13th International Conference on Fracture June 16–21, 2013, Beijing, China -4- • to cross-correlate the test piece temperature, measured by a spot-welded thermocouple (e.g. TC7 in Figure 4) with the controlling sheathed thermocouples and ensure that the thermal and load cycles were correctly applied and synchronised. 3.3 Test Set-Up The test piece had a square section (7 x 7 mm), with a corner notch introduced into the centre of a 20 mm parallel gauge length. The sheathed thermocouples were attached as shown before and ~63µm diameter Pt wires were then spot-welded to the test piece surface to monitor the PD across the crack. The sample was pre-cracked with load shedding to reduce the stress intensity factor below that of the desired starting stress intensity factor. Once the thermal performance on the two sheathed thermocouples matched that previously observed during the calibration, then it was possible to commence the test. 4. Crack Growth Rate Measurements and Facility Demonstration To calculate the crack lengths and growth rates it was found that fitting a polynomial to the PD measurements provided the best way to generate reasonable data. A simple linear relationship between the crack length and the PD measurement is assumed for this procedure. The stress intensity factors were calculated using proprietary fits for the “geometry dependent compliance factor” (Yn) within the basic equation: ΔK = Δσ.Yn.(π.a) 1/2 (1) - where ΔK = change in stress intensity factor, Δσ = the change in applied stress, Yn= a geometry dependent compliance factor and a = crack length. Only the tensile portion of the loading was used in the stress intensity factor calculation and hence a maximum stress intensity factor is shown in the plot against the normalised da/dN (Figure 5). 1.E-01 1.E+00 1.E+01 1.E+02 10 100 Kmax (MPa.m-1/2) Relative crack growth rate (/cycle) RR Validation data - fine grain RR1000 (conventional resistance furnace) Fine grain RR1000 TMF Coarse grain RR1000 TMF Figure 5. Plot of relative crack growth rate again the maximum stress intensity factor for two microstructural variants of RR1000 The final stage of the facility development was to demonstrate its operation and compare it with existing data obtained using a conventional resistance furnace. This demonstration was undertaken using an advanced nickel base superalloy known as RR1000, used for gas turbine disc applications and manufactured using a powder metallurgy and forging process. These initial trial tests showed reasonable agreement with existing data, although somewhat higher rates were measured with the
RkJQdWJsaXNoZXIy MjM0NDE=