Problem: Computer Turning Itself On And Off

Associate
Joined
29 Nov 2010
Posts
196
I was woke up in the middle of the night to my computer repeatedly turning itself on and off a month or two ago. At the time of course I just turned it off and went back to sleep, concerned that I was was being haunted by some tech savvy ghost, possibly for ticking 'I confirm that I am 18' boxes before I was. But when I woke up in the morning and turned on my PC, the problem persisted. I assumed this could be a few issues; overheating, faulty motherboard, faulty PSU. Before doing any testing I thought my best option would be to give my computer an overdue clean out of everything, so that is what I did. I took absolutely everything apart, including the fans and the case, and cleaned any dust off of it all. When I put it all back together and turned the power on, it worked! And it did so for the last couple of weeks. Unfortunately this did not last as today the same thing has occurred.

Mid browsing, it flashed that there was an error with some application and then quickly turned off. I think the error occurred because of the fault though and not the other way around as the problem is clearly bigger than a simple faulty application. I wasn't doing anything intense, literally just had a few tabs open on Chrome. I turned it back on and I got to the dashboard where I was able to load Chrome, and then it crashed again. Now if I try to turn it on it is crashing before it even allows me to press 'Start Windows Normally'. I wanted to test if the CPU was overheating, which seems unlikely as it is crashing before it even has time to warm, but I don't have enough time on my computer before it crashes to do that now, unless there is some quick way through BIOS, but I thought I should consult you wizards on here first.

Any help or advice on how to fix this would be greatly appreciated, thank you in advance.
 
Whats the spec? Tried swapping ram about/different slots/just one stick? Also tried taking everything out the case to power on just to rule out a short?

If it was me id suspect the PSU but things like this can be anything :/ I'd try unplugging everything that wasnt necessary and see if it can boot to bios and sit there ok.
 
If attachment of CPU cooler has failed CPU could overheat in seconds.
Though all not prehistoric CPUs have very good temperature protections.

Could be also failed PSU, maybe also motherboard.
15 years ago during capacitor plague friend had PC's motherboard failing, which started causing progressively more and more errors and crashes.
(though that progression of symptoms happened over couple months)

Memory wouldn't be likely to go out in this kind of progressive way, unless motherboard's VRM for memory was failing.
 
Motherboard or maybe PSU, or even CPU is not 100% out of the question.

If I were a betting man though I'd say motherboard. They have tons of components on, which can start to fail, it wont necessarsarily kill your PC straight away, but cause "weird" things to happen.

However, I would go back in, and re-seat everything first, the fact that you dusted it out and after it worked, may have been you slightly jiggled something (RAM etc) that eventually uniggled itself again. Its worth a go in the first instance.

Failing that, try unplugging any hard drives/SSD's etc, see what happens, it might be more stable if its not actually trying to boot anything.

I reckon do the above first, report back.
 
Hey guys, thank you very much for your help. I had a go at getting it running again this morning, again with only temporary success. Firstly, I tried turning it on as normal and it was turning itself off and turning itself back on shortly after, as it was yesterday. So I opened up the case and made sure absolutely every connection was properly seated, tightly secured the CPU fan and removed the case restart button from the motherboard as I had the idea that maybe this was faulty and the cause. When I turned my PC back on it stayed on for about 5-10 minutes, during which I assumed I had fixed it by removing the case restart button from the motherboard as in the short period of time I had on my computer I downloaded an application to test my CPU core temperatures which were very comfortable, just below 40 degrees Celsius and only peaking to about 45 straight after start up. But then, of course, it restarted itself despite the case restart button being unplugged.

Next thing I did was remove all 4 sticks of RAM which total to 8 gigs. Surprisingly when I turned it back on it stayed on, although for some reason nothing was being displayed on my monitor, at all. It wasn't even registering a signal, rather idling like there was nothing connected. So I tried to turn the computer off at the motherboard and it seemed to refuse first time... don't know if that was me holding down the power button for too long or an actual rebellious mobo? I pressed down the power button again on the motherboard and it turned off. So I put the RAM back into their slots, properly secured and turned it back on. This time it started and then turned itself off, again before BIOS even loaded, and then didn't even turn back on. I tried this 3 times with the same result each time. So I took one set of RAM out, tried again, it started, turned itself off, didn't turn itself back on, same problem. So then I tried removing absolutely everything from the motherboard and PSU, keeping only the two power cables connected to the motherboard of course as well as the CPU and CPU fan. Again, it was just turning on and turning off straight away. Right now I have everything unplugged, and I'm 90% sure from my testing that it's the motherboard. What do you guys think?

Specs:
Foxconn P67A Series - motherboard
Antec TP-650 - PSU
Intel i5 2500K Sandy Bridge LGA1155 socket (not overclocked) - CPU
Corsair XM53 4GB (2GBx2) x2 (8GB total) - RAM
AMD HIS Radeon HD 6870 - Graphics Card
SanDisk Extreme 120GB - SSD
1TB Western Digital Black - HDD
Antec 905 (stock fans, big top fan not working) - case
Stock CPU fan
 
Could certainly be motherboard.
Heck, would consider pretty much whole PC having given good service.

You could basic test PSU by trying "jumpstarting" it by connecting green wire to ground (black) from 24 pin motherboard wire.
It should be old enough to have proper colour coding in wires, instead of modern fashion BS of all black Bs wires.
 
Cool, looks like I'll be getting a new motherboard then! Only problem is nobody really sells them any more, especially not OverclockersUK. I don't know if this is allowed, but could somebody suggest where I may be able to buy one? There's a few rubbish ones on Amazon and there's some used and refurbished ones on eBay. How do you guys feel about used or refurbished motherboards, especially ones this old? I'm extremely sceptical...
 
Cool, looks like I'll be getting a new motherboard then! Only problem is nobody really sells them any more, especially not OverclockersUK. I don't know if this is allowed, but could somebody suggest where I may be able to buy one? There's a few rubbish ones on Amazon and there's some used and refurbished ones on eBay. How do you guys feel about used or refurbished motherboards, especially ones this old? I'm extremely sceptical...

I hate to be the bearer of bad news but you will struggle to get one these days that'll work on a Sandybridge, your probably looking at used eBay jobs at best.

I'm in a similar (ish) position, fortunately my motherboard is still going, but recently the primary PCI-E lane died, so I'm having to put the GPU in the lower lane, but yea, it's early signs that it's going.

Loads of people still using Sandybridge CPU's cus they were/are just so good, funny that the death of the Sandybridge CPU won't be the CPU itself, but the motherboard to support it.
 
I've found some new ones on eBay which seem quite good. There's a company from China who's selling 'brand new' Gigabyte GA-Z77P-D3 motherboards. Is this a good board? Should I be concerned about fakes or anything else?
 
I'm in a similar (ish) position, fortunately my motherboard is still going, but recently the primary PCI-E lane died, so I'm having to put the GPU in the lower lane, but yea, it's early signs that
PCI-e lanes of x16 slot come from CPU.
So might be even CPU itself wearing down and not just motherboard.
 
Back
Top Bottom