Residual stress measurements from endmilling of annealed AISI 4340 have been made at multiple points within the cut geometry to investigate the effects of location on the machining-induced residual stresses from endmilling. In the same experiments the effects of axial depth of cut and feed on the residual stresses induced in the machined surface have been investigated in a design of experiments framework. The experimentation demonstrates that location, feed and axial depth of cut have strong influences on the machining-induced residual stresses from end-milling when expressed in a workpiece coordinate frame. In addition, expression of the residual stresses in a coordinate frame fixed in the tool demonstrates a simplification in the residual stresses from endmilling in that the stresses at multiple locations within the cut geometry show strong similarities when expressed in this coordinate frame.

A model to predict the full biaxial surface and subsurface residual stresses from the turning process is presented. The model formulation includes thermomechanical coupling, plastic heating, frictional heating, convection, conduction, thermal softening and strain hardening. Calibration and validation of the model are undertaken. The predictive model stands as a tool both to optimize the turning process based on the resulting residual stresses and also to gain an in-depth understanding of the physics of residual stress development from the turning process. In addition, thorough analysis of the experimental results reveals insight into the effects of feed and depth of cut on the surface and subsurface machining-induced residual stresses.

Controlled orthogonal and controlled oblique machining of annealed AISI 4340 have been undertaken in a design of experiments framework to investigate the machining-induced residual stresses resulting from these processes. The experimentation demonstrates significant simplifications in the machining-induced residual stress problem when the stresses are expressed in a coordinate system fixed in the tool and also indicates that the directions along the cutting edge and normal to the cutting edge of the tool are principal directions of the machining-induced residual stresses. Based on the experimental results, a plane strain thermoelastoplastic model of metal flow under the flank of a cutting tool is developed to predict the full in-plane biaxial residual stress profiles existing at and beneath the newly created surface. Calibrated results show favorable agreement with the experimental machining-induced residual stresses in annealed AISI 4340. [S1087-1357(00)00201-X]

The effect of different uniaxial and triaxial stress states on the stress-induced martensitic transformation in CuZnAl was investigated. Under uniaxial loading, it was found that the compressive stress level required to macroscopically trigger the transformation was 34 pct larger than the required tensile stress. The triaxial tests produced effective stress-strain curves with critical transformation stress levels in between the tensile and compressive results. It was found that pure hydrostatic pressure was unable to experimentally trigger a stress-induced martensitic transformation due to the large pressures required. Traditional continuum-based transformation theories, with transformation criteria and Clausius-Clapeyron equations modified to depend on the volume change during transformation, could not properly predict stress-state effects in CuZnAl. Considering a combination of hydrostatic (volume change) effects and crystallographic effects (number of transforming variants), a micromechanical model is used to estimate the dependence of the critical macroscopic transformation stress on the stress state.

The effect of stress state on the character and extent of the stress-induced martensitic transformation in polycrystalline Ni-Ti shape memory alloy has been investigated. Utilizing unique experimental equipment, uniaxial and triaxial stress states have been imposed on Ni-Ti specimens and the pseudoelastic transformation strains have been monitored. Comparisons between tests of differing stress states have been performed using effective stress and effective strain quantities; a strain offset method has been utilized to determine the effective stress required for transformation under a given stress state. Results of the tests under different stress states indicate that (1) despite the negative volumetric strain associated with the austenite-to-martensite transformation in Ni-Ti, effective stress for the onset of transformation decreases with increasing hydrostatic stress; (2) effective stress vs effective strain behavior differs greatly under different applied stress states; and (3) austenite in Ni-Ti is fully stable under large values of compressive hydrostatic stress.