IEEE 1560: “IEEE Standard for Methods of Measurement of Radio-Frequency Power-Line Interference Filter in the Range of 100 Hz to 10 GHz”
IEEE 1560 was first published in 2005, then reaffirmed in 2012. It has been inactive since 2023. It’s available for purchase here. Inactive doesn’t mean that there’s anything wrong with it, it mostly means that people got behind on the paperwork. We were able to approve a renewal effort to move forward at the EMC Symposium in the summer of 2024, and it will be up for IEEE SA approval in the Fall. Once that comes through, a working group will be organized under the leadership of Dasha Nemashkalo of the University of Twente. If you would like to be involved in that effort, please contact me at standards@emcunited.com.
Once upon a time, a customer asked for help evaluating the effectiveness of certain power line filters they had procured. They were building a secure shield room and wanted in situ measurements to determine if the filters were performing as advertised. I looked up the filters and the marketing copy said: “Our shielded room filters are designed to meet or exceed the requirements of MIL-STD-220 with 100 dB attenuation from 14 KHz to 10 GHz.”
So I looked up MIL-STD-220 (download here) and found a test method that looks like this:
Which assumes 50 Ohm coaxial connections throughout. This makes perfect sense for the kind of RF filters that MIL-STD-220 was written for (“Method of Insertion Loss Measurement”, freely available here). It does not make sense for systems that look like this, with not a coax connector in sight:
It turns out that prior to the first publication of IEEE 1560 in 2005, there was no widely accepted standard document governing the testing of power line filters, so the industry picked up the closest related standard, MIL-STD-220 for RF filters. Now, I have to admit, the proof is in the pudding: manufacturers have been testing to this method for decades and obviously haven’t had any massive customer revolts over inadequate performance. So something is working right-most probably some excellent design practices by the suppliers in question. But in any sane universe, we would have switched to the more tailored and rigorous IEEE 1560 document by now. (I went on a bit of a rant about this during a talk I gave to the SE Michigan EMC chapter on how different standards stack up to each other.)
The following test methods are included in IEEE 1560:
Quality assurance--No load (10 kHz to 1 GHz)
Quality assurance--Loaded (10 kHz to 20 MHz)
RF characteristic mismatched impedance--No load (100 kHz to 30 MHz)
Variable source impedance attenuation measurement (100 Hz to 100 kHz)
Attenuation measurement (100 kHz to 30 MHz)
Aperture leak test by electric fields (1 GHz to 10 GHz)
Voltage drop and waveform quality test (linear/non-linear loading)
S-parameter measurement (100 Hz to 30 MHz)
Eagle-eyed readers will notice that the only test covering 30 MHz - 1 GHz is “Quality assurance--no load”, which is described as a workmanship test more than a performance characterization measurement. As we move forward with the 1560 renewal effort, this is the kind of thing we’ll need to decide how to address.