On the uncertainty of porosity measurements of additively manufactured metal parts

Equations and a methodology are presented for calculating the uncertainty of porosity measurements of addi-tively manufactured (AM) metal samples. The bulk porosity of six metal AM samples are measured using Archimedes' principle; the density of two metal powders are measured using a gas pycnometer. The AM sample geometries include a 10 x 10 x 10 mm (H x W x D) cube, a 10 x 10 mm (empty set x H) cylinder and a bracket with a complex freeform geometry. Each geometry is fabricated in both titanium and cobalt chrome. The AM fabrication settings are modified to print samples with bulk porosities of less than 2%. The expanded uncertainties of the porosity measurements are evaluated and found to range from +/- 0.14% to +/- 0.18% (with 95% confidence). The dominant uncertainty contributor is found to be the measurement repeatability. The equations presented in this work can be used by researchers and technicians to identify the key uncertainty contributors in their own porosity measurements.