I assembled this system 4.3y ago and have a ton of hours on it.  [email protected]/2400mHz 32G DDR3/GTX Titan at stock speed.  Never had any issues with this build.  Core temps are always under 60C at peak load.

I was flying a couple of days ago when all of a sudden the screen just went blank.  No BSOD, just blackness.  I turned off the power supply and let it sit for a 10 min, turned it back on and then hit the PC's start button and the case fans started for maybe 5 seconds, then off it went again.  So I had a look inside and noticed while the Titan was seated, the right hand side where the little plastic retention tab helps lock it down was not in the locked down position.  So I took the Titan out, blew out its socket and reinserted and secured it, plugging in its two power cord manifolds.   I repowered on, hit the start button, and this time it stayed on (case fans & LEDs) for maybe 20 seconds then shut itself down again before any beeps, no POST screens, in fact nothing on the display monitor.  Next I blew out the whole box, checked for any other accessible parts (2 of 4 dimms I can get to to push down a bit, the other 2 are under a Noctua air cooler.   Finally, I try one more start button press and this time nada/zip.  The power on and reset buttons on the mainboard were illuminated demonstrating power got thru some element of the power supply.  But this time and future tries--no activity whatsoever, only the power on reset LED buttons on the mainboard remain lit when the power supply toggle is turned on.  

Anyway, you know where I'm going:  power supply, versus CPU, versus some mainboard component failure.   So what to look at first, power supply?  Some power gets to the mainboard LEDs that are within the power-on and reset buttons on the mainboard.  Don't have any diagnostic tools but maybe I can take the PS somewhere and have it tested, or buy an analyzer, or perhaps a new power supply for my next build which may be right about now!   I doubt it's the Titan, as I say, never overclocked, but can't tell.

Question 1:  how does start up behave when the CPU is completely dead?  

Question 2:  how does start up behave if a video card is removed and there is no mainboard GPU built in?  I tried starting with Titan removed, and it's dead there as well.

 

There is a graphics chip built into the CPU so if you take out the GPU it should start up and give you a picture.  I was thinking boot order but that also gives you a screen image saying it can't boot up.  It sounds to me like its the motherboard but I think you need to take it in to a shop to run some diagnostics and determine exact cause.

When the problem initially hit, you said there was just blackness (no BSOD). So the screen went black with the system was still running, or did the system shut down ?

17 minutes ago, Captain Z said:

When the problem initially hit, you said there was just blackness (no BSOD). So the screen went black with the system was still running, or did the system shut down ?

The system had shutdown (power off) in midflight, suddenly w/o warning.

I would try first resetting to factory default no OC, take the PM battery out and disconnect the power cord let sit for a few hours put everything back in the bios and try it again.

1 hour ago, angeli662 said:

I would try first resetting to factory default no OC, take the PM battery out and disconnect the power cord let sit for a few hours put everything back in the bios and try it again.

I agree, resetting your BIOS is never a bad idea for issues like this.

I've never heard a CMOS Battery referred to as a PM Battery... new to me!  However, many motherboard have a CMOS reset switch so you don't have to remove the battery.  If you're has one (likely does) then it will either be located on the back of the computer, or (more likely) you can find it on the motherboard.  In both cases, it will be a push button (push and hold for 5 seconds then release) and the button should be lit.

If this doesn't work, then boot your computer and look for an alpha numeric display on your motherboard. During boot you'll see several different codes, and when the computer completes boot OR turns itself off you should then see an error code, which you can look up in your motherboard manual or online.  That should lead you to the issue.

Best wishes!

 

Very good guys I will try this tomorrow and report back.  I don't feel good about it though--dead completely  when the start button is pressed.

I would unplug graphics card and drives then boot up to bios. If it boots up then plug things back one at a time. I had something similar the other day and this did the trick.

And if you remove the battery remember to set the date and time correctly otherwise you will have problems with the OS. 

There is no integrated graphics on an extreme-series CPU like the 3930K, so if you think the GPU is the issue, you'll need to swap in, at least temporarily, another video card.

The mobo may have some subsystem status LEDs or perhaps display a series of 2-digit state codes as it boots...watching those as you power it up could give some idea of where it's failing.

In my own experience, this sort of failure has been power-related...either a bad PSU or a failed component in the mobo's VRM circuitry.  PSUs are one of the most common failure points, so starting there makes some sense.  If replacing the PSU produces the same issues, then I'd go to only 1 DIMM, and unplug all the other components and see if I can get to the BIOS.  If that fails, swap in another DIMM to eliminate a bad memory stick as the problem...then if it's still no-go, it's either the mobo or CPU, in which case I'd seriously consider just punting and starting a new build.

Good luck!

4 hours ago, w6kd said:

There is no integrated graphics on an extreme-series CPU like the 3930K, so if you think the GPU is the issue, you'll need to swap in, at least temporarily, another video card.

The mobo may have some subsystem status LEDs or perhaps display a series of 2-digit state codes as it boots...watching those as you power it up could give some idea of where it's failing.

In my own experience, this sort of failure has been power-related...either a bad PSU or a failed component in the mobo's VRM circuitry.  PSUs are one of the most common failure points, so starting there makes some sense.  If replacing the PSU produces the same issues, then I'd go to only 1 DIMM, and unplug all the other components and see if I can get to the BIOS.  If that fails, swap in another DIMM to eliminate a bad memory stick as the problem...then if it's still no-go, it's either the mobo or CPU, in which case I'd seriously consider just punting and starting a new build.

Good luck!

Thank you Bob.  I do have another GPU I can use out of another older box.  I tried everything else today, meaning removing the battery, clearing RTC jumper and it did spin up and I even got a peak at Q-codes until it shut itself off again maybe 20secs after pressing the start button, but it didn't show any codes that I could look at the table in the manual to see if any clues existed.  Doesn't that imply the CPU still works, to get that far into a boot up?  The Corsair HX850 has a 7y warranty and it's 4.3y/o, but I don't have another PSU to swap in if the graphics card substitution doesn't pan out.  I will have to pay a shop to test it.  Maybe I could just take the PSU in to test it in a PC shop--is that a reliable test approach for a PSU in this scenario?

R&R DIMMS will be a pain to test as two of the 4 are obscured by the Noctua cooler so I'd have to remove that to get to them, then reapply the cooler for each test.  Maybe I can just do this and put one DIMM in and see.  Does that really anything though?  Seems plausible the mobo can't manage what it needs to do w/ all 4 DIMMS installed, even though it can handle any one of them?   I'm thinking w/ the number of hours of use on this platform if the PSU or GPU isn't the problem it's time to rebuild.  The DIMMs run at their default mHz w/ normal voltage.

I don't know what to make of it.  I put in a functioning GTX 280, cleared the RTC jumper again, took the battery out and back in, then hit the start button and this time I got BEEPS--yes!  But then it shut itself down after the last beep.  What I got was one continuous beep followed by four short beeps, and the ASUS manual says this is a 'Hardware Component Failure'.  Sounds terribly generic, but does it get us more or less off the hook as far as the PSU goes?  Again, it's a 7y warrantied Corsair HX850 that was that had ample headroom for this system according to the most conservative PSU calculators.  I can't imagine it's the memory either.  It's been running at its stock frequency w/ no overvolting.  So it sounds more and more like something in the motherboard to me.    What happens when you power up a mobo and there is no CPU at all I wonder?

If you suspect your PSU, rather than taking it to a tech to check, maybe buy a power supply tester (always handy to have around): https://www.amazon.com/HDE-Power-Supply-Tester-PCI-E/dp/B005UZHB6G/ref=sr_1_4?ie=UTF8&qid=1498520127&sr=8-4&keywords=pc+power+supply+tester.

Also worth checking that your CPU cooler is seated properly - sudden, unexpected shutdowns can be caused by an overheating CPU.

Thanks I probably should do buy a PSU tester though I tend to think it's not the PSU, but that is based on the fact the POST beep routine called it a 'Hardware Component Failure', which doesn't sound like a PSU, but who knows, certainly not me!   The CPU is screwed down by virtue of the Noctua DH-14's hold down design.  This being said, the case sits vertically and has for 4.3y.  Maybe all that weight physically stressed some parts?  

I've pretty much given up at this point I think it's time to build a new box.  I will continue to use the Titan which still works it appears, at least there is evidence now it was not the cause of whatever has gone awry now.  So I need mobo, CPU and memory.  I'm intrigued by SkylakeX, but not quite sure.  I will end up going to some kind of closed loop water cooler just to lose the weight and volume of the big air cooler.

I did order the PSU tester thanks again for that suggestion.  The price was certainly right.