sunkindly wrote on 2025-11-26, 18:50:I finally got a chance to open it up: […]
Show full quote
I finally got a chance to open it up:
The attachment IMG_8120.jpg is no longer available
The attachment IMG_8123.JPG is no longer available
The attachment IMG_8124.JPG is no longer available
The attachment IMG_8128.JPG is no longer available
In my ignorance I don't see anything obviously wrong. The exposed wire ends don't seem ideal though?
Sorry it took me a while to get back to you and thanks for posting the pictures.
I'm actually pleased to say this PSU looks pretty well-built. If anything, it appears to be a true AT PSU from back in the days and not a cheapo ATX gutless unit that has just been converted to AT by not using the 5VSB.
But here is where it gets a little strange: I see a brown United Chemicon KY series capacitor on the output side. I suppose if this AT unit was built really late in the 90's (like in '99 or perhaps in 2000), then I can *somewhat* see such a cap used in the PSU (Chemicon KY was available by then)... though certainly not widely used in PSUs. Most PSUs of that era did not use such low(er) ESR caps. UCC LXY, Nichicon PL, Rubycon YXA, and Panasonic HFQ were more the norm if a manufacturer went with Japanese capacitors. The reason I mention this is because the UCC KY cap really sticks as out-of-place for an AT PSU. Moreover, I can't seem to recognize the brands of any of the small caps. That along with the black anodized heatsinks (which wasn't common in PC PSUs either) and one of the ICs with sanded off markings makes me suspect this could be an old AT PSU that was reworked/e-waste upcycled by some cheap Chinese no-name factory... and if really so, then no telling what else they might have modified / goofed up. It's not something to run into commonly, but I have seen it with laptop AC adapters a few times. Usually the cheapo factory will replace the electrolytic caps, just in case there was anything wrong with the old ones... but they usually end up putting either whatever they find (sometimes used caps too) or much cheaper (but new) and crappier caps. On your particular AT PSU, though, only the UCC KY cap appears to be out-of-place. Otherwise, the rest of the big output caps appear to be original - at least judging by the hot glue between them (again, something that was commonly done on old PSUs that is now usually done with silicon glue.) So I have a slight (but not big... like maybe 10-15%) suspicion you might have an e-waste upcycled PSU here. Can you by any chance see/show what the rest of the output electrolytic caps look like? (Try bending the output wires away from them.) I see "CEHTR" written on two of the caps, but can't quite make out what brand that is. And perhaps a picture of the PSU's label too? Sometimes, the design of the label can hint at who made it.
Of course, it could also be just a less-than-common build too... hence the above irregularities with the caps and heatsinks.
In any case, though, the PSU looks well-made. The fuse in the fuse holder, along with the AC connector on the PCB are tell-tale of something that was done on really old PSUs (AT and early ATX era) only from the better manufacturers. The twisted AC wires (to reduce common mode noise and EMI/RFI) is another sign that whoever (originally) built this knew what they were doing and put good design effort into the PSU. And moreover, both the driving transistors and the output rectifiers appear to be quite beefy (the 5V rectifier appears to be TO-3-cased, so the 5V rail is very likely good for more than 20 Amps... which should be plenty for any AT PSU.)
As far as the "exposed" wires you mentioned - this is pretty common in PSUs. If anything, the manufacturer actually did a pretty good job to cover them as best as possible with heatshrink tubing. On cheap PSUs, this is almost never the case. So this is definitely not a cheapo PSU. Speaking of wires, I see the wires all have proper names and labels on them, and appear to be of proper size (thickness)... so again, this is likely not a cheapo PSU.
With that said, I don't really see anything wrong with this PSU. In regards to the issues you are describing (with the monitor "flashing" and CMOS resetting), are you sure that's not caused by something else? I've repaired a lot of PSU over the years - both cheapo units and well-made PSUs, and I have never seen a PSU send a "power surge" through components and not cause them to die. If/when a PSU causes something in the PC to "bug out", it's always due to poor output filtering (be it due to bad capacitors or just not adequate filtering from undersized output capacitors).
The LEDs on the HDDs flashing during power up could be due to the PG rail taking a bit longer to signal that the PSU voltages are ready (or alternatively, on shut down, signaling to the PC that AC power is lost and thus telling PC components to shut down properly ahead of time before the PSU looses power.)
And lastly, regarding the monitor "flashing"... do you by any chance have the monitor's power cable plugged into the outlet of the PSU? (Like was done back in the day.) It appears that on this PSU, the power switch on the PC cuts out power not only to the PSU itself, but also to the monitor power plug on the PSU... so that could be why you see the monitor flash. But not 100% what you mean by "flash" with your description.
All in all, this PSU is based on a pretty standard, tried and true, half-bridge design. So if its output voltages are in spec, and since it does appear to have a good amount of output capacitors on the output, I don't see how it can be causing all of those other issues you mentioned. Not saying you're making stuff up, but rather just trying to see if you can confirm / repeat the results you saw from using this PSU.
Anyways, that's all from me on this one. Hope someone finds this info useful.