PSI - Issue 19

Fabian Becker et al. / Procedia Structural Integrity 19 (2019) 645–654 F. Becker et al. / Structural Integrity Procedia 00 (2019) 000–0 0

647 3

Fig. 2. (a) Manufacturing of specimen bars with molding tool; (b) Final specimen geometry and used coordinate system

2. Materials and Methods

The material used in this research is a unidirectional E-glass fiber reinforced epoxy, which is manufactured by pre-preg laying in a hot molding press. The tool has three molds to produce unidirectional bars. After demolding, each of the bar is sawn into three specimens, that are further prepared manually by deburring the edges on the compression side. After this process step, the specimens are cured. The final specimen geometry has a non-rectangular cross section based on the work on Feldten (2014). The fillet radius on the tension side of the specimen is 5 mm and causes a shift of the neutral axis during bending to guarantee a gradual damaging behaviour on the tension side during fatigue testing without damages caused by stress inhomogenities near the edges (Figure 2). The specimens have a thickness of h = 5 . 64 mm ± 0 . 12 mm and a width of w = 20 . 12 mm ± 0 . 03 mm). Their length was not measured individually, but larger than 170 mm, since the stress state depends on the support span in the three point bending. For the scope of the research presented, the coordinate system is defined as ”1” in fiber direction, ”2” in the lamina plane and ”3” in thickness direction (Figure 2). To introduce transverse stresses during the three point bending fatigue test, a clamping system is developed. The clamping system consists of an upper and a lower steel plate, which are identical in the geometric dimensions (Figure 3). The upper half is connected with an adapter to the 25 kN servohydraulic testing rig from Instron, Darmstadt, Germany for fatigue testing. Between the two steel plates, a 30% glass fiber reinforced polyamide layer is placed with a radius of 100 mm and a flat section in the middle to allow for the desired pressure distribution as defined by means of finite element analysis. The flat section is scaled geometrically, so that the ratio of the length of the flat section and the support length on specimen level is equal to the ratio of the clamping section and support length on leaf spring level. However, the thickness of the specimen is not scaled geometrically from leaf spring level. Since only the pressure distribution on the surface of the specimen was investigated, a final conclusion on the scaling approach for the internal stress distribution cannot be drawn within this research. The upper and the lower half of the clamping are connected by two M12 bolts (one for each side) made of Ti–6Al–4V alloy for lower mass and necessary durability of the bolts. The endings of the bolts are supported by spherical bearings, which avoid any bending moment acting on the bolts guaranteeing a pure tensile load application. The load on the bolts is achieved by tightening the two nuts at the end of the bolts. The bolt loads are measured with applied strain gauges on the bolt shafts in Wheatstone bridge configuration. To protect the strain gauges, they are covered with 3D-printed parts of PLA plastic filled with 2.1. Clamping system to introduce transverse stresses