Why at all do you want to route PE across the board?
Apart from this point, EMC considerations often suggest to connect a decoupling capacitor between circuit ground and PE, but I don't know if it's appropriate for your design.
Why at all do you want to route PE across the board?
Apart from this point, EMC considerations often suggest to connect a decoupling capacitor between circuit ground and PE, but I don't know if it's appropriate for your design.
There might be reasons to route PE over PCB, but without knowing the particular requirements, it's hard to discuss.
I remember a design, where low inductance decoupling between PE (metallic case) and circuit ground was intended, the solution was a PE copper pour overlapping the ground plane. But in another design, you may want low capacitance between ground and PE...
Personally, all the designs I have seen in the last several years have usually had the board 'ground' connected to the PE by a largish SMD capacitor (large to get the spacing between the pads). And there is a distinct isolation area between any PE connections on board and the rest of the circuitry. Having a PE plane will couple this capacitively to all sorts of signals, or the board ground if that is the nest plane, not a good idea for EMC. I recently did some boards that had to be immune up to 18GHz susceptibility and had an RF engineer advice on all aspects of the design from layout to filters chosen.
Don't forget some EMC engineers call PE the noise distribution network.
Talking about "PE" without referring to a real instrument design is about to meaningless, I fear. In so far I won't contradict your comment, just say, there are different cases as well.
I agree, one of Henry Otts papers refers to a screen plane under the ground plane. But with todays high frequency noise having two planes (PE and GND) is going to couple HF RF noise capacitivly.