I have a computer running Windows 7 (64-bit), with an Asus P8B75-M LX Plus Motherboard, Intel Pentium G860 @ 3.00GHz CPU.

Here's all the details of what I've done, but the short version is the original question.

This computer started having issues with random shutdowns about a week ago. They would usually happen 30-45 minutes after the computer was turned on, but sometimes it would take up to 6 hours to die. The VGA port also wasn't working and occasionally the Ethernet port will have lights on (flashing yellow left, solid orange right), but not connect to the internet. I was able to remote into the computer and see that windows loaded and programs could run. I checked the device manager and the only thing that showed under the Display Adapters was the LogMeIn driver.

After some trial and error I got the computer to load properly by unplugging the power cable and holding the power button for 30 seconds to try and clear any extra charge on the MB. Also leaving it unplugged overnight. Both of these options have worked, but not 100% of the time, sometimes I have to try it 4 or 5 times to get a proper start.

When I get a "proper start" the computer does have the Intel HD Graphic Driver in the Display Adapters section of the Device Manager and it doesn't have any issues with random shut downs. I can restart the computer as much as I want without any problems as well, but when I completely shut down the computer it will eventually revert to the "bad start"

I have-

  • Replaced the PSU with a confirmed working PSU
  • Tried a new SSD with a fresh install of windows and a cloned version of current SSD/Put the current SSD into another computer (worked fine)
  • Reseated RAM/Tried confirmed working different RAM
  • Applied new thermal paste to CPU/Ran a temp monitor on overnight/multipule times, CPU never got higher than 55°C
  • Removing all peripheral devices
  • Started in Safe Mode
  • Updated Drivers
  • New Sata Cables/Power cords
  • Ran CPU/RAM/Chkdsk/sfc scans/tests

Eventually the "bad start" always comes back at some point when the power is completely shut off or when the power switch is used which leads me to believe there's an electrical issue in the MB itself. So coming back to the original question - Can bad capacitors on MB overload and cause boot sequence issues/random shutdowns?

I have removed the motherboard and checked for blown/warped capacitors or burnt connections of any kind, and I couldn't see anything wrong with any of it.


As gregg wrote, the direct answer to what you asked is "yes". However, it's "yes, but", with a big but (just so your computer isn't offended, I'm not claiming your computer has a big butt). Let me provide some perspective and context.

The type of capacitor that's prone to failure is the little electrolytic cans. They do fail, and when they do it's visible, usually on top because the tops are scored to allow them to fail "safely". A capacitor can be in the process of failing, but not have visible signs, yet. They do age and change internally until they reach a point where they die and are physically obvious.

A capacitor is part of a circuit designed to operate in a certain way; the designer wouldn't spend the money to put them there if they didn't serve an important purpose. Where electrolytic capacitors are used, the circuit typically has a lot of tolerance because the capacitors do age and their characteristics change a bit over time. But at some point, they will be out of tolerance and then dead.

Good quality capacitors should last for many, many years. But there have been problems in the industry periodically where batches of crappy capacitors got used, and failed early in the computers' lives. So computer age isn't a reliable basis for considering or excluding capacitors as a potential cause. But if the computer is very old, the chances of capacitors being an issue is higher.

When capacitors go out of tolerance or fail, it changes how the circuit works. Depending on what specific function the capacitor is part of, that function is likely to not work as designed. So a wide range of symptoms are possible from failing or failed capacitors, including boot failures and shutdowns.

But boot failures and shutdowns can be caused by endless different problems. You've already looked at a number of the possibilities. Blown capacitors are an easy thing to check because they are readily visible. Ones that have just aged out of tolerance are a different story. That's the kind of thing that's the province of refurbishers; they have test equipment and detailed circuit specs to track down board-level failures in a practical way. Very few computer repair shops attempt board-level diagnostics or repairs because they just aren't equipped to do it. They swap boards with a new or refurbished one and send yours off to a refurbisher. So if you don't see visible failures, there isn't much you can do to determine whether capacitors are the problem.

So the answer to your question is yes, it could potentially be bad capacitors on the motherboard. But if they aren't visibly blown, that factoid won't be very helpful. Unfortunately, so many things could be the culprit that a Q&A site can't be a lot of help in solving the problem.


TL;DR Yes to your main question. I've got numerous Dell desktops at a business where you can see visual defects at POST/BIOS, sometimes just black screen, sometimes frozen at POST/BIOS with nothing attached, but monitor & power. One of the capacitors actually popped while I was checking the status of them.

  1. When it shuts off randomly: are we talking a reboot or power-off-stays-off?
  2. The 'good' vs 'bad' start is a bit odd for sure. So the 'bad' start can take up to 6hr till incident happens, whats the longest length of time its worked fine in the 'good' start (logically it should be more than 6hr, but I'm curious if it can go longer)?
  3. Update the BIOS & edit your question to show you tried that & the version you upgraded to
  4. To isolate OS/driver issues possibly causing crash I'm curious if you could let it sit in BIOS or boot to live-linux or diagnostics tool for 6hr+ a couple separate times (to isolate the good/bad start)?
  5. A diagnostics/stress tool COULD help you push the hardware to isolate if that is at fault. Best to boot to one so you can eliminate the OS/drivers as possibly culprit. My preferred tool is UBCD: Ultimate Boot CD & it has tools to stress CPU (Prime95 is a tool people like I think UBCD has), RAM, GPU, etc.

Regarding the display adapter switcharoo: depending on how you remote it it may use a mirror display-adapter-driver so not seeing the Intel one might be totally normal. Internet not working & the power-off issues could be related to OS crashing. For NIC issues you describe I usually start with a TCP/IP reset (KB299357)

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.