Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Radiated emissions problems

Bjtpower

Full Member level 5
Joined
Dec 24, 2015
Messages
280
Helped
2
Reputation
4
Reaction score
2
Trophy points
18
Activity points
2,423
I have designed Power supply for Quasi Flyback converter.
Input voltage-230VAC
Output Voltage-17VDC at 4A
Switcher frequency at full load is 48KHZ.

While radiated emissions test, I am failing by 2dB at 60MHZ and 120MHZ.

I did analysis with spectrum analyser and observed that at Bulk capacitor -ve and Shunt resistor below source of mosfet, I could see worse results.
Can anyone help with how we can mitigate this issue?
Regards,
Marx
 

dick_freebird

Advanced Member level 5
Joined
Mar 4, 2008
Messages
7,126
Helped
2,088
Reputation
4,180
Reaction score
1,934
Trophy points
1,393
Location
USA
Activity points
57,169
Probably should show the PCB layout (at least
the close-in, power switch, transformer and
bypass / filter loops), the waveforms (e.g. is
60MHz really the half-sine phase of the power
pulse, or is it really the reciprocal or risetime?)
and schematic (e.g. is there or is there not a
snubber, what traces might take off from hard
switched nodes as antennae, etc.?).

When you say "Bulk capacitor" does that mean
the input filter, output filter, ???
 

dick_freebird

Advanced Member level 5
Joined
Mar 4, 2008
Messages
7,126
Helped
2,088
Reputation
4,180
Reaction score
1,934
Trophy points
1,393
Location
USA
Activity points
57,169
Input filter loop contains a large discontinuous
current-toggle, which wants loop-area minimized.

The ESR/ESL of the filter in aggregate, can also
create a voltage-toggle which yields an E-field.
This wants higher quality (but not too high...)
capacitors. Too high Q can make every output
switching edge into a "tone burst".

If you can determine whether the EMI is coupled
magnetically (current toggle) or electrically
(voltage toggle) then maybe the solution will
reveal itself.

You might inspect the board with a real time
AC coupled 'scope looking for places where
60MHz can be found. It may be "bursty" or
continuous (following output switching as
a high harmonic). Might want to see what
you may have for peaks below 60MHz (they
may exist, but fall just within the EMI fail-mask).
Do you have 20MHz, 12MHz, 9 MHz, ... (odd integer
divisions of the trouble tones) visible? The
fundamental is what needs to be run to ground
(heh) to figure out its source and mitigation.
But also accept the possibility that this is not
fundamental-synchronous but rather self-
generated (Q of filters, as tanks). The nature of
the 60MHz (burst or continuous) is the clue for
that.
 

Bjtpower

Full Member level 5
Joined
Dec 24, 2015
Messages
280
Helped
2
Reputation
4
Reaction score
2
Trophy points
18
Activity points
2,423
Hi Dick..
I checked today with E-Probes..
Yes.. I can see peak at 2MHZ,7MHZ and 13MHZ which is more than 67MHZ and 118MHz.

Regards,
Deepak
 

Bjtpower

Full Member level 5
Joined
Dec 24, 2015
Messages
280
Helped
2
Reputation
4
Reaction score
2
Trophy points
18
Activity points
2,423
Hi..
We can see noise by electrical field..
Could you please explain where we need to check apart from Layout.
I wish to pass without PCB because it is well designed and still I don't see any opportunities to look into PCB.
 

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top