3
\$\begingroup\$

Here is my PCB drawing that couldn't fix EMC problem. There are peaks at 120-130 MHz at radiated measurement. Are there anyone who have any opinion to fix this?

Note: My circuit behave as led driver which drives at 500 kHz and output current is nearly 210 mA. R14 and R13 set output currents, these two resistors have 100 mA voltage drop and we can set output current by changing these resistors. At input stage I have EMC filter but there is still something I couldn't suppress.

pcb

result_updated_01

schematic

\$\endgroup\$
  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. Any conclusions reached should be edited back into the question and/or any answer(s). \$\endgroup\$ – Dave Tweed Jan 14 at 13:29
  • \$\begingroup\$ Why do you appear to have random inductors sitting in the middle of current return paths? A large part of EMI is having parasitic inductance in the current return paths and in your circuit you have actually placed inductors there (namely L3 and L6, but also perhaps L5) \$\endgroup\$ – DKNguyen Jan 14 at 17:36
2
\$\begingroup\$

Because this is a switching device, care must be taken in proper decoupling of the input.

You have a wrong order of your input caps C15 and C13. You must connect C15 (100nF) as close as possible between Vin and Gnd inputs of the switcher.

Just try it out, you will see much difference in EMI measurement.

BTW: In my experience, this is one of the top 3 reasons/causes why designs fail at EMI tests.

\$\endgroup\$
  • \$\begingroup\$ Dude you completely right. Today i redraw pcb and putted a 100nf as close as IC. I increased output inductor to the 150U and it passed. Thank you all btw. \$\endgroup\$ – IHK Jan 15 at 17:08
2
\$\begingroup\$

While I can't give you a definitive answer on what will fix your issue, three suggestions jump out immediately...

  1. Treat LED+ and LED- like a differential pair. I suspect this to be the source of most of the radiation, since it has the largest loops. The current should flow from C14 through the LEDs, and then back to C14 as close as possible to the outward flow...preferably on the bottom side, under the LEDs and at least the center of the thermal pads.

  2. Is L5 shielded? If not, consider a shielded replacement.

  3. Increasing the value of C14 will keep more RF current close to the switcher and out of the large LED loop. Moving C14 down around L5, LED3, and LED4 will minimize your smaller switching loop.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.