Abstract :
A source of error in microwave attenuation measurements, not previously evaluated, is treated in an original analysis. The error is caused by effects of deviations of connectors or adapters from standard specifications. For example, if the coaxial connector at the insertion point in an attenuation measuring system deviates from standard specifications, the measured attenuation of a coaxial pad inserted in this system may differ from the measured attenuation of the same pad when inserted in a system having a standard connector at the insertion point. In the analysis, an attenuator installed in a system is represented by three cascaded two-ports, the central one representing the core or kernel of the attenuator, and the others representing connector pairs. Three general cases are considered: the waveguide component under test has 1) either sexless or mating connectors, 2) nonmating connectors of the same type and sex, or 3) nonmating connectors of different types. The results indicate that significant errors are possible with present connectors, such as the type N, but that these types of errors become negligible when high-precision connectors are used. The analysis permits one to tell how good a connector must be for a specific attenuator application.
