HIP vs as-built: when post-processing pays back

Hot isostatic pressing applies isotropic gas pressure (typically 100–200 MPa) at elevated temperature (typically 0.6–0.8 of T_solidus) for 2–4 hours. Internally connected porosity is collapsed by creep and diffusion bonding. Open-to-surface porosity is not closed — gas pressure equilibrates across the surface.

L-PBF Ti-6Al-4V with predominantly gas porosity: fatigue allowable typically rises from ~280 MPa as-built to ~480 MPa HIP'd (10⁷ cycles, R = -1, machined). The cost of one HIP cycle on a 30 kg batch amortises to about $50–80 per part for typical aerospace bracket geometry — almost always worth it for fatigue-critical hardware.

L-PBF Inconel 718 follows the same pattern; the HIP also helps homogenise the dendrite spacing before solution + aging. We see fatigue allowables roughly double, and elongation roughly triple, vs as-built.

If the defect population is dominated by lack-of-fusion (sharp, planar voids from un-melted regions), HIP is far less effective. The void surfaces oxidise during build and the oxide layer prevents diffusion bonding; HIP closes the void geometrically but the interface remains a crack-initiation site.

On AlSi10Mg, HIP can soften the fine eutectic microstructure if not paired with a re-aging cycle (T6). The typical recipe is HIP + solution + age, not HIP alone.

On binder-jetted 17-4 PH, HIP is typically embedded in the sinter cycle (sinter-HIP) rather than a separate step. A separate HIP after a fully sintered part is usually not cost-effective.