Computer won't POST/boot

[angelkiller]

Computers fucking hate me!!

Brand new build. 4 Days old. Everything stock.

Has worked flawlessly up until now. I turned it on and it booted for about 20 seconds and then shut off. I tried to turn it on again and it wouldn't boot. I turned the PSU off and let it sit for ~5 min then tried to power it on again. It boots, but again, only for about 20 seconds.

I have tried resetting the CMOS. Makes no difference. I have tried each stick of RAM individually. No go. What gets me is that nothing has changed. I hadn't touched the thing in about a day and I sit down ready for some BC2 and this happens.


Help please?

 

[JATownes]

Sounds like the PSU went south to me. How about some system specs?

 

[streetfighter 2]

Sorry to hear that.

PSU is bad would my first guess. Could be mobo too.

 

[angelkiller]

Sounds like the PSU went south to me. How about some system specs?

Totally forgot about the specs. I posted too quick.

Gigabyte X58A-UD3R
Ci7 920
6GB DDR3-1600 Corsair Dominator
XFX 5870
60GB OCZ Vertex
1TB Samsung F3
LG Dvd burner
Silverstone ST-1000



And I'm off to bed now because I'm thoroughly exhausted at this point. I'll try anything you guys suggest in the morning. Thanks.

 

[Jakeman97]

Sorry to hear that; especially on a new build. Just happened to me on a year old build with the exact same symptoms you describe. Did all the trouble shooting ....it was the PSU. Hope you can RMA.
Best of luck to you

 

[angelkiller]

Ok so get this.

Today, I give it another shot. The thing won't turn on. I flip the PSU off, wait 3 min, flip it back on. Press the power button and it comes to life. Just like last time.

But then it behaves differently. Instead of turning off after 20 seconds, it stays on. Says the BIOS is corrupt, yada yada. I fix it, load defaults, and it boots into Windows. I'm currently in windows running a Linpack test via OCCT. Been running for 4 minutes now with no issues.

WTF??

If it were a PSU issue, I shouldn't be able to run the CPU at 100%. On top of that, the whole computer is working again. But nothing has changed!! I should note that it only has 2GB of RAM in it because I took the other 4GB out when I was testing it....

Anybody have any ideas what's going on? This isn't following any patterns at all. At least it's working. :/

 

[Fourstaff]

The motherboard might be faulty or something is short circuiting inside. Try running again with all the ram?

 

[angelkiller]

I just put all the RAM in. It's turning on....

And it's booting into Windows, which is longer than 20 seconds. So far so good.


I though about a possible short. But why would it go away? How would it go away? The only thing I've done is swap out RAM sticks.


Edit: And shit, the thing turned off after about 2 min at the Windows desktop. Square 1.

Edit 2: Alright, with all 6GB back in, it's doing the 20 second thing. Lemme go back to 2GB.

Another wierd thing is that after the computer shuts off, I have to flip the power switch on the psu, wait about 2-3min and turn it back on. Otherwise, the computer won't turn on at all. The problem seems to be either the motherboard, RAM or the PSU, or worse, some combination of the three. (which is next to impossible to diagnose.)

Edit 3: With 2GB in, it booted past 20 sec, but turned off right at the Windows desktop.

Edit 4: Now if you'll excuse me, I'm off to look for a good cliff to throw this off of.

 

[angelkiller]

Any more ideas? I just turned it on again with 2GB. It booted fine and stayed at the Windows desktop for 10 minutes and I just restarted it. And it booted into Windows fine again.

I'm going to see what happens if I turn it off. Edit: I shut it down and I was able to power it on via the power button on the case. Usually, I have to turn the PSU off for a few minutes and turn it back on for it to work again.

I'm also gonna try to memtest all 6GB of RAM. Edit 2: All 6GB memtested fine. 1 pass, no errors, 1067MHz, cas 8. I haven't done any tweaking in the BIOS yet so it's at defaults.

My guess is some kind of short that only shorts sometimes.

Edit 3: Turned off about 20 min into BC2. I was doing really good too.

Seriously, any suggestions?

 

[angelkiller]

No bites??

I took the mobo out and I'm running it on a box of Cinnamon Toast Crunch.

No fires yet.

 

[newtekie1]

Psu?

 

[angelkiller]

I guess what I'm asking is if there's anything else I can do to to make sure it's the PSU.

Not to discredit what you said, but I'm not totally convinced it's the PSU and here's why. I don't see how the PSU could be bad if it can pass stress tests and 20min of a game. It would make sense if it crashed right when I started the stress test or game, but right now it completed both tasks. I'm playing BC2 right now.

 

[westom]

Not to discredit what you said, but I'm not totally convinced it's the PSU and here's why. I don't see how the PSU could be bad if it can pass stress tests and 20min of a game.

Routine is for a defective supply to boot a computer. Then (intermittently) cause a system crash. Nothing you have done will give a definitive answer.

How to know - without any doubt - whether the power system is good or bad. Power 'system' because the supply is only one component of that 'system'.

First multitask so that all peripherals are being accessed. For example, the game would execute complex graphics. But also searth the hard drive, while downloading from the internet, while reading a CD-Rom, while powering something via a USB port, while ... only then are you truly stressing the relevant parts. Every component must do something that consumes power.

Locate cables from power supply to motherboard. With a digital multimeter on 20 VDC, measure any one purple, red, orange, yellow, green and gray wires. Post those numbers (to three digits) here to learn about things you did not even know were part of the power 'system'.

Only way (short of $thousands in test equipment) to identify a power supply as good or bad requires those numbers. This is your only useful solution to learn if the power 'system' is good or has a defect.

 

[p_o_s_pc]

a easier way to know if it is the PSU is to switch it out with a spare if you can and see if it works.
with those specs i would say ~600+w psu would be needed

 

[westom]

a easier way to know if it is the PSU is to switch it out with a spare if you can and see if it works.

First, you have assumed the entire power 'system' is the supply. Supply is only one component of the entire power 'system'. Replacing a supply takes maybe five times more work compared to the meter. A meter also says - without doubt - if a failure exists and where that failure is located.

Normal is for a good supply to fail in another computer. Normal is for a defective supply to boot a computer. The OP is seeking a definitive answer. An answer without doubt is only possible with the meter.

Meanwhile, better techs always confirm a replaced supply with the meter. Again, a defective supply may still boot the computer. Only a multimeter would identify that defect long before it causes intermittent computer crashes.

An example of how a meter defined three supplies - without doubt:
http://forums.techguy.org/hardware/941654-custom-built-computer-problems-3.html#post7640024

 

[angelkiller]

The computer hasn't crashed since I took it out of the case. That was about 2 hours ago. This is the longest it has run since I began having this issue. I think there's a connection.

About the PSU. When I was gaming, I checked one of the PCI-E plugs with my (very cheap) multimeter. It said 12.25. Looks good to me. And I know that not everything was being stressed. So my measurement doesn't mean much.

And I have zero extra parts with me right now. So unfortunately I can't swap out anything.

 

[angelkiller]

This makes me want to cry.

After running flawlessly for over two hours, I leave to take a shower and when I come back, the computer is off. A PSU reset and and it boots up, but only for like 2 minutes at the Windows desktop.


I was measuring the 12v rail the whole time and I didn't notice anything strange. 12.25v the whole time until it powered off, and of course the voltage dropped.


?????????????????????????????

My patience is now at 0. What should I do next? I'm gonna do whatever you guys say to. Really

 

[Fourstaff]

If you can, unassemble everything and start again. Make sure you take extra care in the little things like putting the correct amount of thermal paste and installing standoffs.

 

[westom]

I was measuring the 12v rail the whole time and I didn't notice anything strange. 12.25v the whole time until it powered off, and of course the voltage dropped.

You did not measure all those wires. Without numbers, then the few who actually design hardware cannot eliminate confusion.

If 12 volts is causing a crash, then numbers from that yellow wire reported a defect constantly during those entire 2 hours. Posted above is how to obtain useful numbers. Essential numbers are from all six wires when a system is fully multitasking to every device and port.

Everything done previously will not provide useful conclusions. At best, can only result in speculation. Measure those six wires during heavy multitasking. Post those numbers here. Then have a useful reply. Posted quite bluntly is how to stop being confused.

 

[bubs]

I am not a computer tech or anything. I studied through A+ for a bit but i didnt pass any tests. I just thought my experience may help. My computer used to do the same sort of thing. I had upgraded my core2duo to a q6600. I kept the same power supply and everything. But every so often the computer would just power off like yours does. Then it would start loading again. It was totally random when it happened. I could be while dloading, browsing the net, Playing games, watching movies. Well one day my computer totally died, it would boot for about 10 seconds then die. I changed PSU and it worked great again. I use cheapo PSU's as i have brought expensive ones in the past and they can be as faulty as the cheapo ones. But my guess and sadly it is a guess, that your PSU is faulty and will eventually die on you

 

[angelkiller]

So nearly three months later and...

I got the PSU RMA'ed, but get what? I plug it back up, and I'm getting the same issues.

If the computer boots, usually it'll be fine for somewhere between 45 min and 7 hours. It's so random. It's turned off after about 45 min of surfing the internet, and then two days later, I play BC2 for ~7 straight hours. (before randomly turning off) Usually it's around the 2 hour mark, and it doesn't seem to be affected by what I'm doing.

But sometimes, after booting, it'll turn off even before it shows the BIOS screen.

A lot of the times it'll be stubborn and not boot. To get it powered on again, you have to, flip the PSU switch off, unplug it, and wait. Sometimes 2 min, sometimes an hour or so. Plug it back up, flip the switch on and see if it boots. If it doesn't, usually repeating these steps and waiting longer works.

WTF?

Seriously.

@westom: Next time the computer boots, I'll get you those numbers you wanted.

 

[westom]

So nearly three months later and...

I got the PSU RMA'ed, but get what? I plug it back up, and I'm getting the same issues.

The previous post was so bluntly clear. " Power 'system' because the supply is only one component of that 'system'." Then described was how to know what the entire 'system' was doing. You did not do that. You simply assumed something was defective. Then replaced that something. Also called shotgunning.

Posted on 17 Oct is what you do now. A good time to read and pay serious attention to those three posts from "the few who actually design hardware".

Normal is for a perfectly good supply to appear defective in some systems. Normal is for a defective supply to appear good even for months; then cause intermittent failures. To see a defect when it originally exists and before it causes failures means numbers from a meter. When a system is fully loaded and stressed as described.

"Posted quite bluntly is how to stop being confused." You did not do that. Now you are confused. Do what was posted. Provide all voltage numbers to three significant digits. Then learn what you really have before trying to fix anything.

 

[angelkiller]

@ westom

Got a new battery for my multimeter, it was giving me weird results before. However, the computer won't boot long enough to get into windows so I can stress the system. So I know you've said the key to determining if the power system is bad or not is with those numbers. As soon as I can get this thing in Windows, I'm going to get those numbers.

That said, I've noticed some other irregularities. Look at this:

Specifically, look at the Dimm section. Notice how only 4GB is detected. Also notice how the 'Dimm slot' and 'Enabled slot' capacities do not line up. Here's a review of my board, and what it should look like. I have installed the RAM in the white slots on my board, as the manual says.

The above picture was taken about a week or so ago. Today, I get into the bios and I snap this pic:

Notice how all 6GB is recognized. I don't have a picture, but I promise that the BIOS reported 4096 MB before today.

Making things worse, after I see all 6GB recognized, I try to boot into Windows. It boots, and about 3 minutes on the internet, I get a BSOD.

I've Memtested the RAM before, no errors. (not recently though) I've tried each stick indivicually and none will boot the system. (that doesn't prove anything though)

Any thoughts? I'm about to go fucking insane here!!!

 

[95Viper]

I am not really up on the i7s and such, will be soon though(gathering parts).

Are your memory settings(timings, voltage, etc.) in the bios matching your ram's(like 8-8-8-24).
Maybe a bump in voltage on the ram. Just asking... isn't that ram rated at like 1.65V and you are running it at 1.52V?

If the ram is checking out ok, try swapping around and see if it is maybe a bad memory slot or check for foreign objects and dust in the memory slots.

Just for grins and giggles, are your standoffs, screws, and phenolic washers properly done.
No shorts or such from an off alignment and the like?

Like I said, just asking 'cause I ain't up on the memory tweaks and all for the i7\x58 platform... yet.

Oh yeah, that error code 0x0000001A is usually memory associated.

 

[westom]

However, the computer won't boot long enough to get into windows so I can stress the system. So I know you've said the key to determining if the power system is bad or not is with those numbers. As soon as I can get this thing in Windows, I'm going to get those numbers.

Getting it into Windows to confirm a supply is good can come later. Currently it will not even go into Windows. So before it even tries to get to Windows or crashes, what are the voltages. If power is the reason why it does not get into Windows, then power is will appear defective before or after it tries.

Provide numbers for those six wires as described earlier. Especially, what happens on any one red, orange, and yellow wire as the system attempts to boot. Strange voltage numbers may explain why some memory is seen and another not. Either some memory is intermittent or motherboard semiconductors act intermittently.