Offline Flyback SMPS: Y2 capacitor from primary to secondary can be high value?

Hi,
We have an offline 60W Flyback SMPS. It has live neutral and earth coming into its mains input connector. There are then two 2n2 Y2 capacitors ….from live to earth and from neutral to earth.
The Flyback secondary is isolated, and not connected to earth.
We also have a Y capacitor between primary and secondary grounds of the Flyback…..the secondary ground is not connected to earth.
As you know, Y2 capacitors connected to earth are limited in value…however, would you agree that the Y2 capacitor which connects between primary and secondary grounds need suffer no limitation in its value whatsoever?.....because it wont contribute any current flow to earth.

Hi,

We also have a Y capacitor between primary and secondary grounds of the Flyback

Click to expand...

What is "primary ground"?
If you don´t mean "EARTH_GROUND" then the impedance of the Y capacitors may enable current folow from secondary to earth.

Thus the bigger the capacitor the bigger the expectable current.

Limiting the Y capacitor value is because of safety.

Klaus

the value of Ycap from primary to sec limited in class II at 2.2n (i think... maybe 2.8...). this is actulal if user can touch the secondary side circuit (usb charger for example)
so if you designing internal ps for something in closed plastic box with only input terminals (energy meter for example), i suppose nobody cares what is your Ycap from input to out

What is "primary ground"?

Click to expand...

Thanks, By this i am referring to the node at the ground terminal of the primary current sense resistor....ie DC Bus negative.

so if you designing internal ps for something in closed plastic box with only input terminals (energy meter for example), i suppose nobody cares what is your Y cap from input to out

Click to expand...

Thanks, and so it follows that having a range of different sized Y capacitors across the isolation barrier, from big to small, would be best in mitigating common mode noise? (a range would be needed because eg a big capacitance woudl have a lower SRF, hence small caps with high SRF are also needed).
And presumably the best mitigation of all for common mode noise would be to de-isolate the Flyback SMPS altogether, and add a shorting piece of wire from primary ground to secondary ground……because then all the noise frequencies could return to the primary via this short piece of wire?

treez said:

so it follows that having a range of different sized Y capacitors across the isolation barrier, from big to small, would be best in mitigating common mode noise? (a range would be needed because eg a big capacitance woudl have a lower SRF, hence small caps with high SRF are also needed).

Click to expand...

i have no experience with paralleling of y caps, so cant say anything

treez said:

And presumably the best mitigation of all for common mode noise would be to de-isolate the Flyback SMPS altogether, and add a shorting piece of wire from primary ground to secondary ground……because then all the noise frequencies could return to the primary via this short piece of wire?

Click to expand...

i dont think it so straightforward... because i think it not just "noise frequencies return to the primary"

Thanks, i postulate that if we "de-isolate" a 60W offline flyback SMPS LED driver, by connecting its primary and secondary grounds together with a short wire piece...then that 60W flyback SMPS led driver gives essentially no more common mode noise problem than an offline 60W buckboost led driver, which uses no transformer and just uses a single winding inductor. Do you agree?

Basically yes, unless you have significant capacitance between switching circuit nodes and earth ground.

Thanks, from this, it follows that if we want to see if its worth adding shield windings to an offline Flyback SMPS, then we should simply short its primary and secondary grounds (if this is possible), and then see if the conducted EMC scan improves….if it does, then its worth adding in the shield windings…would you agree?