PSI - Issue 5

David Palousek et al. / Procedia Structural Integrity 5 (2017) 393–400 Vendula Kratochvilova et al. / Structural Integrity Procedia 00 (2017) 000 – 000

395

3

Han et al. (2013) studied effects of microstructure on fatigue crack propagation and its AE response. After the AE analysis, the same 3 fatigue stages were determine for different materials structures. However, the differences in the AE signal from material with large amount of martensitic phase were not as significant as in other structures (ferrite and pearlite, ferrite with carbide precipitates and widmanstatten structure). Stage 1 corresponded to the crack initiation, stage 2 to the plastic activities in plastic zone ahead of the crack tip and stage 3 to ligaments shearing between micro voids, micro cracks and dimples. Fatigue behaviour of steel using AE method has been studied by Mentl et al. (2014). Unlike Han et al. (2013) his team used four points symmetric bending test. The three fatigue stages were also shown in AE signal. The changes of mechanical properties (cyclic hardening / softening), dislocation movements and slip plane formation were assigned to stage 1, the crack nucleation to stage 2 and micro / macro crack propagation to stage 3. Presented paper extends our previous studies and investigates the fatigue behaviour and crack propagation of new SLM materials using AE method, Kratochvilova et al. (2016).

2. Experimental methodology and used materials

2.1. Aluminum alloys

Three sets were produced from SLM material AlCu2Mg1.5Ni and the results were compared with conventionally produced (extruded) material before and after T6 heat treatment. The difference between the sets are in used metal powder and in distribution of samples on the building platform. Samples of 1 st and 2 nd set were made from new powder, but samples of 1 st batch were made one by one and 2 nd batch in one build. The material quality of this batch was really unstable. Only 4 of 12 pieces were possible to fatigue tested (green on the Fig. 2), because the quality of the rest was not sufficient. The worst specimens were located near the edge of the building platform (red on the Fig. 2). Samples of 3 rd batch were produced one by one, but recycled powder was used.

Fig. 2. Layout of 3rd batch samples and the surface of the worst (red) and the best one (green)

As it is known from previous study of Brandl et al. (2012), the material AlSi10Mg is suitable for SLM technology. One batch of this material was produced in order to evaluate the suitability of material AlCu2Mg1.5Ni. All samples were produced by machine SLM Solution 280HL. The production parameters were determined on the basis of previous studies of Koukal et al. (2015) and were the same for all batches of Al alloys:

 Laser power: 200 W  Scanning speed: 100 mm/s  Hatch distance: 110 µm  Beam diameter: 82 µm