A novel scaling law for the tip vortex cavitation (TVC) noise was determined, employing the Rankine vortex model, the Rayleigh–Plesset equation, the lifting surface theory, the boundary layer effect, and the number of bubbles generated per unit time . All terms appearing in the final derived scaling law are well known three-dimensional (3D) lifting surface parameters, except for . In this study, the dependence of with inflow velocity and hydrofoil dimension is investigated experimentally while trying to retain the same TVC patterns among different experimental conditions. Afterward, the effect of on the TVC noise is analyzed. Optimal TVC observation conditions are determined from consideration of cavitation number and Reynolds number of two comparable conditions. Two geometrically scaled hydrofoils are concurrently placed in a cavitation tunnel for the hydrofoil size variation experiment. Wall effects and flow field interaction are prevented with the aid of computational fluid dynamics. Images taken with a high‐speed camera are used to count by visual inspection. The noise signals at all conditions are measured and an acoustic bubble counting technique, to supplement visual counting, is devised to determine acoustically from the measured noise data. The broad-band noise scaling law incorporating and the International Towing Tank Conference (ITTC) cavitation noise estimation rule for hydrofoil are both applied to estimate the TVC noise level for comparison with the measured noise level. The noise level estimated by the broad-band noise scaling law accounting for the acoustically estimated gives the best agreement with the measured noise level.