Custom built PC turns off for reasons I do not know why...

June 21, 2014 at 10:04:02
Specs: Windows 8, Dual Xeon X5680

I have been struggling with a system a good friend built. The first set-up, meaning pre-new mother board, new ram, new graphics card, deteriorated quickly and we, essentially, replaced all things hardware...except...the power supply.

Now, the same issue still exists and what I am deducing is that the only element that carried from the original set-up was the power suppler and the CPUs, although, I do not know how to test this to see if it so...the power supply or even if I can.

I felt this was a good place to go considering I am seeing similar inquiries in other areas and hopefully the experts deem this issue worthy enough to address, provide me insight as to my situation and walk me through in maybe resolving this.

I am looking forward to any guidance that anyone can share as this is disrupting my livelihood workflow and calling upon my good friend has been exhausted.

Thank you kindly for reading through and any direction that can be provided to resolving this issue will be enormously appreciated as it is critical to my financial world.



See More: Custom built PC turns off for reasons I do not know why...

Report •

June 21, 2014 at 12:33:28
Often, but not always, due to overheating. When you fitted the CPU did you remove the thermal paste/grease then refit it with new in accordance with the instructions for the CPU?

Always pop back and let us know the outcome - thanks

Report •

June 21, 2014 at 14:06:10

Thank you for responding.

I am not sure, I am not the one who assembled it, although, I would have to say the person that did reviews hardware and I feel pretty confident that however the CPU needs to be prepared for setup, it was done.

You mention overheating, is there a way to check the temperature?

Thank you again, for responding.

Report •

June 21, 2014 at 14:20:51
Assuming sensors are present you can find the information in BIOS. Alternatively there are small freebie programs that do this, such as this one:

Always pop back and let us know the outcome - thanks

Report •

Related Solutions

June 21, 2014 at 17:33:04
The right way to a new build, rebuild, or diagnosing one that was just put together as a whole unit is to do a bench test and do it step by step:
How to properly apply thermal compound:
There is no other real way to determine what component or assembly is the issue.

You have to be a little bit crazy to keep you from going insane.

Report •

June 22, 2014 at 06:59:58

Thank you for sharing that information, greatly appreciated.

I have the CUPID Hardware Monitor Pro installed, although, I am not sure what I should be looking for though?

Would providing information from the Event Viewer help better assist?

Report •

June 22, 2014 at 07:07:55
It's difficult to offer any help without specs. Does your system have dual Xeon X5680's? Pretty pricey build. Please post the complete specs - CPU, motherboard, RAM, graphics, hard drive, etc. Having an adequate power supply is extremely important, especially considering the amount of money it appears you have invested. A single X5680 is rated at 130W which is about 11A from the +12v. Double it & factor in the rest of the hardware requirements & you'll see you need a damn good power supply.

Start by posting the make/model/wattage of your power supply along with the amperage specs for the +3.3v, +5v, & +12v rails. You'll find that info on the label on the side of the PSU. You'll have to open the case to see it.

Report •

June 22, 2014 at 08:46:38
Understood Riider, that makes perfect sense and thank you for replying...

I pulled the information from CPUID.

CPUs: Dual Intel Xeon X5680
Motherboard: Supermicro X8DT3
RAM: DDR3 24gb
Graphics Card: AMD Radeon HD 79000
Drives: Intel SSD 500gb
ST32000641AS Seagate Technology 2tb
Western Digital WD Black WD1001FALS 1TB
Plextor DVD

Power: Corsair 1200

AC Input 90-264V ~ 15A 50/60Hz
DC Output +3.3V +5V +12V -12V +5Vsb
Max. Current 30A 30A 100.4A 0.8A 3.5A
Max Combined Wattage 180W 1204.8W 9.6W 17.5W
1204 @ 50°C ambient temperature

Report •

June 22, 2014 at 11:32:52
Re #5
CPUID should show temperatures, as per the picture in my link at #3.

Always pop back and let us know the outcome - thanks

Report •

June 22, 2014 at 12:19:56
Screen Capture of CPUID
My apologies Derek, new to this...sorry.

I have the total capture of all hardware components being tracked, here they are.

Report •

June 22, 2014 at 14:38:45
I don't believe any of the temperatures given are high enough to cause restarting.

Does it restart automatically after shut down?

Are there any bleeps?

Does it have to run a fair while before it shuts down and is there anything particular you are doing at the time?

Best check for firmness of all connections, just in case something has worked loose.
You could also clean all edge connectors with a soft pencil eraser and pop the devices in and out a few times to ensure there is no oxide on the sockets themselves. In particular check RAM, HD connections (power and signal), and video card.

Always pop back and let us know the outcome - thanks

Report •

June 22, 2014 at 15:46:15
2nd capture : Computer on load.
Understood, temperatures still below critical.

Thank you kindly for replying Derek.

I put a load on the computer:

REDCINE-X Pro 64-bit and looped a 4k video at 1/2
Opened Photoshop and randomly worked an image
Opened a browser and played a 720hd video

At the same time and the screen capture is the temperature results.

Do you feel at those temperatures an issue could arise where it shuts down do to overheat or are they still some degree away from it being that critical?

Yes, it automatically starts when it shuts down, normal boot...just about 14 seconds.

It seems to happen randomly, when I am working on it and when I am not.

Thank you Derek and I am looking forward to your thoughts on those load temperatures.

Thank you kindly for responding, greatly appreciated.


Report •

June 22, 2014 at 16:28:14
It gets a tad warm but I've rather discounted restarting due to overheating from what you have given, especially as it also happens when you are not running it hard. Does it always take a while before it shuts down?

See if you can disable "restart on failure" - see this:
This might result in a blue screen or other error report when it shuts down and any information given could give a clue about where the trouble lies.

Also you could look in Events (hold Windows key down then tap X - select from popup). Look under Windows logs, particularly System and Application. Only consider errors that coincide with the computer shutting down. There are often many errors showing so don't be too alarmed about what else you see. Mostly they are minor Windows bugs that make no difference to operation.

Always pop back and let us know the outcome - thanks

message edited by Derek

Report •

June 22, 2014 at 17:34:57
Hmm...what is the possibility of ambient room temperature...

This might sound crazy...but, it just dawned on me.

I live on the 12th floor, under a tar roof that radiates heat into my apartment, I also work close to 7 foot wide blinded windows that face an obstructive view where sunlight hits it during most of the day and heat pushes in from there as well as heat rising from lower floors.

If I do not run my air conditioning for at least 5 hours to evenly cool the room where it is the essentially the same temperature through-out my work space, there are hot pockets of air that sorta get trapped under my desk where the back of the system pulls air from...I have been running the air conditioning for many hours, room is evenly cool...if I do not or even run it for a little while before it causes the overall room temperature to gets really, really hot...80 to 90 degrees...where some of those hots spots, like the one under my desk is probably around it if not cooled down from hours of running the air conditioning.

Is it possible that room temperature and those pockets of heat that sit close to the back of my system get pulled in 70/80/90 degrees and heat up the system causing it to shut off?

Is this totally out of line or is there plausibility here?

I've been working on the system for the past 4 hours, no issues...where the room it totally and evenly this a possibility?


Report •

June 22, 2014 at 19:19:07
Along those lines, please list your case and the amount, placement, and orientation (intake/exhaust) of your fan(s). Under high stress conditions, these can matter as too few, too many, or the wrong orientation of fans can have an adverse effect on the proper cooling of components. Even the position of the power supply within the case can effect the proper set up of your fans. With a top mounted power supply, a rear fan of 120mm set to exhaust is often enough and a front intake fan low down in the front can be helpful for cool air to the graphics card and hard drives. More than that is usually detrimental with side fans especially bad for proper air flow. With bottom mounted power supplies, inverting the power supply so that it draws air from inside of the case instead of from the bottom below the case is actually helpful, a Rear fan 120mm to 140mm high up in the case for exhaust, a front fan low down for intake, and an optional top fan 120mm to 140mm for exhaust where the case has the option is useful when additional heat stress is present (high ambient temps, overclocking, or high powered components running full tilt for extended periods). Side fans should be avoided. Secondary front fans (intake) are only needed when running multiple graphics cards with multiple hard drives, and the like.
The air needs to enter low in the front, go back and up past all components and exit high in the rear, removing as much heat as possible. This smooth flow is essential which is why I say to avoid side fans which disrupt this even flow and create turbulent eddies that actually trap heat in areas of the case.

You have to be a little bit crazy to keep you from going insane.

Report •

June 22, 2014 at 19:46:10
Thank you Fingers for taking interest in my issue.

The case is a Lian-Li PC-A76

The power supply is bottom mounted.

The two fans in the front are pulling air in.

The two fans, top of the case, are pulling air out.

The two fans mounted on the CPUs are pulling heat away.

The fan, back of case, is still and not rotating.

The fan in the power supply is pushing heat out.

Does this help at all?

Report •

June 23, 2014 at 08:34:28
Yes, agreed, the room ambient might push the temperatures up further.

"The fan, back of case, is still and not rotating".
That sounds odd - is it plugged in correctly?

Always pop back and let us know the outcome - thanks

Report •

June 23, 2014 at 08:47:57
I honestly believe it is the ambient temperature.

I bought a thermometer and placed it near my work area at it read 75 degrees in early morning.

I am curious to see where it is now...coming on to about 12pm.

I ran a short load test.

Did the same thing, 4k loop, youtube video and PS and the temperatures were higher in a much quicker amount of time just at 75 degrees than the other setting of the room being evenly cooled.

I think working with everyone here and thank you everyone for all of your help thus far, that I am finally closing in on the problem that has eluded two people for so long.

Going to do more tests to see how the system reacts.

And, Derek, I am going to check that high back fan to see what is going on with it.

Good point.

Report •

June 23, 2014 at 09:29:08
OK, maybe it's just a bit borderline on temperatures - thanks for feedback.

Always pop back and let us know the outcome - thanks

Report •

June 23, 2014 at 13:34:50
Hi Derek,

Thanks again and everyone else for contributing...your guidance is priceless.

How do I check that one fan? From what Fingers described, my system has to be perfectly tuned...if that fan is not operating in combination with what you just mentioned, "a bit borderline on temperatures" maybe, just maybe that one fan not operating, to create that perfect stream of cooling air, is the culprit?

Is there anyway to diagnose that one particular fan?

Also, is there a set temperature, perhaps in the BIOS (a amateur speaking here) which is set to a temperature to shut down on critical heat? To get a better sense of when immediate "shut-down" should occur?

Thank you Derek for sticking through this with me as well as Fingers and riider, you are really, really appreciated for this.

Talk soon,


Report •

June 23, 2014 at 14:37:44
Case fans are really inexpensive so I would first check that it is plugged into the motherboard correctly (power down and unplug the system first), if it is, I would try replacing it ($8.00 to $20.00 US typically). If you want, you can also check in your machine's BIOS to see that the fan plugged into that motherboard connector is set to run (it may be tied to the chipset temperature but should have a minimum fan speed of about 30% to 35% and climb to 100% as the temperature rises).
Also note that case fans come in 3 pin and 4 pin varieties so get what best matches your motherboard's connector. Also note that fans are offered in different CFM's (cubic feet per minute of air) as well as sizes so you might look for one which moves more air but try to find one that does not become too loud (rated in DB (Decibels)).

You have to be a little bit crazy to keep you from going insane.

Report •

Ask Question