VOGONS


First post, by Ozzuneoj

User metadata
Rank l33t
Rank
l33t

I just bought this PSU on ebay since it was a great name brand, and the seller said that was fully tested with a PSU tester... and it was reasonably cheap ($15 shipped).

As it turns out, when I hook it up to my tester (which is a modern digital ATX tester) it has several faults on it. I opened the unit up to see if there were any obvious problems and there is one 47ohm resistor that looks like it may have gotten a bit hot at some point (the yellow glue stuck to it has turned brown) but it still tests at 46.5 ohm, so it seems to be fine. But if I do a simple continuity test on the ATX connector, the 3.3v, 5v and ground wires are all completely shorted! On my other PSUs I can test these connections and they briefly show as a short, then close off... I assume this has something to do with a capacitor being charged. This unit however shows open shorts across several connections inside with no change no matter how long I leave the tester on them.

To my utter disgust and astonishment, when I looked at the capacitors (which are all in perfect shape externally), there are three large caps that are made by... *drumroll*... Fuhjyyu. I know these are notorious for failing, but I've never seen them fail so completely without any visual indicators. I honestly don't know if I should blame them yet.

This looks like a quality unit, but I don't know how to diagnose the cause of such a problem any further than I already have.

Now for some blitting from the back buffer.

Reply 1 of 3, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie

What faults did the tester indicate?

Without any further knowledge (pictures etc), the 47 ohm resistor could be a dummy load, so it is meant to just sit there and consume power in order to keep the output voltages stable. Sometimes a minimum current draw is needed to keep feedback loop happy and the internal resistor guarantees it.

How did you do the continuity test? Was there a resistance reading near zero? Some multimeters just beep if the resistance is below 100 ohms or something similar, so for a random multimeter whose workings are unknown, I would not trust the continuity test mode but just resistance reading.

My suggestion is to just connect the PSU to mains and see if 5V standby is there (purple wire). Then to start the PSU you connect the green wire (PSU_ON) to ground (black), and the rest of the voltages should appear. And again, sometimes the PSU needs some minimum load (hard drive or two) to get correct output voltages.

Reply 2 of 3, by Ozzuneoj

User metadata
Rank l33t
Rank
l33t

Thanks for the tips!

My tester shows +5v=HH, -12v=LL, and +12v1=LL (+12v2 is also LL, but it is an old unit with no 4pin 12v connector). PG is fine.

I tested the purple 5vsb and it is 4.95v, so that seems fine.

My "method" for testing the continuity was nothing special (or even accurate apparently). I'm using a Fluke Model 75 multimeter. With the unit unplugged I merely tested the continuity at the ATX connector. All of the black wires, the red wires and the orange wires would cause it to beep when any of them were connected. I'm an electrical newbie so I see now that they do have a resistance of between 14 ohms and 40 ohms roughly... I'll try to remember to pay more attention to the readout when testing for continuity problems in the future.

I just tested it manually using the paper clip and testing the voltages with my DMM (all testing done at the ATX connector). +5v reads 5.55, +3v reads 3.33, +12v reads 10.85 and -12v reads -10.48.

It seems the tester is right in that the +5v is quite high and the +12v is too low... I guess I could use an old bad drive to see if a small load brings these numbers up to spec.

Now for some blitting from the back buffer.

Reply 3 of 3, by Ozzuneoj

User metadata
Rank l33t
Rank
l33t

Hah!

Thank you Jepael...

Adding a hard drive got the tester to stop complaining, and adding a second brought the voltages even closer to spec. With two drives connected, when I measure with my DMM I get +12v=11.76, +5v=5.24, 3.3v=3.35, -12v=-11.76.

Much better! A couple of my old drives are apparently more dead than I thought though. The PSU would squeal significantly with them connected and one of them actually made it shut off after a few seconds. My other ones don't do this at all.

I intend for this PSU to be used for some kind of beefy older system since it can supply 30a on the +5v and 12a on the +12v rail.

Now for some blitting from the back buffer.