PC shuts down, won't turn on - bad PSU?

[Colonel Travis]

Built a new desktop system about two months ago, everything's been working fine. Today it shut down immediately and will not turn back on. I left it unplugged for about 20-30 minutes, thought maybe it had overheated. When I pushed the power button the fan LED flashed and the fans spin for about one second, then everything quit. Tried again later. Same thing. Then I unplugged everything inside, took out the CMOS battery, put it back in, reconnected the PSU to the motherboard (but no components) and tried to turn it on. Again, same thing.

It had done this a couple times over the weekend - I mean a dead cold shut down, not the normal way - but it always turned back on. Not any more.

Bad PSU? It's a 700W OCZ ModXStream Pro. Been looking at the OCZ site for info., too. Looks like mine is dead, but wondering if it could be something else?

 

[Mellon Head]

It could be something in the power circuitry on the motherboard. OCZ supplies are usually pretty reliable. I've been using them for years, but it's not unheard of to get a bad piece of any electronic equipment.

Try swapping out the supply and see if that helps. If it doesn't boot, start looking at the board.

 

[Colonel Travis]

Unfortunately, I don't have a spare PSU. I had thought about buying a new one, trying it out, then taking it back.

 

[Colonel Travis]

OCZ's site has a jumpstart test, and I hooked the PSU to one hard drive and nothing else, and the hard drive fired up and stayed on. So it sounds like a component problem, not a bad PSU. Two things I did not mention earlier:

Thing 1 - when my PC shut down it was only during playing Modern Warfare 2. At the time I thought it was problem with the game, because it has had issues.

Thing 2 - One time playing MW2 I noticed a pixel static, artifacts, whatever on the screen. They only appeared one time. I rebooted, played the game and they were gone.

Bad video card?

 

[HerrKaLeun]

it still can be the PSU since the HDD doesn't use all voltages (only 5 V, I believe). However, also could be the motherboard.

shutting down during playing could be several things:
- insufficient cooling
- too weak PSU (doubt that in your case since you have 700 W)

the artifact could be the same as above plus a simple software error, especially since it didn't re-appear. you might be looking at several issues that are not related.

 

[Mellon Head]

it still can be the PSU since the HDD doesn't use all voltages (only 5 V, I believe). However, also could be the motherboard.

shutting down during playing could be several things:
- insufficient cooling
- too weak PSU (doubt that in your case since you have 700 W)

the artifact could be the same as above plus a simple software error, especially since it didn't re-appear. you might be looking at several issues that are not related.

Spinning drives still use 12V, SSDs don't. Other than that, I agree.

It could be excessive droop on the 12V rail. That will starve your CPU from its' 12V P4/P5 connector (The 12V connectors by the CPU), and that might make the CPU misbehave. Voltage problems on the motherboard can also cause that.

@Colonel Travis:

Try hooking up the supply again only to the motherboard and GPU if necessary. Pull out the RAM sticks and then try to power up. The PC should beep at you. I can't remember how many times for bad RAM, but it will chirp if the board is more or less okay. If it does, shut it off and put in a RAM stick and repeat. If it doesn't show the BIOS screen or chirp, power off and replace that RAM stick with the other one and try again.

If you can get to the BIOS screen, look at your Hardware Monitor screen and see if the voltages are okay. If they are, power off and plug in your system drive. If you can get Windows to boot again, try adding components until you find the one that's failed. Hopefully this will work.

Please post back with the results.

 

[Colonel Travis]

Well, I tried all the components with the PSU and they all worked except the video card. MicroCenter is about 15 minutes away from me, so I went down there and exchanged the old card for a new one, same kind. Now my system boots and I can look into it in more detail and see if everything is in good order.

One thing I noticed with the new video card - there is no fan on it any more. They replaced it with a huge heatsink. When I opened the box I thought I got the wrong card. Nope - right model. Box was sealed properly, which meant I hadn't gotten someone else's return. Wonder if the manufacturer realized I wasn't the only one with a cooling problem (if that's, in fact, what the problem is/was)?

Appreciate the feedback.

 

[Colonel Travis]

Jiminy freaking Christmas, typing this on my laptop because the desktop just abruptly shut down after about 5 minutes of minimal use. Wasn't killing Russian terrorists in MW2, just working on getting dump files.

 

[HerrKaLeun]

do you have a volt-meter to measure the voltage if it is high enough? i don't think that is the problem since your PSU is large, and in idle you wouldn't have problems assuming it worked a while under load. just to exlcude that easy test.

my tip would be mainboard or RAM. Since your system shuts down quickly, there is no testing that with common software. I'm hope someone comes up with a simpe, way of figuring this out short of you buying all new components to trial-and-error which one is the cause.

i assume you already triple-checked all cables etc. so that none is broken

 

[Mellon Head]

Is your CPU heat sink on good and solid? It's starting to look like thermal shutdown to me. Like it's taking five minutes to get too hot, and shutting down. That could be the CPU, or the MOSFETs on the mother board.

The next time you get it powered up, take the side panel off and touch some of the components on the motherboard (A shock is very unlikely if you don't touch any of the electrical connectors, and even then it's remote, but do it carefully), and see if anything is screamingly hot. The stuff between the CPU and the rear exhaust fan is the most likely culprit, as well as the NorthBridge heat sink. Make sure that all of your fans are turning.

 

[merkat106]

I had a problem similar to yours with my build powering off randomly. Turned out that the mobo was defective (either a short or fuse issue). Took me 6 months to nail it down and exchange the board for another. The defective board was a Gigabyte board with an AM2 socket. Got an Asus board, no problems since...

 

[Colonel Travis]

might be my situation, merkat. Everything in the computer works when the video card is out. I've got 2 PCIe slots and when I plug the video card in either one, fans spin for less than a second and it shuts down.

 

[Mellon Head]

Does the video card require a PSU connection? Some do, and that will definitely send your board into conniptions if it isn't plugged in...

Edit: NVM, I just looked it up. No extra power required. My guess is that it's either a bad card, or a problem with the PCIe bus, and the motherboards power to the bus. That would be my semi educated guess.

 

[Colonel Travis]

Mellon, thanks for your input. Exchanged the motherboard today. MicroCenter let me do so after about 60 days, so big thumbs up for that. Good Lord, so glad I didn't mail order this stuff, I'd be here until the sun swallowed the earth.

I've got everything out of the case, started completely at square one. Only thing hooked up is the PSU, CPU & fan, 1 stick of RAM, 1 case fan. About two hours ago they all worked. I ate some dinner, came back, now nothing works. I'm at my wit's end...

Maybe it is the freaking PSU after all. Unfortunately, don't have a spare to test, or a voltage meters, etc.

 

[Mellon Head]

It is starting to look like the PSU, isn't it? Have you got a buddy that can lend you one for an hour or so? If it works, it will put the issue to bed, once and for all.

I know it's rough, I just went through three weeks of troubleshooting with my new build. I do sympathize.