Check out my LISN design. It includes a wideband current probe that CORRECTLY measures common-mode EMI, something that has never been properly addressed in the measurement standards. They all ignore the role of the grounding conductor as a path for conducted emissions...?
90% of the EMI cases we all deal with can be completely solved with a properly-designed common-mode choke encompassing all AC line conductors. Packaged EMI filters often will not solve the problem because the EMI current is carried via the grounded conductor, which is simply carried through the filter. That being said, installing such a filter inside the offending equipment, with its case bonded to the equipment case, is more likely to be effective than placing the filter in an external enclosure.
As for LED bulbs, I have characterized a number of them and detailed reports are on my RFI web page. There are many screw-in bulbs that are indeed compliant. The ARRL RFI lab has similar data which they will eventually publish. In contrast, I have yet to find an LED fixture (with their ghastly switchmode converters) that comes close to being Part 15 compliant. ?
Gary NA6O
