Robust Finish Machining Induced Residual Stress Characterization via DIC and Statistical Averaging for Incremental Hole Drilling Technique According to ASTM E837
摘要
Existing methods for quantifying machining-induced residual stress (RS), can be constrained by uncertainties in subsurface stress profiling arising from non-standardized material removal procedures, mechanical gauge misalignment, and long measurement times, limiting the availability of reliable RS data and reducing confidence in subsurface stress evaluation.
ObjectiveThe goal of this work is to develop a digital image correlation (DIC)-based virtual strain gauge (VSG) framework integrated with incremental hole drilling (IHD) to improve fidelity, repeatability, and statistical robustness of full-depth RS measurements in finish-machined Ti-6Al-4V.
MethodsOrthogonal finishing cuts were performed under controlled cutting speed and tool-wear conditions using a custom in-situ testbed. Full-field DIC strain data were processed to generate VSGs emulating ASTM-defined rosettes, enabling strain averaging over large pixel domains and multiple symmetrical locations. Strain–depth profiles from multiple holes were converted to RS using H-Drill software and averaged via the Student’s t-distribution to achieve 95% confidence intervals. Dual-level averaging across strain fields and replicated measurements suppresses DIC noise and measurement scatter inherent in physical gauges.
ResultsThe averaged RS–depth curves vary with tool flank wear, producing tensile or compressive surface stresses and differing magnitudes and depths of subsurface compression. Combined strain and stress averaging produced smooth, repeatable profiles with accuracy exceeding conventional IHD. The process requires less than 1 h per sub-surface profile.
ConclusionsThe DIC-integrated, statistically averaged IHD approach provides a minimally destructive, high-precision, and high-throughput method for quantifying machining-induced RS, enhancing surface integrity characterization and model validation reliability.