Strange rebooting issue

[Tech_Triage]

I have an issue that I have never encountered before. When I press the power button on the desktop pc it turns on for 2 seconds, then shuts off, then automatically turns back on for another 2 seconds and then back off again.

I first thought it was a over heating issue, and or a processor failure. But I believe it is neither.. I finally decided to take the power button jumper wire off and jump it with a flat head screw driver. It boots fine..

So I grab a few extra power buttons from my junk pile. Using all three buttons one after another and each display the same issue. So I know it isn't just the one power button. It displays the same issue with each different power button.

But yet it works fine if I jump the pins with a flat head screw driver. What could it be? I have never seen this issue before.. it is very odd. Any help and or suggestions are greatly appreciated.

 

[Tech_Triage]

I will go nuts if it is something simple lol :banghead:

 

[SuffolkComputers]

Wow, that is weird! Are the pins bent in anyway? Maybe the power connector isn't getting full contact

 

[4ycr]

I know it boots fine with a screwdriver but try a spare power supply. I have seen something similar and the PSU was the problem

 

[Tech_Triage]

I know it boots fine with a screwdriver but try a spare power supply. I have seen something similar and the PSU was the problem

I forgot to add that I have tried 3 different psu's also.

 

[Tech_Triage]

Wow, that is weird! Are the pins bent in anyway? Maybe the power connector isn't getting full contact

Not that I can tell, they do not look damaged or bent or scratched in any way. It could be a poor connection but the reboot cycle is the same pattern of off for 2 seconds and on for 2 seconds then off for 2 then on for two in infinite consistent loop until I either flip the switch on the psu or pull the plug on the psu

 

[Tech_Triage]

I am sure this doesn't make any difference but it is a socket 775 @ 3.06ghz using ddr2 @ 2gb ram 1600mhz. Psu is a 650w.

It doesn't throw any audible codes nor does it even load the boot screen in dos.

I can also use a switch, push it, it turns on, quickly remove the switch's connector from the pwr jumper pins and it will fully boot and be fine, until I shut it down. Although a reboot does not affect it, it stays on once it is on. But if I shot it down, it is back to the off/on cycle behavior again... so wierdweird


Maybe the machine is Cursed!

 

[Tech_Triage]

I also tried booting it up out side of the case just laying on the table. Still same crap... ugh!

 

[GeeksInKhakis]

I forgot to add that I have tried 3 different psu's also.

Damn, that was my bet...those are the same symptoms I've seen with two systems with bad power supplies.

 

[JustInspired]

Think about it logically. Whether you use a push button switch or you bridge the two prongs with a screwdriver, you are creating a MOMENTARY electrical connection.
The only difference (providing the switch is not faulty) is the length of wire.
Perhaps the board is faulty and the wire is picking up some stray electrical or wireless interference that is not apparent without the antenna effect if the wire.
This is the only rational explanation I can think of.

That's not to say I don't believe in possessed computers

 

[ComputerRepairTech]

I have an issue that I have never encountered before. When I press the power button on the desktop pc it turns on for 2 seconds, then shuts off, then automatically turns back on for another 2 seconds and then back off again.

I first thought it was a over heating issue, and or a processor failure. But I believe it is neither.. I finally decided to take the power button jumper wire off and jump it with a flat head screw driver. It boots fine..

So I grab a few extra power buttons from my junk pile. Using all three buttons one after another and each display the same issue. So I know it isn't just the one power button. It displays the same issue with each different power button.

But yet it works fine if I jump the pins with a flat head screw driver. What could it be? I have never seen this issue before.. it is very odd. Any help and or suggestions are greatly appreciated.

i've been thinking about this and thinking about it and no matter how much of a silly explanation I come up with I can not explain the highlighted part.

Edit: Just to confirm you have tried this without ANY of the front panel connectors attached aside from the power switch?

 

[nonchalant]

Remove the battery & clear the CMOS.

 

[Tech_Triage]

Remove the battery & clear the CMOS.

Did that also. I am thinking it may be resistance in the wire to the pwr button.. that is the only thing I can come up with

 

[djyox]

First its not the ram right?

When you replaced the buttons, did you not replace the wire too? I suppose you could go grab some wire from an old case and try it out.


Unless there is a short in the wire, or maybe the coating is thin in one area and the board is picking up on the signal from another wire? You could add some black tape anywhere the wire comes into contact with any other wire (or just replace the wire too, I suppose that would be easier)


I used to work on RF in trucks, and if two RF cables would touch, it would make static in the one radio. So we would keep them apart when we could, otherwise we would put more plastic shielding around them.



Because correct me if I'm wrong, if you close the circuit, it turns on. Open it and it does nothing. Close the circuit again, and it will shut off, open it will do nothing again. So no matter how it gets closed, it will do something: IE: Wire shorting out, button being pushed, or screw driver being applied to pins...

 

[LifelineIT]

I'd bet you a 99cent box of nuggets that it's a ground fault on the board.

I dunno your exact scenario, but perhaps the switch housing is ground-faulting, resulting in an always-engaged arrangement of the power button, but this fault is only occurring when the button is on and smooshing the actual apparatus into the board itself. No button = no squish.

OR there's some sort of overvoltage situation happening that's causing it to arc through the switch (unlikely)

Use your screwdriver method and keep it engaged and see if you can force the behavior to repeat, this will tell you if it's a constant-on fault. I suspect that the apparatus of the physical switch is staying in contact with something it's not supposed to, and it's fault looping.

 

[B Trevathan]

Have you checked the battery or better yet replaced it?

Have you tried using a "ATX Control Kit" or a short wire to jump the pins?

Have you tried powering up with no drive cables attached?

What happens when you hold the power button down?

 

[Eagle21]

I had a very similar problem last year. It was an intermittent problem which occurred mostly from a cold start.

Replacing the memory fixed the problem (even though the existing memory had tested OK). So it must have been a memory-motherboard incompatibility problem, or the motherboard was starting to fail and no longer worked with the current memory.

Try a google search for your motherboard model number - it may be a known problem for your particular motherboard.