Naked PC power supply

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
jalbers
Posts: 18
Joined: Tue Dec 18, 2001 1:01 am
Contact:

Naked PC power supply

Post by jalbers » Mon Sep 19, 2005 7:56 am

I am trying to get some PC power supplies to work with not computer attached. I know they won't start without a load so I have tried connecting up a floppy drive, hard drive, and CD drive at the same time but the power supply still won't come on. It is one of the newer types of supples with no on/off switch and a 20 pin connector that attaches to the mother board. To be specific it is a Power Tronic PK-6145DT3 145W power supply from a gateway computer. ANyhelp would be greatly appreciated. Thanks

User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

Re: Naked PC power supply

Post by philba » Mon Sep 19, 2005 9:15 am

google for one of millions of pages on this topic.<p>you will need to do 2 things, put a load on one or more of the outputs (specific PS dependent) and assert "Power Good" or what ever its called. google is your friend.

User avatar
Edd
Posts: 885
Joined: Sun Jan 27, 2002 1:01 am
Location: Dallas Tx
Contact:

Re: Naked PC power supply

Post by Edd » Mon Sep 19, 2005 7:41 pm

Considering the unit to be of the ATX family with that typical 20 pin block connector and its
sometimes stuck /cramped/and/ sluggish plastic snap in locking tab….Grrrrrr.
Initially, I will give the four corner posts of the plug, the corners with orange 3.3VDC and 3.3VDC orange wires are
pins 1 and 11, on the opposite corner ends are pin 10 yellow wire (+12VDC) and pin 20 red
wire(+5VDC). That should let you figure in the other number terminals if you cant read them by
now for confirmation.
An initial test of the power supply that has its IEC power cord installed will be the testing of
the individual little internal switch mode design stand by power supply and that voltage can be
measured at pin 5…Burple. … and referenced to pins 15-16-17…all being black wired ground
busses.
Now to fully power up that unit will require a loading of the +5VDC supply, and the other one I
use is the +12VDC supply as I have a connector made with two 1157 auto tail light lamps that go
to a common (universally used) Molex 4 pin power connectors end pins and their grounds to the center two pins. This gives adequate minimal loading to a unit, if it is a properly functioning unit.
A jumper wire between pin 14 green wire (power on) and an adjacent pin 15 ground buss should bring
a working unit up. Then the supply voltages outputs can be confirmed.<p>Addenda:
BTW the most common failure mode leading up to a final mass destruction and meltdown of semis
is the handful of electrolytic capacitors mounted within the unit. As a responsible preventative
maintenance of the PS drop the two or three sided metal cover to gain visual inspection of the units for
any “pooching” out of the tops of units…or if a rupture of a topside X cross stress slit has occurred , you are on or past borrowed time. A final test would be to power up the unit and warm it up and then take two hands and put them in your pockets for electrical shock hazard safety and then take the third hand and use its fingertip to enact a temperature test atop each electrolytic . A unit starting to develop excess ESR failure mode will be quite warm to the touch…time for a change out with a like unit of a 105deg temp rating.
<p>
73's de Edd
[email protected] .........(Interstellar~~~~Warp~~~Speed)
[email protected]........(Firewalled*Spam*Cookies*Crumbs)
;) ;) <p>Driving a crappy car changes your entire mind-set. If someone cuts me off on the freeway, I can't flip them off because I may need that same guy to jump-start me in a few minutes. <p>Bonus ….. Ka …..ching !!!<p>I've finally got myself a car that really turns heads……. that being mostly because of its knocking, rattling, squeaking and backfires. .<p>[ September 19, 2005: Message edited by: Edd Whatley ]</p>

rshayes
Posts: 1286
Joined: Tue Mar 04, 2003 1:01 am
Contact:

Re: Naked PC power supply

Post by rshayes » Tue Sep 20, 2005 2:58 am

I would emphasize turning off and disconnecting the power supply before attempting to determining the temperature of the electrolytic capacitors by touching them with a finger.<p>The aluminum cans of these capacitors are usually connected to the negative terminal of a bridge rectifier that is in turn connected to the AC line. For one half of the AC cycle, they are connected to the neutral lead of the AC supply, which might be connected to ground. It should be on a normal 110 volt system. On the other half cycle, it is connected to the hot lead lead of the AC supply which will have a peak value of minus 170 volts or so with practically no current limiting.<p>These power supplies are controlled by a logic signal. These probably switch off the main power supply outputs, but there is no guarentee that they shut off the internal rectified power.<p>Basically, before you stick a finger in, DISCONNECT the power supply from the power source. Even then be cautious, there may be a substantial charge left on filter capacitors.

User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Re: Naked PC power supply

Post by jollyrgr » Thu Sep 29, 2005 4:00 pm

I read this a few days ago and it got me to thinking about how many of the ATX power supplies from used computers there will be available for "free" and how few of the AT style will remain in a couple years. This week I noticed a coworker putting a beefier power supply in his computer. And the manual came with pin outs and instructions! Here is a scan of the image:<p>Image<p>
Someone check my logic, but this is what I see from this graphic:<p>Pin 14 is the control lead from the power supply to turn it on or off. Pulling this lead to ground (low) turns on the power supply. Pulling the lead to high +5V turns it off. These ATX power supplies have +5V going to the mother board at all times; they call this +5VSB which is found on Pin 9.<p>Maybe my logic is flawed here but let's say we find pin 14 and pull it high via 5K resistor. The power supply will be off. Now, lets pull the pin 14 to ground and hold it there via closed switch, the power supply will turn on.<p>I have not tested this yet but think I have it right. The schematic for the resisitor would be:<p>PIN 9 (+5V)----/\/\/---PIN 14-----/ -----Ground (PIN 15)<p>
Did I miss anything? I won't be offended by any corrections. Basically put a 5K resistor between pin 9 and pin 14. Put a switch between pin 14 and ground (15). Turn the switch on, ground pin 14, power supply turns on.<p>(DO NOTE: The colors listed may NOT be what is used on every ATX supply. Pin out your supply with a volt meter to be absolutely sure.<p>[ September 29, 2005: Message edited by: Jolly Roger ]</p>
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!

User avatar
Edd
Posts: 885
Joined: Sun Jan 27, 2002 1:01 am
Location: Dallas Tx
Contact:

Re: Naked PC power supply

Post by Edd » Thu Sep 29, 2005 8:11 pm

Sir Jolly Rojerewski:<p>I'll just bet I can clarify the situation for you if I inject in an actual schematic of the typical ATX unit instead of
your thought structuring necessitating the encountering of an unfathomable 'black box' situation.
Re:<p>;) ;) <p>I'm as frustrated as an arsonist in a petrified forest.<p>[ September 30, 2005: Message edited by: Edd Whatley ]<p>[ September 30, 2005: Message edited by: Edd Whatley ]</p>

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Naked PC power supply

Post by Robert Reed » Thu Sep 29, 2005 8:53 pm

Why is this page so wide--can't look at one thing without losing another.

User avatar
philba
Posts: 2050
Joined: Tue Nov 30, 2004 1:01 am
Location: Seattle
Contact:

Re: Naked PC power supply

Post by philba » Thu Sep 29, 2005 10:30 pm

It's because images that are wider than your display are contained in the page. Basically, it's the web browser's fault. <p>Actually, I believe that the forum software could compensate for this problem but don't hold your breath for an update.

User avatar
jollyrgr
Posts: 1289
Joined: Thu Jan 03, 2002 1:01 am
Location: Northern Illinois
Contact:

Re: Naked PC power supply

Post by jollyrgr » Fri Sep 30, 2005 3:52 pm

Thanks Edd! Based on the schematic I see that the pull up resistor I mentioned (the 5K resistor) is already in the supply and is a 4K7 resistor, R23. If I had thought and put a standard value, I'd have gotten it exactly right. But somewhat amazing that I came as close as I did to the actual value used as the internal pull up resistor, isn't it? :-) <p>To whom ever wants to activate ATX supplies: You won't need the resistor I suggested. Simply ground the PS-ON connection and you are in business. Thus connect a switch between pin 14 and something like 13, 15, or another ground point and you are in business. <p>The manual I had did not include a full schematic and I didn't bother searching one out. Just looking at the sections you pointed out gave me kind of what I was expecting to find. But I would have thought the pull-up resistor would have been on the mother board. It's nice to know they spent the other one cent to put the resistor in the supply.<p>In the various models listed in the manual, there are some extremely powerful units out there. The model my coworker was installing, a 480 Watt unit, put out a massive 22Amps on the 12V rail and something along 30Amps on each of the 3.3V and 5.0V rails. Too bad switching power supplies are as noisy as they are, or do I have that wrong? In other words, can these be used to power an HF radio without introducing too much hash? I know that older AT supplies have been used in such a manor but I've only heard complaints. Any votes here?<p>And it might be cheaper to use these for the outdoor "Malibu" lights as long as the power supply remains indoors (as mine are) with a feed to the outside. A 300 Watt Malibu transformer is over $100. A 300 ATX supply is only $20 new. If you rescue one, it is "free". <p>(Note: the correction was revising a typo of 40Amps to 30Amps).<p>[ September 30, 2005: Message edited by: Jolly Roger ]</p>
No trees were harmed in the creation of this message. But billions of electrons, photons, and electromagnetic waves were terribly inconvenienced!

Post Reply

Who is online

Users browsing this forum: No registered users and 36 guests