Computer turning off, not overheating

SetentaeBolg

Prominent
May 14, 2017
5
0
520
Hi

For the past couple of days, my computer has been turning off without warning and attempting to reboot immediately thereafter. This started happening every few hours or so but now it can barely be on a minute before the restart occurs - sometimes it tries to restart immediately on powerup. My PSU is a Corsair RM1000 (more than I need, I think) that has been owned for around 2.5 years, my motherboard is an Asus Rampage V something something leet extreme turbo deathknob. Originally it was restarting with a message of "Overclocking has failed" (I haven't done any overclocking, just left things on auto). Recently it's reporting a detected power surge.

I suspect the PSU is at fault. Does anyone have any other ideas? I know the PSU is still in warranty.

The only change in behaviour I've done recently, a few months ago I got into the habit of leaving the PC on all the time - probably not a wise move, in retrospect...
 
Solution
I apologise if I came across defensively, entirely my fault. I have solved the problem! I got a PSU to test things with and while connecting it up, found that my motherboard 12v power connector was loose. I have re-affixed it and it seems to be fine now (after some *hilarious* Windows 10 Automatic Recovery black screens, for some reason the boot just... started working again). I forgot the first rule, always check the connections! (I didn't actually forget it, but obviously I didn't check very carefully the first time!)

Thank you for everyone who took the time to post suggestions. My next step is to reconnect the old PSU to confirm that it definitely was the loose connection at fault and not the old PSU.

One question which I am...

SetentaeBolg

Prominent
May 14, 2017
5
0
520
Hi, thank you for the swift reply. I've been watching the CPU/motherboard temps in the BIOS monitor during a reboot and all appeared normal (circa 30 C if I remember correctly). Restarts have happened during boot, as well as during normal operation, no load on the system. Apologies for not including my specs, they are below:

Intel 5930K 3.50GHz (Haswell-E) Socket LGA2011-V3 Processor
MSI GeForce GTX 980 Gaming Edition 4096MB GDDR5
G.Skill Ripjaws 4 32GB (4x8GB) DDR4 PC4-19200C15 2400MHz
Asus Rampage V Extreme Intel X99 (Socket 2011) DDR EATX Motherboard
Samsung 750GB SSD 840 EVO SATA 6Gb/s Basic
Corsair RM Series RM 1000 '80+ Gold' 1000W Power Supply
Be Quiet! Dark Rock Pro 3 CPU Cooler
 

Paperdoc

Polypheme
Ambassador
Check the CD that came with your mobo for a utility that you can install and run under Windows that will display several system items for you continuously as you work. That probably will let you see the CPU and case (mobo) temperatures, so you can check during normal use. You MAY also find a tool for that in the controls for your graphics card to check its temperature.
 

SetentaeBolg

Prominent
May 14, 2017
5
0
520
Hi, sorry I should have been clearer. It has restarted while I was in the BIOS, looking at the temperatures and everything was ok. I have installed CoreTemp and it reports nothing unusual. I really don't think it's an overheating issue - all fans are working away, the case feels cool, no warm air.
 

Paperdoc

Polypheme
Ambassador
My suggestion was just a way to verify your belief that overheating is not the problem. Sometimes temperatures shown in BIOS Setup right after a reboot are lower than they are during actual operations. That's why I suggested trying to find and use the utility that came with your mobo to do that.

I agree with your first thought - that the source of this is a failing PSU. My suggestion was just to be sure you are right that overheating in NOT the problem, before spending money on a new PSU. By the way, is that PSU still under warranty?
 

SetentaeBolg

Prominent
May 14, 2017
5
0
520
I apologise if I came across defensively, entirely my fault. I have solved the problem! I got a PSU to test things with and while connecting it up, found that my motherboard 12v power connector was loose. I have re-affixed it and it seems to be fine now (after some *hilarious* Windows 10 Automatic Recovery black screens, for some reason the boot just... started working again). I forgot the first rule, always check the connections! (I didn't actually forget it, but obviously I didn't check very carefully the first time!)

Thank you for everyone who took the time to post suggestions. My next step is to reconnect the old PSU to confirm that it definitely was the loose connection at fault and not the old PSU.

One question which I am curious about, with the old PSU, the 3.3V reported in the bios was 3.05V - is that a problem?
 
Solution

Paperdoc

Polypheme
Ambassador
Glad to hear you found the problem and solution.

RE: the voltage reading in BIOS - you need to check carefully what reading that is. There are several voltages shown there, and some of them are AFTER the mobo's own voltage regulators have exercised their control for the specific needs of that mobo and CPU chip. So unless this is really supposed to be what the PSU is supplying BEFORE the mobo does its thing, don't worry.
 

SetentaeBolg

Prominent
May 14, 2017
5
0
520
The specific figure I was looking at shows as 3.3 V (or almost exactly that) with the new power supply and 3.05 with the old one - I'm assuming that they are close enough that no real problems would result? Is that wrong?
 

Paperdoc

Polypheme
Ambassador
That does not seem right to me. I would expect a 3.3V DC supply to be VERY close to that, not low by 0.25 V or 7.5%. That PSU may be under warranty still - 5 year warranty, according to the Corsair website. Check with Corsair Tech support for that voltage reading and warranty status.
 

TRENDING THREADS