18V@4A secondary transformer with less than 100pF between secondary and mains ground.

IMO primary-secondary capacitance is much more important.
Most of noise from the mains spreads through it.

Two bobbin/splitted bobbin transformers has enough low p-s capacitance.

Shielding (tape or one layer wire between primary and secondary) also gives a good result.
 
Hans, I have a similar requirement for my eDD NextGen design, but I also need 5kV isolation between prim and sec.
I spoke with Menno van der Veen and it's a tough target. It can easily be more than a nano-Farad.
You could talk to him.
At any rate, it will come with very much reduced efficiency.

Jan
 
IMO primary-secondary capacitance is much more important.
Most of noise from the mains spreads through it.

Two bobbin/splitted bobbin transformers has enough low p-s capacitance.

Shielding (tape or one layer wire between primary and secondary) also gives a good result.
Yes, that’s the capacity I’m referring to.
Do you mean two Transformers one after the other, both with EI cores and each having two separate bobbins ?

Hans
 
Yes, that’s the capacity I’m referring to.
Do you mean two Transformers one after the other, both with EI cores and each having two separate bobbins ?

Hans
No.
I thought such transformer, when primary part and secondary part are separated:
1.) separated bobbins for pri and sec on the far legs of EI transformer;
2.) or splitted bobbin, when primary and secondary are in separated bobbin part.

I prefer the latter.
Sample:
https://www.reichelt.de/de/en/mains...86a52bb061f180924cf94a811132&LANGUAGE=EN&&r=1
 
The R-core may not help anyway. At least it will depend on how the transformer is made.
I just made a quick measurement on an R-core transformer I have. It measured something like 110pF between primary and secondaries. It was a different transformer with 2 x 15V and 2 x 9V secondaries.
But I think that the primary is split between the two bobbins and the same for the secondaries, probably to get a good coupling/regulation. For lowest capacitance the primary should be on one bobbin and the secondary on the other one. This will probably sacrifice some of the coupling, but I think the capacitance should be lower.