Dead PC

Associate
Joined
29 Dec 2012
Posts
109
Location
England
Hi All

Not sure if this is the correct location, but the PC bought from here died last week and looking for suggestions

1 x OcUK GeForce Titan X 12 TB
1 x Acer Predator XB270HU 27" G-Sync Monitor
1 x Intel 5820K 3.30GHz (Haswell-E) Socket LGA2011-V3 Proces
1x Asus X99-A Intel X99 (Socket 2011) DDR4 ATX Motherboard
1 x Kingston Predator 16GB (4x4GB) DDR4 PC4-21300C13 2666MHz Quad Channel Kit -
1 x Samsung 500GB 850 EVO SSD 2.5" SATA 6Gbps 32 Layer 3D V-NAND Solid State
1 x EVGA SuperNova G2 1000W '80 Plus Gold' Modular Power Supply
1 x Corsair Hydro H100i GTX 240mm High Performance Liquid CPU Cooler
1 x BeQuiet Silent Base 800

Extras:
1x 256 GB samsung evo,
Seagate 8TB BarraCuda 5400RPM 256MB Cache Internal Hard Drive (ST8000DM004)
AIO GPU cooling loop from EVGA

built in 2015, recently tried to turn on the pc and the cpu, psu and gpu fans span for a few seconds and nothing happened.
removed all external cables and onto the desk where I could see 00 on the led motherboard area. Seems CMOS was the issue, so removed the CMOS battery replaced the jumpers to 2-3 for a few mins. Jumpers back to 1-2 and a new battery inserted.
Now I get a red PWRSW and RSTW green on, no movement of any fans.
Removed ram, gpu no effect. then removed and replaced the cables from the PSU, also no effect

Drives removed and replaced with one from another pc that failed to work.

Writing this on a passive linux box.

Anyone got any suggestions as to the likely cause of failure and how to fix.
Could be a dead motherboard, psu or cpu ?
replaced the power lead from a working one and no issues there

Took a quick look on motherboard pages and the X99 does not seem to be sold any more which means that if I got someone local to take a look, they are unlikely to be able to fix with the same,

Due to the position of the CPU cooling fan, the ram sticks are not in optimal location, but nothing that could be done at the time.
So are there any checks that I can do this end with a multimeter that is somewhere in the house and if as I suspect the pc is dead, can I re-use the ram at least ?

And would it simply be a case of a new motherboard, CPU plus thermal paste ?

Not into overclocking, but still into gaming :)

edit: would be interested in an M2 boot drive, but not sure if any way to get the licence details off the old drive other than plugging it into this pc ?
 
Last edited:
unable to boot the pc to even get to the bios, if I could get to the bios that would be nice, but when the pc has power, all I get is
red PWRSW and RSTW green

Went to the evga support forums and followed out the checks suggested on https://www.evga.com/support/faq/FAQdetails.aspx?faqid=59582 which suggests that the PSU is not the issue as when the paperclip on the motherboard is in the holes, the fan runs.
So the problem appears to lie with the motherboard or the cpu

when the power button is pressed there are no beeps, the mouse and case lights such as around the power button are not illuminated either
 
Last edited:
yes, i did look into it:

How do I recover from dual BIOS?

  • 6. Turn on the motherboard by pressing the case power on button (this is as far as i got). Once the board starts this time you should see the Gigabyte splash screen, or POST page, then the Auto-Recovery from Dual BIOS will kick in. You will see a checksum error, and then recovery from BACKUP BIOS will begin.

    no splash screen, no sounds that the drive or cpu fans usually make when starting up but will try monday
 
if this in an OcUk pre-built you'd be better off moving this to the customer service forum so they can investigate directly, anything you do yourself has a chance to invalidate any warranty.
 
Suspect too late for that, recall it was a 3 year warranty and the local guy has confirmed that either the mobo or the cpu is dead, he was going to hand it back to me, but suggested he might like to price up to get a new mobo and cpu.
found out the issue with the dual bios, the board got changed to an asus x99 which only has the single bios, no idea why it got swapped and the copy of the pdf invoice is on the dead pc and on the external drive that does not work with linux

my building skills have not been used for several years and would prefer someone who knows what they are doing and can fix it if it does not work when they build it
 
stock cooler ? it came fitted with water cooler
the power supply seems to be ok based on the paperclip test, will have to see if my "spare" psu can power the board, but probably not
there is no on board graphics and the screen is hdmi / 4k
 
thanks for the comments, checking the psu out in more detail this afternoon, got the old gaming pc working, no sound and no login for here, bookmarks and mobile bookmarks not the same

got a multimeter and so need to get that sorted and check the power rails, will be reading the evga site in more detail, would sooner take the time to check the stuff correctly than order stuff that I thought was broken only to find it is ok and it is something else
 
If the paperclip test works, I would assume there would be enough stability/power output at least for it to POST.
Especially if you have disconnected everything except for the motherboard/CPU.

To my knowledge, the only way to distinguish a motherboard failure from a CPU failure, is to put the CPU in a known working motherboard.

I had a gigabyte board (Gigabyte Z97P-D3) inexplicably die a couple of years ago. Was ok powering off but would never turn back on again. The PSU/CPU worked fine on another (ASRock) Z97.
 
Take the H100i cooler's fan and pump headers off the motherboard, and plug in a normal fan into the main CPU fan header. If it starts up, turn the system off before a minute is over. Don't worry it won't be able to heat up dangerously in under a minute even with the pump and radiator fans not working.
 
edited the original post for motherboard, unable to find the receipt, but have the old gaming pc back which is compatible with the backup drive.
being a stock ocuk built pc there is no spare bits of pc's and no spare coolers, the cpu coolers attached to the other working pc's are for different sized cpu's and if I start removing working bits, I mght end up with non working bits

so my problem is being able to diagnose what is wrong with the psu or motherboard / cpu.

Took the psu out again and got the mobo cable attached where I could work on it, with a multimeter, I got 10 volts from pin 16 (PS-on), 25 Volts from pin 9 (+5vSB) and the paperclip trick still worked

made an account on the evga support site, so will have a look at that and may well take off the cpu cooler on sunday


https://www.tonymacx86.com/attachments/atx24-pinout-jpg.197622/
 
had an interesting afternoon, did the sensible step of changing the psu.
Seems my old gaming pc has a 700 watt psu and used the graphics card.

the cpu fans were not removed from the motherboard,. the power supply was the only change

https://thumbsnap.com/s/8bbN48Ss.jpg shows that I can boot and the psu seems to be the issue, not the mobo/cpu that got charged £30 for diagnostics

the cpu reseating would need more thermal paste ideally, will contact staff to see about an rma
 
Back
Top Bottom