[Jacg]

Hi, I noticed many people have a similar problem with their PC turning off after a few seconds. I have been unable to find an answer in the forums so if you have any additional suggestions I'd be grateful.

Asus M2N SLIdeluxe, AthlonX2 6000, Antec TP3 650watt, Geforce 8600GT, 2x 1GB corsair, 500GB SATA, windows 7.

- Green LED on the mobo remains lit as it should.
- The computer powers on, power LED light up, fans spin, then the system shuts off. Then it will not respond to pressing the power button.
- My first attempt to move the computer case resulted in this powering on and off by itself. In the end I cannot explain this occurrence.
- I suspected the power switch wiring. After I metered out the switch and confirmed the switch and wiring was intact, I plugged it back into the Mobo and as soon as the plug touched the pin it powered on and off again by itself. I reaffirmed the power switch wiring was okay, yet it still will not respond to pushing it.
- I gave a single tap on only one power switch pin on the mobo (did not short the 2 power switch pins) with an isolated screwdriver tip and it again powered on and off. This is electrically equivalent to plugging in the open power switch which I did before. Maybe it responds to capacitance? Maybe I don't know what I'm talking about.
- If I try to power it on again right away it does not respond. I realized the system actually needs about 5 minutes or more before it will again respond to the screwdriver with powering on and off. If the power switch remains plugged in it never respond to pushing it.
- I can also get it to power on then off by connecting AC power to the power supply. As soon as power is connected it powers on then off by itself with no input from me. This method does not require the delay.
- Tried reseting the CMOS, did not work.
- Tried removing all RAM to get an error code, beeped once quickly before shutting off again.

- I read online topics and decided to check the PSU. I jumpered the power on pins and check every pin, every harness. Fan stayed on and all the voltages were in range of perfect.
- I tried minimal system build with either stick of RAM, no chassis fans. This did not fix anything.
- I then removed everything and visually checked for burns or damage. No apparent damage.

- This computer is a few years old. No previous symptoms before this problem.
- 500GB HD and Win 7 were both installed 6 months ago with no issue.
- The power switch is guaranteed not the problem.
- I don't think it's the power supply (650w), but is it possible that the voltage will drop under load and shut it down? I'd be surprised if my system used more than 350watts at any given time.
- I think it might actually be the motherboard, perhaps the circuitry is failing. Yet I find this strange because I generally don't hear of a motherboard suddenly stops working. My computers have always been reliable until now.

Thanks for reading, I'd appreciate any suggestions.

[Advertisements]

The switch on the front of the case is a momentary contact switch, so it only jumpers the pins when you press the button. It's not like it continuously carries current while the machine is on. It's possible that the switch is getting stuck resulting in the computer thinking you are holding the power button in an attempt to turn the computer off.

But if that were the case, then by removing the leads to the motherboard and jumpering the two pins with a screwdriver, the computer would stay on. You can try that to eliminate the switch as a potential problem.

On a side note, I ran into this before with a buddy's computer and it turned out that it was the 4-pin connector that supplies power to the CPU. I checked it with my power supply tester and it tested good. I noticed that if I fired it up without the 4-pin lead connected that it would stay on. I'm not sure if it was a good idea to try that or not, but it didn't do any damage to his system. We swapped in a known working power supply and it fired right up.

Can you borrow a power supply from a friend, or perhaps pick one up from Best Buy? You can return it to the store if it ends up not being the problem.

Edited by Spyderturbo007, 04 March 2011 - 08:41 AM.

[Spyderturbo007]

The switch on the front of the case is a momentary contact switch, so it only jumpers the pins when you press the button. It's not like it continuously carries current while the machine is on. It's possible that the switch is getting stuck resulting in the computer thinking you are holding the power button in an attempt to turn the computer off.

But if that were the case, then by removing the leads to the motherboard and jumpering the two pins with a screwdriver, the computer would stay on. You can try that to eliminate the switch as a potential problem.

On a side note, I ran into this before with a buddy's computer and it turned out that it was the 4-pin connector that supplies power to the CPU. I checked it with my power supply tester and it tested good. I noticed that if I fired it up without the 4-pin lead connected that it would stay on. I'm not sure if it was a good idea to try that or not, but it didn't do any damage to his system. We swapped in a known working power supply and it fired right up.

Can you borrow a power supply from a friend, or perhaps pick one up from Best Buy? You can return it to the store if it ends up not being the problem.

Edited by Spyderturbo007, 04 March 2011 - 08:41 AM.

[iammykyl]

After I metered out the switch and confirmed the switch and wiring was intact, I plugged it back into the Mobo and as soon as the plug touched the pin it powered on and off again by itself.

Were you working on a live system?

- I gave a single tap on only one power switch pin on the mobo (did not short the 2 power switch pins) with an isolated screwdriver tip and it again powered on and off.

Just tried that on my bench test rig, nothing happens, which makes sense as no circuit is established. Computer only starts when shorting out 2 pins.

Remove all the leads from the front header block and then short out the 2 pin power on pins?

Does your Mobo have on board power and reset buttons? Could be a fault there.

[Jacg]

Thanks for replying. I checked the system with a brand new PSU, no fix, same problem

I initially didn't test the PSU because I didn't have a PSU to borrow but luckily the folks at Future Shop said buy one and return in 30 days, open box, they don't care.

I do understand the power switch pins on the mobo require only momentary contact to power on. I reaffirm the power switch and wiring are not at fault.

[Jacg]

Yes, you may think I'm nuts but I almost always work on my PCs live unless I'm disconnecting the PSU harness from the motherboard. I know this is a no-no practice. What I mean is I'll add cards and fans and drives-and certainly the power-on switch wires; without shutting off power. My old PC from 2002 still works great and this problem-PC is a few years old. I'm always careful to frequently create a static discharge path when handling components, but I don't maintain a steady path.

However, I know damage to circuitry can take time to show symptoms, so I'll quit being so careless. I might even start using a grounded wrist band.

Back to the topic, I posted a minute ago that the PSU was not the problem.

The Asus board has all power on, reset, pw LED, IDE LED, and speaker pins in typical configuration with spares omitted, requiring external power and reset buttons. The package even came with a neat labeled pre-plug block to make wiring it dummy proof.
I know the 2 pins require a momentary closed state to power on the system, that's why I found it so strange that touching a single pin will power it on and I posted it as a detail. That is also why I posted that lame suggestion that the motherboard is hooped and is responding to capacitance in the open circuit. Maybe it's true.

I'm pretty sure it's the motherboard. Any other thoughts?

[Jacg]

And yes I did try to manually short the power-on pins. If it just finished powering on then off, it does not respond until after a 5-10min delay as I described in my first post. If the power wires are connected it will never respond to closing the circuit. If the power wires are off and the delay has lapsed, and I try to manually short the pins it does power on then off, but it's not actually responding to the momentary closed state; rather it powers on as soon as one side of the jumper touches a pin, same as with the screwdriver.

[iammykyl]

And yes I did try to manually short the power-on pins.

Did you do that with all plugs removed from the front header block? Thought that without those wires connected would rule out the possibility of a fault in Reset, Power LED, IDE LED.

I know the 2 pins require a momentary closed state to power on the system

My understanding as well, but in some way the button must function in another way as if you hold it down for 4 seconds, the computer will shut down.

That is also why I posted that lame suggestion that the motherboard is hooped and is responding to capacitance in the open circuit. Maybe it's true.

Not lame at all. This sometimes works for all sorts of problems, (don't understand why)

Computer off. Hold in the power button for about 20 seconds,(with it still held in_ switch on the PSU)

Result from another post.

Posted 08 February 2011 - 03:58 PM

Wow... here is what I did so far I followed your instructions almost.. I did not plug in all my ram yet but with what you suggested by holding the power button down i was able to actually get the computer to boot.. Albeit i had already unplugged the CPU cleaned it off and also cleaned off the CPU fan and put it all back in and i got it to boot up.. gonna continue to further test out and make sure everything works one piece at a time..

[Jacg]

Roger that, I'll simplify what I described earlier; every plug has to be removed before the PC will respond at all, and will respond only with a single pin touched, or by itself when turning on the AC power switch on the PSU. All responses remain the same, on then off.

I didn't think holding the power button would do anything when switching the PSU on, but I tried anyway. No fix, same result when switching on the PSU before, on then off.

I assure everyone my power switches work fine. *** In fact I actually cut out my factory button & LED block on my case and mounted it to a blank CD drive cover because I cut a 4" hole to the front of my case where the buttons used to be located-to facilitate a large fan. I actually need to transfer the power buttons to my old computer because the old computer case has an unusable, archaic power scheme. In this way the power buttons can be transferred like a CD drive into an empty CD drive slot.

*** Anyway I moved the power buttons to my old computer until my newer one is fixed, and the computer works normally, power buttons are good.

I'll summarize the findings:
- Power supply, power switch, hard drive, RAM, CPU cooler fan are all okay.
- The computer turns on then off by itself when AC power to the PSU is switched on.
- The computer turns on then off when a single power pin is touched with an isolated conductor then requires a delay to repeat.
- Leaving the power switch harness plugged in renders the computer completely unresponsive.

Based on this would anyone conclude the motherboard is at fault? It sure seems like to me.

One last thing - the only devices I have not tested thoroughly are the CPU itself and the video card, though I did try booting it up with no video card, no fix.
I read it's common to do a bench test outside the case to check for grounding problems, so I'll try that next. Does anyone know if a new motherboard will at least boot up and give error codes with the CPU removed? I'd like to try that but-is that safe for the board? Or for that matter, will the board boot up and remain on if everything except power is removed?

[iammykyl]

I think you have covered ever thing and would agree that the Mobo is the culprit.

This information is only from other forums and not from a professional source.

No motherboard will post without a CPU present.
On some motherboards nothing will happen at all without a CPU present. (possible because the board has some sort of protection)
On some motherboard with no CPU present, lights come on, beeps may be heard, (can find no information for that type of beeping) the CPU fan will spin if connected.

This still will not tell if the Mobo has a fault. Can only be tested by the manufacturer of the board.

If you are going to do a bench test, can you source a Socket AM2 to try on the board in case it is the CPU.

Also give the board a good shake over a piece of white material to see if anything, like piece of wire, falls out. A good blow with some compressed air in all the slots.