• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Power Supply Topology

A

aaqubit

Member
What topologies do computer power supplies use?

I read somewhere that most computers use a Forward topology, but also read in other places that this topology is good only up to ~200W, which obviously isn't enough.

 
Pretty much most are forward topology.

Most use a conventional single transistor forward, into a multi-output transformer. The converter is powered by a rectifying (optionally voltage doubling) the mains voltage to generate approx 330V DC. There is conjoint regulation of the 5 and 12 V lines (the average is taken, and regulated), with the 3.3 V line generated by mag-amp regulation of the 5 V line.

Active PFC PSUs use a high-power factor boost regulator before the reservoir caps to generate a semi-regulated 385 V for the forward converter.

High power PSUs often use 2 forward converters. 1 for 12V and 5V, the other for 12V and 3.3V. Some manufacturers will leave out the post-regulators and provide conjoint regulation for each converter. Some have opted for single output transformers, with the 5V and 3.3V supplies derived by post-regulating the 12V.

The most recent '80+' supplies may replace the forward converter with an active-clamp converter and replace the mag-amp post regulator with a synchronous buck. However, this type of design is uncommon, because the extra MOSFETs are expensive - many 'high efficiency' designs are simply tweaks to the old inefficient forward/mag-amp topology.
 
woa, thanks for the informative post

Can you provide any voltage/current/power limitations of the forward topology? I'm interested in making a power supply. I'm reading up a book ("Switching Power Supply Design" by Abraham I. Pressman) and it tells me that the forward topology is could above 60V, below ~200W. Even if you say that high power PSU often use 2 forward converters, does that imply a maximum power of ~200Wx2 = ~400W? This is still not a "high-end" supply. Am I missing something here?
 
Most PSUs use a single forward converter. It's only the really high power ones, 650 W and up that use two.

The idea of specific topologies having a 'power limit' is a bit inaccurate. There's no limit as such, instead there are power ranges where particular topolgies are more suitable - the low efficiency of the forward converter, and the high stress on its switch (single transistor version) are offset by its simplicity and cost. However, as power increases the stress on the switch and need for heatsinking becomes excessive. Higher efficiency designs with components operating under lower stress, e.g. half or full bridge are more complex and expensive, but are better able to handle the high powers.

In practice, some of the benefits of the bridge designs are obtained by using a 2 transistor forward - and this may be used in the higher power PSUs.

A 2 transistor topolgoy may be better than a single transistor one, and a bridge may be better than a forward converter, but the need for more swicthes and more complex control has to be balanced against the benefits. Evidently, the PSU manufacturers have found that a bigger heatsink is cheaper than a more efficient design.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top