the gutless, bloated, and fried power supply hall of shame

[mariushm]

Re: the gutless, bloated, and fried power supply hall of shame

Here's article discussing it:

http://www.eetimes.com/design/power-...ork-in-the-US-

You're right, seems PFC is not required but it really makes it easy to obide by all the rules if active pfc is used and possibly also passive pfc.

The European Union (EU) has developed a current harmonic content specification (EN61000-3-2) to improve line power utilization. Meeting these harmonic current criteria can be easily accomplished by using power factor correction (PFC) boost pre-regulators. However, achieving this specification does not necessarily require a power factor controller or a high power factor.

So those harmonics have to be met, pfc passive or active can be used for it but it's not required.

To reduce design costs, some power-supply designers found they could meet EN/IEC61000 requirements by using other means with some filtering. The designs could pass the harmonic requirements with a PF as low as 0.8.

However, Energy Star (Reference 1) is proposing a new PF legislation of 0.9 at full load in 2008 that will become part of the 80 Plus initiative (Reference 2). As this becomes a reality, some of the old IEC 61000 designs may not meet this new stringent requirement. In order to have a PF greater than 0.9.PFC, pre-regulators will need to be added to these designs

So if you want to label the psu as 80plus you need to use some form of pfc


later edit: this url seems to be a powerpoint presentation discussing the IEC 61000-3 standard : https://cdn.badcaps-static.com/pdfs/...53d00d3604.pdf

this also has some good info : https://cdn.badcaps-static.com/pdfs/...62c0d0ae47.pdf

hint : type in google "IEC 61000-3 filetype:pdf" and you get more.

<- this seems like the official standard 61000-3-3

[TELVM]

Re: the gutless, bloated, and fried power supply hall of shame

Pics of another gutless wonder from a friend:

After just a year of use this wonder went and he was lucky the attached comp survived the ordeal. Keeps blowing the fuse.

I told him to forget about repairing and don't waste any effort on it.

[Pentium4]

Re: the gutless, bloated, and fried power supply hall of shame

Maybe the NTC Thermistor is shorted? And man that thing sure could use some input filtering! Could you post a picture of the label?

[Behemot]

Re: the gutless, bloated, and fried power supply hall of shame

I'd say 300-350 W?

[tom66]

Re: the gutless, bloated, and fried power supply hall of shame

Not with the fake 330uF. If they are real, I put continuous power at 150W; but if they are fake 220uF (from the size they look to be around that, and I've seen it before in cheapies) I'd say squeezing any more than 100W would be asking for trouble.

Problem is with those caps being so small, the transistors have to work in overdrive to put more current through the transformer to keep the output stable. They usually don't manage it so you see 600~1000mVp-p of ripple on the output in a 100Hz sawtooth wave. See this picture from a "430W" PSU with 350W load. It used 680uF primaries, predicting 340W max output... and the ripple is already through the roof:
http://www.hardwaresecrets.com/artic...y-Review/332/7

Notice how the ripple peaks every 10ms -- this is no coincidence. It is at 100Hz, so is due to lacklustre primary bulk caps.

And of course, the transistors don't last as long... probably one of the contributions to them going bang!

[Behemot]

Re: the gutless, bloated, and fried power supply hall of shame

U think I am stupid? Was guessing label power…

[tom66]

Re: the gutless, bloated, and fried power supply hall of shame

U think I am stupid? Was guessing label power…

Sorry, that wasn't obvious from your post.

[Wester547]

Re: the gutless, bloated, and fried power supply hall of shame

It used 680uF primaries, predicting 340W max output... and the ripple is already through the roof:
http://www.hardwaresecrets.com/artic...y-Review/332/7

Notice how the ripple peaks every 10ms -- this is no coincidence. It is at 100Hz, so is due to lacklustre primary bulk caps.

And of course, the transistors don't last as long... probably one of the contributions to them going bang!

I don't think the primary capacitors are the reason for it being so inefficient. It only has a single TO-220 switcher rather than a TO-247 or TO-3P switcher - that's going to hurt it badly. This review of the Antec/Delta VP350 shows it doing 350W (though it exploded at 475W) with only a pair of 470uF/220V/105C CapXon capacitors, and the voltage regulation, noise, and ripple suppression was excellent, though the efficiency as an overall was only 76%. With very good cooling, superb AC/DC filtering, and enough of an overspec'd secondary and primary otherwise, I think you might be able to get away with smaller primary capacitors, though certainly not with fantastic efficiency. Granted, I also know that Delta design their PSUs such that their components aren't stressed, so that also helps, but even then...

[goodpsusearch]

Re: the gutless, bloated, and fried power supply hall of shame

This is an exception of the rule.

[tom66]

Re: the gutless, bloated, and fried power supply hall of shame

Yes, it uses a different topology though. The coincidence with capacitance =~ output power only appears to work for half bridge and flyback designs. With a different transformer design, you can get away with smaller capacitors.

[TELVM]

Re: the gutless, bloated, and fried power supply hall of shame

No pic but the man tells me the label pretended to be 500W, no less .

[tom66]

Re: the gutless, bloated, and fried power supply hall of shame

Oh my. It can't even do a third of that. My favourite is this PSU claiming 1050W, but only capable of a mere ~600W...
http://www.jonnyguru.com/modules.php...Story&reid=187

And again, 1200uF, so should be 600W capable, and it just starts topping out then with ripple through the roof.

[TELVM]

Re: the gutless, bloated, and fried power supply hall of shame

In fact we can see the label with a remarkably mendacious '500W' at the top of the second pic .

[Behemot]

Re: the gutless, bloated, and fried power supply hall of shame

It says 500W MAX But not for how big fractions of a second

[Wester547]

Re: the gutless, bloated, and fried power supply hall of shame

Yes, it uses a different topology though. The coincidence with capacitance =~ output power only appears to work for half bridge and flyback designs. With a different transformer design, you can get away with smaller capacitors.

Even more confusing, though, is that this power supply manages 300W in what appears to be half bridge topology with only 330uF/200V/85C OST SPS input capacitors (link)! Granted, the medium speed, 80mm ADDA fan in there seems to be running at about full speed the whole time by comparison to the more temperature controlled fans seen in other PSUs, and the +12V rail's voltage regulation is somewhat dismaying (though still within spec especially since the whole PSU is at 77% efficiency), but even then....

[tom66]

Re: the gutless, bloated, and fried power supply hall of shame

I can definitely see what looks to be a sawtooth wave on the output of the 12V, which looks too clean for normal switching noise so it's probably 100Hz. There's no obvious scale but once that starts exceeding normal ripple (~50mVp-p?) you really have a problem...

[Behemot]

Re: the gutless, bloated, and fried power supply hall of shame

For +12 V/-12 V its 120 mV. Man, at 50 mV, 90 % of market would be out of spec

[Wester547]

Re: the gutless, bloated, and fried power supply hall of shame

Yes, it uses a different topology though. The coincidence with capacitance =~ output power only appears to work for half bridge and flyback designs. With a different transformer design, you can get away with smaller capacitors.

I also wonder if it was because of the 20AWG wires by comparison to the 18AWG wires used - if I'm not mistaken there's about a 60% difference between what they can handle in amperage per wire (though it varies per wire even then, IIRC). Could that be the reason, rather than diverging topologies making it easier for primary capacitors to "store" energy?

EDIT: Oh, well, I would think the efficiency is lowered even with a voltage drop. ^^;

[Behemot]

Re: the gutless, bloated, and fried power supply hall of shame

If the wire cannot handle it, it just creates voltage drop, how can it affect power? Only way I can think of is the consuming device taking even more current when voltage goes down (that's what VRMs do though).

[PeteS in CA]

Re: the gutless, bloated, and fried power supply hall of shame

A voltage drop across a wire is wasted power: P = V * I. So too-small wire gauge can affect the over-all efficiency. Theoretically, the safety agencies made sure the wire gauge was adequate for the current. But if the agency marks are fake, or the P/S mfr used a smaller wire gauge than what they submitted to the safety agency ...