This paper presents the first comparative study of fabricating reference porous media with three different additive manufacturing (AM) techniques, namely Fused Deposition Modelling (FDM), Stereolithography (SLA) and HP Multi Jet Fusion (MJF). Various parameters in each step of these 3D printed reference media that contribute to model inaccuracy, including the intrinsic limitations of each printing technology are examined. The analysis of repeatability and dimensional accuracy of 3D printed parts shows that SLA yields the highest integrity in 3D printed reference medium with tolerances well below 2% of the nominal thickness of the mould. Hence the reference medium processed via SLA is tested for the permeability study. Statistical analysis based on 20 measurements of permeability using the same set-up shows a coefficient of variation of less than 2%, confirming the elimination of the inherent source of variability in real textile. © 2020 Elsevier Ltd