Discussion: Figure 1 shows two rubber bands stretched over two signal paths of a test board that is used in several Technical Tidbit articles on this website. In some sense, the rubber bands serve to illustrate how a signal loop is affected by a break in a ground plane. The signal path at the bottom of the board is over a solid ground plane. All signal paths define a loop, from source to load and back again. In the lower path in Figure 1, the signal path extends from the BNC connector to a 47 Ohm resistor on the right (hidden by the rubber band). The signal returns by the path of least impedance, which at frequencies much above tens of kHz is the path of least inductance. This is right under the signal path in the ground plane. The current in the ground plane spreads out somewhat, but not much, being confined mostly to a few widths of the signal path in the ground plane.

The two sides of the rubber band loop also are very close to each other because force is needed to stretch the bands further apart, much like the electrical signal path.

In the upper path, the signal loop extends from the BNC connector to the 47 Ohm load on the right (partially obscured by the rubber band),  The signal returns over the path of least inductance as before, but this time it is forced around the end of the ground plane break.  The returning signal current will stay near the signal to load path until close to the break and then follow the break closely around its end and back to the signal path, opening a much larger loop between the two sides of the signal loop than without the break. The signal to load path and the returning path in the ground plane now define a good sized loop which has all manner of unpleasant effects on circuit operation. Five Technical Tidbit articles on this site describe some of these effects and are linked at the end of this article. The unintended effects of the ground break include: crosstalk, lowered immunity to EMI, increased emissions, and slowing of the signal risetime.

The upper rubber band in Figure 1 acts in a similar manner. The two halves of the rubber band want to be close to each other. Force is required to separate them and pull one half of the rubber band to the end of the ground break. The rubber band would like to snap into place again, but is being held apart, defining a large loop. The force holding the rubber band at the end of the ground break is like the increased loop inductance in the signal loop of the electrical signal. The current really wants to have the two parts of its loop close to each other to minimize inductance. The difference between the two cases is that the current hugs the ground break whereas the rubber band forms a triangle. But the analogy is good enough for a manager.

Figure 2 shows another example of forcing a signal return away from the signal path, a pigtail connection of a shield. If you connect a coaxial cable using the device in Figure 2, it is not difficult using a three Volt, 30 MHz square wave with a two nanosecond risetime to exceed emission limits by 40+ dB. If the device of Figure 2 is inserted in a coaxial cable of any cost, the shielding performance of the coaxial cable will likely be worst than the cheapest cable you can buy without the device inserted.