Publication

Recently, multiple evaluation systems (MES) that allow for a large number of tapping torque tests (T-3) to be performed on a single workpiece have been gaining in popularity for the evaluation of metalworking fluids (MWFs). However, MWF formulators have had difficulty obtaining statistically significant results or results consistent with experience in the field, raising questions about the efficacy of MES. This paper develops statistical and experimental design considerations for MWF evaluations by MES that aim to maximise the sensitivity of T-3 to MWF performance and to improve the correlation between laboratory and field performance. Towards this end, a metric of resolving power is developed that quantifies the ability of a T-3 operating condition (speed, material, tool size, etc.) to discriminate between MWFs. It is shown that as resolving power increases, the correlation of T-3 response to expected field performance increases. The paper concludes with a discussion regarding economic trade-offs between increased cost, resolving power, and statistical significance of T-3 results.