Faulty power button or sign of something more serious?

Associate
Joined
8 Jul 2010
Posts
833
Location
Staffordshire
Sometimes when I turn my computer on, nothing happens. It's as if I haven't fully pressed the power button but after two or three presses, my computer will start. I've just tried turning it on and, again nothing happened when I pressed the power button but this time I left the computer alone as I was moving something from my (actual) desktop. After one or two seconds, the PC powered up, as if it's reaction to the power button was delayed.

Could this be a sign of something serious or is it more likely to be related to the power button itself?

NOTE: The computer was fully shut down before I pressed the power button, it was not in a low power state or any form of hibernation/sleep.

EDIT: I've just been doing a bit of research and feel it's necessary to pre-empt some questions and elaborate on what's happening as it seems that the most common problems when it comes to a delayed startup invariably involve symptoms that I do not have. After pressing the power button, nothing happens. The computer does not briefly boot only to lose power for a period of time before rebooting. Non of the fans or HDD's in my system come on. Next time I boot up, I'll use the power button on the mobo to see if the same thing happens. Although, as the issue only seems to happen intermittently (most commonly after the computer has been off since the previous day), there's no guarantee that I'll be able to reproduce the issue.
 
Last edited:
Associate
OP
Joined
8 Jul 2010
Posts
833
Location
Staffordshire
I would first check the power connector from case to motherboard. Is it seated correctly?

You can try shorting the power pins directly on the motherboard to turn the computer on. If it works fine every time doing this then it's the case/power button.

If not, power supply is the most likely culprit, but could also be motherboard.

Perhaps it is just a loose connection. I had checked the connections to the mobo but hadn't though about the other end of the wires and I was fiddling around in that area recently.

I'll check a few things when I get time but I'm gonna be a bit busy tonight.
 
Associate
OP
Joined
8 Jul 2010
Posts
833
Location
Staffordshire
(Why didn't I just leave it alone so I could enjoy my Sunday night?).

I just took the top panel off to check that I hadn't knocked any of the wires when I was doing some work in that area recently and everything looked fine. I double checked the connections to the mobo and they also appear to be fine. I thought I'd take the opportunity and observe what happened when I pressed the power switch. For a brief moment I though I'd gotten lucky when I pressed the power button as nothing happened (at least I was able to replicate the problem... or so I though). Five or ten seconds and still nothing. I pressed the case button several times, tried turning the PSU off and on. I tired reinserting the PSU power lead. I tried using the power switch on the mobo itself but no matter what I do, my computer shows no signs of life other than the "POWER" and "RESET" buttons on the mobo lighting up and the lights on my GPU being on (which they are even when to computer is off).

So, I guess now the question is, can it be anything other than a PSU issue and if so, what?
 
Associate
OP
Joined
8 Jul 2010
Posts
833
Location
Staffordshire
I'm just testing the PSU with a Thermaltake Dr Power II and it does pass but not always. The voltages are all within tolerances (but not as close to spec as the reviews I've read said they'd be). Where the unit occasionally fails is the PG signal. The reading is a little high at just over 500, around 530 or so but sometimes the PG test fails. I've got a sneaking suspicion that it may have been down to the way I operated the PSU tester.

With that in mind, am I right to concluded that the motherboard is the likely culprit at the moment? And, if so, how should I proceed?
 
Associate
OP
Joined
8 Jul 2010
Posts
833
Location
Staffordshire
Check ALL the connections on your motherboard

My brother's PC has this issue occasionally with the GPU coming slightly loose - but it can be nearly any connection: as soon as the motherboard decides the voltages somewhere aren't right, it will simply refuse to start.

Before testing the PSU (which I took out of the case for ease), I double checked all the connections and tried to boot but again nothing happened. The lights on the GPU and mobo (which are on when the system isn't powered up) were on but nothing powered up, no fans, no HDD's. There were no signs of life at all other than the lights.

I will reinstall the PSU and check again but in the mean time, how would I go about narrowing down the fault to the mobo?
 
Associate
OP
Joined
8 Jul 2010
Posts
833
Location
Staffordshire

Thanks for the thorough response. I'll work through a checklist when I've got time (possibly tomorrow) and report back.

Id be inclined to pop the PG pin out of the 24 connector block and just temporarily rig up one of the 5v lines into it's place eliminating the actual PG signal. Quite fiddly but ultimately will tell you if this PG thing is causing an issue.

That's beyond my level of expertises at the moment and I don't have to tool to remove the pins. Thanks for the response though.
 
Associate
OP
Joined
8 Jul 2010
Posts
833
Location
Staffordshire


I had some time yesterday to fiddle around and I was able to get the computer to boot with every non-essential piece of hardware removed. Again though, I'm really busy at the moment so I might not be able to start the elimination process of adding hardware for a few days.

Thanks for the checklist. It's made this whole thing a little less stressful.
 
Associate
OP
Joined
8 Jul 2010
Posts
833
Location
Staffordshire
First it's the weather, then it's visiting relatives, then it's work...

I've finally got around to trouble shooting the problem.. well, I say trouble shooting. I decided to take the opportunity of upgrading my case to a 900D (relevant gif below) and after taking everything apart (including removing the CMOS battery), everything seems to be working. I'm guessing I've just learned a CMOS related lesson.... anyway, thanks for the help everyone.


http://latimesblogs.latimes.com/.a/6a00d8341c630a53ef015391422b9f970b-600wi
 
Back
Top Bottom