Spring-in and Warpage – Progress in Simulating Manufacturing Aspects

Kurzfassung

Increasing application of polymer matrix fibre composites in lightweight constructions has brought about considerable effort in modelling the material and structural behaviour. Besides laminate analysis and failure prediction major attention is turned towards simulation of the manufacturing process aiming at models and means for producing structures ‘right first time’. Several different aspects like fibre cutting, draping, filling pattern and online thickness measurement have to be observed in order to end up with a high-quality product. After demoulding plane structures often show a warpage and angles do not reach the designed values (spring-in). Svanberg [1] has listed a number of factors with different influence on the shape distortions. At the DLR considerable effort has been put into the development of methods for simulating the distortions in order to avoid them by counter measures. Wille et al. [2] pointed out that besides an empirical, trial and error based procedure there is a choice between simulation-based and semi-analytical approaches. This paper focuses on the latter ones as being fast and sufficiently accurate for the initial design. Spring-in is observed at curved structures. Radford and Diefendorf [3] have developed a simple formula where the amount of spring-in depends on the difference between in-plane and transverse properties. Spröwitz et al. [4] have shown that an equivalent coefficient of thermal expansion can be determined with L-shaped coupons. With the aid of this coefficient the spring-in of more complex structures can be calculated with sufficient accuracy. Experiments showed considerable warpage of plain structures especially for slender plates. A simple formula based on the classical lamination theory was developed by Kappel et al. [5]. Measurements with test specimens determine the unknown parameter. Comparisons with detailed FEM analyses as well as with experimental results will demonstrate the accuracy and the possible range of applicability of the proposed methods. REFERENCES [1] Svanberg, J.M.: “Predictions of manufacturing induced shape distortions – high performance thermoset composites.” PhD thesis, Lulea University of Technology, Sweden, 2002. [2] Wille, T., Freund, S. and Hein, R.: “Comparison of methods to predict process induced residual stresses and distortions of CFRP components”. NAFEMS Seminar Progress in Simulating Composites, Wiesbaden, Germany, 2011. [3] Radford, D.W. and Diefendorf, R.J.: “Shape instabilities in composites resulting from laminate anisotropy,” Journal of Reinforced Plastics and Composites, Vol.12, pp. 58-75, 1993. [4] Spröwitz, T., Tessmer, J. and Wille, T.: “Process simulation in fiber-composite manufacturing – spring-in.” NAFEMS Seminar Simulating Composite Materials and Structures, Bad Kissingen, Germany, 2007. [5} Kappel, E., Stefaniak, D., Spröwitz, T. and Hühne, C.: “A semi-analytical simulation strategy and its application to warpage of autoclave-processed CFRP parts.” Submitted for publication in Composites Part A, 2011.