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

[Pentium4]

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

Yeah nice chunks of metal on there! I might throw it in my "recap later" pile, it's pretty well built for its time period. I find it interesting that all the Jamicon's are fine (even the 12V filtering one) and ALL of the Teapo caps are bulging

[ben7]

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

The metal thickness barely contributes to continuous running power. The thick heatsink is nice for quick bursts of high load, but not for long term. Also depends on how efficient the thing runs.

[Pentium4]

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

True...And it has a thermal fan controller so it would probably run warm, guessing low 70's for efficiency

[Wester547]

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

The metal thickness barely contributes to continuous running power. The thick heatsink is nice for quick bursts of high load, but not for long term. Also depends on how efficient the thing runs.

So a heatsink this thick and huge (courtesy of larrymoencurly) doesn't contribute much to running power at all (though it does result in less room for airflow)?

[ben7]

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

LOL

[momaka]

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

So a heatsink this thick and huge (courtesy of larrymoencurly) doesn't contribute much to running power at all (though it does result in less room for airflow)?

A thick heat sink by itself without any fins won't do much, yes. But a thick heat sink with lots of fins like the one you showed above will. The number of fins and their size/area dictates how much heat the heat sink will dissipate. Of course, if the heat sink is not thick enough in the middle, then not all of the heat from the silicon will be able to get to the fins. So the heat sink also needs to be thick in the middle in order for all of the heat from the silicon to quickly reach the fins.

So a thin heat sink with many fins = not optimal
A thick heat sink with very few fins = no optimal as well
A thick heat sink with lots of fins = most optimal design
And a thin heat sink with very little fins = cooling disaster (cheap Sun Pro PSUs come to mind here )

[Wester547]

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

A thick heat sink by itself without any fins won't do much, yes. But a thick heat sink with lots of fins like the one you showed above will. The number of fins and their size/area dictates how much heat the heat sink will dissipate. Of course, if the heat sink is not thick enough in the middle, then not all of the heat from the silicon will be able to get to the fins. So the heat sink also needs to be thick in the middle in order for all of the heat from the silicon to quickly reach the fins.

So a thin heat sink with many fins = not optimal
A thick heat sink with very few fins = no optimal as well
A thick heat sink with lots of fins = most optimal design
And a thin heat sink with very little fins = cooling disaster (cheap Sun Pro PSUs come to mind here )

Oh, I know that finless heatsinks get too hot to touch (80+ efficient PSUs IMO are the only excuse to use them, as it were). Base thickness definitely matters too... but what I did notice is that the primary heatsink in the image of larry's Hipro PSU is less impressive. It still looks thick and like it has decent surface area but it only has 3 (large and thick) fins and doesn't seem very close to the fan at all (to top it off a lot of those Hipros have a large Passive PFC mounted on the primary side which means less airflow to that section...)... I think every Hipro of that design (or remotely similar) used the same primary heatsink (even the 200W models), though in varying sizes. Maybe that's why it's rated at 250W but it could probably do 300-350W fine given how brutally overbuilt it is. Most Hipros of that time have a very nice main switcher (being a single forward design), even the 280W models, known as STW9NB90. Its on resistance isn't as low as the parts you find in 80+ PSUs but it is very overspec'd for 250W...

[shovenose]

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

Solid Gear "480W" (more like 200W) power supply, came out of an iBUYPOWER PC (now running a 650W Antec happily)...
http://www.psudatabase.com/solid-gea...-power-supply/

[c_hegge]

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

^
Leadman. That would be a 0W PSU, as I doubt it would do anything at all in spec

[ben7]

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

That look as like a perfect candidate for modding, it has places for more capacitors, and some pi filter coils. The transformer is of a decent size too. The primary capacitors have the stripe from rubycon, but the text is wrong - not sure if they are fakes or just a different brand, I can't see their brand on them.

[PeteS in CA]

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

Unless you could find a more substantial pair of heatsinks with exactly the right mounting hole placement and footprint - not necessarily impossible - and change to higher current I/P rectifiers, recapping and adding ferrite rod inductors (to reduce spike noise, IIRC) would probably just result in a fairly good 230W-250W P/S.

Judging from the material used for the O/P inductors, my guess is that it's a TL494 type PWM, ~30KHz, transistor switched half bridge design. So to do much better you'd have to change to I/P 'lytics (currently 470uF), change the switch transistors (currently probably MJE13007 type) to MJE13009 type (or similar current capability), and possibly the O/P rectifiers. If you have a really great boneyard of non-repairable P/Ss for the parts and lots of time on your hand, it might be an interesting/fun exercise.

[PeteS in CA]

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

The metal thickness barely contributes to continuous running power. The thick heatsink is nice for quick bursts of high load, but not for long term. Also depends on how efficient the thing runs.

The thickness of the metal affects the thermal resistance from the heatsinked device to the rest of the heatsink. Thin metal = high thermal resistance; heat can't get out of rectifiers and switch devices, causing them to run hotter. There is a point of diminishing returns beyond which increased thickness just adds weight and cost, however.

Fins increase surface area, reducing thermal resistance from the heatsink to air. So more fins means better heat dissipation until the fins start significantly impeding airflow and actually hurt the heat dissipation. You can, however, also have fins that are skinny enough and long enough that the increased surface area toward the ends of the fins are meaningless for heat dissipation and only impede airflow.

So, "Heavy metal is goodness," is an oversimplification, looked at hypothetically. But with real-world P/Ss you are much more likely, IMO, to find wimpy and inadequate heatsinks than to find overweight heatsinks or heatsinks so over-finned as to screw up airflow. Over-sized heatsinks equals extra weight; extra weight equals higher price for the heatsink and higher shipping cost for the finished product. Wimpy is cheaper for getting product built and shipped. So I think my oversimplification is pretty much true for the real world.

[shovenose]

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

Unless you could find a more substantial pair of heatsinks with exactly the right mounting hole placement and footprint - not necessarily impossible - and change to higher current I/P rectifiers, recapping and adding ferrite rod inductors (to reduce spike noise, IIRC) would probably just result in a fairly good 230W-250W P/S.

Judging from the material used for the O/P inductors, my guess is that it's a TL494 type PWM, ~30KHz, transistor switched half bridge design. So to do much better you'd have to change to I/P 'lytics (currently 470uF), change the switch transistors (currently probably MJE13007 type) to MJE13009 type (or similar current capability), and possibly the O/P rectifiers. If you have a really great boneyard of non-repairable P/Ss for the parts and lots of time on your hand, it might be an interesting/fun exercise.

I have an entire bin of case-less old PSUs - but not gonna wast e any time on this thing

[shovenose]

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

Next gutless wonder is this iMicro "450W"
http://www.psudatabase.com/imicro-hp...-power-supply/

[jamesbo]

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

The UL E Code says its made by Sirfa. But it looks like a Leadman.

[shovenose]

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

Didn't notice that... but it does look a bit different from the Solid Gear which was a Leadman.

[PeteS in CA]

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

"I Micro" ... how appropriate. 250W I might believe. At 450W it might glow ... briefly.

[shovenose]

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

"I Micro" ... how appropriate. 250W I might believe. At 450W it might glow ... briefly.

It was actually able to power an overclocked LGA775 with a 9800GX2 for a few minutes then I got distracted

[PeteS in CA]

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

You'll have to translate that into Analog, sn.

[shovenose]

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

You'll have to translate that into Analog, sn.

What do you mean??