Abstract
A methodology is described for measurement of environmentally assisted small crack growth rates in a steam turbine blade steel in simulated condensate environment. Using a droplet technique a single corrosion pit was grown to a controlled depth in a specific location on the gauge length of a tensile specimen of a FV566 blade steel. Methods of measuring crack development from this precursor were then developed based on high resolution optical measurement and digital image correlation (DIC) in combination with direct current potential drop (DCPD) measurement. The optical method developed enabled automated high resolution measurement of crack length that could detect cracks from a pit with a surface length extending from the pit greater than about 1 μm and with very small measurement uncertainty, reflected in a standard deviation of only 0.41 μm at a crack length of about 100 μm. The use of surface crack opening displacement (SCOD), derived from DIC analysis, to estimate crack depth has some promise but further study is needed to assess the dependence of SCOD on the crack shape so more accurate crack depth determination can be obtained. In contrast, there was very good agreement between the crack depth calculated from DCPD and that actually measured on the fracture surface, once a continuous crack front had been established.
