Accelerating MRO procedures for composite materials using innovative detection techniques
In the past three decades composite materials have gained an increased popularity, especially within the aerospace sector. In this time-frame, innovation moved slowly away from the aerospace industry towards other sectors such as the information technology sector and the renewable energy sector. Aerospace innovation slowed down - nowadays other sectors are even surpassing the aerospace industry with the use of composite materials, such as the wind energy sector, the automotive sector (exclusive sports cars, and nowadays even for the general consumer market products) and the civil engineering sector (composite bridges). One of the factors in this, is the strict policy regarding certification for procedures and the use of materials and systems. This factor however can also be used in favor of innovation, by applying a consistent and long term research and development policy. One of the main technological bottlenecks for the use of composite materials, is related to maintenance, repair and overhaul. The introduction in aviation of composite components (A330/340, A380, Boeing 777) and fuselages (Boeing 787, A350) has not led to a similar growth in the use of composites for repairs. Repairs in composite parts are either prohibited (leading to extensive replacements) or executed with metals (aluminum or titanium). One of the limiting factor for widespread use of composite materials in repairs is accurately and reliably locating a damage. Therefore composite parts are generally over-dimensioned. A fast and efficient detection method for barely visible damages on large surfaces may mean a huge leap in the adoption of composites for repairs, and an improvement of design parameters for composite parts . This lack of an appropriate detection method is also a limiting factor for other industries and could be used by a large community and bring the aerospace sector one step ahead. Carbon and glass fibre reinforced polymers consisting of thermosets and thermoplastics are light in weight, they show an increased strength in comparison to conventional materials and a reduced sensitivity to corrosion and fatigue. A downside is their susceptibility to impact damage. When this occurs it is often difficult to identify: it either affects the reverse side of the material or causes an internal delamination. Conventional detection techniques either fail, or are rather labor intensive (such as Rontgen monitoring) and expensive. Especially the anisotropy of the material is causing problems in the conventional detection techniques. In the current study we are investigating whether the physical principle of repeated reverberating signals allows us to generate a fast and efficient detection method for barely visible damages on large surfaces. The method we are investigating is based on detecting changes in the material using these signals. After performing a reference measurement on an undamaged material, a second measurement of the same material at the same location and a comparison of the signals should show that they are similar. However, when damage occurs the signal will change. Our method is not based on the analysis of the first arrival or an specific alteration in phase or time, but rather we are considering the entire reverberation of the signal. By analysing the cross correlation comparison of both signals, together with the autocorrelation function of the original signal, it is possible to distinguish changes in the occurring wave pattern over very large areas with a minimal number of sensors. Supporting analyses techniques, obtained from control theory, are also considered. The underlying physical principle are the changes in the wave pattern. When a pulsed signal is created by one piezoelectric transducer and is received by another transducer, it will register so- called surface waves traveling through the material. When use is made of low frequent signals Lamb waves are generated. These waves have different propagation modes, which can be used to our advantage. Their wave behavior makes them sensitive to either surface effects (surface damage) or internal changes (delamination). This way we are capable of detecting different types of damages. At these low frequencies, waves attenuate much less than the more conventionally applied high frequency (MHz) waves used in NDT, so that large areas are covered. The described method compares favourably to current thickness, flaw and bond test ultrasonic test equipment. This new approach could also work on aluminum alloy materials (and other traditional materials), and simplify the maintenance of conventional aircraft. The system, which is under development at the Amsterdam University of Technology in collaboration with Delft University of Technology and the Van der Waals Zeeman Institute of the University of Amsterdam, comprises of individual transducers that are permanently attached to the top surface. By testing samples under different environmental conditions, a varying working environment can be mimicked. In the current set-up the number of transducers is minimized to two sensors. However, when use is made of more transducers, it is already possible to make a simplified surface mapping of the investigated object, which makes the data analysis afterwards far easier. In contrast to other methods the number of sensors for a tomographic approach is reduced. Due to the enhanced 2D images generated by this approach, the operator needs a much lower entry level to use the equipment and is able to perform a much more detailed analysis in a fraction of the time. Our aim is to attain Technical Readiness Level 5 within three years to show feasibility and to entice companies to invest in the further development. We envisage that the described techniques will first be used for maintenance purposes, after certification. Over time, it will influence the design of new aircraft it may be used as an onboard Structural Health Monitoring system.