[RFI] RFI Digest, Vol 241, Issue 21

[Jim Brown]

Excellent post, Kenneth. I've been involved in EMC for more than 25 
years as Vice-Chair of the EMC WG of the Audio Engineering Society 
Standards Committee, largely composed of members from the UK and North 
America. We first became aware of serious RFI issues from variable-speed 
motor controllers for elevators in buildings with large performance 
spaces, and all of the issues were the result of massive failures by 
system designers. We learned then that controllers were usually widely 
separated from their motors, and wiring practices were ignorant of the 
fundamentals of preventing both magnetic and electromagnetic coupling.

The same failings are present in consumer products like HVAC systems, 
and in industrial control systems. I'm encouraged that you and your 
colleagues are doing your part to educate designers of these systems.

One of the most important principles that we learned is that avoiding 
the creation of the fields is at least as important as shielding. For 
example, all current-carrying conductors MUST be twisted pair, and 
circuit layout, including board layout, is critical. In the NJ 
Performing Arts Center, where Neil Muncy first encountered the issue 
with elevator motors, it was the magnetic field that was coupling into 
audio system wiring. Since current-carrying conductors are nearly always 
bypassed to "ground," it is critical that ground conductors be run WITH 
and tightly couples to current carrying conductors so that they function 
as transmission lines.

In his lectures, Henry Ott has emphasized this principle, noting that 
when "ground" layers on boards are interrupted under signal, the traces 
no longer form transmission lines, wreaking all sorts of havoc. He also 
observes the invisible schematic hiding behind the ground symbol.

73, Jim K9YC

On 2/14/2023 12:20 PM, Kenneth Wyatt via RFI wrote:
> Steve is correct that it’s possible to purchase large appliances that are designed with EMC in mind (mostly for the European market). As an EMC consultant for the past 15 years, I’ve helped several appliance clients with their washing machines, air conditioning systems, etc. achieve EMI compliance. The problems I’ve seen include EMI mains filters located too far inside the works, which allows noise coupling directly into the mains wiring. Also the VS drive electronics is not fully enclosed with shielding. But the biggest problem is that the internal cable routing is haphazard, allowing multiple coupling between noisy circuits and mains wiring, which can lead to both radiated and conducted emissions.
> 
> I, and a handful of others, in the world also teach product design for compliance to EMC and product safety, but these topics are largely ignored in college and university. The result is that designers end up learning the hard way about EMC and the end users bear the consequences.
>