Can't tell a thing by looking at the outside.

Timecop1983 wrote:

Maybe I shouldn't bother with any of them and buy an ATX->AT adapter?

This is actually not a bad idea btw. These adapters don't ever get rotting caps and give you a lot of flexibility, I'd recommend you get a couple of those. The amount depends a bit on how extensive your collection is of course, but I'd say get a couple so you can save some money on shipping costs)

PCBONEZ wrote:

Can't tell a thing by looking at the outside.

Don't you know whether the PSUs of these producers were good quality? I do not have the possibility to look inside...

CWT is at least a known PSU manufacturer, the other PSU is held together with chassi screws... 😜

I kind of doubt CTW used the highest qualiy caps back in 1998 so I would look inside the PSU before using it.

Timecop1983 wrote:

Don't you know whether the PSUs of these producers were good quality?

No. I don't know.

There were way to many companies selling AT PSUs back in the day and a lot of them have gone *poof*.
Also many were rebranded something else because companies wanted to give the impression they were building there own.

The demands on PSUs weren't so great back then so even crap brands could last a few years, but now they are 15, 20 or more years old.
One may have been used a year and the other 20 years.
One may have crap caps and the other high quality.
You just have to look inside to see how they held up.

If you are lucky searching for a blog or forum article that looked at one you are interested in might find something.

In more recent history CWT is hit and miss. Probably was back then too.

The Emiter Co one is a brand I dunno but please note: there is no model number on it and the two phone numbers don't have the same number of digits.
Pretty sure that's a generic rebranded one (use your own sticker) and they apparently didn't even proof read their own labels.
.

Does anyone know when short circuit protection came in? With regard to using ATX supplies with an AT converter, I am much more comfortable with this because whenever I have accidentally caused a short circuit with an ATX power supply, it shuts off immediately. But the times I have had a capacitor short on a motherboard while using an AT power supply, it does not shut off, it just delivers max current, usually with the voltage sagging so much the PSU fan doesn't start. This keeps going until something goes bang - in my case the tantalum capacitor on the motherboard.

So in my mind, no AT power supplies have short circuit protection, and all ATX ones do. How (in)accurate is my assumption? I'm sure I remember seeing something about this being required in the ATX spec but I figure at least a few manufacturers skipped that part...and I don't know whether any high quality AT PSUs had it already.

I'd like to think that having short circuit protection might help save some aging parts as components start to fail and go short-circuit. Although then again it's a lot easier to find the part that needs replacing when it looks like a small piece of charcoal soldered onto the PCB 😉

It seems that the best idea would be to buy ATX->AT adapter...

Timecop1983 wrote:

It seems that the best idea would be to buy ATX->AT adapter...

I can highly recommend it. It's cheap and I agree with PCBONEZ fully as from my own experience, the quality of PSUs in those AT cases was almost always very poor! I found one Seasonic AT PSU and the rest were kinda "made" by companies kinda like "Happy Hippo" (this name is fictional btw) or some other vague name I never seen or heard about again.

Only good thing about those AT PSUs was that they were made before the cap plague but most AT PSUs are really very empty inside. The better ones tended to be non-clone proprietary ones. This all changed when ATX started taking off

Malvineous wrote:

Does anyone know when short circuit protection came in?

Day one. - It's called a fuse.

Over-current protection is something else.
I dunno if/when AT PSUs had it.
.

Using an ATX PSU is not guaranteed to give over-current (or over-voltage) protection to you.
All you have to do is look at PSU reviews to see those protections aren't always there and/or often don't work despite being required.

Old ATX PSUs are going to have the same issues as old AT PSUs.

Using ATX PSUs that are too new may cause problems with load balancing. (Nothing on 3.3v or not enough on 12v to regulate properly.)

If you are going to be in retro electronics of any kind you should at least learn to solder and to recognize and fix the common problems.
.

Tetrium wrote:

by companies kinda like "Happy Hippo"

I have a Happy Hippo alarm clock. - Only one I could find annoying enough to wake me up.
When the alarm goes off she starts yelling calisthenics like an aerobics instructor or something.
ONE ... TWO ... ONE TWO THREE ...
.

PCBONEZ wrote:

Day one. - It's called a fuse. […]
Show full quote

Malvineous wrote:

Does anyone know when short circuit protection came in?

Day one. - It's called a fuse.

Over-current protection is something else.
I dunno if/when AT PSUs had it.
.

Well according to the design guidelines (page 22) it's called short circuit protection. Over-current protection is more for exceeding the limitations of the power supply, without necessarily causing a short circuit. And in my personal experience with AT supplies, a short circuit in a motherboard component won't blow the fuse. It might shut down eventually once the PSU gets hot enough, but whatever is causing the short seems to heat up and burn/blow before the PSU gets hot enough for anything like that.

So it should definitely be there in ATX supplies (depending on how closely they follow the specs) but AT supplies is a bit of an unknown. I suspect it was easy to implement in ATX since they already have a soft-off control circuit, but since that's not present in an AT supply it would mean more expense adding some kind of shutdown circuit, so it wouldn't surprise me if most AT supplies don't have any sort of short circuit protection (perhaps instead relying on presumably slower overcurrent protection.)

I'm going to be a bit pedantic here, then I'll stop.

Fuses are specifically designed to trip if the energy into the system (I^2 t) exceeds the damage threshold the fuse is designed for. It doesn't provide short circuit protection in the same way a mag trip breaker will (those open at a specific current through the trip coil). Also, neither of those protections provides a limit of the short circuit current. Current limiting fuses and breakers are another thing.

Malvineous wrote:

PCBONEZ wrote:

Day one. - It's called a fuse. […]
Show full quote

Malvineous wrote:

Does anyone know when short circuit protection came in?

Day one. - It's called a fuse.

Over-current protection is something else.
I dunno if/when AT PSUs had it.
.

Well according to the design guidelines (page 22) it's called short circuit protection. Over-current protection is more for exceeding the limitations of the power supply, without necessarily causing a short circuit. And in my personal experience with AT supplies, a short circuit in a motherboard component won't blow the fuse. It might shut down eventually once the PSU gets hot enough, but whatever is causing the short seems to heat up and burn/blow before the PSU gets hot enough for anything like that.

So it should definitely be there in ATX supplies (depending on how closely they follow the specs) but AT supplies is a bit of an unknown. I suspect it was easy to implement in ATX since they already have a soft-off control circuit, but since that's not present in an AT supply it would mean more expense adding some kind of shutdown circuit, so it wouldn't surprise me if most AT supplies don't have any sort of short circuit protection (perhaps instead relying on presumably slower overcurrent protection.)

Regardless fuses are short circuit protection and what you asked is when it was first used in PC PSUs.
That's what I told you.

You did not specify "Output" short circuit protection which is apparently what you meant.
If you actually read through that section in your link it is intended to protect the PSU from damage, not the motherboard.
Same purpose as the fuse, and not what you (and others including some review writers) are thinking it's for.
It also states the maximum short circuit energy as 240 VA - on any rail. If it gets that high the motherboard is likely already shot.
(To restate. If there is a short on the mobo the mobo is already toast. Their objective is to keep that fault from damaging the PSU too.)
.
One of the earlier Design Guides (ver 0.9) actually says modifying a PS/2 PSU and calling it an ATX is just fine, so evidently PS/2 PSUs have this Output Short Circuit Protection you are asking about.
I haven't looked at these side-by-side in eons but as I recall PS/2, XT and AT are indistinguishable electrically so it would follow that versions that existed at the introduction of ATX all would have it.
.
Because of their age, to get a solid 'for sure' answer about AT you will probably have to resort to interpreting schematics.
I'm a little too busy now to get that involved with it. I come here when I'm 'on break' lately.
.
I can offer some tips/hints/clarifications.
.
That it's included in some ATX PSU Design Guide does NOT mean it's excluded in AT designs nor does it mean it's included in marketed ATX PSUs.
Unlike a 'Standard' or 'Specification', compliance with 'Design Guides' is optional. - It says so right in your link in the very first section under "Scope".
That's why they are called "Guides".

So just because you have an ATX PSU there is no guarantee it will behave any differently than you experienced with that AT PSU.
Manufacturers are not obligated to follow guides so there may be no "Output Short Circuit Protection" at all or it may not work.
That has been demonstrated time and again in PSU reviews.
.
Also note the v1.3 PSU Design Guide you chose states it is to support the Ver 2.03 ATX Motherboard Specification.
To my knowledge that Specification was never published or made official.
There is no hint of it online at all so if it existed Intel never shared it with anyone outside of Intel.
Apparently they skipped publishing Motherboard Specifications from 2.01 to 2.10 and it gets confused as a PSU Spec which it isn't.
The Specification is what manufacturers are obligated to follow to call their part ATX compliant.
I will use the Ver 2.1 ATX Mobo Specification as the ver 1.3 DG you called up says "ATX Motherboard Specification 2.03 (or later)".
That is here. http://www.formfactors.org/developer%5Cspecs%5Catx2_1.pdf
There is no mention of Short Circuit Protection of any kind being -required- by the ATX Specification.
.
I know that's not how it should be but that's how it is.
.
~~~~~
Now rambling. (Some more?)

The Design Guides for ATX PSUs are pushed by Intel.
The ATX Specifications are supposedly handled by FormFactors.org but I suspect it's Intel in the drivers seat.

XT, AT and PS/2 were all IBM Standards/Specifications that date back as far as 1983.
Finding detailed info on those is tough because they are so old.
You would probably have to look at schematics to determine if they have the protection you care about and even then you can't be 100% certain because of all the IBM clones going on back then.

If you look here there are actually at least 40 different Form Factors and over a dozen "Authorities" on their Specifications.
https://en.wikipedia.org/wiki/Computer_form_factor
A lot of them get lumped together and are thought of as part of the same Spec but that isn't actually true.
.