PSI - Issue 19

Sara Eliasson et al. / Procedia Structural Integrity 19 (2019) 81–89 Sara Eliasson / Structural Integrity Procedia 116 (2019) 000 – 000

86

4. Effect of specimen parameters

It is important to choose an adhesive with the right characteristics for fatigue testing. Three adhesives were tested. One adhesive was ruled out early due to too low shear strength, but two of the adhesives have a sufficient shear strength, and the final choice was based not only on shear strength but also on the ductility of the adhesive. In Figure 5a, results from the adhesive study are presented. The adhesive study is of a comparative nature and the specimens tested were manufactured in the same way with the same dimensions, clamped with the same clamping force and loaded with the same cyclic load. Aluminum tabs are used for the test and a minimum of three specimens are used for each adhesive. The number of cycles to failure is the best value reached for each adhesive. The fatigue testing is performed according to testing procedure presented in section 3. All failures were adhesive failures. This investigation clearly shows that the 3M DP420 adhesive withstands a cyclic load much better than a glue with equal shear strength from Araldite.

Figure 5: To the left (a) adhesive study results; No. of cycles to adhesive failure for fatigue specimens, to the right (b) temperature evolution study results; temperature in specimen for different tab material.

The temperature in the tab area of the specimen can be greatly affected by the choice of tab material. In Figure 5b, a temperature investigation is presented which indicates the importance of the choice of tab material for transferring heat away from the specimen. The temperature increase is presented and measured in the tab area of the specimen. The value for the temperature is extracted when the temperature is stabilized. Aluminum has a much higher temperature transfer coefficient, and this makes it the better choice for keeping a stable temperature and maintaining the properties for the adhesive. The choice of adhesive is assumed to have a small influence on the temperature compared to the choice of tab material and the same adhesive is used for all specimens in the test. The fatigue testing is performed according to test procedure presented in section 3. The tabs are made 5 mm wider than the specimen. This extra space works as a tolerance for slippage of tabs and excessive adhesive during the manufacturing process.

5. Failure modes

An unwanted failure mode is, as mentioned, tab region failure; this is presented in Figure 7. Failure in the tab region can be due to many factors, too high clamping force can be one and in Figure 8 it is clearly seen how the material is crushed and the adhesive is creeping out on the edges. It is also seen how damage grows along the fibers following fiber bundles and stitching. Figure 9 shows a micrograph for a typical crack growth occurring growing along the fiber bundles following the stitching.