Strange PC problem

Recommended Videos

dfcrackhead

New member
Apr 14, 2009
1,402
0
0
thenumberthirteen said:
Hi guys I need your help to diagnose a problem with my PC. It seems to randomly turn itself off. No BSOD or any other indication just all of a sudden *pop* no power. It's not a problem with the house's electricity or the socket as my monitor is also plugged into the same power strip and it's fine.

It may have something to do with overheating as if you try and turn it back on straight away it only lasts a minute or two before it crashes. The thing is it happens when I'm not even taxing it and the temperature is normal (between 30-45c). I got it a few months ago and it's been fine until now.

I've done virus checks and restored it to before the problems happened, as well as changed the power strip, but no joy. Any ideas? I don't have much money so I'd like to try and fix it myself, or at least find the exact problem before I take it to be fixed.

Any help will be appreciated. Thanks!
Definitely overheating, my fathers computer had the same problems, open it up and turn it on and watch the fans and just kind of feel around inside(like the air over the parts, you do not want to touch the parts if they overheat) and feel for excess heat. The problem my dad had was one of the fans wasn't working and the power source or something was overheating so bad it almost burned to touch it.
 

RoboGeek

New member
Apr 3, 2010
128
0
0
i had a similar problem with a pc before it just turned off randomly in the end we couldn't fix it and got it replaced.

may i ask is the computer gradually slowing down before it dies because my graphics card used to overheat because it doesn't have an on-board fan it would gradually slow down and then eventually die
 

Gasaraki

New member
Oct 15, 2009
631
0
0
It's because your super computating computator isn't geforcing enough RAMs into the hard spec fanning raid mobile for it to properly ddr3 your PCI-E...
Either that or it's just overheating...
 

dfcrackhead

New member
Apr 14, 2009
1,402
0
0
armageddon74400 said:
It's because your super computating computator isn't geforcing enough RAMs into the hard spec fanning raid mobile for it to properly ddr3 your PCI-E...
Either that or it's just overheating...
I feel extremely nerdy because that made me lol and then I reread it thinking about how wrong all of that terminology was...
 

Hawgh

New member
Dec 24, 2007
910
0
0
I concur with the overheating diagnostic. Move it somewhere cooler, get better cooling or play at night. Would probably also help to clean it, but if you can't get into the case, that'll be a toughie.
 

thenumberthirteen

Unlucky for some
Dec 19, 2007
4,794
0
0
RoboGeek said:
May i ask is the computer gradually slowing down before it dies because my graphics card used to overheat because it doesn't have an on-board fan it would gradually slow down and then eventually die
There is no indication, notice, lag, or anything. It works perfectly fine and just shuts off. This is one of the reasons why I wasn't 100% sure it was overheat as in the past my computers have had some symptoms before crashing. The Graphics card does have a fan on it, but that doesn't rule out the possibility that it may have something wrong with it (I don't thinks so as it crashes before Windows even starts sometimes). Anyway it's getting fixed, but that means no PC for at least a week :( Thank god for my iPhone.
 

Cid Silverwing

Paladin of The Light
Jul 27, 2008
3,134
0
0
Everyone keeps saying it's an overheat issue, but I'mma throw my 2 cents into this.

Try running the comp with the one side panel taken off so the ventilation is completely free. Also try cleaning out the fans as much as you can.
 

Virus0015

New member
Dec 1, 2009
186
0
0
Sounds quite like overheating, although the only experience I have had with it was on an old factory-built machine where the GPU overheated. I experienced artifacts, to a point where all the textures in my games went black.

Seeing as graphics cards usually give warning signs before they conk out completely I would place the fault in either the PSU or CPU. The PSU could not be providing enough power (very unlikely if it was pre-built and no components were modified), or it may be overheating.

The best you can do to diagnose the problem on your own is:

-Clean dust out of case to aid airflow.
-Run with as many case panels off as possible.
-Swap out components if you have a willing friend, or use some old ones that are still compatible with the motherboard (one at a time to identify the fault).
-Cry.

If these suggestions don't help, give it to the seller.
 

RoboGeek

New member
Apr 3, 2010
128
0
0
thenumberthirteen said:
RoboGeek said:
May i ask is the computer gradually slowing down before it dies because my graphics card used to overheat because it doesn't have an on-board fan it would gradually slow down and then eventually die
There is no indication, notice, lag, or anything. It works perfectly fine and just shuts off. This is one of the reasons why I wasn't 100% sure it was overheat as in the past my computers have had some symptoms before crashing. The Graphics card does have a fan on it, but that doesn't rule out the possibility that it may have something wrong with it (I don't thinks so as it crashes before Windows even starts sometimes). Anyway it's getting fixed, but that means no PC for at least a week :( Thank god for my iPhone.
in that case it doesn't sound like the graphics card is overheating i would suggest taking it back to where you got it as if you go playing around inside they could claim that you broke the computer yourself and not fix it or charge you
 

cuddly_tomato

New member
Nov 12, 2008
3,404
0
0
See if there is any dust or dirt build up on any of the heat sinks...


...also, if you know what you are doing get some thermal paste, take the heat sink off the processor, clean up both the heat sink area and the processor head so they sparkle, then put the thermal past on and then the sink back on.
 

DSK-

New member
May 13, 2010
2,431
0
0
I would also like to suggest it could be a PSU problem, but you haven't mentioned whether or not your computer powers off when it is idle, so that might not be the problem.
 

Arcticflame

New member
Nov 7, 2006
1,063
0
0
A few suggestions.

1. Open up your case and check if your CPU fan is firmly on place, if it can lift up even slightly then it's not on properly, and CPU's overheat ridiculously quickly, and shut down with no warning. You might have to re-seat your fan.

2. Check if any capacitors on your mobo are bulging or look funny.

3. Check if there are any loose bits of metal, or perhaps a screw resting somewhere on your mobo, it can short and shut down your PC.

4. make sure your cat isn't being a bastard.
 

CultistRat

New member
Sep 18, 2009
168
0
0
Your power supply is possibly shot, I had a similar problem; my computer would take a long time to start up, and wouldn't turn on at all after a while.
 

The Lost Big Boss

New member
Sep 3, 2008
728
0
0
Just a couple of weeks ago my computer did the same thing. It turned out to be my Power Supply. So if you can go and find an old computer with a PSU and switch it out and test the results.
 

Abedeus

New member
Sep 14, 2008
7,412
0
0
Overheating, PSU problem or somehow the "shut down temperature" is a lot lower than it should be. Check in BIOS if you have that option enabled. Temperature at which most CPUs turn off is about 65-75'C, while for GPUs it's more in the area of 90-100'C.

However, my money is on temperature.
 

cerebus23

New member
May 16, 2010
1,275
0
0
overheating or power supply are the top 2 most likely causes.

get a temp program like others have said, monitor the temps of the cpu, that would be the mostly likley cause of a shutdown. take the side panel off the case, make sure the case heatsinks fans are all functioning, maybe stick a desk or floor fan in front of the open side blowing air in see see if the temps go down that way which they should to some degree try playing games etc whatever is causing it to shut down, if it works fine like that then its is overheating.

if it still fails even though things are running cooler and there is no difference, then your mostly likely looking at a cheap psu that is underpowered for the system or is failing itself.

If it is under warranty then take it back to the store and yell loudly, about their crappy workmanship and dont muck with it too much. they can work out the problem and fix it, since that is their job.

another side note if they did not put the thermal paste properly, meaning they not put enough on there and never got it spread out and did not get air bubbles etc out of it, on the cpu when they attached the hs that can cause massive differences in cpu temps, or they did not clean the cpu hs faces off well enough and there is dirt or dust in there that can cause massive thermal problems also. Or they went really cheap and just used a stock intel hs/fan with that gawd awful therma tape junk.

With a little know how, you can get some artic silver thermal compound, some acetone, some coffee filters, and a can of compressed air and apply your own thermal compound after cleaning and dusting out all the surfaces thoroughly. This works for gpus, cpus and anything that has a heatsink with a surface to surface meeting that is meant to draw out heat. And is almost always much more efficient than anything companies use when you get video cards and prebuilt pcs in general and can lower temperatures to a noticable degree. all these items can be had for 20 to 30 dollars also so it makes it a cheap "upgrade" if you will.

Its not terribly complex you can probably find how tos on net for just about any video card or cpu. But if its under warranty it can void it if you do any kind of "modifications" like this.

Take it farther and you can get aftermarket hs/fans that will move more air, good cpu hs/fan can run you anywhere form 20 to 50 dollars, case fans anywhere form around 3 to 15 dollars or so depending on if you want efficient and super quiet or you just want something that will move a ton of air and dont mind the noise. case fans are pretty easy to swap out few screws find the connector swap it done.

cpu hs/fan naturally is more involved and be a royal pita if your got a tight case, or the mb tray does not have a opening in the back for hs/fans since some cpu ones have back plates that attach to the back of the motherboard. SO in some cases you need to take out the motherboard entirely to put one on and that can involve taking apart the entire pc most likely.

probably more information tan you needed but just take it back maybe some of this will be useful down the road, have the people sold it to you fix it give you a new one that works.
 

Wicky_42

New member
Sep 15, 2008
2,468
0
0
My mum's started to do that. Figured it was a connection somewhere, quite likely to be the power supply, though we never did bother getting it working again...
 

DarthFennec

New member
May 27, 2010
1,154
0
0
Might be overheating, might be overdraw on your power. Might also be something wrong with your motherboard, but probably not, or let's hope it's not. Motherboards are expensive shit.
 

Laughing Man

New member
Oct 10, 2008
1,715
0
0
It isn't overheating, the big give away was when he said it crashes whilst loading Windows, no computer on earth should be shutting down due to overheating issues after only 30 seconds of activity. The nature of the shut downs is so random that it is not overheating. Over heating is a shockingly easy thing to diagnose.

Ask this question, does the computer shut down or BSOD whilst doing intensive task but works normally all other times? If the answer is yes, then you can think overheating, if it is anything else then it is something else and in this case it is something else.

The most likely causes are

a). Piss poor PSU causing Vcore overvolt which is triggering the motherboards protective CPU overvolt shut down protocol.
b). Fault RAM, faulty RAM is hard to spot as it can manifest at any time whilst you are doing anything, it is however very easy to test for.

The reason why you are not getting an BSOD is most likely because Windows isn't set to display the BSOD, you get the error the BSOD flashes for a split second and Windows launches in to an instant attempt to reboot. This feature can be turned off, google how because I can't be bothered listing every version of Windows and the process you need to go through to disable the auto reboot on each one.

Anyway resolutions

If it's a). you need to replace the PSU with one that is not a generic off the shelf POS, as most pre built systems come with. Testing for it is next to impossible without the correct software and even then the only software I have seen that does a half decent job and saves result is the stuff bundled with Asus motherboards.

So you're best start is to test for b). Download a program called memtest (google it) follow the instructions and it will test your ram for you. If you can't be bothered with that you can do the hard and fast rule of pulling all but one stick of ram and seeing if the computer will boot with that and test them in turn until the computer dies.

Some additional notes

If overheating is your issue then opening up the computer and cleaning it is a massive waste of time and will actually increase the risk of you killing your components by doing something stupid like blowing on them or worst still using a vacum cleaner. If dust is causing overheating then your computer has a fundamental cooling problem that needs to be addressed well before you think about wasting your time with cans of compressed air. I've seen computers run with literally a vacum bag worth of dust in them without ANY significant overheating issues.

Overheating is piss easy to check and requires NO software to diagnose. Firstly CPU overheating is the only real cause of sudden shut down and even then most CPUs have self throtting safety so sluggishness is often seen well before a sudden shut down. Almost all motherboards from the last 2 years have their own on board temp monitoring system built in to them which can be accessed through the BIOS. If your PC shuts down suddenly, access the BIOS if the temp says it's anything over 55 then you've got an overheating CPU.

GPU overheating shouldn't cause a sudden shut down but it will cause

a). A massive increase in noise as ALL fan cooled GPUs that are powerful enough to suffer from overheating issues have fans that are close to that of a jumbo jet on take off and are temp throtted and will increase to 100% fan duty cycle as the temp rises. If you don't notice this then you're more than likely too stupid to work the software needed to monitor GPU temp anyway.

b). An overheating card will cause noticeable graphical errors to appear on the screen, of course a faulty card will cause the same but a little consideration of weather the card is also doing a). will tell you if it is a faulty card or overheating.