• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

So uh....what could have caused my gaming PC to spontaneously combust?

dLMN8R

Member
My gaming PC is getting kind of old at this point - built it in early 2012 - but it still worked well enough to play most games at low/medium details. And when I built it, I didn't skimp out on part quality. Here's what's in it:


  • EVGA GeForce GTX 680 2GB
  • Intel Core i5 3570k (3.4GHz) - not overclocked
  • ASUS P8Z77-V Pro LGA 1155
  • 4x4GB G.SKILL Ripjaw DDR3 1600 (PC3 12800)
  • Crucial M4 256GB SSD
  • 600W PSU - OCZ ModXStream 80 Plus

A few weeks ago, I was playing a game and my PC suddenly turned off in the middle of it. When I tried to turn the PC back on, everything spun up just fine, but I had no picture coming out of my GPU. I tried doing a hard-power off, unplugging the PSU for a while, changing out cables, nothing helped. Looked inside the case, everything seemed fine.

I gave up at the time, completely unplugged the PC, and left it there for a few weeks.


Today, I plug it back in, turn it on, and BZZZZZTTTTT I see a jolt of electricity inside my case for a half-second and I smell the delightful smell of burnt electronics. I unplugged the PC and everything else from the case immediately, and here we are. I want to let it sit for a while before I try to examine it further.


So, WTF?? How could something like this happen? It might have had some dust in there but certainly not enough to cause something like this. I hadn't touched anything inside the case either, this all just started when my PC rebooted while playing a game a few weeks ago.

Furthermore, I'm sure some parts in the PC are completely fried. Probably the motherboard, maybe the GPU/ram/etc.? How could I go about diagnosing what is and isn't completely busted without annoyingly working through each part? Especially if I don't want some bad part to go and fry other good pieces?

I was planning to upgrade for VR sometime in the next 6 months anyway, but I didn't exactly want it to happen this way...
 
Sounds like a PSU issue. I once had a PSU straight-up just explode when I was playing the first STALKER game. Luckily nothing else was damaged, was able to swap it out within an hour or two.
 

dLMN8R

Member
Seems like a power supply problem that might have taken out everything else with it.

That's what I'm worried happened :-/

It's a great PSU that lasted me a while, but looking back, it's far older than 4 years. I actually bought the PSU nearly 7 years ago - November, 2009!
 

Rellik

Member
Everyone else pretty much said what i was thinking. Sounds like the PSU took some other hardware down with it.

You may get lucky and it's just the PSU. Fingers crossed.
 

Mohasus

Member
My friend had a problem like yours.

He was playing and suddenly the PC turns off. Next time he turns it on, it is a lighting show inside his case. The mobo and the CPU were fried, since there was visible burnt marks on the mobo, it wasn't covered by warranty/RMA. The PSU worked fine during the paper clip test and then later in his new build.

His fan was making strange noises, so it might having been that causing his AMD CPU to overheat (at least the first time it turned off).
 

Duxxy3

Member
Time to start replacing shit. Start with the power supply. If that fixes it, great. But chances are it took out your motherboard. If you lost your CPU as well it might just be time to upgrade.
 
Time to start replacing shit. Start with the power supply. If that fixes it, great. But chances are it took out your motherboard. If you lost your CPU as well it might just be time to upgrade.

If it smells burnt, his mobo is definitely wrecked AT LEAST. A dead PSU would just refuse to start.
 

Eusis

Member
That's what I'm worried happened :-/

It's a great PSU that lasted me a while, but looking back, it's far older than 4 years. I actually bought the PSU nearly 7 years ago - November, 2009!
Yeah, this is it for sure. Just hope there weren't more casualties!
 

WaterAstro

Member
Yeah, your PSU could overload shit like that. Always check the capacitors inside to see if they're leaking. Capacitors can blow up.

Get a Cosair PSU, havent had any issues with them for years.

You have to research beyond just "get a Corsair". Corsair actually has several different manufacturers making their PSUs. There's also the problem of finding out which PSU has Japanese capacitors or not. A lot of research needs to be done for picking out PSUs to ensure getting the best quality possible.
 
Start by replacing the PSU to see if it took anything alongside it.
I suspect that at least the motherboard was fried as well. In that case your only choice is to find another mobo with a 1155 socket to test the CPU,RAM,GPU and the SSD.
I am not sure its worth the trouble though. If you can't get it to work with just a PSU change its probably better to get it to a computer store to see if they can salvage anything to sell.
Keep the case and the SSD if you can and go for a full rebuild.
 

Zenaku

Member
If it smells burnt, his mobo is definitely wrecked AT LEAST. A dead PSU would just refuse to start.

Yeah, I've had a PSU die on me before, it wouldn't even power itself on, let alone send power to the rest of the PC. If things are spinning back up/lighting on then the PSU should still be working.

Doesn't mean it's not the problem, but try and check other things first.
 
Had a somewhat similar experience last week with mine. Was watching a youtube video and the screen goes to "no signal" and I look down to see no power on my tower. No noises or anything, just like it shut down. Nothing denotes power is getting to the PC. No whirring, no lights, nothing. Pretty stressful. I still don't know what's going on with it. Thought it was PSU but i picked a new one up and had it plugged up with the same. No power at all. Don't know where to start here. So any help would be appreciated here as well.

Intel core I7
asrock z77 extreme
Gigabyte radeon 7970

Best of luck to you too OP.
 

dLMN8R

Member
I could just replace the PSU but I'm worried a new PSU could get ruined too in case it was actually something else. Seems unlikely, though.

Regardless, I value my time too...I don't want to replace things piecemeal just to find out over a matter of days/weeks what might or might not be busted.


I almost want to replace the entire thing wholesale, since I was planning on doing so anyway to build a VR-capable machine. So the challenge would be in finding out what I need to recycle vs. what I can sell.

The big problem is that if the motherboard is wrecked, how can I find out if the CPU is wrecked too? I'm not going to buy a new LGA 1155 motherboard just to test this CPU....I'd get a new mobo which supports a new CPU chipset. So the old CPU is in limbo pretty much.

Luckily, all my data is backed up (OneDrive + Carbonite), but hopefully my SSD wasn't fried too.
 
600w isn't too small for a 680? Those old cards are power hungry. Maybe your PSU got old and just shit the bed cause it was being pushed


Get a Cosair PSU, havent had any issues with them for years.

Not all corsairs are high quality. They have different manufacturers making different models. Id personally recommend Seasonic
 
The big problem is that if the motherboard is wrecked, how can I find out if the CPU is wrecked too? I'm not going to buy a new LGA 1155 motherboard just to test this CPU....I'd get a new mobo which supports a new CPU chipset. So the old CPU is in limbo pretty much.

As I said go to a computer store and ask if they have a spare 1155 mobo to test your remaining components on.
 
You have to research beyond just "get a Corsair". Corsair actually has several different manufacturers making their PSUs. There's also the problem of finding out which PSU has Japanese capacitors or not. A lot of research needs to be done for picking out PSUs to ensure getting the best quality possible.

Corsairs have come a long way on the low end since the garbage CX series units they were pumping out at the start of the decade and are reasonably safe across the line now. The CXM series is a really high quality CWT unit with Japanese capacitors. But the Seasonics are and Antecs are just as serviceable and even high quality units are quickly becoming commodity items.
 

Yokai

Member
Check for bad capacitors. My graphics card blew a capacitor once and a similar event occured. When most electronics fail, it usually involves a cheap cap bursting.

Of course, don't poke around the inside of your psu though -- just check the major boards and see if you can identify any.
 
Similar thing happened to me a couple of years back. Same issue with things turning off, powering back on just fine, and then the same sound/sight/smell as you described. Was the power supply as everyone else is indicating. Could definitely tell once I pulled it out as I could smell the "electrical" burning smell by breathing in from the exhaust fan area.

All my parts were fine except for the PSU, so I hope you have similar luck.
 
Seems like a power supply problem that might have taken out everything else with it.

Yes. Just shutting down or restarting spontaneously is a sign of PSU issues. CPU/GPU errors usually cause a restart (BSOD) but would allow stuff to show up on screen, hard drive issues cause Windows not starting / error codes while starting computer. Memory issues cause blue screens of death. Or used to, dunno how that looks in Win8/10. Overheating is a restart when you put strain on computer.

Restart and the "aftershock" is a clear power supply failure.

In all my time, I have never seen a power supply cable fail.

Anybody see one fail?

I had a power supply fail. Computer would be in a restart loop. Would simply refuse to go longer than a second or two before starting again.
 

rtcn63

Member
Black/brownout possibly. Get an uninterruptible power supply if you don't have one, they help lower the chances of shit getting mysteriously fried.
 
Nah, people always overestimate how much power they need.

http://www.anandtech.com/show/5699/nvidia-geforce-gtx-680-review/19

Entire PC load is well under 350W. The GTX 680 may be power-hungry by today's standards, but it was pretty efficient when it first came out.

The GTX 680 requires 38amps on the 12V rail. Check your PSU specs and see how many amps it provides. Might not be an issue when your system is new but as your PSU gets older it can cause issues.

I do agree that people over estimate when they buy a 1000w PSU for a single GPU but people also don't do enough research and under estimate. Especially with older PSU's the wattage doesn't mean much if the power is split across rails.


Anyways, I hope you get things up and running again with minimal loss.
 

dcx4610

Member
Sounds like a power supply for sure. If you smell something burning, it's likely a capacitor that blew.

I would buy either a EVGA or Corsair power supply and give that a try. They both have great warranties and will ship you brand new replacements if something goes wrong.

It's unlikely anything else was damaged on the PC. The power supply failed but it did its job and blew a capacitor instead of ruining your components.
 

dLMN8R

Member
Well, I think it's dead :'-(

I replaced the PSU and carefully took out the old one and hooked up the new one. When I turn it on, the LED on my motherboard lights up, but when I hit the power button the GPU and CPU fans just spin for a split second before turning back off again.

Does that give anyone indications of what might be the issue? There's no beeping error codes or anything, just a quick spin of the fans and that's it.
 

dLMN8R

Member
One final bump in case anyone has any ideas about what could be causing this:

When I turn it on, the LED on my motherboard lights up, but when I hit the power button the GPU and CPU fans just spin for a split second before turning back off again.

I can try to find a spare 1155 motherboard as suggested earlier in the thread, but still curious if people have other ideas in the meantime.
 

Duxxy3

Member
One final bump in case anyone has any ideas about what could be causing this:



I can try to find a spare 1155 motherboard as suggested earlier in the thread, but still curious if people have other ideas in the meantime.

How do the capacitors on the motherboard look? Any leaking or bulging?
 
Another PC killed by an OCZ PSU.

The only thing worse than OCZ PSUs was OCZ SSDs. Good thing they went bankrupt a few years back and can't kill people's PCs anymore with their junk components. Today's "OCZ" is a brand of Toshiba SSDs.
 

Weevilone

Member
As an aside, when I have something like this that seems to be an electrical failure, I'll shy away from pushing my luck before I swap out the power supply or get it tested. I actually bought a load test device for this at one time because this happened to me with brand new parts.
 

galv

Unconfirmed Member
1. Remove the GPU
2. Plug HDMI into motherboard

If boots: GPU is dead
If not: Motherboard/CPU/RAM is dead

3. Try a new CPU - It's cheaper than buying a board to buy a $15 CPU, like a used Celeron/Pentium. If it works, you're all good. If not, it's only 15$.

If boots: CPU is dead
If not: Motherboard/RAM is dead

4. Try new RAM - Get a 1GB/2GB stick, they're also cheap

If boots: RAM is dead
If not: Motherboard is dead


The best way is to make your way to a Microcenter and pay them the money to find the problem - they have all the parts you'd need.
 

Aureon

Please do not let me serve on a jury. I am actually a crazy person.
Realistically, a power spike fried the PSU.
(It could also be capacitors on the mobo.)

The PSU is 99% dead.
RAM \ GPU \ HDDs are very likely okay.
CPU and motherboard are what's most likely to have been fried, especially the motherboard.
I'd replace the PSU first, and test out everything else, possibly with the help of somebody else's working PC.

Nevermind, update:
Well, I think it's dead :'-(

I replaced the PSU and carefully took out the old one and hooked up the new one. When I turn it on, the LED on my motherboard lights up, but when I hit the power button the GPU and CPU fans just spin for a split second before turning back off again.

Does that give anyone indications of what might be the issue? There's no beeping error codes or anything, just a quick spin of the fans and that's it.

Motherboard is likely fried.
When that happens, it's generally the failsafe on the CPU cooler that turns it off because it doesn't detect a PSU connected to the 4-pin mobo slot. Doublecheck that cable first - it's a tricky cable and a common point of failure.
Also, get the GPU off that thing. There's no need to add another possible point of failure to the mix.
 
Top Bottom