Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

On the springback-effect of CFRP-stringers. An experimental, analytical and numerical analysis

: Stephan, A.; Schwinge, E.; Müller, J.; Öry, H.

Society for the Advancement of Material and Process Engineering -SAMPE-:
28th International SAMPE Technical Conference 1996 : November 4-7, Seattle
Seattle, 1996
Society for the Advancement of Material and Process Engineering (Technical Conference) <28, 1996, Seattle>
Fraunhofer IFAM ()
residual stress; springback; thermal distortion

The cure process of curved and symmetric composite stringer induces residual stresses. These stresses lead to a thermal distortion of the stringer (springback effect). An initial angle of 90 degree between the T-stringer upper and lower part is primarily reduced by the anisotropic thermal behaviour and the matrix shrinkage. To compensate for this effect the tool angle is adjusted accordingly (trial-and-error approach). A new manufacturing concept for composite structures currently under investigation at Daimler-Benz Aerospace Airbus requires the exact determination of the thermal distortion. Besides the primary parameters anisotropic thermal expansion and shrinkage, other variables may influence the angle change. These secondary parameters e.g. stacking sequence, cooling rate etc. are a function of the chosen manufacturing technique. The paper describes the experimental investigation as well as the analytical and numerical simulation of the springback effect for a typical CFRP stringer used in aircraft structures (Fibredux 6376-HTA). The T-shaped stringers were manufactured in an autoclave. A complete and intensive variation of all previously determined variables is conducted in order to verify the simulation methods. The analytical simulation is based on the classical lamination theory and a compatibility analysis. The finite-element calculation is conducted with the software package (PATRAN/ABAQUS). The results indicate that the influence of the primary parameters can be well implemented into the simulation. The secondary parameters are included by the introduction of a correcting factor.