Bundled cable parameters and their impact on EMI measurement repeatability

EMI (electromagnetic interference) test procedures specify that long cables should be bundled at their center in some circumstances. The author investigates the effect of cable bundling using analytical models and measurements. He examines how cable bundle parameters such as length and `tightness´ can affect the repeatability of the measurement. Simple models of a bundled cable suggest that relatively small changes in the geometry of the bundle can significantly affect the common-mode cable current. Parameters such as length, tightness, location and the number of turns determine the impedance of the bundle. As the impedance of the cable bundle changes, the resonant frequency of the system shifts. This can result in large changes at the very frequencies where EMI problems are most likely to occur. It is also shown that lossy cables or cables with a lossy common-mode termination are less likely to be sensitive to minor changes in cable bundle parameters. The resonant peaks in a lossy system are smaller and cover a wider band of frequencies. Small shifts in the resonant frequency do not have as much of an impact on the currents induced at any one frequency