1.5 year old gaming pc suddenly fails to POST most of the time

bibitaka

Commendable
Apr 26, 2016
4
0
1,510
Okay, so I have read ALOT of posts on this forum about the same issue, and I believe I have gotten a bit on the way to solving my problem, but now I'm out of ideas.

Problem: When trying to boot up the PC, all fans start spinning, LEDs light up, for a few seconds before it shuts off. Then it tries to boot itself again, over and over, until (very rarely) that it actually manages to POST successfully.

Timeline: I bought the components in august 2014, built it myself, and it has been performing almost flawlessly until now. It happened every now and then that my PC would just freeze, I didn't think much of it, I thought it had something to do with software.
But late in last december or so, I started encountering problems to POST. It would only start on like the 5th boot sometimes. Over time, it grew worse, until two weeks ago when I gave up.
I hear you ask: Did I use thermal paste? Yes, a drop about the size of a pea, like the internet told me.

What I tried: Cleared CMOS, did the paper clip test on the PSU (Corsair RM750), it failed, sent it in for repairs, got a brand new one instead (Corsair RM750x). But get this: I STILL fail to POST most of the time. That makes me wonder, did something other then my PSU also break?

So I started breadboarding. With only the PSU, motherboard, CPU and CPU watercooler connected, I still fail to POST. That rules out the RAM, the case and the GPU, right?
Now, interesting part is: I get this strange sound coming from the watercooler pump unit, mounted right ontop of the CPU. I'll admit, I would describe the sound as something close to eggs in a frying pan. Not a very comforting sound, but I can't exlude the possibility that it's meant to sound like that. Either way, a faulty CPU cooler shouldn't cause an immediate boot failure, right? Using fans, I can see when the system reboots itself over and over, even when I disconnect the strange sounding CPU cooler. That sortof exludes the CPU cooler from causing the boot failures.

I should add that during the entire time I have had this PC, I have never gotten any motherboard beep codes. I can't seem to find any internal speaker on the motherboard, so I assume it doesn't have one, which sucks.

What I did the few times I successfully did POST: Went to BIOS, set it to use optimized defaults. Checked system information from msconfig, checked cpu temp using realtemp. CPU temperature seems ok at 29-34 celcius, CPU load seems okay aswell during idle. If I kept the system on I could probably keep it going for several hours if not a few days. It's the initial power-up process that is the problem I think. So I went into BIOS again, ticked "enable dummy load", which should stop the system from powering down due to consuming too little power, apparently. Still, no game.

Synopsis: With my limited expertise, I am 98% sure that neither my new PSU, my GPU, some sort of case grounding, or my RAM is causing these boot failures. What I DO think is, it's either my motherboard or my CPU. How they sometimes work but most of the time don't, that is a mystery to me. I hope you experts here can help me out.

Obligatory component list:
PSU (old): Corsair RM750, 750w
PSU (brand new): Corsair RM750x, 750w
Motherboard: Gigabyte GA-Z97X-UD3H
CPU: Intel Core i7-4790K
CPU cooler (watercooler): Corsair H100i
GPU: EVGA GeForce GTX 780 3GB
RAM: Crucial DDR3 BallistiX Tactical 16GB
Case: Fractal Design Define R4
Thermal paste: Noctua NT-H1
 

bibitaka

Commendable
Apr 26, 2016
4
0
1,510

With an eraser? Wouldn't that create static electricity? There is another forum topic about it that kindof scares me:
http://www.tomshardware.co.uk/forum/253501-30-cleaning-memory-eraser-chips

Anyway, I carefully cleaned the sticks by blowing air on them. Same for the ram slots. There wasn't much dust on them to begin with though. I can usually tell when there's dust clogging up by looking at the case fans. If there's a thick brown layer of dust, I give the whole case a dust check and clean the filters. Anyway, on the case of the RAM, I'm using the reccomended slots from the motherboard manual (1 & 2). Moving them around didn't have any effect. Interesting note: When I left only one ram stick in (into slot 1), I did manage to POST TWICE in a row! Then it went back to the normal bootup failures. Suffices to say, I'm not convinced that the ram stick caused the successful boots, it could have been something else like cpu temperature (it had been off for a while, and my room is cold).
 

bibitaka

Commendable
Apr 26, 2016
4
0
1,510


Good question, I should have answered that in the OP.
I never fiddled with overclocking anything. I'm using all the parts as they came out of the box. I know, I know. It's an awful waste of that beautiful 4790k ;)

Though that reminds me, I did find some post on this forum about someone with the same problem as me (searched for it but couldnt find it again). He eventually discovered his BIOS had started feeding 1.5v to the CPU, and when he turned it down to 1.3v it started working fine again. I did a check through my BIOS, but I couldn't find anything related to tuning cpu power. Then again, I didn't know what to look for. And I decided on just reverting the BIOS settings to the optimized default anyway.
 

bibitaka

Commendable
Apr 26, 2016
4
0
1,510
Both the ram and the ram slots look very clean now. Part of me still think it's the ram acting up, that might explain why my pc used to freeze when I was using 3D animation software.

But the fact remains, whether or not I leave one, two or zero ram cards plugged in, my pc still reboots after 1-2 seconds after I start it. Then it continues to reboot itself until I shut off the power. If the ram was the problem, wouldn't the pc stay on for long enough to do a "missing ram" beep code or something, even if a motherboard speaker wasn't plugged in?

It's the 1-2 seconds, sometimes 1-5 seconds of booting up that has me confused. It's not at all regular. If the motherboard knew what the specific problem was, wouldn't it shut down at EXACTLY the same time on every bootup? Why the randomness? Not to mention why it sometimes boots up perfectly and pretends nothing is wrong. Does it want me to worship it or something!? ;)

And another mystery plagues me, what are the chances that I had both a PSU failure AND another faulty component, both causing the EXACT same symptoms? My last PSU clearly failed the paper clip test, so I was sure that was the problem. Could my bad PSU have damaged the CPU, RAM or motherboard somehow?

I should add, that on the rare occasions that my pc keeps itself powered on, there are times when my screen just stays black. The screen has been connected to the GPU, but since I'm pretty sure the GPU is fully functioning, I'll leave the screen plugged into the motherboard, and see if that has any effect.

At this rate, I'm considering again contacting the website I bought the parts from. Though I bet they'll be suspicious as to why I'm returning so many parts.

Edit: here is a short video of the problem. I only pressed the case power on button once.
https://youtu.be/hO1m7Bd-DOk