How to measure test repeatability when stability and constant variance are not observed

Abstract

Passive intermodulation (PIM) is a critical measurement for radio frequency (RF) communication networks. Yet PIM measurement inherently has very poor repeatability, which makes product assessment unreliable. The RF industry struggles with the issue since there are no known solutions. With the increasing demand for low-PIM performance, there are pressing demands to address the challenge. Two fundamental problems make traditional gage R and R study invalid for PIM: (1) PIM in nature is unstable and unrepeatable; (2) PIM measurement has inherently inconstant variance at different PIM levels, which is primarily due to limited capability of PIM analyzer. This resulted in several less-known issues significantly impacting the estimation of PIM test repeatability, including sample selection, one-sided spec and differences between test R and R and gage R and R. The paper proposed two fundamental changes when studying R and R of PIM test or tests in general violating constant variance assumptions: (1) sample selection; (2) what measurement to use to better estimate and represent the test repeatability. Special sampling is proposed to minimize the impact of inconstant variance. A more direct R and R measurement, margin of error (MOE), also known as study variation, is proposed to replace traditional gage R and R metrics to more meaningfully represent PIM test R and R. Several statistically based techniques to improve the repeatability and reliability of PIM measurement are also discussed. The study and proposed solutions apply to not only PIM test but also tests in general violating constant variance assumptions.