Brief overview About a couple weeks ago, my custom-desktop (built 2015) started hard crashing and auto-rebooting. The computer has crashed so many times at this point that I can barely keep track (like 50 times in 2 weeks). When I start it up in the morning, it works fine for a few hours, then it will suddenly crash, reboot, and then wait 5-10 min before crashing again and starts to enter the death-spiral of repeated crashes that become increasingly more frequent. Last time this happened (yesterday), it couldn't even get past the BIOS splash screen before another crash. Seems like a short circuit? I have tried a lot of troubleshooting steps so-far, and will list that out in detail below - but for those TL;DR, I am 98% sure it is hardware related, the question is, how to isolate what piece of hardware is at fault? How to test? motherboard or power supply or something else?

Computer Specs:

PS: EVGA 750W GOLD Power Supply

MB: Asus Z170-A, running American Megatrends BIOS 3802 3/5/2018 (UEFI)

CPU: Intel i5-6600K Skylake CPU @ 3.50GHz, 3504 Mhz


Troubleshooting steps taken thus far:

  1. I dis-enabled auto-reboot (per suggestions online), and then when machine crashed again, it automatically rebooted, which clearly seems to indicate that the problem is not triggered by Windows.
  2. I checked the Windows Event Viewer, and here is what it looked like: Kernel-Power Event log There is no detail available beyond this event ID, because the machine apparently doesn't have a chance to write a dump file prior to crashing, which strongly suggests that it is a hardware problem. The crashes happened several times with no software running, so that confirms that hardware is likely suspect
  3. I opened up machine and thoroughly cleaned it and checked pin connections between power supply and the motherboard. I re-seated these connections, and it seemed to stop the crashes for 5-6 days, but then it started happening again.
  4. I updated BIOS to most current build available
  5. I installed a CPU temperature reading program to see if the i5-6600K is overheating. Here is a normal read-out (cold start) CPU Temp cold start followed by a read-out I was able to capture shortly after a crash. I don't know about you, but 57°C seems awfully hot for a processor. Makes me start to think that the processor is overheating? Which gets to the next thing CPU Temp after crash
  6. I have tried tweaking the BIOS to disenable overclocking but I am thus far unsuccessful - I don't know why the ASUS Z170-A motherboard is so hard to throttle down, the 'EZ menu' doesn't provide an easy option to just shut it off, and in Advanced mode, even with most settings in the 'AI Tweaker' panel disenabled, it still overclocks. Still looking for a good guide out there that can show me how to do this.

Can you provide any insight? I am at my wits end

  • You may also like to mention detailed specs of your computer including the Power Supply model and its rating as well.
    – patkim
    Commented Nov 9, 2018 at 16:38
  • I don't see you've made sure your CPU cooling solution works. Too much thermal paste can actually degrade heat transfer; a little dab'll do ya! Suggest you remove the CPU cooler from the CPU, wipe both surfaces clean, and then apply a thin layer of thermal paste or use thermal tape specifically for CPUs before reseating the CPU fan. Then, monitor the fan in the CPU cooler, perhaps with openhardwaremonitor.org
    – K7AAY
    Commented Nov 9, 2018 at 16:38
  • Did you remove any hardware to see if that helps? See if you can borrow another PSU. Are the fans working? To be honest it seems like you need to take it to a computer shop
    – Dave
    Commented Nov 9, 2018 at 16:39
  • @K7AAY Thanks for the link to that software - using that now. It looks like my GPU is right at 50°C despite not running any graphical apps - that seems pretty high doesn't it? I hear what you saying about re-seating the CPU, but I don't understand why it would thermal paste would suddenly be running hot if it's been fine for 3 years Commented Nov 9, 2018 at 17:07
  • Your manual says 75 is high, and I work with hotter CPUs every day which are stable. Your i5-6600K is designed for it. Whoever assembled the system could have been a member of the 'more is better' school and may have put too much thermal compound on, or not spread it around. Removing and reseating memory, and removing the heatsink and putting down new compound are #1 and #2 in stability fixes in my shop.
    – K7AAY
    Commented Nov 9, 2018 at 17:23

1 Answer 1


What can be overclocked, can be underclocked.

First, as per p. 1-20 of the manual, confirm jumper CPU_OV is Disabled which prevents overclocking.
CPU_OV jumper and settings

Retuning the memory as per page 1-36 can also help, but before you do that, please remove then reseat the RAM.

Page 2-61 shows how to reset the BIOS settings to Optimized Default. Do it.

Pages 2-7 through 2-10 deal with Fan Speed and Overclocking. Set the fan speeds to maximum, and then when the system is stable, start dialing fans back, in small increments, until you get your first hint of misbehavior, then speed up a little and lock in that setting.

Set for Daily Computing in EZ Tuning, not Gaming; in general, anything which boots performance in the EZ Tuning section and other sections you don't want; stable (which you want) is slow.

  • 1
    Hi. That's not what my BIOS looks like - installed motherboard is Asus Z170-A series, not Z170, so there is no "Five Way Optimization" setting Thanks for your comments on CPU power control - will look for those settings Commented Nov 9, 2018 at 16:59
  • Corrected; see above.
    – K7AAY
    Commented Nov 9, 2018 at 17:24
  • 1
    Thank you! I forgot about the motherboard overclock disable switch. I will try that Commented Nov 9, 2018 at 17:29
  • 1
    I opened up the case and moved the physical motherboard pin-cap to disable overclocking, and then set things to "Optimized Default" like you suggested. I also toggled off the 'EZ XMP switch' to disenable overclocking the DIMMs. Rebooted the machine, and would you believe the ASUS splash screen still says Overclocked at 22% (???) I don't know how that's possible Commented Nov 9, 2018 at 18:03
  • 1
    Then it's time for a chat with ASUS support to ask why it shows overclocked when you theoretically disabled those with CPU_OV and EZ XMP: icr-am.asus.com/webchat/…
    – K7AAY
    Commented Nov 9, 2018 at 18:08

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .