3 ms·
The description of how SMPS works in thei article is shockingly inaccurate! > To simplify, an SMPS essentially > takes the AC input and turns it on > and
by dmitrygr 8y ago
The description of how SMPS works in thei article is shockingly inaccurate!
> To simplify, an SMPS essentially
> takes the AC input and turns it on
> and off very quickly (between 10,000
> and 20,000 times per second), with
> the output voltage being determined
> by the duty cycle - the amount of
> time it is on for each flipflop. The
> shorter it is on, the lower the
> voltage. This is fed into regulators
> and capacitors to produce a stable
> AC voltage, which is then converted
> into DC.
Literally no part of that is accurate. SMPSs generally rectify first,then PWM, then smooth
- Animats 8y agoRight. But not the real issue. See [1][2]. Most switching power supplies have a small capacitor between the AC line and output sides of the transformer, to provide a somewhat tenuous path to ground for RF noise on the output side. That's a fundamentally bad idea, but cheap. Better approaches involve either a real ground, with a 3-prong plug, or some new designs which handle RF noise in a way that doesn't need a path to ground.[3] Here's a patent (in Chinese) for a solution without a bypass capacitor.[4] [1] https://i.stack.imgur.com/fV15S.png https://i.stack.imgur.com/fV15S.png [2] https://electronics.stackexchange.com/questions/216959/what-does-the-y-capacitor-in-a-smps-do https://electronics.stackexchange.com/questions/216959/what-... [3] https://ieeexplore.ieee.org/document/7488992 https://ieeexplore.ieee.org/document/7488992 [4] https://patents.google.com/patent/CN202818110U/en https://patents.google.com/patent/CN202818110U/en