Power supply build quality pictorial. part 2

[Behemot]

Re: Power supply build quality pictorial. part 2

…

Had a look at 10 ms and lower, there was some ripple at this frequency, but not that high, around 40 mV amongst both +12 and +5 V. The scope also measured the same ripple no matter what time I have set so it has to measure the highest peaks on the whole interval, you just see smaller part when you "zoom in". I use integrated peak-to-peak meter, you don't think I read it manualy, do you?

[Pentium4]

Re: Power supply build quality pictorial. part 2

Got a Wintech here. Not too bad and looks pretty honestly rated for the most part. Checked and 3.3V = 20A, 5V = 30A, 12V = 16A. I posted some before and after pics. It had all Canicon/Seacon caps. Surprisingly they all tested good capacitance and low ESR too. And after a long life in a smokers home...I replaced all the caps except the primaries (and -12V/-5V filtering caps), added some input filtering, and relubed the fan. I know it isn't a full input filter but the 2 X caps, 2 Y caps and coil should help quite a bit. After recapped and loaded up, the 5V measured 5.01V and the 12V measured 12.02V, gotta love it Didn't measure the other rails. With those heatsinks I would trust this thing up to 300 continuous. What do you guys think?

[Behemot]

Re: Power supply build quality pictorial. part 2

I can see no varistor, there is also something what looks like free spot for choke?

[ben7]

Re: Power supply build quality pictorial. part 2

The real question is - how does it hold up to short term loads above its max rating (a few seconds), and also if the protections are working (if it has any!)

[c_hegge]

Re: Power supply build quality pictorial. part 2

What brand was the fan in that?

[Pentium4]

Re: Power supply build quality pictorial. part 2

The fan is "Zonsu" not sure if this thing has any protections. If it does, probably not much haha

[Shocker]

I tried to post here, but I couldn't get the attachments to work properly.

[Behemot]

Re: Power supply build quality pictorial. part 2

Did you select them in the form and than upload before posting? For me it opens in smaller window so the Upload button is not visible until I scroll right.

[Pentium4]

Re: Power supply build quality pictorial. part 2

Yay a Leadman...Not sure if I want to put any time or effort into getting this thing usable. The heatsinks are just plain pathetic and that transformer is so small, and no silkscreening to add input filtering. 3.3V = 20A, 5V = 30A, 12V = 16A. At least the coils are good size and this LP actually has PI filter coils

[Behemot]

Re: Power supply build quality pictorial. part 2

Re-mark it as 350 W and you are OK

[lti]

Re: Power supply build quality pictorial. part 2

For some reason, the ripple on those power supplies is extremely high, even with pi filters.

[Behemot]

Re: Power supply build quality pictorial. part 2

Low capacity input cap, low capacity output caps? Not mentioning not that good/efficient primary driving, too.

[c_hegge]

Re: Power supply build quality pictorial. part 2

The primary sides on those leadmen are often pretty anemic. I'd give it maybe 200W max before the switcher goes kaboom.

[tom66]

Re: Power supply build quality pictorial. part 2

Re-mark it as 350 W and you are OK

I doubt it will even do 350W, the primary caps are too small, maybe a low 300W but ripple would be significant.

[Wester547]

Re: Power supply build quality pictorial. part 2

As for the primary capacitors, if I'm not mistaken that looks like forward topology and I have seen 560uF input capacitors do 350W before in that topology (A 350W Hipro, doesn't say so in the review but elsewhere I have found it uses 560uF input capacitors) and it managed to do 420W with no problem within 4 hours of testing, albeit at 230V/50Hz and only at a room temperature in between 26-27C. But you can see that the URLed PSU is designed much better than the Leadman you see here, though the one in the review is only rated up to 40C room temperature/90% relative humidity. The only thing decent in that Leadman, as far as I'm concerned, is its secondary toroidal coil.

[Pentium4]

Re: Power supply build quality pictorial. part 2

It powered a Sandy Bridge i3, 8GB RAM, and a GT 430 for about 2 years and it was one of the dustiest computers I've ever seen which doesn't help. The motherboard and RAM died, and I told my co worker "I BET YOU ANYTHING the ripple from the PSU killed them" You can't see in the pic but there is also a MOSFET on the primary heatsink. Should I try to help this thing or chuck it?

[c_hegge]

Re: Power supply build quality pictorial. part 2

^
Can it. The whole design of the primary sides on these units just makes them produce excessive ripple, even with good caps and up-to spec parts.

[Pentium4]

Re: Power supply build quality pictorial. part 2

^
Can it. The whole design of the primary sides on these units just makes them produce excessive ripple, even with good caps and up-to spec parts.

What if I were to take the input filtering attached to an AC receptacle off a different power supply and wire it to the PCB, replace the only coil it has with a bigger one, and put some higher quality input caps in there, think that would make the input adequate? I probably wouldn't do it just because I don't trust this thing but still curious if that would help it enough

[tom66]

Re: Power supply build quality pictorial. part 2

Ripple kills caps on motherboards, but rarely anything beyond that.

[c_hegge]

Re: Power supply build quality pictorial. part 2

What if I were to take the input filtering attached to an AC receptacle off a different power supply and wire it to the PCB, replace the only coil it has with a bigger one, and put some higher quality input caps in there, think that would make the input adequate? I probably wouldn't do it just because I don't trust this thing but still curious if that would help it enough

No, that wouldn't help. It's more to do with the topology. It's a flyback converter, not a forward converter, which produce far more ripple.

The input filtering will improve the interference that the PSU generates, and the primary caps will improve the holdup time.