Computer Randomly Shuts Down

August 20, 2018 at 16:03:05
Specs: Windows 10, Intel i3 3225 3.3GHz/ 8 gigs
Lately my computer has been having issues performing. At random times it shuts down and all the fans stop spinning and the LED's turn off. When I press the power button it will turn on for a second, then turn back off for 1-3 seconds, then start normally from there. Well...I wouldn't call it a normal boot up either because when it does this, most of the time my monitors screen still remains black when this happens and I have to press the reset or power button over and over for it to start booting normally again. If my computer goes into sleep mode it also does the same thing, when I wake it up, it powers on for a second, then turns off, then starts and then the black screen still being there on the monitor and to resolve this I have to press the reset button over and over it start functioning normally again. I recently did a fresh install of Windows 7 with formatted drives then did a fresh install of Windows 10 and during this process, it shut off a handful of times randomly. All drivers have been installed and are the latest ones. I've been getting an error during booting which says "ME FW update failed" so I ran SFC scannow in CMD and updated the BIOS version. Now here's where the problem came in once again, when I was updating the BIOS, I noticed when it finished updating the entire thing would freeze and shut off on its own, then start back up on its own and the screen would go black again and would not go into the BIOS screen unless I pressed reset numerous times. Out of suspicion that the BIOS may not have updated properly, I kept using the BIOS update file and it would freeze up, shut down, and restart into a black screen almost every time until I thought to myself maybe this is enough and manually checked the BIOS version in CMD to make sure it updated to the right version (which it did). I have not seen the "ME FW update failed" error during booting yet. But the problem with my computer randomly shutting down, struggling to restart and giving a black screen still remains. I'm suspecting either the motherboard or the PSU, or if it even is at all hardware related at this point. I have cleaned all the fans and there is no dust and all connectors are firmly in place. What are your guys thoughts?

Hardware:
Intel i3 3225 @ 3.3GHz
Corsair Vengeance low profile 8GB DDR3 Ram (2 4GB sticks)
Seasonic G Series 500 watt power supply
Gigabyte GTX 660 OC edition
Asus P8Z77-V
WD black 500GB hard drive


See More: Computer Randomly Shuts Down

Reply ↓  Report •

#1
August 20, 2018 at 16:29:48
Shutdowns are often, but not always, due to overheating...

Have you checked to see if “all” air vents are clear and free of dust? Pay special attention too to the cpu cooling fan assembly; is it also clear of dust? And is that cooling assembly securely mounted, is the cpu still adhering to the cooling assembly?


Reply ↓  Report •

#2
August 20, 2018 at 16:59:03
Yes, all the air vents are clear and free of dust including the CPU heatsink and fan. It is secure to the CPU and appears to be adhering. Idle while surfing on Google and on YouTube, the temperature appears to be hovering between 52-56 degrees Celsius with the GPU at 32-36 degrees Celsius. I have used the stock cooler for the CPU and this entire build is 5 years and a couple months old. I wonder though, does the thermal paste dry up after a while and need to be reapplied?

message edited by BlueGamer123


Reply ↓  Report •

#3
August 20, 2018 at 18:24:28
"does the thermal paste dry up after a while and need to be reapplied? "

No.

"Idle while surfing on Google and on YouTube, the temperature appears to be hovering between 52-56 degrees Celsius with the GPU at 32-36 degrees Celsius"

Where are you getting those temp readings? They seems backwards. The CPU should be in the 30's at idle & the GPU in the 50s. https://ark.intel.com/products/6569...


Reply ↓  Report •

Related Solutions

#4
August 20, 2018 at 19:23:09
I use CPUID HWMonitor.

Reply ↓  Report •

#5
August 20, 2018 at 20:37:36
It is possible that you applied a little too much thermal compound and it is not transferring the heat properly. Here is the right way in case you want to review it and try again:
http://www.arcticsilver.com/methods...

If by fresh install you actually deleted all partitions, created a new partition and began the installation, then it cannot be in the software. Since it also did it during the BIOS update (or right after it), you have a hardware problem.
First I would rest the heat sink by cleaning off all of the old thermal compound and replacing it as per instructions. Then I would run Memtest86 to test your memory for problems. If the memory proves good after running through all tests at least once, I would test the hard drive. Seatools by Seagate has been a popular one for a while and works on all conventional drives.
If you get crashing during running Memtest86, remove memory and test one stick at a time.
Cleaning the contacts on the memory with a clean soft pencil eraser followed by popping them in and out 5 or 6 times may help memory issues if the problem is only poor contact with the sockets. Test again of course. Replace the memory in matched pairs if this is necessary to make use of dual channel mode and reduce conflicts.

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


Reply ↓  Report •

#6
August 20, 2018 at 22:29:47
If you have an on-board GPU, to eliminate a faulty GPU board, connect your monitor to the that for a few days and remove the GPU board.

Reply ↓  Report •

#7
August 21, 2018 at 00:47:55
"It is possible that you applied a little too much thermal compound and it is not transferring the heat properly. Here is the right way in case you want to review it and try again"

I used the stock CPU cooler which came with the CPU, it came with thermal compound pre-applied on under the heat sink.

I ran Memtest86 for almost 3 passes, none of those passes came up with any error with the memory. Prior to this test however the computer did shut down on its own and did the on-off thing and only booted normally when I hit the reset button over and over again, after this happened I decided to run Memtest86. Now running SeaTools, I ran S.M.A.R.T Check, Short Drive Self Test, Short Generic - without any problems or error being displayed. When I ran the Long Generic test, the entire computer shut down on its own again after a while of running the test and did not complete it. I ran it again and the same thing happened mid test. The second time it occurred however, I was away from the computer for a good while and when I returned I saw that the computer was infact self-shutdown and did not go into sleep mode, this time when I hit the power button it turned on in one go. It makes me think something may be heating up enough to the point on doing a shut down like this and due to me not turning it on right away it had enough time to cool and booted up in one go. But I could be wrong.

"If you have an on-board GPU, to eliminate a faulty GPU board, connect your monitor to the that for a few days and remove the GPU board. "

No on-board GPU.


Reply ↓  Report •

#8
August 21, 2018 at 05:40:30
"No on-board GPU"

Sure there is, HD Graphics 4000 integrated in the CPU. And there are ports on the rear of the board.

Try doing a stress test with OCCT & see if the CPU overheats & system shuts down.
http://www.ocbase.com/


Reply ↓  Report •

#9
August 21, 2018 at 07:23:53
Not getting a blue screen? hmmmmm

Have you tried runing something like intelburntest & msi-furmark at the same time to see if its the powersupply?

i5-6600K[delid]@4.8GHz/4.3GHz@1.4v LLC=6 | 2x4GB Crucial-DDR4-2133CL15@14-14-14-28 1T 2700MHz@1.35v
MSI Armor RX 570 4GB@1340Mhz core@1.110v/1865MHz BiosModded


Reply ↓  Report •

#10
August 21, 2018 at 12:59:53
Tried OCCT, and got the test self aborted with the messege "Core #1 over maximum value! Value Reached: 86, Max Value: 85." The CPU temps shot up within 2-3 minutes during the PSU and CPU test. The computer did not shut down however.

UPDATE:
Re-tried the hard drive scan (Long Generic) for the third time, it went smoothly and came up with no errors.

message edited by BlueGamer123


Reply ↓  Report •

#11
August 21, 2018 at 21:19:32
All you have said and tested points to overheating being the primary issue here. It is possible that one or more of the hold downs for the heat sink have popped out or is not completely seated. If this is the case then removing the heat sink, removing the old thermal compound and replacing it with new thermal compound, making sure that it seats properly. Please see application instructions linked to above for details. Use rubbing alcohol and an old credit card to remove the old compound, do not use any metal to help scrape it off from the heat sink and the CPU. Reseating it without replacing the compound will not help.

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


Reply ↓  Report •

#12
August 24, 2018 at 00:46:49
So picked up a tube of Arctic Silver, cleaned the CPU and heatsink with isopropyl alcohol, applied with the pea method and reassembled. At start up it did the on, off, on thing again but there was no black screen this time and went right into a normal boot. The temperatures did seem significantly lower so I disregarded the on,off,on procedure it did on startup and thought it was fixed...nope, after a couple hours of just surfing on the web it randomly shut down again. At this point I'm suspecting the motherboard.

Reply ↓  Report •

#13
August 24, 2018 at 03:59:49
It might be the motherboard; equally it "might" be the psu... Any chance of borrowing a known good one to test with?

And another possible - albeit it a very.... l-o-n-g shot...

RAM issues.

Remove all RAM sticks; clean the edge connectors with soft pencil type eraser; wiped them clean with a soft lintless cloth (even a kleenex tissue). Re-insert one stick firmly and then remove. Re-insert a second time and then a third time and then leave inserted. Power up and see what happens. Then power down, Insert the second stick - three times as above and power up again etc...

Also any plugin boards do the same cleaning routine.

Do not use anything abrasive for cleaning or wiping the edges. Insert/re-insert three times ensures the sockets are also to cleaned to some degree.

Edge connectors on RAM and boards can often oxidise over time.. The cleaning process is to remove as much of that as is possible.

As I say - it's very.... l-o--n---g shot...

message edited by trvlr


Reply ↓  Report •

#14
August 24, 2018 at 11:50:42
Since Memtest came out clean, I would go with the most likely being the power supply being either under sized or more likely just going bad (very common with cheaper power supplies but possible with any). As stated, try borrowing a known good one or purchasing one where you can return it if it turns out that you do not need it.

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


Reply ↓  Report •

Ask Question