Flyback Converter Primary to Secondary Capacitor

[Henneshoe]

Hi All, I am looking for a sanity check here. (And maybe a lesson in SMPS design.)

I have a Flyback converter that was generating a lot of RF emissions. We placed a 4.7nF/500V capacitor between the primary and secondary windings and it greatly reduced the RF emissions (-15dB). I am a little fuzzy on what is happening.

My theory is during operation a common mode voltage/current is coupling from the primary to the secondary of the flyback transformer. Without the capacitor there is no convenient return path for the energy and it radiates RF emissions. The capacitor provides a return path for the energy. I have read a few online forums that support this theory.

From your experience, is this the accepted way of handling this problem?

 

[Comcokid]

On any SMPS with a transformer isolating primary and secondary, a small high-voltage capacitor between the primary and secondary side is typically needed to allow the induced common mode current a path to close-the-loop. Not only does the transformer couple magnetically, but the close proximity of the two windings has a parasitic capacitance and the high dV/dt in one winding couples to the other setting up current that will want to close-the-loop back to the source. Otherwise, this 'common mode' current will use any path that might be available, and this means that during conducted emissions testing it will circulate back through the LISN(s) and show-up as an emissions problem.

The capacitor may not necessarily be located from winding-to-winding but might be located from the ground on one side to the ground on the other side, or may be located from each ground side to a chassis ground, or any number of other places.

If a capacitor between primary and secondary sides is not acceptable, an metal shield internally in the transformer between the windings to a ground might be used.

 

[grigson]

i agree with that.

Also, i am sure you know that its quite common to have to dotted ends of the transformer coils "Lining up" so that you get less dv/dt between them.....ie when one end of one coil goes positive, the end of the adjacent coil also goes positive so the delta v between them stays the same, or at least does not change so drastically.

 

[Henneshoe]

Thank you Comcokid and Grigson for your input.

For clarification, we placed our capacitors on the "ground" of the secondary and the "ground" of the primary. Does 4.7nF seem like a reasonable value? We have more testing scheduled this week and as part of that we want to try larger and smaller value capacitors. Do you think we will see any difference in our emissions?

Grigson, for this design we used a transformer from Wuerth (see attached datasheet). N1, N2,and N3 are in series for the primary, N4, N5, and N6 are the secondary. It does not appear the dots are on the same side of the part. I am a bit confused by your explanation, so in your opinion is this not optimal?

 

[MacGyverS2000]

The dots are simply a convention to help the engineer recognize what side of the input/output coils are in phase from a voltage standpoint. How you connect to those coils is entirely up to you.

Dan - Owner

http://www.Hi-TecDesigns.com