28 Stress level as a function of failure cycle for the un-patched aluminum, composite patch repaired aluminum, and the oven sensitized composite patch aluminum for the three stress levels are shown in Fig. 3.7. Sensitized plate was “as-delivered” aluminum conditioned at 121 °C for 10 h, which resulted in a DoS of 47.3 mg/cm 2 as measured by the G67 test method. The data points for each plate condition are fit with a power law to highlight the general trend of the data and to show the difference between the plate condition groups. Each data point represents one specimen. Application of a composite patch repair to a cracked aluminum plate subjected to a cyclic tensile load increased the cycle count to failure for the three stress levels tested. The benefit of the composite patch repair system increased with the level of applied stress. When subjected to a far-field tensile stress level of 34.5 MPa the number of load cycles increased by four to five times for the sensitized and as-delivered plate condition. Similarly, for the 75.2 MPa stress level the cycle counts increased 6.5 to 7.5 times. When the stress level was increased to 100.3 MPa the cycle count increased between 12 and 15.7 times the baseline cracked plates [31]. Composite patch repairs increase static plate strength, fatigue life, and large displacement capacity of the crack plate. Additionally, to demonstrate performance under high stress, a large composite patch reinforced aluminum edge notch test specimen (0.8 × 1.0 m) was tested in the custom designed fixture shown in Fig. 3.8. The fixture consists of four floating load arms and two compression struts. With load applied to the arms a large region of high stress forms ahead of the blunt notch that typically produces highly controlled ductile fracture of the aluminum plate [32]. The reinforced specimen consisted of a 28 ply E-Glass epoxy laminate with a nominal thickness of 15 mm and an axial elastic modulus of 18.6 GPa bonded to a 5083-H116 aluminum plate 6.35 mm thick. Uni-directional reinforcement was inserted between plies of the typical quasi isotropic laminate to increase patch modulus and strength across the crack plane. Initial quasi static analysis of the specimen indicated grip failure prior to laminate failure. Mitigation of this risk involved extending the 50.8 mm V-notch into a 127 mm blunt notch and removal of 152 mm of composite at the notch tip as shown in Fig. 3.8. Dominant failure modes observed during testing are shown along with the test fixture in Fig. 3.8. Aluminum fracture and laminate failure were observed, however disbonding of the composite did not occur. Visible aluminum bonding surfaces include multiple composite plies starting with the interface ply, which indicates the composite to aluminum bondline remained intact. Laminate failure propagated through the bottom half of the laminate until testing was terminated but the patch above the crack remained intact and securely bonded to the aluminum. The load-deflection response of the composite reinforced plate is compared with baseline aluminum and welded aluminum plates in Fig. 3.9 [32]. Under load the aluminum crack propagated approximately 30° to the crack plane and then arrested 2 cm from the composite, which is the location where the crack direction changes in Fig. 3.8. Loading increased until aluminum plate failure, shown in Fig. 3.9, occurred within the grips. Loading continued to increase until composite 0 2000 4000 6000 8000 10000 12000 14000 16000 1 10 100 1,000 10,000 100,000 1,000,000 Far-Field Stress (psi) Cycles CCT Specimen Fatigue Cycles to Aluminum Failure 5 Inch Crack Length Aluminum Only Patched Sensitized Patched Fig. 3.7 Aluminum specimen and series far-field stress level versus cycles to failure for 2a 0 = 5 in. fatigue testing D.C. Hart
RkJQdWJsaXNoZXIy MTMzNzEzMQ==