Crashworthiness of Corrugated Composite Shells

The increase of number of vehicles on the road has led to an increase in the traffic accidents. Consequently, there is also an increase of deaths and serious injuries. The desire to improve the crashworthiness of automobiles cannot be overestimated. It has been estimated that the annual financial loss from traffic accidents is equal to 2% of the GNP of the USA. Together with a range of environmental concerns and social pressure backed by legislation have led and continue to lead to highly innovative designs, involving lighter materials such as composite materials. During the last decades, researchers' interest have been directed towards the use of composite structures for enhancing crashworthiness due to their superior properties, in particular, strength and stiffness to weight ratios and their ability to be tailored in composition and shape. Of particular interest of this study is the use of corrugated composite shells in the automobile industry as a crush energy absorber device. Extensive experimental and computational programs have been conducted in order to investigate the crushing behaviour of corrugated and non-corrugated composite tubes. In the experimental program, innovative mandrels and filament winding machine were designed and fabricated. The investigated corrugated composite tubes consist of a number of similar circular cones connected together by circular tubes in the order: conecylinder- cone-cylinder and so on. The outer diameter of each tube is fixed at 100 mm and the number of layers is maintained 6 layers. Two material types, namely, filament wound carbon/epoxy and woven roving glass/epoxy were investigated. A comprehensive quasi-static crushing test program was performed to examine the influence of the tube length and the corrugation angle on the energy absorption capabilities. The load-displacement curves and typical load paths together with deformation histories were presented and discussed. The specific energy absorption, energy absorption per unit length, crushing force efficiency and stroke efficiency were calculated, analyzed and discussed. Comparisons in terms of specific energy absorption and/or energy absorption per unit length capabilities between corrugated composite and non-corrugated tubes were also presented and discussed. Macroscopic photos were taken during the tests to help understanding the failure modes and the failure mechanisms were analyzed microscopically. The crushing load-displacement behaviour, energy absorption, and the observed mechanisms and failure modes were found to be sensitive to the change in the corrugation angle, the tube length and the fibre type, and distinct collapse modes were observed. The most dominant observed failure modes are: catastrophic and brittle fracture modes in filament wound carbon/epoxy tubes, progressive folding mode (in the corrugated tubes) and splaying modes (in the non-corrugated tubes) of woven roving glass/epoxy tubes. Also splitting mode was observed in both filament wound carbon/epoxy and woven roving glass/epoxy tubes. The results have shown that corrugated composite tubes are efficient impact energy absorbers. Based on tube length, GL3-0 and GL1-20 tubes experienced the highest energy absorption per unit length (19.85, 18.89 kJ/m, respectively). Based on material type, filament wound carbon/epoxy tubes exhibited higher specific energy than the corresponding woven roving glass/epoxy tubes where 15.7 kJ/kg was absorbed by CL4-40 tube. Over all, GL1-20 tubes could be recorded as the best choice for crashworthiness applications (moderate load carrying capacity, high absorption energy, high CFE and high SE). Concurrent with the experimental work, numerical analyses was carried out using commercially available Finite Element Software (LUSAS). Three-dimensional Linear buckling Finite Element was conducted for the corrugated and non-corrugated composite tubes to predict the critical failure load, the deformation at the critical load, and the stress concentration contours. Both the experimental and numerical results were presented for different reinforcements, different corrugation angles, and different tube lengths. Relatively, a reasonable agreement between the experimental initial failure load and the computational critical load was obtained, specially for corrugation angles = 30 and 40 degrees. Knockdown factor is used to compare the results and found to be varying in the range from 0.259 (CL4-10) to 0.998 (GL4-40)