Thank you for being a valued part of the CNET community. As of December 1, 2020, the forums are in read-only format. In early 2021, CNET Forums will no longer be available. We are grateful for the participation and advice you have provided to one another over the years.

Thanks,

CNET Support

Question

need help with pc that wont power up

Jan 30, 2013 12:54PM PST

Hi all... I have had an issue with my desktop for the last 7-8 months... Its driving me nuts and costing me money... Ive been building pcs for several years and never came across this issue ever... Heres whats happening:

My pc normally works fine... But for some random reason, (sometimes) when i shut the pc down, it will not power up again... (everytime) i shut the pc down and switch off at the mains it will not power back on...

Now.. The first time this happened i checked everything and rushed out and bought a new psu thinking the one installed had died.. On fitting the new psu the pc would not power on... I went through lengthy trouble shooting and could not find a solution... So i took the desktop to my local repair shop they diagnosed it as a dead motherboard...

The motherboard was only a couple of months old so i took the board back to the shop i bought it from looking for a replacement and explained my issue and diagnosis from my loca technician.. Staff in the shop i bought it from plugged it into a test set up and it powered up... Redfaced i left the shop with the board... Plugged it back into my desktop... And it wouldnt power up... I phoned the shop... They said bring the whole desktop in... I brought it in and said they would call me... They called me ten minutes later saying that the pc powered up no problem after i left the shop... Redfaced again but happy Happy i went home.. Plugged in and the pc did power up lol...

Few days later i finished working on my project... Shut the pc down.. Switched off at the mains... Got up the next day and it wouldnt power up.... Took it back to the shop for diagnosis... They couldnt power it up... They recased it and it powered up... Then they put everything back into my case... And it powered up... They called me to come and collect it.. I went in and the guys handed me my desktop and it should be ok now... I asked them to pkug it in and show me... They plugged it in and it would not power up hahaha.. Crazy... They came to the conclusion that there was maybe a fault in the psu but i explained that i had already replaced it... But they replaced it anyway... And it worked... I took the desktop home and all was fine... But only fine if i do not switch off at the mains.. Was talking to my project partner about it on the phone one night after the pc not powering up for a couple of days... I said to him.. 'i will try it just now' and guess what.. It powered up...

Ive never experienced anything like this before... Please help me out guys lol

heres the specs:

Motherboard: Gigabyte 990XA-UD3
Processor: AMD FX 8120 8-core black edition
RAM: 2 sticks of 4GB Corsair DDR3
Graphics: ATI Radeon HD5450
PSU: Powercool Extreme Mod 850W
OS: Windows 7 Ultimate 64bit

forgot to mention.. I changed the RAM too as i heard there was issues with ripjaw ram and the cpu. Thought it might have been what was causing the issue... Didnt make a difference.. Except for the increase in performance of course... When the pc does power up i have absolutely no issues at all its very rubust and fast... Any help would be great thanks in advance Happy

Discussion is locked

- Collapse -
Answer
also recased it
Jan 30, 2013 1:01PM PST

Ive also recased it twice thinking it was faulty powerswitches

- Collapse -
And how about no case?
Jan 30, 2013 1:34PM PST

When I can't get it to power up I start with no case and about 4 parts. It does not take many parts to power up.
Bob

- Collapse -
not tried that
Jan 30, 2013 11:55PM PST

not tried that... however the system has been recased a few times to no avail...

the psu replaced twice...
the ram replaced...
system recased three times...

all made no difference...


says to me that the motherboard is at fault... didnt want to say in my original posting.. wanted it confirmed by someone else on here...

thing is i have been back and forward to the shop i bought the board from 3 times over the past 6 months or so... they know my face when i walk in... because this has been an ongoing issue without solution despite all the hardware replacements etc... am i in my rights to request a replacement board? bear in mind the timescale.. but also bear in mind my reccurrant visits to their shop and paying diagnosis fees... they should take this into consideration??

your thoughts please....

- Collapse -
That's why we try it with the basic few parts and no case.
Jan 31, 2013 12:10AM PST

There is no reason to consider timescale since all it takes is one mounting stud in the wrong spot, a jumper in the clear position or a documentation error. It wasn't long ago a motherboard suffered a 50% return rate as the printed doc showed the HDD LED on a pair of pins. It actually was a reset pin and 50% of the time the LED would be on in a way to hold the board in reset.
Bob

- Collapse -
been through this
Jan 31, 2013 1:40AM PST

yeah i have been on the look out for those kind of manufacturer problems... not found any evidence online... mounting studs and jumpers have all been thoroughly checked both by me and by the shop...

when the system powers on (has not done for the last 3 days) but when it does... this rules out any misplaced jumper or mounting stud holding back start up...

this ALWAYS happens when i remove mains power for anytime... it SOMETIMES happens when i shut down the pc but leaves mains on...

basically if i remove mains power from the system and reapply it... the motherboard seems to be unable to send the signal to the psu to start the system up when i press the powerswitch.... which i think is a fault on the board...

i could try what you are suggesting... power up outside the case... never done this before because i dont have the means of a powerswitch...

whats the best way to power up outside the case without a powerswitch?

- Collapse -
About the power switch.
Jan 31, 2013 2:02AM PST

I looked at this board and it's a simple push button to turn on. On the bench I set the board on a square of cardboard and connect up the basic parts. No more than are needed for power up. We don't even need a CPU for this round.

I'll use a metal pen tip, jumper or small metal flat screwdriver to complete the power on jumper circuit. I am guessing that new technicians would be looking for a push button switch.
Bob

- Collapse -
Look at...
Jan 31, 2013 2:10AM PST

The power sw. as found on a PC case is a "momentarily one" it needs only a initial contact to work. So, a small screwdriver placed on the pins, short it and it powers-ON. A quick touch of the pins and it should power-ON, Same to power-OFF only keep the screwdriver present until it starts the shutdown.

With all your problems, I don't see where the mtrbd. got replaced. Understand, PCs now power-ON because of a logic power circuit. So, if you pulled the AC cord for 5mins. it may repower-ON again after returning it. Why, because you drain the mtrbd. of minor power in caps, etc..

I would closely inspect all of the mtrbd. with a mag. glass to see better and under light, a possible ding of contacts or runs could be had. But again, why noy just replace mtrbd., or flat out try yet a different brand.

tada -----Willy Happy

- Collapse -
it might.. it might not...
Jan 31, 2013 3:00AM PST

yeah Willy it might power up... it might not... this is the most frustrating part of the issue... i can pull the ac cord and reapply it several times with no joy.. then out of the blue it will power up... the issue is so random that i cant (and the shop cant) find out what is causing it... the shops atual wording was "i dont know what it is your doibg with the desktop thats causing this" to which my reply was "im not doing anything, there is a fault with this board and i want it replaced" their response was that "the board would be sent away to gigabyte to be tested as its under manufacturer warranty... and if it came back two weeks later as a faulty board.. then it would be replaced but if gigabyte found that i have done something to the board that has caused this then it wouldnt be replaced" my response was "this board has been behaving like this since i bought it"....

i dont have another board to test thats compatible with the cpu and ram i have... but with everything else already been replaced it pretty much leaves the motherboard to blame as its the only part of the set up that has not been replaced (apart from the cpu) and the same issue has been recurrant through all testing and hardware replacement

- Collapse -
out of case test
Jan 31, 2013 3:03AM PST

thanks for all your replies i will test the board outside of the case later tonight and post back results... thanks again Happy

- Collapse -
Ahh, the story is clearer now.
Jan 31, 2013 3:05AM PST

I was under the impression you built this machine. I will share that we had an old machine that would not reliably power up at the office. We found that it would if we pulled the AC and stuck it back, most of the time.

It was a mystery why it did that and since it worked we carried on. A few years passed and so did the machine. A new motherboard and it's working fine. This does not mean much but it's a common failure.

Sorry that I thought this was a machine you built.
Bob

- Collapse -
i did
Jan 31, 2013 3:17AM PST

yeah i did build it mate... the cpu, ram, motherboard and psu were all bought from the same shop...

- Collapse -
Then the shop is, well.
Jan 31, 2013 3:23AM PST

Not on the hook to make it all work unless they build it and deliver it. I know folk that think otherwise but so far if someone selects the parts then they build it, there is no warranty or even implied warranty.

It's been quite a long discussion and we've yet to cover the basic no case test.

It sounds like the bad motherboard issue.
Bob