One of the PC enthusiast websites took it upon themselves to do a real load test across a range of PC power supplies. It turns out that ALL of the PC supplies dropped their power good signal before reaching their rated output. Most of them recovered when the load was lowered, but one actually failed completely and was unusable thereafter. Good ATX supplies follow Intel's spec, which includes several interesting requirements. Failing power supplies must not emit flame or smoke (bad smells ok, but not visible smoke). Failing power supplies must not damage the load (ie, crowbar type protection required). I've never had a PC power supply fail and destroy the PC. (I've had several where an external event destroyed both such as lightning, but never the supply's failure itself). So, all that being said, a simple power supply tester with minimal load will be useful, but power supplies are cheap and unless you have a good reason not to you should always replace the supply if you think it's a problem. Even good PC power supplies degrade quickly over time, and the quickest way to figure out whether the PS is the problem is to connect a new one. If you really want to test it first, and prefer not to buy a tester, then place 2W loads on 3.3V, 5V, and 12V lines (most power supplies are regulated at either the 3.3v or 5v output and all other outputs are relative to the main regulation, so you might only need one load). Then short the power on pin to ground and check the output and power good signal. If power good does not become active then the power supply is bad regardless of the voltage output. If power good is active, and the votlages look good under the small load then you can be reasonably certain the power supply is good. The power good pin is actually quite picky - I don't usually test for it, but when I have I've never seen a bad power supply output active power good. -Adam On 1/27/08, William Chops Westfield wrote: > > On Jan 27, 2008, at 2:02 PM, peter green wrote: > > > To thorougly test a power supply you need to test it with both the > > stated minimum and stated maximum load on each rail. > > heh. While this is theoretically true, I think testing a typical > consumer computer power supply at its stated maximum load is just > asking for trouble... (and even if it's accurately rated, aren't > there weird co-dependencies between 5V and 3V, or similar?) > > BillW > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Moving in southeast Michigan? Buy my house: http://ubasics.com/house/ Interested in electronics? Check out the projects at http://ubasics.com Building your own house? Check out http://ubasics.com/home/ -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist