Weird power problem

[Joel Kleppinger]

I'm completely baffled by this. What happens is that this PC really does work fine, at least as long as the power supply isn't where it's supposed to be. Check out the pics:
http://home.earthlink.net/~jkleppinger/power1.jpg
This pic is taken while the computer is on and working fine.

http://home.earthlink.net/~jkleppinger/power2.jpg
This is right after shutting down, and carefully inserting the power supply, being very careful not to touch or bump anything in the PC. Now, when I push the power button, nothing happens (and it is plugged in).

The only thing I can figure is the magnetic fields in the power supply (which is an Enlight 250W) is somehow causing problems with the CPUs/motherboard. That's about the only thing I can figure since the heatsinks of the CPUs don't touch the power supply. There's about an inch of clearance.

Anyone that knows what is going on or has any ideas, I'd certainly appreciate it. I'd like to get this system running ok with the power supply in so I can get it out of the way.

[Joel Kleppinger]

---UPDATE---
I decided to try an experiment. I pulled the power supply back out, and fired the system back up. Then, I took the power supply and inserted it as far as it would go into its spot, and everything worked ok. Then, I turned off the system and turned it back on with no problems at all.

Here's what it looks like

http://home.earthlink.net/~jkleppinger/power3.jpg

[TigerStrike]

Thats pretty **** weird....

Maybe a frayed or broken power cable?

[MrEd]

Frayed cord could be the problem. It's probably grounding out against the frame.

[Joel Kleppinger]

I think grounding has something to do with it, but I'm not sold it's really the power supply causing it. I've had problems with motherboards grounding out before in this particular case, and so that's been a suspect all along.

I just seems odd that it would be a frayed cable because all the insulation on all the cables appears fine visually. A shorted cable doesn't totally jive either because the current position of the power supply is nearly identical to the normal one (the one that doesn't work).

I continue searching, seeking.......

[This message has been edited by Joel Kleppinger (edited 08-07-99).]

[BBA]

Conductor strands will sometimes break inside wire insulation. Look for signs of bent/kinked spots in wire or move wires back and forth to see if you can induce the problem. Chances are one of the wires to the MB connector from the PS is the culprit, and to make it worse, it could be one of the sensing wires that turns the PC on or off, and once it's on, it wont matter if it has continuity or not. Try to turn the pc on while moving the wires in that case.

Pain in the ****, better you than me!

BBA

[This message has been edited by BBA (edited 08-07-99).]

[Joel Kleppinger]

Hey thanks BBA...... especially that last comment. http://www.sysopt.com/forum/smile.gif

[skywalker[TSG]]

next time when this happends put the PS back in the case when it is still turned on

did u not like my idea of having pics on the user reviews or what u havent even replyed ?

[Joel Kleppinger]

Well, the problem with putting it back in while it's on is that the cord must be pulled out to pass all the way into the case.

Sorry about not replying... I remember reading it, but just got tangled up in my brain with the host of junk that's already there. http://www.sysopt.com/forum/smile.gif

[Ygor]

Time to defrag?

[Nathan Gleit]

I dont think it was grounding of PS case to chassis because your operational pic has it resting on chassis (and it should be that way). Could be a bad crimp on one of the connector pins On the PS wiring harness, most likey MB PS connect. Did you wiggle the wires where they go through the PS chassis(by grommet) to see if it would fail? wonder if one of the cables was stopping one of the fans. Nice dually rig!!!


[This message has been edited by Nathan Gleit (edited 08-07-99).]

[800XL]

I think you win the prize for "Computer problem most likely to involve VooDoo, Demonic Possesion, or Haunting". I think I'd have to go with the 'loose wire' guess on this one also. Could be on the connector to the board, the connection to the PCB in the power supply, or anywhere along there. Another thought would be a screw, metal shaving, or small critters floating loose in the power supply itself. If you can move the wires around with it running and it stays running, it could be something that only affects things at bootup like a power-good line, or something else, but its a start. You would think a loose wire would be a little more random than this.

You could fish the power cord through the hole in the case, get the thing to boot up, and then put it back where it belongs to see if it continues to work. I'd be on power supply number 2 or 3 by now though, or maybe getting out the chicken blood, bones, and working up some mojo against da curse.

[Joel Kleppinger]

Voodoo nothing. No Possession or ghosts either. I have determined the nature of the problem with certainty. It is the motherboard is grounding out through the case... but barely. What I did was install the power supply while it was plugged into the wall and then plug it into the motherboard. The fans on the motherboard twitched like they should, but the system wouldn't power up.

So I decided that it was something related to the act of plugging the power connector into the motherboard. The only force I was using was downward, so I figured that it must be forcing the motherboard down as well. Well, if it forces it down, I'll force it up. http://www.sysopt.com/forum/wink.gif

I proceeded to lift the part of the motherboard next to the power connector (this is not recommended for children at home) and push the power switch. The reason I was able to lift the motherboard is that I had loosened two screws earlier because I always suspected grounding, but couldn't put a finger on it. When I pushed the button, it powered up just fine at 595 MHz. http://www.sysopt.com/forum/wink.gif

So I released the motherboard and everything continued on as normal. Then I turned off the machine to run a test. Sure enough, when I pushed the power button again, it fired right back up.

Here's my theory: Something the motherboard was grounding out was related to the first time it gets power and turns on the power supply. If the current isn't broken between the time the power supply is first plugged in and the motherboard turns on the power supply, then it bypasses the circuit somehow and always powers up. At least, that's my best guess.

So for all you techs out there, always remember this: There's nothing that can't be solved by knowledge, experimentation, and prayer. http://www.sysopt.com/forum/smile.gif

[BBA]

Plus a little luck.