CISPR 17 Measurements

The use of power line filter insertion loss/attenuation curve data (as published by filter manufacturers) has been a frustrating experience to less experienced EMC engineers for a very long time. Engineers with extensive experience in EMC consider these curves, generally prepared from data taken in a 50??test setup, to be of extremely limited value. In spite of this, filter manufacturers continue to publish 50??data, since popular 50??equipment, connectors and test cables make these the most easily taken measurements. Attenuation curves using 50??impedance are severely criticized in many books and technical papers as well as in insertion loss measurement standards such as Mil Std 220A and CISPR 17. CISPR 17 gives several alternatives to 50??insertion loss measurement curves. These alternatives are aimed at showing the filter effectiveness in real situations rather than in an artificial situation. The cause of the problem is that in real life situations, a power line filter is not terminated with 50??impedances. In fact, the filter termination is usually an unknown value that often changes with frequency. Since the filter performance is largely dependent upon termination impedance, the curves given in 50??can never represent the real live situation.