·Table of Contents
·Methods and Instrumentation
Optical fibres for in situ monitoring the damage development in composites
L. Rippert, M. Wevers
Katholieke Universiteit Leuven, Faculty of Applied Sciences, Department of Metallurgy and Materials Engineering (MTM),W De Croylaan 2, B-3001 Leuven (Heverlee), Belgium.
S. Van Huffel
Katholieke Universiteit Leuven, Faculty of Applied Sciences, Department of Electrical Engineering (ESAT-SISTA/COSIC), K. Mercierlaan 94, B-3001 Leuven (Heverlee), Belgium.
The complex damage development process in composite materials demands a system that can continuously monitor the damage state in particular structural applications. Fibre optic sensors embedded in the composite material can offer an alternative for the robust piezoelectric transducers used for acoustic emission (AE) monitoring. For simplicity and robustness reasons, intensity-modulated optical fibres were chosen to detect damage in CFRP composite laminates based on the microbending concept.
Advanced signal processing techniques based on time-frequency analysis have been applied on the signals collected during loading of the CFRP composites. The Short Time Fourier Transform (STFT) has been computed and noise reduction algorithms (adaptive filtering and spectral subtraction filtering) have also been used. The transient signals being detected can be correlated with acoustic emission signals, analysed by a classical AE parameter study and with a modal acoustic emission (MAE) system. The signals are attributed to the initiation of damage in the materials and thus the optical signal contains, besides the level of overall strain, information in the elastic energy released whenever damage is introduced in the host composite. Tests on the sensitivity of this in situ technique have also been performed and will be documented; the distance between the optical fibres to be embedded in a real structure will depend upon this.
This research shows the possibility and advantages of embedded optical fibres to monitor in situ the damage development in composite materials.
The emergence of optical fibre communications technologies in the 1970's has enabled the development of embedded optical sensors for process condition monitoring and for smart materials/structures applications. Meanwhile, thanks to the evolution in computer technology, powerful data analysis tools appeared. With optical fibres embedded in composite materials and advanced data processing of the optical fibre signals, a non destructive testing (NDT) system can be integrated into this complex material, component or structure similar to the neural system in a human body. In this approach fibre optic sensors  will offer an alternative for the robust piezoelectric transducers used for acoustic emission (AE) monitoring.
Indeed fibre optic sensors have several advantages compared to the electronically based sensors like piezoceramics such as light-weight, all passive configurations, low power utilisation, immunity to electromagnetic interference, high sensibility and bandwidth, compatibility with optical data transmission and processing, long lifetimes and low cost (as long as using silicon fibres). The difficulties to overcome to obtain reliable data are the complex signal processing and the embedding procedure of the optical fibre (influence on the host material properties and connections with the outside world) [2, 3].
Different kinds of optical fibre sensors can be used: intensity-modulated sensors, phase-modulated sensors (interferometers), and Bragg grating sensors.
Different configurations for phase-modulated sensors can be set namely the Michelson interferometer, the Mach-Zenhder interferometer and the Fabry-Perot interferometer. They have been successfully used for acoustic wave detection [4, 5, 6] and damage detection in various materials  and in particular composite materials [4, 8, 9]. Phase-modulated sensors are usually very sensitive but also very complex and fragile.
Bragg grating sensors are capable of very accurately measuring physical quantities like strain and temperature.
Intensity-modulated sensors detect variations in the intensity of the transmitted light caused by a perturbing environment. The main causes for intensity modulation are transmission, reflection and microbending. The major limitation for these sensors is that any intensity fluctuations in the output not associated with the measurand produce erroneous results. Various schemes can be used to self-reference sensors and so correct this problem . Several intensity-modulated sensors have been successfully used to measure damage but they usually rely on the optical fibre fracture [11, 12]. The first intensity-modulated sensors developed, used the microbending concept to detect pressure, acceleration, displacement, temperature and strain [10, 13, 14, 15]. Intensity-modulated optical fibre sensors require only a simple and robust sensing system.
In this study, it will be shown that the optical signal, collected from an intensity-modulated sensor based on the microbending concept, contains information not only on the strains in the composite due to remote loading but also on the elastic energy and hence strain released whenever suddenly damage is being introduced in the host material. Advanced signal processing techniques based on time-frequency analysis are applied to obtain these results which are compared with those obtained from an acoustic emission monitoring system.
2- Principle of operation
In the microbending concept (see figure 2.1), the physical property to be measured is converted into a displacement which bends the fibre at certain locations. If an optical fibre is bent, small amounts of light are lost through the cladding because the condition of total reflection is violated. The amount of intensity loss depends on the amount of bending and thus on the amount of displacement . The stress field in a damaged layer is not the same as in an undamaged layer. This may cause the optical fibre to bend in the material. So the initiation of damage should correspond to a decrease in the intensity of the transmitted light as it was shown by M. Surgeon .
Fig 2.1: The microbending concept.
This concept can be explored further. Damage can also be characterised by mechanical waves propagating in the material. When a wave hits an optical fibre, the stress bends it locally and so some light might also be lost (coupling between propagating and radiation modes). A high sampling rate (SR) can be used to detect transient signals which could be stress waves released by matrix cracking, delamination or fibre fracture phenomena. This requires signal analysis tools such as filtering, time analysis and time-frequency analysis (STFT). This paper will focus mainly on the signal analysis tools.
An HeNe laser source is used with a multimode optical fibre embedded in a carbon fibre reinforced composite material. The output signal light intensity is collected by a photo-diode, and is sent to a computer via an oscilloscope card as seen in figure 3.1. The signal treatment was done using MATLABâ
software. The Short Time Fourier Transform (STFT) is computed, time analysis and time-frequency analysis are performed.
Fig 3.1: The sensing system
An Acoustic Emission (AE) system was also used to detect damage inside the material: Wave Explorer from Digital Wave Corp. It needs broadband sensors (Digital Wave B1025) with a nearly flat frequency response in the 50 - 3000 kHz frequency range. It uses the plate wave theory as a theoretical background and analyses the waves according to their mechanical nature, namely extensional and flexural waves. It is called a Modal Acoustic Emission (MAE) system [17,18].
MAE allows a more convenient way to identify damage mode by looking at the frequency content of the acoustic waves produced by damage initiation. The presence (or absence) of extensional and flexural modes is the key to the damage mode characterisation. It has been proven to work in an efficient way for matrix cracking and fibre fracture . It also allows a clear recognition of noise grip and EMI.
A multimode optical fibre (FIP100110125 from Polymicro Technology Inc.) was connected to a 10,9 mW laser source (LGK 7654-7 from LASOS) via an optical coupler (HUC 13-633-M-2.6 GR-2 from OZ OPTICS Ltd.). The optical fibre diameters were 100 µm for the core, 110 µm for the cladding and 125 µm for the polyimide coating. The optical fibre characteristics have been chosen to maximise the light intensity and the strain/stress transfer from the composite material to the optical fibre, and to minimise the influence of the optical fibre embedment [20, 21, 22, 23]. The small diameter difference between the core and the cladding increased the loss of light due to bending. Great care was also taken with the different optical connections.
Laminates were produced from a Vicotex 6376/35/137/T400 C/epoxy prepreg. The prepreg was cut and stacked into a [02°, 904°]s lay-up. The optical fibre was embedded in the 90° direction in the middle plane of the specimen. A polymeric bore tube was put around the optical fibre at its exit point from the composite specimen. It shrank around the fibre during the cure and so protected this weak point. The samples were cut from the plates at a length of 150 mm, a width of 25 mm and a thickness of 1.2 mm.
The optical signal was collected in a photodiode (D-series pin photodiode from UDT Sensor Inc.), further amplified (UDT-1200A Medium Speed Amplifier from UDT Sensor Inc.) and sent to an oscilloscope card (DAS-1800AO board from Keithley) with a 12 bits A/D Converter.
Three channels were used; the first one to collect the optical signal, the second one to collect the optical signal that has been filtered via a capacitor to get only the AC component and amplified by a 11.5 factor. The last one collects a trigger signal sent by the AE system each time an acoustic event is detected.
The sampling rate (SR) was set to 10 kHz.
The optical signal post-processing was done on a SUN workstation using MATLABâ
software, in particular the Signal Processing Toolbox. A program was written to filter the signal, to compute its STFT and visualise the changes in its power spectrum over time. Additional tools were developed to extract damage related information from this time-frequency analysis.
Tensile tests have been performed on the 4505 Instron testing machine with a 100 kN loadcell. To be certain that the specimens didn't slip in the grips, aluminium end tabs were bonded to the specimens using a two components Araldite 2011 epoxy glue. The Instron machine was operated at 0.5 mm/min displacement rate. A homemade Labview program drove the tensile machine and the oscilloscope card. The load was applied continuously.
4- Results and discussion
Ten specimens were tested. At the beginning of every tensile test, pencil lead break tests were also performed for calibration purpose.
4.1 Low pass filtering
The figure 4.1 shows the loading curve got from the Instron tensile machine. From 0 to 90 seconds, several pencil lead break tests were performed to calibrate the AE system. Then, the loading was applied until the specimen final fracture. After 261 seconds of test, there was a sudden decrease in the applied strain due to some damage near one of the aluminium tab.
Fig 4.1: The loading vs. time
Fig 4.2: The optical signal filtered with a low pass Butterworth filter
The figure 4.2 shows the optical signal for the same test. A 4th order low pass (LP) Butterworth filter has been applied with a 5 Hz cut-off frequency. The curve has been reversed for better comparison with the preceding one. The optical intensity starts decreasing (see as an increase in the figure) when the loading starts to be applied. A low frequency oscillation can be seen on the curve and comes from vibrations produced by the Instron machine. The final fracture is clearly seen and the strain release at time 261 seconds also appears as a change of slope on the curve. It is thus shown that the optical signal contains information on the external strain.
Some sharp spikes can also be seen on figure 4.2. To see them more clearly, the same low pass filter was applied to the AC component of the optical signal (amplified by an 11.5 factor). As can be seen on figure 4.3 the signal is constant with some big spikes. The time instant at which those spikes appear is the same as the time of occurrence of some AE events.
Fig 4.3: The AC component of optical signal filtered with a low pass Butterworth filter |
These AE events can be related to damage inside the material, so those spikes can also be related with damage.
The main limitation is the sensitivity; the events detected by this method are only the most energetic ones. The less energetic events detected by AE can not be seen on the optical signal or are smaller than the curve oscillations and then can not be detected by the threshold technique. Removing the polyimide coating from the optical fibre may increase the system sensitivity.
Most of the high-frequency components are filtered out so the signal can not be used for damage identification. Another drawback of low pass filtering is that changes are less sharp and some delay may also appear. Figure 4.4 shows the optical signal (whole signal and AC component) for a 1-second time window. The AE system detects an event during this period. The lower curve shows the filtered signals using a LP Butterworth filter (25 Hz cut-off frequency). The signal is sharp enough to allow a good time location but damage identification is still not possible.
Fig 4.4: The optical signal and the amplified AC optical signal (top left and right) after a 25 Hz LP Butterworth filter (bottom). The AE system detects an event in this time window (middle).
This method shows that damage detection is possible with intensity modulated optical sensors based on the microbending concept, but some more advanced signal analysis techniques are required to proceed further.
4.2 Noise reduction
We look for small effects on the optical signal. So, to increase the Signal to Noise Ratio (SNR) several steps were taken.
The biggest noise source is from the laser power supply (50 Hz and the harmonics from the net). An adaptive filter (LMS method, 2 taps) was used to remove this noise up to 1 kHz . Analysing this filter was time consuming (computation time). Therefore, since this noise above 1 kHz is small enough we stopped the filtering at this frequency.
Then, spectral subtraction, a filtering technique used in speech processing , was applied to see more clearly the expected 'optical events'. This technique requires that the background noise environment remains locally stationary to the degree that its expected spectral magnitude value just prior to an expected event equals its expected value during the event. It is also assumed that significant noise reduction is possible by removing the effect of noise from the magnitude spectrum only.
Fig 4.5: The optical signal during four successive pencil break tests on the surface of the specimen: the original signal (top) was filtered with an adaptive filter (middle) and the spectral subtraction method (bottom).
These techniques were applied to pencil lead break tests (figure 4.5) and to tensile tests (figure 4.6). They allow a good noise reduction without reshaping too much the signal. New kinds of events (not detected with the low pass filtering technique) can be sensed (figure 4.7). They also correspond to AE events linked to damage.
Fig 4.6: The optical signal (left) and the AC component (right) for the same time window as in figure 4.4: the original signal (top) was filtered with an adaptive filter (middle) and the spectral subtraction method (bottom).
Fig 4.7: The optical signal (left) and the AC component (right): the original signal (top) was filtered with an adaptive filter (middle) and the spectral subtraction method (bottom).
4.3 Time-frequency analysis (STFT) and damage identification
In this study the signal was non-stationary (and transient signals were also expected), so its frequency content, visualised by the Fourier transform, varied in time. The STFT can be used to visualise the frequency content of a signal over time.
Fig 4.8: The optical signal and the STFT corresponding to a pencil break test event.
Modal acoustic emission analyses acoustic waves according to their nature, namely flexural and longitudinal waves. The signal spectra show some wave packages in the low frequency range. According to M. Surgeon and al. [26, 27], a flexural mode is typically found below 140 kHz and an extensional mode between 400 - 800 kHz.
Waves produced by a pencil lead break test are quite similar to those produced when damage occurs in a composite material. Time-frequency analysis has been performed on pencil break event (figure 4.8) and on real damage events (figure 4.9 and 4.10). These signals show some waves packages in the very low frequency range (in a similar way to what can be found with modal acoustic emission). These events can be clearly localised and characterised in the time domain and in the frequency domain. but no truly conclusive results could be produced so far to correlate the waves packages with the kind of damage.
Fig 4.9: The optical signal and the STFT corresponding to a real damage event.
Fig 4.10: The optical signal and the STFT corresponding to a real damage event.
It has been shown that an intensity modulated optical sensor based on the microbending concept can be used for continuous damage monitoring. It is simple and robust but requires some advanced signal analysis tools like adaptive filtering, spectral subtraction filtering, and time-frequency analysis (STFT).
The principle has been proven to work. The sensor can detect damage initiation and characterise its frequency content. The similarities between optical and MAE signals should permit damage identification. To achieve this, another technique (like edge replica) may also be useful for better damage characterisation.
It will also be required to use other signal analysis tools. For instance, the STFT is not the best tool to study low frequency transient phenomena. The Wavelet Toolbox might offer an alternative.
The authors would like to thank Ing. J. Vanhulst for his precious help with the data acquisition system, and K. Eneman for providing the code for the applied filtering techniques.
S. Van Huffel and M. Wevers are senior research associates of the F.W.O. (Fund for Scientific Research -- Flanders).This work was supported by the F.W.O. Project no. G.0200.00, by the Belgian Programme on Interuniversity Poles of Attraction (IUAP-4/2 \& 24), initiated by the Belgian State, Prime Minister's Office for Science, and by a Concerted Research Action (GOA) project of the Flemish Community, entitled ``Mathematical Engineering for Information and Communication Systems technology''.
- B. CULSHAW, J. DAKIN, "Optical fibre sensors: systems and applications", vol. II, 1989, pp. 727-745.
- D.A. KROHN, "Fibre optic sensors: Fundamental and applications", Instrument Society of America, 1988, pp. 1-61.
- M.V. GANDHI, B.S. THOMPSON, "Smart materials and structures", 1992, pp. 217-288.
- R.M. MEASURES, "Advances toward fiber optic based smart structures", Optical engineering, vol. 31, nb. 1, January 1992, pp. 34-47.
- I.B. KWON, C.G. KIM, C.S. HONG, "Simultaneous sensing of the strain and points of failure in composite beams with an embedded fiber optic Michelson sensor", Composite Science and Technology, vol. 57, 1997, pp 1639-1651.
- V. BHATIA, C.A. SCHMID, K.A. MURPHY, R.O. CLAUS, T.A. TRAN, J.A. GREENE, M.S. MILLER, "Optical fiber sensing technique for edge-induced and internal delamination detection in composites", Smart Mater. Struct., vol. 4, 1995, pp 164-169.
- T.A. TRAN, W.V. MILLER III, K.A. MURPHY, A.M. VENGSARKAR, R.O. CLAUS, "Stabilized extrinsic fiber optic Fabry-Perot sensor for surface acoustic wave detection", proceeding of SPIE vol. 1584, Fiber optic and laser sensors IX, 1991, pp 178-186.
- R.O. CLAUS, M.F. GUNTHER, A. WANG, K.A. MURPHY, "Extrinsic Fabry-Perot sensor for strain and crack opening displacement measurements from -200 to 900°C", Smart Mater. Struct. vol. 1, 1992, pp 237-242.
- H. TSUDA, T. IKEGUCHI, J. TAKAHASHI, K. KEMMOCHI, "Damage monitoring of carbon-fiber-reinforced plastics using Michelson interferometric fiber-optic sensors", Journal of materials science letters, vol. 17, 1998, pp 503-506.
- J.W. BERTHOLD, "Historical review of microbend fiber-optic sensors", J. Lightwave technology, vol. 13, nb. 7, July 1995, pp 1193-1199.
- N.D.W. GLOSSOP, S. DUBOIS, W. TSAW, M. LEBLANC, J. LYMER, R.M. MEASURES, R.C. TENNYSON, "Optical fibre damage detection for an aircraft composite leading edge", Composites, Vol. 21, Number 1, January 1990.
- S.R. WAITE, G.N. SAGE, "The failure of optical fibres embedded in composite materials", Composites, vol.19 number 4, July 1988.
- C.M. LAWSON, V.J. TEKIPPE, "Environmentally insensitive diaphragm reflectance pressure transducer", Proceeding of SPIE, vol. 412, 1983, pp96-103.
- J.W. BERTHOLD, W.L. GHERING, D. VARSHNEYA, "Design and characterization of a high temperature fiber-optic pressure transducer", J. Lightwave technology, vol. 5, 1987, pp 870-876.
- S.K. YAO, C.K. ASAWA, "Microbending fiber optic sensing", Proceeding of SPIE, vol. 412, 1983, pp 9-13.
- M. SURGEON, "Continuous damage monitoring techniques for laminated CFRP composite material", Ph.D. thesis, KU Leuven, 1999.
- S. ZIOLA, I. SEARLE, "Automated source identification using modal acoustic emission" Review of Progress in Quantitative Non-destructive Evaluation, 1997, Vol. 16A, pp. 413-419.
- M.R. GORMAN, "Progress in detecting transverse matrix cracking using modal acoustic emission", Review of Progress in Quantitative Non-destructive Evaluation, 1998, Vol. 17A, pp. 557-564.
- M. SURGEON, M. WEVERS, "Continuous damage monitoring techniques for laminated composite materials", Proceeding of ICCM-12 Europe Conference, July 5th - 9th 1999, Paris.
- N.C. EATON, M.J. CURRAN, J.P. DAKIN and H. GEIGER, "Factors affecting the embedding of optical fibre sensors in advanced composite structures", AGARD Conference Proceedings 531, 75th Meeting, Oct. 5-7, 1992, pp. 20.1- 20.14.
- S.S.J. ROBERTS, R. DAVIDSON and R. PAA, "Mechanical Properties of Composite Materials Containing Embedded Fibre-Optic sensors", Fibre Optics Smart Structures and Skin IV; Proceedings of the Meeting, Boston, MA, Sept. 5-6, 1991, Bellingham, WA, SPIE vol. 1588, pp. 326-341.
- D.W. JENSEN, J. PASCUAL, J.A. AUGUST, "Performance of graphite/bismaleimide laminates with embedded optical fibres. Part I: uniaxial tension, Part II: uniaxial compression", Smart Material Structure, 1992, vol. 1, pp. 24-30, pp 31-35.
- L COENEN, "The influence of embedding optical fibres on the mechanical properties of a carbon fibre composite and monitoring the damage development using the acoustic emission technique", Master thesis, KU Leuven ,1995.
- M. SURGEON, M. WEVERS, "Static and dynamic testing of a quasi-isotropic composite with embedded optical fibres", Composites Part A, 1999, vol. 30, N°4, pp 317-324.
- M. WEVERS, "Listening to the sound of materials: Acoustic emission for the analysis of material behaviour", NDT&E International, Vol. 30, N°2, pp99-106, 1997.
- J.G. PROAKIS, D.G. MANOLAKIS, "Digital signal processing: principles, algorithms, and applications", 3rd ed., Prentice Hall, Upper Saddle River, NJ, 1996.
- S.F. BOLL, "Suppression of acoustic noise in speech using spectral subtraction", IEEE Transactions on acoustics, speech, and signal processing, vol. ASSP-27, N° 2, April 1979.