Hi
I have following remarks regarding your flyback circuit:
1) You need to know the leakage inductance for proper dimensioning of your RCD clamp. Leakage inductance is completely determined by construction of your transformer!
2) Does your product belong to class 1 or class 2 safety class?
3) As always with reference circuits, you need to spend additional thoughs about EMI filtering. There is a big chance that this design will fail conducted emission tests. There is a big chance that the 0.33nF capacitor is not sufficient to pass these tests.
4) I don't see any surge protection at the input. All EMI standards have some form of surge testing at mains.
the design was automatically done by power.com using innoswitch3 after i gave them the specs needed. however, please look at my circuit in attachment for input protection for a Mornsun (or Hi-link) power module. it has all recommended input protections done.
as i mentioned, i have very simple way of doing stuff. i designed this awesome power supply around recom power module (+ my own buck regulator circuits after it) and had only 5 in stock as test batch. all sold instantly even without me advertising it. i made a quality power supply for the market which demands it. it is for sega dreamcast console... a true better replacement of original power supply which excels at everything and provide better performance.
i plan to make this and sell it via my personal website + 2 big retailers in the small niche market (retro game consoles). absolutely no certification or paid tests will be done... i just don''t have enough money to do them. these retailers never asked for anything either.
maybe what you guys say apply when i want it to be soled at a bigger level or more mainstream... but this niche i don;t think it would matter.
You spend so much money on the AC power supply and then you go with some SamYoung electrolytic caps ... at least that's what I think they are... can't even read a series on them, so who knows what they are. Maybe NXP because it's the only yellow series on lcsc ... but they print the series on the label on that one.
Why would you even need 3300uF worth of capacitance, aren't you using buck regulators running at high switching frequency to produce 5v and 3.3v? Not sure it makes sense to have so much capacitance...
1200-1500uF 16v solid capacitors from good brands are like 30-50 cents each
caps are samyoung NXH series:
https://www.lcsc.com/product-detail/Aluminum-Electrolytic-Capacitors-Leaded_SamYoung-Electronics-NXH-25V3300-16-25_C359208.htmlnot bad ones, 105 degrees 10k hours low esr. i needed big caps to eliminate last bit of noise. i saw them being used in apple chargers + did research and read about their specs before using them. my go-to for good quality and cheap were Lelon RZW but these were better.
i think if i enhanced the layout just a little more i will get better result without big caps... someone told me so after giving him the layout and a unit, he said i need to let the return current be forced to return to output caps negative side first then all the way to source...instead of having ground vias everywhere like i do now. plus i have 2nd stage filtering which i think can be eliminated as well.
i will try his suggestion in my next project which is saturn psu, and if it worked better then i will revise this design to make it better.
i did use some good low esr solid polymer caps but weren't enough for 3.3v rail at 3amps due to layout not fully optimized. thus bigger caps were the compromise i did to solve it for that version and it did.