Portable Refractometry for Measurement-Enabled Moka Coffee Brewing: A Factorial Study of Roast Level, Grind Size and Initial Water Temperature

Portable coffee analytics are already present in professional and enthusiast coffee practice. The open question is how such measurement can be translated into decision-relevant knowledge. This study addresses this question through moka brewing and low-cost digital refractometry. A full-factorial experiment was conducted with a one-cup Bialetti Moka Express, varying roast level, grind size and initial water temperature in a 3 × 3 × 2 design with five replicates per condition (n = 90). Total dissolved solids (TDS) were measured in triplicate with a DiFluid R2 Extract refractometer, and extraction yield (EY) was calculated from TDS, beverage mass and coffee dose. EY was analysed using a linear mixed model; dispersion was compared with Brown–Forsythe tests on within-condition deviations. Within the tested ranges, grind size was the strongest lever for EY (coarse 13.5%, fine 21.7%). Roast level had a smaller, directed effect, mainly separating the light roast from the medium and dark roasts. The initial water temperature did not measurably affect EY, although pre-heated water shortened the preparation time. Dispersion did not differ between factor levels once mean differences were accounted for. Read through a craft-consumption lens, the study illustrates how portable analytics can support measurement-enabled practice by making brewing questions experimentally testable.