PSI - Issue 28

J. Weiland et al. / Procedia Structural Integrity 28 (2020) 1249–1257 Weiland et al. / Structural Integrity Procedia 00 (2019) 000–000

1255

7

which corresponds to a light transmission of 50 % (-3 dB) at roughly a length of 20 m. This is well suited for many relevant lengths of adhesive joints. The optical power of the LED can be varied via an adjustable current source. At the POF end the light power is converted into electrical voltage by means of a commercially available photodiode and an electronic amplifier; a data logger records the voltage. For the tests, the tensile testing machine Zwick RetroLine ZO10 from ZwickRoell GmbH & Co. KG Ulm Germany was used. The test setup with measuring equipment is shown in figure 6. The single lap joint was cyclically loaded by eccentrically acting forces within a tensile shear test. The cycle is force guided and consists of the approach of the force levels 0 N – 1000 N – 2000 N – 3000 N – 2000 N – 1000 N – 0 N, each force level is held for 10 s. This cycle is repeated 3 times, until the joint is tested to failure. The testing speed was 1 mm / min. Figure 7 shows the force curve with the 3 cycles over the testing time

Fig. 7: Force-guided cycle loading and failure

5. Results Figure 8 compares the cyclic stress curve for one representative test specimen with the sensor signal of the POF, which is normalized and applied in percentage. Clearly seen is that both signals are oppositely correlated. This means that the measured optical power decreases with increasing stress in the bond. This is due to the increasing strain in the adhesive layer and the associated increasing changes in the fiber cross-sectional shape. Due to the cyclical force progression of the tensile testing machine, it can also be seen that the transmitted light power of POF increases again when the sample is unloaded. As soon as the load has dropped back to zero, the light power restores again its initial value of about 100 %. During one cycle, the integral light output decrease about 15 % under load. After a test time of about 660 seconds the load reaches its global maximum with about 7.9 MPa. At the same time, a global minimum of about 68 % in the course of the light power is record. After the subsequent failure of the test specimen, the mechanical stress decreases while the light output increases again. This confirms the basic phenomenon of the cross-sectional shape dependent light conduction equivalent.