Subjecting failure modes in immunity tests to uncertainty analysis

A good deal of effort has been expended in attempting to understand and reduce uncertainties associated with emission testing. Uncertainties associated with immunity tests are more difficult to quantify. This paper identifies some of these problems, and proposes some solutions. The uncertainties arise in several areas. The standards may not define the test set-up or the failure criteria in sufficient detail; the calibration of the disturbing signal will have an associated uncertainty; the response of the apparatus may be difficult to measure or assess, or may vary according to the mode of operation of the product. The principle of immunity testing is straightforward. An EMC immunity test applies an electromagnetic disturbance of a particular type and amplitude to an equipment under test (EUT), and the effect on the EUT is observed. The observation is compared with the requirement in the test specification to determine whether or not the EUT has passed the test. The electromagnetic disturbance applied to the EUT may be continuous or of a transient nature, although in current standards many continuous tests are subject to time-varying amplitude (e.g. 80% amplitude modulation), and the transient tests can involve repetitive disturbances (e.g. bursts of fast transients)