was then used to quantify numerically the expected accuracy of the measured strain under various loading conditions.

Paper G

The effect of interfacial debonding between an embedded Bragg grating sensor and the composite on the response of the sensor was investigated. The sensor response associated with the strain state in the debonded Bragg grating sensor was compared to that for the sensor with an undamaged interface. A finite element model was used to determine the three-dimensional strain field, including contact analysis for various thermo-mechanical loading. Different debonding locations and sizes were analysed to determine the maximum change in the sensor response. The results showed that debonding could have a significant influence on the sensor response. In the case of transverse tensile loading, the stress redistribution in a debonded sensor resulted in an error in the strain measurements that was less than 54%. Loading in the axial direction may result in strain errors on the order of 1%. The investigation also showed that thermal loading could further increase the error in the strain measurements. Furthermore, a non-linear response from the debonded sensors was observed.


16

REFERENCES

1. Kudva, J.N., Grage, M.J., Roberts, M.M., ”Aircraft structural health monitoring and other smart structures technologies-Perspectives on development of future smart aircraft”, Proc. Second International Workshop on Structural health monitoring 2000, Stanford, CA, Sep. 8-10, 1999, pp. 106-119

2. Ikegami, R., ”Structural health monitoring: assessment of aircraft customer needs”, Proc. Second International Workshop on Structural health monitoring 2000, Stanford, CA, Sep. 8-10, 1999, pp. 12-23

3. Johnson, G.A., Pran, K., Wang, G., Havsgård, G.B., Vohra, S.T., ”Structural monitoring of a composite hull air cushion catamaran with a multi-channel fiber Bragg grating sensor system”, Proc. Second International Workshop on Structural Health Monitoring 2000, Stanford, CA, Sep. 8-10, 1999, pp. 190-198

4. Read, I.J., Foote, P.D., Roberts, D., Zhang, L., Bennion, I., Carr.M., ”Smart carbon-fibre mast for a super-yacht”, Proc. Second International Workshop on Structural Health Monitoring 2000, Stanford, CA, Sep. 8-10, 1999, pp. 276-286

5. Mita, A., ”Emerging needs in Japan for health monitoring technologies in civil and building structures”, Proc. Second International Workshop on Structural Health Monitoring 2000, Stanford, CA, Sep. 8-10, 1999, pp. 56-67

6. MONITOR, Monitoring On-line Integrated Technologies for Operational Reliability, Brite-Euram project BE1524, 1996-1999

7. Greene, J., Tran, T., Gunther, M., Murphy, K., Claus, R., “Damage analysis during full-scale testing of F-15 airframe using arrays of optical fiber sensors”, Smart Structures and Materials: Smart Sensing, Processing, and Instrumentation, Proc. SPIE Vol. 2191, Orlando, FL, Feb. 14-16, 1994, pp. 494-498

8. Trutzel, M.N., Wauer, K., Betz, D., Staudigel, L., Krumpholz, O., Muehlmann, H-C., Muellert, T., Gleine, W., “Smart sensing of aviation structures with fiber-optic Bragg grating sensors”, Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials; Proc. SPIE Vol. 3986, March 6-8, 2000, pp. 134-143

9. Kudva, J., ”Application of smart structures concepts to military aircraft”, Presentation at Second US Army Workshop on Smart Structures and Materials, University of Maryland, MA, Sept. 5-7, 1995

10. Semjonov, S.L., Bubnov, M.M., Dianov, E. M., Kurkjian, C. R., Breuls, A. H., “Mechanical behavior of low- and high-strength carbon-coated fibers”, Fiber Optic Materials and Components, Proc. SPIE Vol. 2290, 1994, pp. 74-78

11. Skontorp, A., “Strength and failure mechanisms of polyimide-coated optical fibers”, Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials, Proc. SPIE Vol. 3986, Newport Beach, CA, Mar. 6-8, 2000, pp. 240-251

12. Bouten, P. C.; Broer, Dirk J.; “Coating composition and fiber lifetime”, Fiber Optics Reliability and Testing: Benign and Adverse Environments, Proc. SPIE Vol. 2074, pp. 59-70

13. Skontorp, A., Cammas, J., “Static fatigue life of silica optical fibers and the significance of fiber coating and handling”, Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials, Proc. SPIE Vol. 4328, Newport Beach, CA, Mar. 6-8, 2001, Paper 4328-11

14. Morey, W.W., Ball, G.A., Meltz, G., “Photo induced Bragg grating in optical fibers”, Optics and Photonics News, Feb. 1994, pp. 8-14

15. Giles, I..P., Mondanos, M., Badcock, R.A., Lloyd, P.A., ”Distributed optical fibre based damage detection in composites”, Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials, Proc. SPIE Vol. 3670, Newport Beach, CA, Mar. 1-4, 1999, pp. 311-321


17

16. Kee, H.H., Lees, G.P., Newson, T.P., ”Simultaneous independent distributed strain and temperature measurements over 15 km using spontaneous Brillouin scattering”, Applications of Optical Fiber Sensors, Proc. SPIE Vol. 4074, pp. 271-279

17. Tran, T.A, Greene, J.A, Murphy, K.A, Bhatia, V., Sen, M.B., Claus, R.O, ”EFPI manufacturing improvements for enhanced performance and reliability”, Industrial and Commercial Applications of Smart Structures Technologies, Proc. SPIE Vol. 2447, San Diego, CA, Mar. 2-3, 1995, pp. 312-323

18. Dakin, J., Culshaw, B., Optical Fiber Sensors: Principles and Components. Vol. 1, Artech House, Boston, MA, 1988

19. Culshaw, B., Dakin, J., Optical Fiber Sensors: Systems and Applications. Vol. 2, Artech House, Boston, MA, 1989

20. Fiber Optic Smart Structures, Ed. Udd, E., John Wiley & Sons, Inc., New York, 1994 21. Kersey, A.D., Davis, M.A., Berkoff, T.A., Bellemore, D.G., Koo, K.P., Jones, R.T., ”Progress

towards the development of practical fiber Bragg grating instrumentation systems”, Fiber Optic and Laser Sensors XIV, Proc. SPIE Vol. 2839, Denver, CO, Aug. 7-9, 1996, pp. 40-63

22. Wosinski, L., Betend-Bon, J. P., Breidne, M., Sahlgren, B., Stubbe, R., ”Quasi-distributed fiber-optic sensor for simultaneous absolute measurement of strain and temperature”, Proc. First European Conference on Smart Structures and Materials, Glasgow, UK, May 12-14, 1992, pp. 53-56.

23. Sirkis, J.S., "Unified approach to phase-strain-temperature models for smart structure interferometric optical fiber sensors: part 1, development", Optical Engineering, Vol. 32, No.4, 1993 pp. 752-761

24. Sirkis, J.S., "Unified approach to phase-strain-temperature models for smart structure interferometric optical fiber sensors: part 2, application", Optical Engineering, Vol. 32, No. 4, 1993, pp. 762-773

25. Peters, K., Studer, M., Botsis, J., Iocco, A., Limberger, H. and Salathé, R.,”Embedded optical fiber Bragg grating sensor in a nonuniform strain field: measurements and simulations”, Experimental Mechanics, Vol. 41, No.1, March 2001, pp. 19-28

26. Erdogan, T., ”Fiber grating spectra”, Journal of Lightwave Technology, Vol. 15, 1997, pp. 1277-1294

27. Udd, E., Schultz, W., Seim, J., Haugse, E., Trego, A., Johnson, P., Bennett, T.E., Nelson, D., Makino, A., “Multidimensional strain field measurements using fiber optic grating sensors”, Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials, Proc. SPIE Vol. 3986, Newport Beach, CA, Mar. 6-8, 2000, pp. 254-262

28. Jin, X.D, Rossmanith, T., Rutherford, M., Sirkis, J.S., Dasgupta, A., DeVoe, D., Rosenberger, F.F., III, Venkat, V.S., Shi, Y.C., Askins, C., ”Progress towards an orthogonal strain state sensor based optical fiber technology”, Evolving and Revolutionary Technologies for the New Millennium, Proc. 44th International SAMPE Symposium and Exhibition, May 23-27, 1999, pp. 1978-1992

29. Measures, R.M., Hogg, D., Turner, R.D., Valis, T., Giliberto, M.J., “Structurally integrated fiber optic strain rosette”, Fiber Optic Smart Structures & Skins, Proc. SPIE Vol. 986, 1988, pp. 32-42

30. Carman, G.P., Lesko, J.J., Case, S.W., Fogg, B., Claus, R.O., “Development of an embedded Fabry Perot fibre optic strain rosette sensor”, NASA-CR-190490, Virginia Polytechnic Inst. State Univ., 1992

31. Wosinski, L., Breidne, M., Stubbe, R., Sahlgren, B., Betand-Bon, J-P., ” Experimental system considerations for distributed sensing with grating Fabry-Perot interferometers”, Tenth International Conference on Optical Fibre Sensors, Proc. SPIE. Vol. 2360, Glasgow, UK., Oct. 11-13, 1994, pp. 146-149

32. David, M., ” Characteristics of embedded Bragg grating sensors in composite”, FFA TN 1998-02, The Aeronautical Research Institute of Sweden, Bromma, 1998

33. Singh, H., Sirkis, J.S., ”Temperature and strain measurement by combining ILFE and Bragg grating optical fiber sensors”, Experimental Mechanics, Vol. 37, No. 4, Dec. 1997, pp. 414-419

34. Jin, X.D., Sirkis, J.S., Chung, J.K., Venkat, V.S., ”Embedded in-line fiber etalon/Bragg grating hybrid sensor to measure strain and temperature in a composite beam”, Journal of Intelligent Material Systems and Structures, Vol. 9, no. 3, Mar. 1998, pp. 171-181


18

35. Sivanesan, P., Sirkis, J., Venkat, V., Shi, Y., Reddy, C. J., Sankaran, S., Singh, H., ”Simultaneous measurement of temperature and strain using a single Bragg grating”, Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials, Proc. SPIE Vol. 3670, Newport Beach, CA, Mar. 1-4, 1999, pp. 92-103

36. Patrick, H.J., Kersey, A.D., Pedrazzani, J.R., Vengsarkar, A.M.,”Fiber Bragg grating sensor demodulation system using in-fiber long period grating filters”, Distributed and Multiplexed Fiber Optic Sensors VI, Proc. SPIE Vol. 2838, Denver, CO, Aug. 5-6, 1996, pp. 60-65

37. Bhatia, V., Campbell, D.K., Sherr, D., D'Alberto, T.G., Zabaronick, N.A, Ten Eyck, G.A., Murphy, K.A, Claus, R.O., ”Temperature-insensitive and strain-insensitive long-period grating sensors for smart structures”, Optical Engineering, Vol. 36, No. 7, July 1997, pp. 1872-1876

38. Haran, F.M., Rew, J. K., Foote, P.D., ”Fiber Bragg grating strain gauge rosette with temperature compensation”, Smart Structures and Materials: Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials, Proc. SPIE Vol. 3330, San Diego, CA, Mar., 1998, pp. 220-230,

39. Xu, M.G., Archambault, J-L., Reekie, L., Dakin, J.P., "Simultaneous measurement of strain and temperature using fibre grating sensors, "Tenth International Conference on Optical Fibre Sensors, Proc. SPIE Vol. 2360, Glasgow, UK, Oct. 11-13, 1994, pp 191-194

40. Jin, W., Michie, W.C, Thursby, G., Konstantaki, M., Culshaw, B., ”Geometric representation of errors in measurements of strain and temperature”, Optical Engineering, Vol. 36, Aug. 1997, pp. 2272-2278

41. Levin, K., Nilsson, S., ”Analysis of the local stress field in a composite material with an embedded extrinsic Fabry-Pérot interferometer sensor”, Second European Conference on Smart Structures and Materials, Proc. SPIE 2361, Glasgow, UK., 12-14 October, 1994, pp. 379-382

42. Leka, L.G., Bayo, E.” A close look at the embedment of optical fibers into composite structures”, Journal of Composites Technology and Research, Vol. 11, Fall 1989, pp. 106-112

43. Levin, K., Sindelar, P. “Assessment of the fibre volume distribution around optical fibres”, FFA TN 1993-12, Aeronautical Research Inst. of Sweden, Bromma, 1993

44. Davidson, R. and Roberts, S.S.J., “Finite element analysis of composite laminates containing transversely embedded optical fiber sensors”, First European Conference on Smart Structures and Materials, ECCM-1, Proc. SPIE 1777, Glasgow, May 12-14, 1992, pp.115-122

45. Jensen, D.W., Pascual, J., August, J.A., “Performance of a graphite/bismaleimide laminates with embedded optical fibers, part I: uniaxial tension”, Smart Materials and Structures, Vol. 1, 1992, pp. 24-30

46. Jensen, D.W., Pascual, J., August, J.A., “Performance of a graphite/bismaleimide laminates with embedded optical fibers, part II: uniaxial compression”, Smart Materials and Structures, Vol. 1, 1992, pp. 31-35

47. Susuki, I., Sindelar, P., Levin, K., ”Static and fatigue strength properties of graphite/epoxy laminates with embedded optical fibers”, FFA TN 1994-08, The Aeronautical Research Institute of Sweden, Bromma, 1994

48. Skontorp, A., ”Structural integrity of quasi-isotropic composite laminates with embedded optical fibers”, Journal of Reinforced Plastics and Composites, Vol. 19, No. 13, Sept. 2000, pp. 1056-1077

49. Skontorp, A., ”Effect of embedded optical fibers on the structural integrity of composites”, Proc. Twelfth International Conference on Composite Materials, ICCM-12, Paris, France, July 5-9, 1999, Paper 518

50. Guemes, A., Menendez, J.M., “Fatigue strength of glass reinforced polyester (GRP) laminates with embedded optical fibers”, Third International Conference on Intelligent Materials and Third European Conference on Smart Structures and Materials, ICIM-3/ECSSM-3, Proc. SPIE Vol. 2779, Lyon, France, June 3-5, 1996, pp. 217-221

51. Chang, C.C., Sirkis, J.S., Smith, B.T., ”Low velocity impact of optical fiber embedded laminated graphite/epoxy panels”, Proc. Fiber Optic Sensor-Based Smart Materials and Structures Workshop, Blacksburg, VA, USA, April 15-16, 1992, pp. 131-136

52. Sirkis, J.S., Chang, C.C., “Low velocity impact of optical fiber embedded laminated graphite /epoxy panels. I. Macro-Scale”, Journal of Composite Materials, Vol. 28, No. 14, 1994, pp. 1347-1370,


19

53. Sirkis, J.S., Chang, C.C., “Low velocity impact of optical fiber embedded laminated graphite /epoxy panels. II. Micro-scale”, Journal of Composite Materials, Vol. 28, No. 16, 1994, pp. 1532-1552

54. Jeon, B.J., Lee J.J., Kim J.K, Huh J.S., ”Low velocity impact and delamination buckling behavior of composite laminates with embedded optical fibers”, Smart Materials and Structures, Vol. 8, 1999, pp.41-48

55. Carman, G.P., Paul, C.A., Sendeckyj, G.P., “Transverse strength of composites containing optical fibers”, Smart Structures and Materials: Smart Structures and Intelligent System, Proc. SPIE Vol. 1917, 1993, pp. 307-316

56. Czarnek, R., Guo, Y.F., Bennett, K.D., Claus, R.O., ” Interferometric measurements of strain concentrations induced by an optical fiber embedded in a fiber reinforced composite”, Fiber Optic Smart Structures and Skins, Proc. SPIE 1170, Boston, MA, Sept. 8-9, 1988, pp. 43-54.

57. Singh, H., Sirkis, J.S., Dasgupta, A., ”Micromechanics of optical fiber sensors embedded in laminated smart structures”, Proc. 1991 SEM Spring Conference on Experimental Mechanics, Milwaukee, June 10-13, 1991, pp.137-144

58. Dasgupta, A., Sirkis, J.S., “Importance of coatings to optical fibre sensors embedded in ‘smart’ structures”, AIAA Journal, Vol. 30, 1992, pp.1337-43

59. Salehi, A., Tay, A., Wilson, D., Smith, D., ”Strain concentration factors around optical fibers by FEM and Moire’ interferometry”, Proc. Fifth AMS/ESD Advanced Composite Conference, Dearborne, Michigan, Sept., 1989, pp.11-19

60. Hadjiprocopiou, M., Reed, G.T., Hollaway, L., Thorne, A.M., ”Optimization of fibre coating properties for fiber optic smart structures”, Smart Materials and Structures, Vol. 5, Aug. 1996, pp. 441-448

61. Nilsson, S., “Stress analyses of embedded Bragg grating-sensors“, FFA TN 1996-57, Aeronautical Research Inst. of Sweden, Bromma, 1997

62. Sirkis, J.S., Dasgupta, A., “Optimal coatings for intelligent structure fiber optical sensors”, Fiber Optic Smart Structures and Skins III, Proc. SPIE Vol. 1370, 1990, pp. 129-140

63. Carman, G., Reifsnider, K., “Analytical optimization of coating properties for actuators and sensors”, Journal of Intelligent Material Systems and Structures, Vol. 4, No. 1, Jan. 1993, pp. 89-97

64. Kim, K.S., Kollar, L., Springer, G.S., ”A model of embedded fiber optic Fabry-Perot temperature and strain sensors”, Journal of Composite Materials, Vol. 27, No. 17, 1993, pp. 1618-1662

65. Kim, K.S., Ismail, Y., Springer, G.S., ”Measurements of strain and temperature with embedded intrinsic Fabry-Perot optical fiber sensors”, Journal of Composite Materials, Vol. 27, No. 17, 1993, pp.1663-1677

66. Case, S.W., Carman, G.P., “Optimization of fiber coatings for transverse performance: An experimental study”, Journal of Composite Materials, Vol. 28, No. 15, 1994, pp. 1452-1466

67. Van Steenkiste, R.J, Kollar, L.P., ”Calculation of the stresses and strains in embedded fiber optic sensors”, Journal of Composite Materials, Vol. 32, No. 18, 1998, pp. 1647-1679

68. Van Steenkiste, R.J., Kollar, L.P., ”Effect of the coating on the stresses and strains in an embedded fiber optic sensor”, Journal of Composite Materials, Vol. 32, No. 18, 1998, pp. 1680-1711

69. Itoh, T., Springer, G.S., “Strain measurement with microsensors”, Journal of Composite Materials, Vol. 31, No. 19, 1997, pp. 1944-1984

70. Matrat, J., Levin, K., Jarlås, R., ”Mechanical and optical behaviour of embedded Bragg grating sensors”, FFA TN 1999-15, The Aeronautical Research Institute of Sweden, Bromma, Sweden, Feb. 1999

71. Jang, T.S., Lee, J.J., Lee, D.C., Seo, D.C., Kim, H., “Fatigue behavior of optical fiber sensor embedded in smart composite structures”, Proc. 11th International Conference on Composite Materials, ICCM-11, Gold Coast, Australia, July 14-18, 1997, Vol. VI, pp. 582-589

72. DiFranca, C., Claus, R., Hellgeth, J.W., Ward, T.C., ”Structure/property correlation of several polyimide optical fiber coatings for embedding in an epoxy matrix”, Fiber Optic Smart Structures and Skins II, Proc. SPIE Vol. 1170, 1989, pp. 505-512


20

73. DiFranca, C., Claus, R., Hellgeth, J.W., Ward, T.C., ”Discussion of pullout tests of polyimide-coated optical fibers embedded in neat resin”, Proc. Fiber Optic Sensor-Based Smart Materials and Structures Workshop, IMSS, Blacksburg, VA, 1991, pp. 70-82


21

Durability of Embedded Fibre Optic Sensors in …8910/...Application of fibre optic sensors have...

Documents

Transcript of Durability of Embedded Fibre Optic Sensors in …8910/...Application of fibre optic sensors have...