PSI - Issue 19

Lloyd Hackel et al. / Procedia Structural Integrity 19 (2019) 452–462 Valentin LOURY--MALHERBE/ Structural Integrity Procedia 00 (2019) 000–000

458

7

6. Fatigue Test Results of AM In718 samples: With strain gauge calibrations confirming the relationship between load and stress in the samples, testing started using a 4-point bending fixture on a 20 kip Instron test rig. Samples were loaded with a sinusoidal of 683 MPa and 68.3 MPa minimum (R=0.1). This represented a peak load of 86% of yield stress. Initial testing included unpeened samples to establish a performance baseline and single shot peened and dual shot peened samples. In this initial sequence laser peened samples and samples where the gauge section was polished were also tested. Figure 5 shows these initial results. Our baseline test of 5 unpeened AM In718 samples (left most blue bar) gave a mean lifetime of 175,928 cycles with a standard deviation of 72,944, that is a standard deviation of 42%. In this work a large number of process variations were tested and, since this was a scoping experiment and due to limited sample availability and the desire to test a range of process parameters only 2 samples per parameter were typically tested. Variation of test results were typically in the range of 20% or less, that is the observed change in performance was always greater than the uncertainty. Testing of shot peened samples (2nd from left, orange bar) gave a 52% lifetime improvement to 268,007 cycles and polishing of the shot peened samples (3rd from left, beige bar) gave a 98% lifetime improvement over the as manufactured AM samples. Dual shot peening (4th from left, gold bar) gave a very significant improvement to 900,870 cycles, a 500% lifetime increase. The initial tests of the laser peened samples (5th from left, green bar) without polish gave only a lifetime of 390,128 cycles but when the gauge surface was polished the lifetime dramatically improved to 1,352,407, a 768% improvement over the baseline AM and well in excess of statistical uncertainty. The laser peening is known to minimally change surface finish so it was expected that the polishing was more important in delaying initiation for the laser peened case than for shot peening, as the sub-millimeter mechanical impacts of the shot seemed to play a significant role in tamping down the rough surface of the as-manufactured AM material. This surface improvement role of the shot peening seemed to be substantiated by the results of shot peening plus laser peening without polishing. For example with no polishing in either case, the test of a shot peened plus laser peened sample (2nd from right, blue bar) gave a very high lifetime of 1,525,211 cycles and the dual shot peened plus laser peened sample (1st from right, grey bar) again with no polishing lasted 1,459,630 cycles. The conclusion from this first phase of work with no post peening thermal exposure is that the shot peening, dual shot peening and laser peening were each sequentially able to provide more and more fatigue life improvement with the laser peening providing the best, approximately 800%, increase.

Figure 5. Fatigue test lifetimes (R=0.1 and 10% compliance) of AM In718 with and without peening.