۱٫ Introduction

۲٫ Material characterization with static tests

۳٫ Self-heating tests campaign

۴٫ Conclusion and outlooks

References

The full estimation of the fatigue lifetime of a given material is needed for the industrial structures sizing. However, it requires time-consuming, expensive fatigue testing campaigns and a huge number of samples. An alternative experimental procedure is based on the monitoring of the “self-heating” properties. The aim here is to correlate the fatigue limit with a change of thermal behavior. The main advantage of such tests is that they only require few samples and are constituted of a limited number of mechanical cycles. Thus, self-heating tests lead to an accelerated estimation of the endurance limit. The purpose of the present paper is to validate this approach for carbon/thermoplastic composite materials.

Introduction

Composite materials are made of at least two components, usually fibers and matrix, which confer to the new material better mechanical properties than those of individual components [1], [2]. The fibers, either glass-fibers or carbon-fibers, confer the mechanical properties to the whole material, while the matrix is used as a resin to link the fibers to each other, and to distribute the loads applied on the material [3], [4]. Composite materials offer a strength/weight ratio better than the one of other homogeneous materials, like steel: this is why they are more and more used in aerospace industry [5]. The structure of composite materials depends on the orientation, the architecture and the size of the fibers, which can be short or continuous. Continuous fibers might be aligned in one preferential direction (unidirectional) or might be woven according to two or three directions. The fatigue phenomenon appears during the lifetime of a structure under cyclic loading. Indeed, during cyclic loading, the material might break even if the maximal load is lower than the damage threshold (thus lower than the failure load). A common way to represent fatigue behavior is to use the Wöhler or S/N curves which provide the number of cycles to failure for each load level [6], [7].