Issues with PC - turn on then turns off immediately

[jjk083]

Hi everyone,

First time poster here, so bear with me.

I built my PC around 6 years ago and it has been running great. Last night watching youtube the PC just cut out, similar to a power outage, but we had no power outage. It seems like it tried to reboot and the system fans were running flat out, but booted no further. The power and reset buttons on the front of the case were unresponsive, so I had to power down at the back of the PSU. After turning power back on, I press the power button and the PC tries to boot, but turns off immediately, then after ~5 sec it attempts to boot again and same thing over and over. For the brief moment it is on, I get "00" on the ASrock on the MB display.

System components are:
Asrock z97 Extreme 6 MB
Intel i7 4770K
Seasonic Platinum PSU
Corsair Hydro CPU cooler (the 110 I think?)
16GB (2x8) Hyper X Fury DDR3 RAM

I've tried reseating all cables, different fans on the CPU cooler, plugging CPU power into different modular socket on PSU, shuffling RAM around, booting with no drives, taking CMOS battery out (5 min) and replace.

Given that the system was running fine then all of a sudden died, I was thinking perhaps an issue with the CPU cooler? I don't think the temperatures prior to the crash should have been too high as the system fans were all running at a low speed, as per the ASrock fan tuning software. Or am I looking more at a CPU/MB failure?

I don't have another PSU to swap in, but I do have the stock air cooler which I will try next, but was hoping to get some expert guidance and opinions on this issue.

Many thanks in advance!

 

[Caring1]

00 usually indicates a dead Motherboard.

 

[Space Lynx]

00 usually indicates a dead Motherboard.

the brutality of such a short sentence. gives me shivers.

this will lighten things up a bit:

 

[QuietBob]

It seems to me that the PSU may be the culprit. Seasonic units (especially the Platinum rated ones) are considered some of the best around, but things happen. My advice would be to swap the PSU for a similarly specced unit, or even try a cheaper one - only to see if the system POSTs at all.

 

[PaulieG]

00 usually indicates a dead Motherboard.

Most likely. However...

It seems to me that the PSU may be the culprit. Seasonic units (especially the Platinum rated ones) are considered some of the best around, but things happen. My advice would be to swap the PSU for a similarly specced unit, or even try a cheaper one - only to see if the system POSTs at all.

This is a much cheaper and less catastrophic rule out unless a bad power supply also took out the motherboard. Either way, the psu swap will tell you what you need to know.

My advice is to always have a backup psu and a stick of ram lying around.

 

[xtreemchaos]

+1 for the PSU, classic kicked the bucket bud.

 

[Chomiq]

+1 for the PSU, check the status of your warranty, it might still be covered.

 

[Bill_Bright]

00 usually indicates a dead Motherboard.

Ummm, according to the User Manual for the OP's motherboard manual, on page 31, it shows for Code 00,

Please check if the CPU is installed correctly and then clear
CMOS.

Of course, if the motherboard is bad, the CPU may not work properly.

If nothing else, since everything inside the computer case depends on good, clean stable power, swapping in a known good PSU to see what happens is where to start.

 

[jaggerwild]

OO borked Bios, Machine's was working fine, hence PSU don't "just die" unless you were running it 24/7?

 

[Bill_Bright]

hence PSU don't "just die" unless you were running it 24/7?

Ummm, sorry but that's totally inaccurate. PSUs can "just die" for many reasons - even if run a couple hours every other day. And frankly, frequent on and off cycles can be much more abusive on electronics, including PSUs, than keeping them on constantly.

Frequent on/off cycling subjects the electronics to frequent and "from one extreme to the other" temperature swings and the associated expansion and contraction of the matter as they warm up and cool down. Frequent on/off cycles also expose the electronics with the abuse caused by "inrush" current extremes that occur every time a PSU is powered up.

When constantly left on, temperatures stay more constant and there are no extra inrush current events.

Also, PSUs don't have to "just die" to go bad. Depending on their design, they could lose one rail, or one of the required output voltages while the others remain. A component in the regulation circuit could fail causing the output voltages to vary or drift out of the allowed ±5% tolerances. A component in the ripple suppression circuits could fail allowing excessive ripple through. There are many ways a PSU can become faulty besides "just dying".

OO borked Bios

I supposed that is possible but BIOS firmware does not suddenly go bad on its own so not sure how it could be borked without some evidence of another problem - like fried components on the board or inside the PSU and pretty sure the old schnoz would be alerted to that.

 

[Jose Jeswin]

I had a similar incident with a corsair vp 550w psu on my fx 8350 system...took the psu off and connected it to a psu tester and the -5v rail and -12v rail leds did not light up....changed the psu to an antec 650w unit and the system works perfectly now fir the last 3 months..

the psu tester i used..
I also used a diagnostic card to test my motherboard..

just a pic...not my system..

 

[Bill_Bright]

and the -5v rail and -12v rail leds did not light up...

But that is not a problem. The ATX Form Factor Power Supply Design Guide does not require -5 or -12V for ATX compliant PCs. Only +5V, +12, and +3.3VDC.

-5VDC has not been used for a couple decades. And -12V has been optional for almost as long - for when the PSU is used in some other application, that is, when used in other electronics besides PC computers.

And note those type testers cannot show if a PSU is good. And that type you showed is even less informative than the better models. I keep one of these PSU Testers in my tool bag in my truck for house calls. The advantage of this model over yours is that it has a LCD readout of the voltages. With an actual voltage readout, you have a better chance of detecting a "failing" PSU, or one barely within the required ±5% tolerances (at least with the tester’s internal load).

But even that tester is not conclusive for a couple main reasons. To properly and conclusively test a power supply unit (PSU), it must be tested under a variety of realistic "loads" then analyzed for excessive ripple and other anomalies that affect computer stability. Those testers only impose a tiny, typically 10Ω load. And they cannot test for ripple.

So you replacing your PSU didn't fix the problem because the Antec provided -5 and -12V. It fixed the problem because something else was wrong with your old PSU. The required voltages were out of tolerance, there was insufficient ripple suppression, or it failed when put under heavier loads, or something else - bad connector, for example.

 

[Jose Jeswin]

But that is not a problem. The ATX Form Factor Power Supply Design Guide does not require -5 or -12V for ATX compliant PCs. Only +5V, +12, and +3.3VDC.

-5VDC has not been used for a couple decades. And -12V has been optional for almost as long - for when the PSU is used in some other application, that is, when used in other electronics besides PC computers.

And note those type testers cannot show if a PSU is good. And that type you showed is even less informative than the better models. I keep one of these PSU Testers in my tool bag in my truck for house calls. The advantage of this model over yours is that it has a LCD readout of the voltages. With an actual voltage readout, you have a better chance of detecting a "failing" PSU, or one barely within the required ±5% tolerances (at least with the tester’s internal load).

But even that tester is not conclusive for a couple main reasons. To properly and conclusively test a power supply unit (PSU), it must be tested under a variety of realistic "loads" then analyzed for excessive ripple and other anomalies that affect computer stability. Those testers only impose a tiny, typically 10Ω load. And they cannot test for ripple.

So you replacing your PSU didn't fix the problem because the Antec provided -5 and -12V. It fixed the problem because something else was wrong with your old PSU. The required voltages were out of tolerance, there was insufficient ripple suppression, or it failed when put under heavier loads, or something else - bad connector, for example.

true...but the symptoms are indicative of a psu problem..

 

[mouacyk]

Check voltage of CMOS battery and replace. An insufficient voltage may lead to BIOS corruption and other issues.

 

[Bill_Bright]

true...but the symptoms are indicative of a psu problem..

 

[mouacyk]

In the pic, the 24pin wasn't plugged in. Just making sure it's plugged in on your board.

 

[RJARRRPCGP]

That's a PATA connector.

true...but the symptoms are indicative of a psu problem..

Likely PSU cap plague or a bad motherboard MOSFET. (or motherboard VRM)

 

[Athlonite]

In the pic, the 24pin wasn't plugged in. Just making sure it's plugged in on your board.

didn't read the bit where it says not his system then

+1 for another psu try out if still a no go I'd remove the HSF and CPU and check the socket for pin damage if nothing looks damaged then reinstall the CPU carefully and replace thermal paste and HSF clear the CMOS and try powering it up again if it still fails then it's a dead mobo

 

[Jose Jeswin]

this pic..?

thats an ide port...not the 24 pin connector...got some other pics for you..

the 24 pin connector is connected..

i do own the graphics card tho...its an ati radeon HD 4650...

 

[mouacyk]

didn't read the bit where it says not his system then

+1 for another psu try out if still a no go I'd remove the HSF and CPU and check the socket for pin damage if nothing looks damaged then reinstall the CPU carefully and replace thermal paste and HSF clear the CMOS and try powering it up again if it still fails then it's a dead mobo

I was aware, but the old IDE connector is in the position of 24pin on new boards. In any case, the led would never have turned on.

 

[Jose Jeswin]

its an old intel G31 ddr2 mb..