Hardware Insights Forum

Wow, there are some crappy PSU's in that round up! It's a shame that the 320W wouldn't deliver its wattage in spec. Are you going to keep and/or fix any of them? The Solytech unit didn't look too bad, although it exploded

Pentium wrote:Are you going to keep and/or fix any of them? The Solytech unit didn't look too bad, although it exploded

No, for 2 reasons.

1. The only two which had somewhat beefy guts (the Thermal Master and the Shaw), all but destroyed themselves in the testing, and repairing them would be more trouble than it's worth. I think the Thermal Master might have even had a melted trace on the PCB.

2. I already have lots of spare 300 Watters which use more modern forward designs (and are more efficient), and can deliver their ratings in spec, and I just don't have the space in my workshop for more.

I do agree, though, that the Thermal Master would probably have done OK in a budget PC if it were re-capped with Japanese caps.

Oh okay I understand. Yeah I'm surprised it did 400W. It'd be a great PSU with OPP and a recap like already mentioned. Will you scrap any of the parts from any of them?

Yes, I did pull the fans and rectifiers from a couple of them. Most of them had blown switchers and other parts on the primary side. I don't totally scrap PSUs unless they are completely trashed and have no useful working parts. Even if so, they are sent off to metal recyclers. I avoid having them end up as landfill if at all possible.

Wow... the ripple is quite low on the Sunpro's DC-DC converted +3.3V rail. That's amazing for such a cheap power supply. Most PSUs with a magamp circuit aren't quite that good. Especially considering it's linear regulation, it makes me happy some of my Hipro 250-300W PSUs use that very method (though they aren't linear regulated from the +5V rail, there's a separate transformer tap/output/winding for it). However, the truly staggering thing is how low the +3.3V ripple is on that Sunpro given how crappy its output filtering is. If its ripple is that low on the linear regulated +3.3V rail, imagine how low it would be for PSUs with much better output filtering and linear +3.3V DC-DC conversion...

^
See http://www.jonnyguru.com/modules.php?na ... 2&reid=123

Wolf found a similar thing with this linkworld. Even though its source rail (the 5V) had 1.3V (1300mV) of ripple, the 3.3V rail was quite clean. Those type of DC-DC conversion circuits usualy do have extremely low ripple outputs.

I just realized that that Coodmax is the same exact thing as my Enyle PSU except mine has a fan controller and a voltage selector switch. http://www.badcaps.net/forum/showpost.p ... tcount=578

I was actually thinking about improving it a little but the transformer might already be at its limit at 260W. At least it doesn't have a 2 transistor 5VSB circuit and room for a full input filter (almost) and it has room for inductors on the output. Also, do you know how good Ever Cool fans are?

Yours looks almost identical. I haven't had much experience with those Ever Cool fans before. I'd say that if you make sure the bearings have plenty of oil, the fan will probably be OK. If you add the PI filtering coils, recap it, and beef up the secondary rectifiers to improve the efficiency, then it might do OK as a 250 Watter.

Cool, I will check the bearing to see if it has oil. I am also going to at least attempt the slap on a better heatsink for the primary...I still can't figure out who the OEM is

And as far as I'm aware Pi coils don't have polarity?...