(a) (b) Fig. 3 (a) Instrumented Test Fixture and (b) Impact Hammer with Load Washer Signal Conditioners Four complete modal impact tests were conducted on the fixture using this test setup. Two variables were adjusted between the tests. The modal impact tests were conducted with the static preload in the bolts tuned at levels between 1kN and 20kN, and the amplitude range of the impacts were held between 10lbf and 20lbf in one set of tests or between 200lbf and 300lbf for a second set of tests. The goal of the test was to characterize the nonlinear response characteristics due to impact amplitude as the preload in the bolts was adjusted. Three impacts were applied at each point on the modal grid where it was determined that the coherence was acceptable. A second experiment was conducted using a Hopkinson Bar to impact the test fixture. The Hopkinson bar applies an impact through a transfer bar. This method was chosen because a one inch diameter aluminum transfer bar could be used in order to prevent deformation at the impact location on the much harder steel test fixture. These tests were conducted by suspending the test fixture vertically from an engine hoist and preloading a transfer bar against the test fixture. The bolts were tightened to a pre-defined torque with a torque wrench, and each load cell’s static value was recorded. The transfer bar was then impacted by a projectile that was fired by a gas gun. The impact force was stepped up by monitoring the air pressure that was used to launch the projectile. For this experiment, six PCB 350C02 shock accelerometers were used to measure acceleration at both ends of each of the three bolts. For each test, the six acceleration measurements were acquired as was the velocity at the center of the square plate using a laser velocimeter. The load cell dynamic data was also recorded. During this round of testing, preload levels of hand tight, 50ft-lb, and fully tight, 100ft-lb, were tested with impacts at 5psi, 10psi, 15psi, and 20psi. The aim of this experiment was to achieve an excitation that better reflected the realistic loading scenario that was expected. The setup for these tests is shown in Fig. 4. Fig. 4 Hopkinson Bar Test Setup 569
RkJQdWJsaXNoZXIy MTMzNzEzMQ==