On 3/26/2015 6:16 AM, Ryan O'Connor wrote:
> I've been designing a smps, and I wonder, is there an elegant way to
> design the input stage of the psu to be able to run from ~85-240VAC? I
> don't fancy using active PFC, and would like to keep it simple.
> Preference is for the voltage to automatically be detected so they
> don't have to switch a switch based on where they live.
>
> Cheers
> Ryan
I think you might want 85-264+. Much of the world is 230-240, so if=20
you're looking for some error margin you'll want 10% higher than the=20
nominal at least.

What is the topology of the supply?
A factor of 2.8 (240/85) without PFC and without switching is very=20
inefficient for medium-high power supplies, if not impossible.

Typical low power supplies use a flyback configuration, where the=20
voltage stress is Vin + N * vout. This can be extremely high, but for=20
low power supplies (5-10w) the primary current is so low that it's not=20
bad to use a high voltage low current MOSFET.

For higher power supplies, you start to look at half or full bridge=20
arrangements, where the combination of potential high current (at 85VAC=20
in) and high voltage stress (at 240+VAC) require fairly expensive=20
transistors OR big heat sinks.

Hence the reason for voltage switches on universal input supplies. They=20
use a capacitor doubler (find a schematic of an ATX power supply to see=20
this.)

But really, active PFC is the way to go here if you have more than ~30=20
watts. It makes things smaller, more efficient (despite the extra stage)=20
and cheaper. PFC chips probably cost about $.25 these days.

-
Martin

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.