Effect of localized thermal history on martensitic microstructure and microhardness variation in wire-arc additively manufactured 17–4 PH stainless steel beads

Wire arc additive manufacturing (WAAM) enables rapid and low-cost fabrication of large steel components. However, cumulative thermal history along a single bead can cause significant variations in microstructure and properties at fixed parameters, a concern for 17 − 4 PH stainless steel, whose martensitic transformation is highly cooling-rate sensitive. This study distinguishes two effects that are often conflated: the global average heat input that differs between beads, and the local thermal history that varies from the start to the end of a single track through progressive heat accumulation. This study combines CFD simulations, thermodynamics, and microstructural characterization to link heat input, thermal history, microstructural evolution, and microhardness in CMT-WAAM-fabricated 17 − 4 PH stainless steel. Three single beads were fabricated under three heat input conditions by varying wire feed rate, travel speed, and gas flow rate, with the arc current and voltage adjusting automatically. The thermal history, microstructure and microhardness variations were examined at the start, middle, and end of each bead. The numerical predictions agreed well with measured bead geometries, while Scheil–Gulliver calculations predicted primary δ-ferrite solidification followed by austenite formation and martensitic transformation upon cooling. Thermal fields showed heat accumulation reduced the local cooling rate from ~ 969–1143 K/s at bead starts to ~ 658–787 K/s at mid-length, driving systematic changes in martensitic morphology. Fine and dense laths were developed under rapid cooling conditions, while coarser and heterogeneous laths developed under slower cooling and higher heat input conditions. Optical microscopy, SEM, and XRD confirmed a predominantly martensitic microstructure throughout, with EDS showing uniform Fe, Cr, and Ni distribution and no detectable retained austenite or quantification of the δ-ferrite phase. Microhardness was highest at start/end regions (~ 398–407 HV) and lowest at mid-length. The microhardness also decreases further with the global heat input and primarily primarily depends on the local cooling rate. These results establish a localized process–thermal–microstructure–hardness relationship, showing that local hardness variation in beads can be predicted from the local cooling rate data, offering a practical basis for heat-input and path-planning control in WAAM.