A question on EMC in product design

[gardenyu]

Hi:

The company was integrating a controller into a lamp fixture. The controller is FCC approved for EMC performance by testing individually. The fixture include a lamp, the controller and a wiring box, and the whole fixture failed the FCC conduction test.

We had to put a metal box around the controller box (originally plastic) to pass the radiation, but of no help to conduction.

In the conduction test report, the EMC looks a lot worse under 1MHz (about 40DB poorer), but in higher frequency looks pretty similar.

Is there any simple solutions? I'm thinking of soldering a film cap across the power feed, or put some shielding materials inside the wiring box (the opening for wiring to the controller is a little big). I also wonder if there is any way to bypass the metal box for radiation test.

The controller box does not have a earth ground input, plus it was potted so pretty much there is nothing can be done inside the box.

Any reply is appreciated.

 

[eem2am]

Hello my freind,

please see



as you will see, fluorescent lights NEVER pass EMC tests.

...but we MUST still use them as they are much more efficient than incandescent bulbs and we would soon burn up all the worlds dwindling oil supply in our power stations if we used incandescent.

so thats why we use fluorescent.....even though they never could pass an EMC test....NEVER.....I assure you of this.

just think of a metre long fluorescent tube with the "go" connection at one end and the "return" connection at the other......as you can tell, you can't even twist the "Go" and "return" wires together as they connect to opposite ends of the fluorescent tube....so therefore, you never could pass EMC as the fluorescent lamp current is horrendously full of harmonics.

if you try to pass EMC with a ballast for fluorescent, you will be there for a long long time.....and you won't manage