appiah4 wrote:

I'd be glad if you can look at these photos and tell me if this PSU is worth salvaging and using.

I'm gonna have to go with "lolno."

CkRtech wrote:

appiah4 wrote:

I'd be glad if you can look at these photos and tell me if this PSU is worth salvaging and using.

I'm gonna have to go with "lolno."

Well 😀 lolno as in its crap and would eventually burn the house or as in its malfunctional and damaged?

It is really low end, both power and construction. I would salvage the cables and heatsink, fuse and toss the rest. Those cables can be soldered into some ATX PSU to convert it to an AT thing.

Tiido wrote:

It is really low end, both power and construction. I would salvage the cables and heatsink, fuse and toss the rest. Those cables can be soldered into some ATX PSU to convert it to an AT thing.

How is this sorcery possible? 😀

appiah4 wrote:

Tiido wrote:

It is really low end, both power and construction. I would salvage the cables and heatsink, fuse and toss the rest. Those cables can be soldered into some ATX PSU to convert it to an AT thing.

How is this sorcery possible? 😀

IMO it is not worth it to do a manual conversion when you can buy an adapter cable with -5v added from eBay for $13.99/$14.00 shipped. I've bought multiple of these adapters and they work great.
(20-pin ATX to AT)
https://www.ebay.com/itm/ATX-20-PIN-to-AT-P8- … DE/251025150787

(24-pin ATX to AT)
https://www.ebay.com/itm/NEW-EPS-ATX-24-PIN-t … SA/252708704697

cyclone3d wrote:

IMO it is not worth it to do a manual conversion when you can buy an adapter cable with -5v added from eBay for $13.99/$14.00 sh […]
Show full quote

appiah4 wrote:

Tiido wrote:

It is really low end, both power and construction. I would salvage the cables and heatsink, fuse and toss the rest. Those cables can be soldered into some ATX PSU to convert it to an AT thing.

How is this sorcery possible? 😀

IMO it is not worth it to do a manual conversion when you can buy an adapter cable with -5v added from eBay for $13.99/$14.00 shipped. I've bought multiple of these adapters and they work great.
(20-pin ATX to AT)
https://www.ebay.com/itm/ATX-20-PIN-to-AT-P8- … DE/251025150787

(24-pin ATX to AT)
https://www.ebay.com/itm/NEW-EPS-ATX-24-PIN-t … SA/252708704697

But this way you don't have the emotion of DIY.

appiah4 wrote:

How is this sorcery possible? 😀

Remove cables from one PSU, solder them to another. You'll want to wire PowerON signal to the mains switch too. Some PSUs have a 3.3V sense wire that must be connected to 3.3V rail or other rails will get way too high voltages. Sometimes dummy load is needed on the unused 3.3V too.

Ebay cables are an option if one lacks skills, time, money or the will to do the conversion. I have the skills, tools, will and time, but not the money, plus nicely custom length cables are visually pleasing too ~

appiah4 wrote:

I bought a rather yellowed AT case lately, and while the front panel and interior has been hosed down in preparation for some re […]
Show full quote

I bought a rather yellowed AT case lately, and while the front panel and interior has been hosed down in preparation for some retrobrighting, I thought I would also check the PSU out.

It was the single most filthy PSU I've ever seen.

So as much as I hate washing down any electronics, I had to hose it down as well.

The funny thing is, the psu's power cable to the power switch had been cut off. Why, I do not know. I am curious whether this was because they didn't want someone to accidentally turn the system on? Could it be something to do with a bad PSU? I can recrimp the PSU's connectors to the cable, but I am afraid it could blow something up?

I'd be glad if you can look at these photos and tell me if this PSU is worth salvaging and using. I have a spare brand new AT PSU but I would rather keep it as just that, a spare.

That's some pretty severe cost down-itis. Coils on the pi filters? Y caps that are tested not to fail short? It's not like anyone needs FM radio, smooth DC output, or a house that doesn't burn down.

Does anyone have one of those PSU-testing machines? They are very expensive.

I'm pretty sure a bunch of resistors between gnd and +5v could be a good test. Just count how many amps the psu is supposed to provide and how many amps the resistor will allow and if the psu blow up with the amps it's supposed to provide, then it's total shit

Actually a part of the supply is missing !!!
But you can complete it installing the missing parts and
replacing the bad ones with quality parts!
Consider it as a KIT for practicing in electronics circuit!

verysaving wrote:

Actually a part of the supply is missing !!! But you can complete it installing the missing parts and replacing the bad ones wi […]
Show full quote

Actually a part of the supply is missing !!!
But you can complete it installing the missing parts and
replacing the bad ones with quality parts!
Consider it as a KIT for practicing in electronics circuit!

Eh? What is missing if I may ask?

The only reason I would take the time to restore a PSU is if it were non-generic / proprietary standard where replacements are either rare or just impossible to find.

If it works and does not behave oddly, then it may serve as a test PSU.

appiah4 wrote:

Eh? What is missing if I may ask?

Pretty much the only thing which looks okay on this supply is the ground point.

Let me itemize the failure:

• One PTC for transient suppression isn't enough.
• Two plate capacitors for line filtration. This is insufficient and unsafe; what's installed doesn't form a bandpass filter and the caps should be safety rated X and Y type for this application.
• Primary rectification? I can't see a FWB module or even a quad diode pack. Are they using half wave rectification here?
• If so, that makes the switcher transistor complement a joke.
• Good thing, though. The transformers on this supply are definitely not sized for a WIN. Nor are the primary caps. Nor are the heatsinks.
• The secondary rectification! Why did they even bother with a heatsink here? It's not like there's a lot of thermal coupling.
• Pi filters? We can't afford those. So what if the supply puts through massive amounts of ripple. We're making money.

Question is, did this supply come with a variety of meaningless power rating stickers so vendors can tune their lies for the customer?

gdjacobs wrote:

Pretty much the only thing which looks okay on this supply is the ground point. […]
Show full quote

appiah4 wrote:

Eh? What is missing if I may ask?

Pretty much the only thing which looks okay on this supply is the ground point.

Let me itemize the failure:

• One PTC for transient suppression isn't enough.
• Two plate capacitors for line filtration. This is insufficient and unsafe; what's installed doesn't form a bandpass filter and the caps should be safety rated X and Y type for this application.
• Primary rectification? I can't see a FWB module or even a quad diode pack. Are they using half wave rectification here?
• If so, that makes the switcher transistor complement a joke.
• Good thing, though. The transformers on this supply are definitely not sized for a WIN. Nor are the primary caps. Nor are the heatsinks.
• The secondary rectification! Why did they even bother with a heatsink here? It's not like there's a lot of thermal coupling.
• Pi filters? We can't afford those. So what if the supply puts through massive amounts of ripple. We're making money.

Question is, did this supply come with a variety of meaningless power rating stickers so vendors can tune their lies for the customer?

Yeah, it was actually rated for 200W. 😵

OK, I've binned this PSU, I'll use my spare good AT PSU in this case.