The 10 th International Conference of the Slovenian Society for Non-Destructive Testing»Application of Contemporary Non-Destructive Testing in Engineering«September 1-3, 2009, Ljubljana, Slovenia, 47-57 ULTRASONIC CHARACTERIZATION OF ADVANCED COMPOSITE MATERIALS B. Boro Djordjevic Materials and Sensors Technologies, Inc. 798 Cromwell Park Drive; Suite C; Glen Burnie, MD 21061 USA bbd@mast-inc.com ABSTRACT With increased use of composite materials in critical structural applications it is more important than ever to independently assure structural integrity. Complexity of the advanced composite materials including layered and bonded structures represents challenges in developing optimized ultrasonic tests. Traditional ultrasonic NDT methods are inappropriate and often misleading when applied to anisotropic and nonhomogeneous composite materials. In advanced technology applications such as aerospace and with industrial emphasis on economics and safety, it is critical to develop robust and practical NDT mehtods. Recent analytical and experimental work with ultrasonic guided wave methods using laser generation and minature ultrasonic receivers enables novel automated and rapid testing of these advanced composite structures.by optimizing and customizing ultrasonic trunsduction proces, measurments of material properties and defect conditions can can be estimated along in-plane sound propagation path. From acoustic response, one can develop meaningful estimate of material modulus, fatigue damge, thermal damge and sense mechnical defect conditions without need to pointvise scan the complete structure. Key words: NDE, ultrasonic, composites, guided waves, structural integrity, laser UT 1. Introduction Composite materials structural integrity can be compromised via many mechanisms including presence of discontinuities or loss of mechanical properties. Ultrasonic methods are directly sensitive to these changes and can be used to assess the integrity of the composite structure.with majority of ultrasonic methods based on and originated from metals experience; the current test procedures are inadequately developed to directly tackle the composite structural issues. Because of composite materials complexity, complexity of the part geometry and often a limited part access, materials damage and materials condition sensing cannot be achieved via conventional ultrasonic methodology. The composite testing requires fresh look at availability of advanced ultrasonic methods better adapted to the unique composite material requirements. 47
Composite mechanical damage is typically in the form of delaminations or disbonds (laminate-tolaminate or laminate-to-core), broken fibers due to impact, fatigue damage that affects the zone of composite material via micro cracking, fiber delaminations, fiber breaks and overall loss of mechanical modulus, or can be caused by thermal damage from prolonged exposure to heat above resin cure temperatures as well as combination of effects due to extreme operational conditions. The detection and evaluation of damage in composites is compounded by the fact that damage is not visible to the naked eye and can occur in many different forms. Methodology used for the guided wave ultrasonic measurements is a new technology that has ability to nondestructively inspect(ndi), evaluate(nde) and characterize(ndc) composite parts integrity not achievable using conventional transducers and procedures. The advanced ultrasonic sensing, sound generation and data acquisition are now available to capture the ultrasonic signal information without requiring point-by-point ultrasonic testing of the composite material. Although not intuitive, signal processing of guided waves is validated technical approach in many remote sensing applications for characterization of targets. Tests were performed on range of solid composites and on the honeycomb structures with a variety of defects and mechanical damage conditions The methodology explores new guided wave laser ultrasonic sources and small aperture receiving sensors configurations that enable testing of advanced composites far beyond conventional ultrasonic capabilities. In contrast to pseudo-imaging methodology such as ultrasonic C-scan, guide wave response to in plane mechanical state of the composite is analyzed in manner similar to processing of sonar signals. This emerging technology explores the ability of the inhomogeneous, anisotropic composite material to propagate ultrasonic guided waves over a range of distances and part configurations. Capturing high fidelity complete ultrasonic waveforms traversing composite material enables automated signal analysis to deduce the material structural condition. 2. Ultrasonic Testing Background Since introduction of bonding and composites into the airframe structures, extensive Nondestructive Inspection (NDI) is often performed using point by point ultrasonic tests that are often automated using complex scanning machinery. Y X C-Scan Fig. 1: Conventional ultrasonic inspection data collection methods require complete part transducer scanning path as in ultrasonic C-scan test. 48
Conventional ultrasonic inspection measurement using compressional waves is usually limited to a single point measurement across the thickness of the composite material and component coverage is achieved via scanning. Ultrasonic C-scan imaging scan process as shown in Fig. 1 is dominant inspection tool in many aerospace composite manufacturing. Modern computerized C-scan has enormous advantages over manual testing in reproducibility and presentation of the ultrasonic tests. However, even for manufacturing subassemblies, ultrasonic C-scan, is often difficult to execute and is not practical in many complex structural configurations such as the leading edges or other complex sections of the flight control surfaces. Successfully implemented for the manufacturing acceptance of large composite wing sections that could be solid, honeycomb or other layered construction, ultrasonic C-scan is very difficult to adapt for field testing of the complex composite parts and composie components assemblies. 2.1 Guided Wave Background Guided ultrasonic waves in the structures and in modern materials such as composites are directly influenced by mechanical condition of the material. Sensing changes in the guided wave propagation characteristics enables direct in plane assessment of the mechanical condition of the material. Ultrasonic test configurations based on traditional ultrasonic transducers and instrumentation cannot observed changes in the guided wave modes due to composite damage and cannot be implemented in complex composite parts [1,2]. Sensing composite mechanical degradation and structural integrity requires a fresh look at the advanced ultrasonic sensing and measurement analysis methods. Guided wave ultrasonic tests includes vide range of mechanical stress wave conditions including: surface waves plate waves Love waves Stonely waves Bulk guided waves and others. Generation of guided waves in composites is best achieved via non-contact laser ultrasonic sources. Fig. 2 illustrates an impulse excitation of a laser ultrasonic source on the surface of the material. Such impulse source generates longitudinal (particle motion parallel to the propagation direction) and shear (particle motion perpendicular to the propagation direction) waves. Head-waves are derived from the longitudinal waves that travel on the surface. Away from the surface, head-wave converts to shear and travel at the shear wave velocity. Rayleigh waves propagate along the solid surface and their wave motion is confined to a region near the surface whose depth is comparable to the wavelength. Surface waves are suitable for detection of near-surface defects. In structures with finite geometry boundaries, particle motion parallel to the propagation direction and particle motion perpendicular to the propagation direction can undergo series of reflections creating complex motion guided waves In plate like geometries, multiple reflection and mode conversion at the plate surface boundary led to Lamb wave generation. Key differences between longitudinal and shear waves (so-called bulk) and surface and Lamb waves (so-called guided) are in media geometry where bulk waves exist in infinite media (wavelength is much smaller than the sample boundaries), whereas guided waves existence requires a boundary (wavelength is comparable to the sample boundaries). In almost all cases, composite structures readily support 49
guided waves although they are not used for inspection purposes. Ultrasonic testing of the structure using guided wavesin in the X/Y plane is at lower frequencies than it is required for (Z-axis) through thicknes longitudinal wave testing. Capture and interpretation of the guided wave propagating signals is more complex but readily processed via modern computers. For example, according to their displacement patterns, Lamb waves are divided in two classes: symmetric (S n ) and antisymmetric (A n ) with respect to the middle of the plate. Each class contains many order modes. Lamb modes are dispersive and phase velocity (c p ) of each mode is a function of a given material elastic constants and a direction of travel in the composites materials. The structure of the Lamb waves in the thickness is different for different modes, and for the same mode it changes as the product between ultrasonic frequency (f) and plate thickness (d). A 0 mode has bigger out-of-plane displacements (along the z direction), whereas S 0 mode has bigger displacements in the plane of the plate (along the x direction). Surface wave Surface source Head wavefront Shear wavefront Longitudinal wavefront Fig. 2: Waves generated by a surface laser impulse source incident on an infinite half-space solid. Fig. 3: Dispersion modes for the plate guided wave. Both A 0 and S 0 plate modes at lower frequencies contain zones of realitivly constant velocity over range of frequencies. 50
Figure 3 is example of possible guided wave dispersions curves in the plate like composite material. It should be noted that A 0 and S 0 modes at lower frequencies have flat zones where velocity of the mode is not strong function of the frequency or thicknes. 3.0 Guided Wave Technology For the composite materials, one must approach the guided wave testing from the fundamental engineering principles that are not biased from traditional materials NDI practices. [1,2] Composite materials exhibit anisotropic and inhomogeneous character. Ultrasonic testing using guided waves is in composite plane (XY direction) and at lower frequencies. Full interpretation of the propagating signals is more complex than traditional assumption for the bulk longitudinal waves. The capability to enable reproducable guided wave measurments requires application of formed laser acoustical sources enabling generation of plate like waves in layered composites structure or surface waves (Rayleigh waves) that cannot be generated using conventional contact transducers [3-7]. Receiving sensor technology is critical to enable the guided wave testing. Laser-optical ultrasonic receivers would be ideal but becouse of complexity and cost, in practiacal tests, we used small aperture unconventional piezoelectric and advanced line element transducers adapted for the guided wave test configurations. Miniature transducer technology used for the guided wave testing shown in Fig. 4 is available but has not been used for the composite NDI. The unique configuration advantage of guided wave inspection is that single test enables the inspection of the entire cross-section of the material along the zone between the source and the receiver transducers as illustrated in Fig. 5. Guided plate wave tests allow inspection of a much larger region, can extend to curved surfaces regions and can effectively test the structure with complex geometries that cannot be examined by the conventional point tests. [8,9] Fig. 4: Miniature transducer receivers adaptable to a guided wave testing of the composites. These transducers were custom build for various advanced ultrasonic NDE applications. Ruler scale in the picture is in inches. 51
Y Guided Wave Zone X Fig. 5: Transducer to transducer zone inspection using guided wave. Distance between transducer can be from one cm to many cm A guided wave signal recorded from a single test as shown in Figure 5, supported with advanced signal capture and modeled response interpretation allows assessments of the structural and material integrity between the test points. Test configuration can be in bi-static (separate transmitter and receiver points) or mono-static (pulse echo with transmitter and receiver collocated). Presence of cracks, fatigue damage, cross-section change, presence of delaminations or disbonds in composites, material change such as thermal damage in composites and other defects or material conditions modulate propagating stress wave and are detected without the need to point-wise scan the complete part. Analysis of ultrasonic signals often as simple as attenuation or mode conversion changes enables deduction of the material condition between the test points. Fig. 6: Guided wave ultrasonic test configuration with laser source illustrating test gauge zone. In comparison to traditional point-wise scanning ultrasonic tests, guided wave ultrasonic methods have: significant faster area coverage, ability for remote sensing of the defects, ability to inspect curved areas, ability to directly sense materials modulus changes, and include significantly increased volume (area) coverage in a single test. The experimental results demonstrate that such testing techniques can perform the inspection from only one side of the structure. 52
Schematic of the guided wave ultrasonic test and data acquisition system is shown in Figure 6 This system is based on laser ultrasonic excitation enabeling non-contact reproucable acoustical source and reproducible measurement of ultrasonic signals time 0 trigger point to better than 5 ns. This unique feature, combined with modern analog to digital conversion, enables very accuret measurment of ultrasonic wavefront travel times. Table 1 summarizes response of the guided wave signals to a range of defects and material property characteristics. Because of the hi resolution time zero ultrasonic signal trigger information, the methodology is sutable to materials chractrization via accurate sound velocity measurments. Table 1: Example of guided wave test response to composite material damage or degradations. Defect/property Characteristics Change in Material Mechanical Property Porosity Wrinkle Heat damage Microcrack Impact damage Delaminations Guided Wave Response Sensitivity Correlation of the square of measured velocity to modulus Velocity is slowed by porosity, frequency shift at certain wavelengths Frequency shift of signals through wrinkle locations Velocity is slowed by heat damage, sever damage induces mode changes and signal attenuation Velocity is slowed and sound is scattered at specific frequencies by microcracks Velocity is slowed and guided modes changed in impact damage regions, signal level attenuated Cause mode conversions and signal attenuation Fig. 7: Comparison of guided wave signals captured via small aperture surface mounted sensors and the conventional wedge ultrasonic transducer. 53
Figure 7 shows guided wave signals captured in the composite plate using conventional wedge receivieng transducers and minature small aperture sensors. The minature and pin transducer aperture is les than 1/10 wavelength of the signals while conventional wedge covers many wavelenghts and creates complex part to transducer interface. It is evident that complex geometry wedges cannot capture true guided wave signature. Small aperture sensors ( Pin and minature transducers) capture relative surface motion component of the guided modes. The wedge transducer complex geometry spatialy and frequency filters the true signature of the waveforms. This effect has been well demonstrated in the work with acoustic emission snesors where conventional piezoelectric transducers could not capture true acoustic emission waveforms. 4.0 Guided Wave Measurements Examples of guided wave results are shown in Figure 8 for the heat damage sample showing the waveform and a wavelet transform. The wavelet transform provides a quantitative interpretation of the guided wave for time (velocity) and frequency [10]. Fig. 8: Example of velocity (time) and frequency shift (wavelet transform) of guided waves in undamaged and thermal damaged regions. Similar detection capability was demonstrate using Lamb waves for the lightning composite damage, Fig 9. Thermally damaged Graphite/epoxy composite material from the simulated lightning strike did not support Lamb wave propagation. Undamaged sections of the panel support 54
strong and distinct Lamb wave in all directions Fig. 4.2a while damaged section of the composite cannot propagate the signals Fig. 4.2b 40 30 (a) Amplitude [mv] 20 10 0-10 -20-30 -40 80 90 100 110 120 130 140 150 160 170 180 Time [ s] Amplitude [mv] 40 30 20 10 0-10 (b) -20-30 -40 80 100 120 140 160 180 Time [ s] Fig. 9 : Diagram of the guided wave test locations for the signal acquired in a damage-free zone (a) and generated on the defect (b). Lightning damaged area cannot support a Lamb wave mode. The ultrasonic tests were performed on the back side of the panel that does not show the damage. Figure 10 is a photograph of the rotary set up to mesure composite directional velocities. The simple test configuration enables very accurate sound velocity measurment in XY plane of the composite plate. Fig. 10: Rotary table enbeling measurents of directional velocities in XY plane of composite. Figure 11 illustrates such measurements for the isotropic and anisotropic composite lay-up configurations. The guided wave velocities can be measured at high angular resolution to characterize composite directional properties. Such directional tests are almost impossible via 55
mechanical test methods. These velocity measurements can be converted to estimated directional elastic modulus of the composite material. In its simplest form, velocity (Vxy) is related to modulus (Exy) and density ( ) by relation: Vxy=K (Exy/ ) 1/2 (1) Fig. 11: Ultrasonic velocities as function of angle for the XY plane of the composite. Thus, simple one-sided guided wave test as shown in Fig. 4.3 enabeling a capture of waveguide signals as shown in Fig 11 provides all the information needed to directly estimate composite in plane modulus. 5. Conclussions Ultrasonic guided wave propagation in composites enables unique test measurments and must be considered for the advanced chracterization of the composite structural integrity. Laser generation, and sub-wavelength aperture transducer sensing is critical. Hybrid formed laser ultrasonic source combined with small aperture receivers enable capture of the complete guided wave signals for a practical experimental mesuremnts. A new methodology approach is feasible for factory and field implementation. The complex composite parts tests can be supported by analysis of acoustical 56
signals and guided wave tests can cover large sections of composite material without a need for the extensive point-to-point scanning. In plane guided wave velocity measurments can directly sense mechanical modulus of composite material. Experimental test confirmed the utility of the composite material properties sensing. 6. References [1] B.B. Djordjevic, Henrique Reis, editors, G. Birnbaum, B. A. Auld, Technical editors, "Sensors for Materials Characterization, Processing, and Manufacturing" ASNT Topics on NDE Vol. 1, published by ASNT, Columbus OH, 1998. [2] B.B. Djordjevic, "Advanced Ultrasonic Probes for Scanning of Large Structures", Proc. Ultrasonic International 93 Vienna, Austria, July 1993, Pub. Butterworth Heinemann 1993 [3] B. Boro Djordjevic, Dr. R. E. Green, Jr., D. Cerniglia, K. Y. Jhang Remote Non-contact Ultrasonic Testing of Aircraft Joints 2000 USAF Aircraft Structural Integrity Program Conference, 5-7 December 2000 San Antonio, TX [4] B. Boro Djordjevic, D. Cerniglia, Remote Non-contact Testing of Aircraft Structures 2001 USAF Aircraft Structural Integrity Program Conference, 10-13 December 2001, Williamsburg VA [5] J. L. Rose, K. M. Rajana, and M. K. T. Hansch, Ultrasonic guided waves for NDE of adhesively bonded structures, J. Adhesion 50, 71-82 (1995). [6] D. Cerniglia, K. Y. Jhang, and B. B. Djordjevic Non-Contact Ultrasonic testing of Aircraft Lap Joints 15 th Worl Conference on NDT,, Rome, Italy, Editor AIPnD, NDT.net October 2000 [7] M. J. S. Lowe, and P. Cawley, The applicability of plate wave techniques for the inspection of adhesive and diffusion bonded joints, J. Nondestr. Eval. 13, 185-200 (1994). [8] D. N. Alleyne, and P. Cawley, Optimization of lamb wave inspection techniques, NDT & E Int. 25, 11-22 (1992). [9] S. G. Pierce, B. Culshaw, W. R. Philp, F. Lecuyer, and R. Farlow, Broadband lamb wave measurements in aluminum and carbon/glass fibre reinforced composite materials using noncontacting laser generation and detection, Ultrasonics 35, 105-114 (1997). [10] D. Cerniglia, B. B. Djordjevic, Analysis of laser-generated lamb waves with wavelet transform Nondestructive Characterization of Materials XI, Berlin, Germany. Jun. 24-28, 2002, Springer, 2003 pp 63-67 57