Author Topic: EMI filtering of LED light strips  (Read 897 times)

0 Members and 1 Guest are viewing this topic.

Offline rhbTopic starter

  • Super Contributor
  • ***
  • Posts: 3483
  • Country: us
EMI filtering of LED light strips
« on: December 24, 2023, 01:26:28 am »
In case anyone else is concerned with EMI.

I'm installing 48x 8' 5500 lumen LED strips in the metal buirlding where I plan to build a screen room.  On advice from a friend who was a TEMPEST engineer I bought CORCOM 6ESB1 & 20ESB6 filters with the intent of cascading them.  I already had some 20VB1s

This shows the 6ESB1 and the 6ESB1-ESB6 pairing.

1961175-0

As I had them on hand I thought I'd see what benefit a 20VB1 might offer.  The results were interesting as the order of filters made a significant difference at low frequencies.

1961181-1

shows all 3 filters in different order with detail to 1 MHz

1961187-2

The ES filters have 330k ohm shunt resistors and the VB has a  1.5 M ohm shunt.  After a series of experiments with a 4395A I found that the 6ESB1-20VB1-20ES6 gave the best attenuation.

These are my general shop lights in a 1530 sq ft building.  It consists of 16 8' banks of lights each individually switchable.  I'm buildnig an 11'x16' screen room in the shop, so keeping noise off the lines at the source is important.

All of the wiring is in EMT and will have hardware cloth screens over the lamps.  The lab itself will be on an isolation transformer with lots of filtering.

Have Fun!
Reg
« Last Edit: December 24, 2023, 01:28:48 am by rhb »
 
The following users thanked this post: boB

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3250
  • Country: de
Re: EMI filtering of LED light strips
« Reply #1 on: December 24, 2023, 08:26:41 am »
Hmm,

on shielded cabins I always see feedthrough capacitors like these:
https://www.pikatron-gruppe.de/fileadmin/downloads/tesch/durchfuehrungsfilter/tesch_db_dff_75a_a14x26_de_en.pdf

And usually halogen incandescent lamps instead of LED lamps.

with best regards

Andreas
 
The following users thanked this post: 2N3055

Offline miro123

  • Regular Contributor
  • *
  • Posts: 206
  • Country: nl
Re: EMI filtering of LED light strips
« Reply #2 on: December 24, 2023, 04:03:56 pm »
what type of LED driver are you using? Good CE UL certified led drivers already offer low EMI at their input and outputs.
Insulation 50/60Hz transformers usually does not provide any advantage and in many cases makes thinks worse.
You can make the calculation for you trnasformer  Xc=1/(2&Pi*F*C)
C - trnaformer leak capacitance
F - the first harmonic on PWM switching is in 20..100KHz range . Unusually frequency to 7 harmonics makes affects the conducted EMI
Frequecy above 30MHZ mostly represent the radiated EMI. Most likely transmuted as common mode noise.
Yes - cascading of many different filters achieve the best results.
Downside of cascading is that you increase  the total leakage current. It could be safety issue is you exceed the maximum allowed leakage current. I did not remembers how much was allowed in US/ UL . In EU is 0.5mA typical 0,75mA max

« Last Edit: December 24, 2023, 04:35:48 pm by miro123 »
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14216
  • Country: de
Re: EMI filtering of LED light strips
« Reply #3 on: December 24, 2023, 07:20:20 pm »
There are quite different LED light. Some are noisy with SMPS type supply and others are low noise with capacitve droppers or a simple linear current limit at rectified mains (no filter).
The should be no problem using one of the quieter LED lamps.
 
The following users thanked this post: 2N3055

Offline mendip_discovery

  • Frequent Contributor
  • **
  • Posts: 851
  • Country: gb
Re: EMI filtering of LED light strips
« Reply #4 on: December 24, 2023, 09:18:59 pm »
I was only talking to a customer about this the other week. They manufactur LED lighting and I was in the R and D area talking to one of the grey beards.

He talked about the manufacturer's of the drivers and how they have very specific requirements regarding how you wire them up, some wires needing to be kept together and some having to cross others at 90 degrees, some needing to be kept between certain lengths etc. Space doesn't always allow for it practically but do as much as you can.
Motorcyclist, Nerd, and I work in a Calibration Lab :-)
--
So everyone is clear, Calibration = Taking Measurement against a known source, Verification = Checking Calibration against Specification, Adjustment = Adjusting the unit to be within specifications.
 

Offline rhbTopic starter

  • Super Contributor
  • ***
  • Posts: 3483
  • Country: us
Re: EMI filtering of LED light strips
« Reply #5 on: December 24, 2023, 09:38:48 pm »
This is the general workshop lighting, not lab lighting.  I'm using the cheapest LED lamps I could find.   I've got wood and metal working facilities which need lots of light.   Expected result is over 150 lumens/sq ft.   I'm  testing a 900,000 lumen spotlight for arc welding shortly.

The 180 sq ft screen room is to be contained within a 1530 sq ft metal building with its own HVAC zone.  There are 2 zones for the building as a whole which has a partition down the middle for a total of 3 zones.  The  thermally insulated shielded room zone will be inside one of the other 2 zones.

My friend with the TEMPEST II cert said the filter arrangement  looks great, but noted reality is not 50 ohms.  Actual mileage may vary.  So I'll sweep it once the filters are installed.

Interestingly, he recommended not having *any* light fixtures in the lab.  Well characterized screened windows e.g mesh or honeycomb screens instead with external lights.  Professionally he used honeycomb, but said that 1/8" hardware cloth was as good if properly bonded.  HVAC is a mini-split which will need it's own cage.

Have Fun!
Reg
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf