[Milsurplus] EMI from LED lighting

Dennis Monticelli dennis.monticelli at gmail.com
Sat Mar 8 00:18:45 EST 2014


My company makes IC's for electronic ballasts so I can provide a little
insight.  Be forewarned that the story isn't a comforting one.

First of all the LED itself is as benign as an incandescent lamp.  The
problem lies with the switching power supplies that feed a constant current
from a high voltage to the low voltage of the LEDs.   All these products
must pass FCC Part 15.  They are not cut any slack with regard to meeting
the spectral mask just because they are "green" and represent a
multi-billlion dollar industry.  Unfortunately, the FCC's spectral mask in
the HF spectrum is insufficient to protect a sensitive receiver from noise.
 Lights are everywhere so unless your antenna is far removed from your home
and fed with well-shielded coax, you will likely experience some noise
proportional to that proximity.  Most of the noise will emanate from the AC
line feed not the small circuit board.  So choking off the currents on the
AC line will help more than extra shielding around the lamp.  The IC makers
are under pressure to make the chips cheap and to require minimal external
components for meeting EMI.  This translates into various PWM schemes that
smear out the noise spectrum so as to avoid high amplitude fixed frequency
spikes.  So instead of getting a few strong raspy tones in our radios we
get a general raising of the noise floor.  The total noise energy is the
same, just spread out.  The more consumer PWM stuff near your antennas, the
higher the noise floor becomes.

I wish I had a happier story to tell.  Unfortunately, PWM devices are
appearing everywhere in the home while the FCC is all about "let the market
decide" and opening up spectrum (including your home power line) for
various digital data formats.  The FCC doesn't really care about AM radio
and the next generation of consumers doesn't either. The BCB is so
"yesterday."  Hams are just kind of caught up in the tide.

Dennis AE6C


*A friend asks:*




























*My biggest question about LED lighting is how much radio interference
itgenerates at frequencies up to about 30 MHz (HF).  Other users of
theradio spectrum are more concerned by interference at higher
frequencies,through VHF and UHF.AFAIK, an LED by itself does not generate
radio interference; it is theelectronics between the LEDs and the AC power
line that concerns me.Fluorescent lights (both CFLs and long straight
tubes) with electronicballasts are bad because an electronic ballast
switches AC line power withvery short rise & fall times, so the ballast
strongly excites the linewith radio-frequency power at whole-number
multiples of the switchingfrequency up to 30 MHz and beyond.It is possible
to reduce this radio ?noise? or interference to a tolerablelevel by means
of shields, L-C filters, and ferrite chokes; but very fewmanufacturers of
fluorescent lighting products do.  If they do anything,it is nowhere near
enough.I?m waiting for the prices of LED-lighting products to drop further
beforeI buy some for evaluation.  Meanwhile, can anyone direct us to
published,quantitative information about the radio noise generated by LED
lightingproducts?Best,-John*


More information about the Milsurplus mailing list