Issue link: https://iconnect007.uberflip.com/i/1467744
62 PCB007 MAGAZINE I MAY 2022 He pointed out examples of different applica- tions with different requirements on a graphic showing transmission length versus transmis- sion speed. Each connector had its own impedance pro- file, which was location-dependent. Influenc- ing factors included transmission frequency, geometry and material, and the arrangement of components. Whereas PCB impedances could generally be considered as two-dimen- sional characteristics, two dimensions were insufficient to describe the characteristics of a connector. Stamm showed the impedance profile of an industrial Category 5 connector as measured with a time-domain reflectometer and explained the deviations from the nominal 100 ohm value. Discussing cause and effect mechanisms, he described how the impedance profile of a connector depended on the transmission fre- quency because relative permittivity was fre- quency-dependent and shorter discontinuities became visible with shorter pulse-rise times, and these caused larger deviations from the nominal impedance. He illustrated the effect graphically. For standard board-to-board connector designs, the exact applications and transmis- sion frequencies were not always known, but the target was to achieve the same impedance profiles within a product family so that the cus- tomer had the advantage of a scalable device design without influencing signal integrity. In the case of wire-to-board connectors, where transmission frequencies and require- ments such as return loss were specified, the target was to optimise impedance profiles until the transmission requirements were satisfied. He took the example of single-pair Ethernet connectors. e impedance profile of a connector was influenced by the geometry of the signal lines and its effect on the capacitance and inductance per unit length, and small clearances between signal conductors or from ground planes could lead to lower impedance values. Sometimes, the impedance value of a connector was too low as a consequence of the small space avail- able for installation. Stamm explained how impedances had been optimised by targeted adjustment of geometry, again with the objec- tive of achieving a consistent impedance pro- file within a product family. He went on to discuss various effects of mate- rial and component arrangement before finally mentioning an investigation during a develop- ment project of the effect of differences in con- tact lengths on run time and propagation delay. In fact, it was observed that the delay skew was 80 times less than the propagation delay, and negligible in practical terms. Aer the formal presentations, Anna Brock- man moderated an open discussion, inviting questions from the audience. From a personal point of view, I found this webinar extremely informative and well-bal- anced in content. I learned a lot. It filled in many gaps in my knowledge and brought the whole topic together in a logical and effective configuration. My thanks to the organisers for a first-class event. DESIGN007 Click here to watch this educational presentation by NCAB Group, Polar Instruments and Phoenix Contact.