atx power supply for hobby electronics

[LucidGuppy]

does anybody know how to wire pc power supplies for use in hobby electronics? google hasn't yet provided the needed information.

[gerty]

Here's something I ran across..
http://www.nfdc.net/home/cbdb/12%20V%20 ... Supply.htm
I just did two of them, I simplified it by putting a 10 ohm 10 watt resistor across the
5 volt output. There was an adjustment pot on one of them for the 12 volt out and I was able to get it up to 13.9 before the crowbar kicked in. You ought to be able to hack one in an hour or two, depending on how you want to terminate. I have a terminal strip on one with 5 and 12 v out, the other has flying leads with 13.5 out.
Hope this helps...BTW, the resistor is needed for regulation, otherwise it'll just oscillate (and raise havoc with a nearby two way radio)...

[LucidGuppy]

Exactly what I need - thanks very much.

[josmith]

It might be a good idea to check these "low" voltages to earth ground. Some pc power supplies switch power directly from the line and may have higher potential to ground.

[JKMADSCI]

If you want your 12 volts rock solid, find the adjustment pot and feed in 12 volts with a 50k pot.
1. remove original pot
2.connect 12 volts to a 50 k pot to the pot trace
that goes towards the 494 controller.
3. most will adjust the 12 volt rail from 8 volts to 14 volts at 10 amps all day long.

[terri]

Many switching power supplies don't oscillate properly or even run without a load. (Some of them have a built-in load to ensure they will start up right.) IBM used to use a 5-ohm 50W resistor in place of the hard drive in systems shipped without hard drives to get the power supply to start up. I think the 10-ohm resistor referred to above is not for "regulating" purposes, but to put a load on the power supply to start it oscillating in known ways, rather than unkown ways. (Cf pp 300-302 "Upgrading and Repairing PCs" Fourth edition.)<p>Makes me wonder how many good power supplies have been chucked because the tech plugged it into the wall, probed the power-out connectors and found no proper voltses or ampses or anythingses. I used to keep an old drive on the workbench just as a substitute load to check out power supplies.<p>[ January 15, 2005: Message edited by: terri ]</p>

[Dean Huster]

By taking a bit of time, I showed students how they can save a tremendous amount of time and aggravation by building a simple resistive dummy load for computer supplies. Since I was teaching in the crossover period between AT and ATX systems, I made a load for both. You just grabbed a supply capable of lower total power output, looked at the label for the various maximum currents for each supply and used Ohm's Law to figure a resistance for each supply output. Connect that resistors from the respective supply outputs to ground, cross-connect the various sense inputs to get the supply to work and add a male mating connector as the "input".<p>With the load, it takes but a minute to check a supply for proper operation even without removing it from the computer. By selecting the load resistors for minimum load, you can check all supplies for operation, although not necessarily at maximum load, of course. It's sure a fast way to eliminate the supply as the problem and to go through all those supplies on the "junk" rack to see if they're basically OK.<p>They're simple and cheap enough to make that you can make several loads to complement various "wattages" (power output capabilities) of common supplies so you can check them properly at their maximum load conditions.<p>And with a dummy load, you don't have to worry about a bad supply wrecking a perfectly good mother board while just trying to check it out.<p>A next step is to make a small box with male and female mating connectors that you install between the supply and the chosen load. By putting one window comparator (each operating from the ±12v supplies) on each supply with LED output, you can get a super-fast and accuratee go-no check of all supplies at a glance. Check twenty supplies during the time it would take you to check one using a DMM.<p>Dean

[gerty]

terri.. you're right..I worded my reply poorly. Without the load resistor the supply won't work properly and the ouputs will fluctuate (osc?).The load resistor stops that, not regulates it..

[terri]

JKMADSCI: Good idea. I'd tried to use a PC supply as a battery charger once, and couldn't get it adjusted high enough to fully charge a lead-acid battery. But that was when I was poorer than now. Next time I'm more poor and I need a charger, I might try your suggestion.<p>(Howdy to Queens, Noo Yawk! I was born in Manhattan and lived in Flushing and LI for the first 1/3 of my life. Then I moved to the mainland. Temporarily. Forty years ago. Still here at a Mile High in CO and lost most of my accent, sawt iv.)<p>JOSMITH: Good safety tip!<p>GERTY: Thanks for confirmation!<p>DEAN: What's the "Requesciat(sp?) In Pace" ("R.I.P.") for in your sig line --if that's what it's supposed to mean. None of my direct business, but every time I see it, I get more curiouserness about it. Or have you commented on it previously?<p>[ January 16, 2005: Message edited by: terri ]</p>

[Dean Huster]

"DEAN: What's the 'Requesciat(sp?) In Pace' ('R.I.P.') for in your sig line --if that's what it's supposed to mean. None of my direct business, but every time I see it, I get more curiouserness about it. Or have you commented on it previously?"<p>Never been explained, just understood in the context of what was then current events. "R.I.P." is for the many hobbyist electronics magazines that have bit the dust over the years, may of which were wonderfully great magazines that every one of us would love to still see in print. Poptronics was the last to die leaving Nuts & Volts as the only remaining U.S. survivor.<p>Radiocraft
Radio-TV News
Popular Electronics >> Computers and Electronics
Radio-Electronics >> Electronics Now >> Poptronics
Radio-TV Experimenter
Modern Electronics
Electronics World
Elementary Electronics
Hands-On Electronics
[the reborn] Popular Electronics >> Poptronics<p>And any others that I may have missed ... and do miss.<p>Dean<p>[ January 17, 2005: Message edited by: Dean Huster ]</p>

[terri]

Ah! Thank you.