tag:blogger.com,1999:blog-5554915078212081470.post6873817798927073190..comments2023-07-31T04:22:23.114-05:00Comments on HiR Information Report: Testing an ATX Power SupplyAx0nhttp://www.blogger.com/profile/12145109647562469601noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-5554915078212081470.post-25642214471940875892009-09-06T11:39:15.947-05:002009-09-06T11:39:15.947-05:00dear sir
i have 24 pin power supply on green cabl...dear sir <br />i have 24 pin power supply on green cable have 5 volt. but this compputer supply do not on computer . i checked out put capacitors, diod's & resistors ,<br />wht should i do plz tell me how to repair this supply this is my e mail id plz reply me on this <br />buttji82@gmail.comAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-5554915078212081470.post-79079162797675022612008-06-13T19:27:00.000-05:002008-06-13T19:27:00.000-05:00Good article; a picture is worth... I'm an experi...Good article; a picture is worth... I'm an experienced electronics technician and what the others have said about the dangers of hi-voltage (120 vac) is VERY true.<BR/><BR/>Using a paperclip to jumper the green to black pins is fine if it is wrapped in electrical tape & nowhere near any circuitry or metal parts. This is not because of any normally present DC voltage present on either of these pins (all black pins are grounded - ZERO volts DC) but rather, if the PSU circuit is defective, a significant stray AC voltage may be present. On electronic equipment, the NESC & IEEE Std 80 industry standards consider 1.0 VAC to be the maximum allowable. <BR/><BR/>To test for dangerous (and damaging) stray AC voltage, simply use a digital volt/ohm meter (VOM); few PC techs have an oscilloscope available. <BR/><BR/>To start, the AC cord should be UNplugged from the wall outlet BEFORE attaching it to the PSU; don't plug the cord into the outlet just yet. Before attaching any jumper, as stated earlier, attach an OLD (as in: you don't care if you let all the smoke out of it) hard drive for proper PSU circuit loading.<BR/><BR/>Insert your jumper between the green pin and any black pin in the PSU-to-motherboard connector. Secure the black VOM lead to the bare metal of the PSU case, then, without touching the PSU case, plug the AC cord into the outlet. Proceeding in this order will prevent you from getting a jolt if the PSU circuit is damaged and a significant AC voltage is on the PSU case.<BR/><BR/>Handling only the insulated wires and connector, touch the red VOM probe to any black pin. A reading in excess of 0.5 VAC between the PSU case and a black pin is an indication of a defective (and unsafe) PSU; DO NOT use it even if the DC voltages appear to be within specs.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5554915078212081470.post-78976273303988940842008-04-20T12:15:00.000-05:002008-04-20T12:15:00.000-05:00Thanks ax0n, I just stumbled on this - nice! The p...Thanks ax0n, I just stumbled on this - nice! The pictures make the difference. <BR/><BR/>And thanks Paul - coming from the same background, I could not agree more. And now I must add my two cents. <BR/><BR/>Generally, is not a good idea to run any PSU without a proper load. If you have reason to suspect a problem, you cannot assume it will behave as expected. Even quality stuff can go bad but sadly, many PSUs are made in child slave labor factories and go through zero quality control. Anything that plugs in the wall can kill - do not trust that your deadly voltage circuits were wired, or are working, correctly. <BR/><BR/>Also, as Paul pointed out, you do not get reliable readings (though no reading is cause for concern when testing) without a load. <BR/><BR/>It is not that it won't fire up with out the load - though you still must fool the motherboard into thinking it is connected to the motherboard, it is that your readings will be inaccurate. You may get a good solid 5V with no load, but then it drops to 2.7V under load - not good. <BR/><BR/>Even the plug-in testers as seen at Newegg, which contain a dummy load, do not present a true load to the PSU (though they are great in a pinch, and I keep one in my bag). <BR/><BR/>The problem is, when using a multimeter as ax0n shows (with or without a proper load), or with a plug-in PSU tester, you still may have a bad PSU as there may be too much ripple (alternating current) riding the DC rails.<BR/><BR/>The only real way to properly test a PSU of any type is with an oscilloscope, and with the PSU under a realistic load.<BR/><BR/>Inserting a probe (multimeter or scope) into the heart of the computer is dangerous, a tiny slip may slice through several critical circuits printed on the board - and IMO, should be left to a trained technician.<BR/><BR/>So what should you do? Swap in a known good PSU. <BR/><BR/>JMHO<BR/><BR/>-billAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-5554915078212081470.post-6667823370899607522008-03-29T13:22:00.000-05:002008-03-29T13:22:00.000-05:00That's cool, I haven't run into a power supply tha...That's cool, I haven't run into a power supply that won't fire up without a load and I probably would have written it off as a bad PS if I did. I have, however, seen voltages stray quite a bit (not what I call "wildly out of spec") from the proper voltage. Thanks for the update on getting more accurate diagnostics.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5554915078212081470.post-78416923443416816872008-03-29T03:57:00.000-05:002008-03-29T03:57:00.000-05:00Great write-up! I would like to point out somethi...Great write-up! I would like to point out something though that may lead the neophyte to improperly deciding a power supply is bad in some circumstances.<BR/><BR/>Switching power supplies are designed to operate at a minimum load rating. This means if you power up the supply without anything plugged in, it's operating out of spec. While it won't hurt the supply, it will definitely give inaccurate readings. In some cases, the power supply won't fire up at all. This is even allowed in the ATX spec: "...The power supply may latch into a shut-down state." What I've seen in this instance is the fan will bump a little bit until you try to restart it. The ATX spec says that if minimum loads of 0.3A for 3.3V and 1.0A for +5.0V exist, all ratings must be met.<BR/><BR/>Usually what I do when I have a known bad power supply is to use one or two 10 ohm wirewound resisters. This is one of the few things Radio Shack still carries. A 10 ohm resistor between red (+5V) and black (Ground) will draw 0.5A at 2.5W. (5V / 10ohm = 0.5A; 5V^2 / 10ohm = 2.5W.) Two of these in parallel will draw a full amp - enough to drive the PS to spec, but usually only one will be enough to allow the PS to kick on.<BR/><BR/>Also, the +5VSB is usually derived from a separate linear regulator, and not the same switching regulator that supplies the main +5V rail. A linear regulator will maintain spec even when no current is being drawn. While it's a nice feature, we don't use a linear regulator for the main supply because it is less efficient (hotter) than a switcher, and certain characteristics of switching power supply technology allow them to be made with much smaller (and cheaper! and more reliable!) components.Paul Rixonhttps://www.blogger.com/profile/00849890722045878120noreply@blogger.com