Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

PC turns on and off every few seconds


  • Please log in to reply
23 replies to this topic

#1 BugsBunny527

BugsBunny527

  • Members
  • 13 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:11:25 PM

Posted 30 June 2017 - 08:01 PM

Hey guys, I've got a problem here that's exhausted my wits. Two days ago I'd been using my PC for about 20 minutes when it froze completely, mouse, keyboard, the works. At first I thought it was just another KDE crash, but when I restarted it, it wouldn't boot anymore. Lights would turn on, the fans would start spinning, and about a second later it would all die down. Then, 2-3 seconds later it would start all over again. The only way to stop this was to unplug it or by flipping the switch on the PSU.

Later that day, when I came home, I booted it successfully and the PC worked for about 10-20 minutes, and then it froze again and the whole cycle started anew. Yesterday and today it didn't boot at all. I turned on the PSU and pressed the power button and it immediately went into the cycle.

I'm thinking it might be the PSU (I hope not, that's a lot of cables), I unplugged the PC at least half a dozen times in the last few weeks because we had a series of nasty thunderstorms. I always took care to shut it down completely and let it rest for a few minutes before unplugging, but I didn't flip the PSU switch.

The PC is more or less new, all components were bought and assembled by me about 1.5 years ago, and it isn't used very often. The PSU is of the bequiet! brand, I don't remember the wattage, but I took care to add an extra 50-100% beyond the max expected power draw.

BC AdBot (Login to Remove)

 


#2 The-Toolman

The-Toolman

  • Members
  • 1,414 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:04:25 PM

Posted 30 June 2017 - 09:38 PM

I would remove the side cover and with a flashlight and magnifying glass look at the motherboard capacitors for swollen tops.

There are ways to test power supplies although the best way is by substituting with a known good supply.

 

Remove all cables going to HDD / CD / DVD Drive and leave all power cables connected to motherboard then power up and see what happens.

 

You could very well have a power supply go bad.

Are you using a GOOD SURGE PROTECTOR or battery back up.

 

I never heard of your brand of power supply.


I'm grumpy because I can be not because I'm old.

 

The world is what you make of it, if it doesn't fit, you make alterations.

 

Under certain circumstances, profanity provides a relief denied even to prayer.  (Mark Twain)


#3 jonuk76

jonuk76

  • Members
  • 2,178 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Wales, UK
  • Local time:10:25 PM

Posted 30 June 2017 - 11:08 PM

This is the worst type of problem unfortunately.  I'd think the PSU is the first thing to check, but there could be a number of causes. In itself a system shutting itself off randomly sounds consistent with a failing PSU. The repeated on/off cycling though could occur, for example, because of the motherboard attempting to recover a corrupted BIOS, which would likely indicate a problem with the board itself, or it could even be RAM or CPU issues.

 

If it isn't a quick fix like the PSU, then it could take a fair bit of fault finding to work out exactly what's wrong.  I would probably start off by resetting the CMOS (unplug from power then short the necessary jumpers or remove BIOS battery for a period) and disconnecting everything bar a single memory module, CPU/cooler, and graphics card (if no integrated graphics) and try to get it to POST. If it doesn't work try alternative RAM slots, a different RAM stick, try removing and refitting the CPU (making sure no damaged pins on CPU etc)..

 

BeQuiet is a known brand, their power supplies are mostly made by FSP, with a few made by HEC.  In Europe I believe they have a 3 year warranty, so hopefully you're covered if that is the problem.


7sbvuf-6.png


#4 ranchhand_

ranchhand_

  • Members
  • 1,670 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Midwest
  • Local time:03:25 PM

Posted 01 July 2017 - 08:31 AM

Boot into safe mode and run it for several minutes. If it still exhibits the same problem, your cause is hardware. If suddenly the problem stops, the problem is most likely software. Then try a clean boot and possibly that will narrow it down.

If hardware problem, I suggest an in-case teardown, then keep only essential components installed and start from there. Too bad you don't have a power supply tester, that would help a lot.


Help Requests: If there is no reply after 3 days I remove the thread from my answer list. For further help PM me.


#5 BugsBunny527

BugsBunny527
  • Topic Starter

  • Members
  • 13 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:11:25 PM

Posted 01 July 2017 - 01:19 PM

 

Boot into safe mode and run it for several minutes. If it still exhibits the same problem, your cause is hardware. If suddenly the problem stops, the problem is most likely software. Then try a clean boot and possibly that will narrow it down.
If hardware problem, I suggest an in-case teardown, then keep only essential components installed and start from there. Too bad you don't have a power supply tester, that would help a lot.

 

I can't boot it into safe mode, it dies before the BIOS splash screen appears. I would expect it to be the PSU, but the strange thing is that it booted normally once after this all started, it even made me believe it was just a one-time problem from the heat or humidity or something.

#6 BugsBunny527

BugsBunny527
  • Topic Starter

  • Members
  • 13 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:11:25 PM

Posted 01 July 2017 - 07:25 PM

Damn, looks like it's the GPU. When I took out the GPU's ATX connector, it rebooted once then seemed to boot fine thereafter, except no signal to the screen, even tho I plugged the monitor into the onboard graphics plug. I've completely removed the GPU and the PC is now working fine since over half an hour, I've even watched videos using the onboard GPU.

 

Do you think this could be due to some BIOS setting that got reset due to power downtime? I almost wish it was the PSU, spent $300 on the card and the packaging is probably gone :( The card is MSI GTX 960 4GB and mobo is Asrock Z97 Killer 3.1



#7 MDD1963

MDD1963

  • Members
  • 699 posts
  • OFFLINE
  •  
  • Local time:06:25 AM

Posted 01 July 2017 - 07:37 PM

The GPU might be fine, given a PSU (and mainboard) capable of supplying it with adequate power....

 

But, running a system without the GPU spares the PSU up to 100-150 watts when/if gaming....

 

Not saying it is the PSU, or the GPU....you won't know until you use a knownb good one, or, test your components in a known good rig under similar conditions...


Asus Z270A Prime/7700K/32 GB DDR4-3200/GTX1060


#8 BugsBunny527

BugsBunny527
  • Topic Starter

  • Members
  • 13 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:11:25 PM

Posted 01 July 2017 - 07:48 PM

Good point. Do you think I could eliminate the PSU by direct testing? Someone above mentioned a power supply tester, I've got a multimeter if that's any good.


Edit: FWIW, the lights and fans on the GPU would turn on normally along with everything else while the PC was power cycling.

Edited by BugsBunny527, 01 July 2017 - 07:52 PM.


#9 The-Toolman

The-Toolman

  • Members
  • 1,414 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:04:25 PM

Posted 01 July 2017 - 08:48 PM

The only true way to test a power supply is to place it under its full rated load.

The power supply tester that is mentioned is not going to do that.

 

I'm inclined to agree with Mdd 1963 that your power supply isn't capable of running your graphic card.


I'm grumpy because I can be not because I'm old.

 

The world is what you make of it, if it doesn't fit, you make alterations.

 

Under certain circumstances, profanity provides a relief denied even to prayer.  (Mark Twain)


#10 BugsBunny527

BugsBunny527
  • Topic Starter

  • Members
  • 13 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:11:25 PM

Posted 01 July 2017 - 09:19 PM

Yeah, you may be right. I've checked again, I've got this PSU https://www.bequiet.com/en/powersupply/518 which has 400W, and according to this https://outervision.com/b/bntqpq that seems to be cutting it a little closer than I thought. Could it have gone bad due to the hot weather? I used to run BOINC on it for hours at 100% CPU and GPU just a few months ago with no hiccups of any sort.



#11 MDD1963

MDD1963

  • Members
  • 699 posts
  • OFFLINE
  •  
  • Local time:06:25 AM

Posted 02 July 2017 - 05:51 AM

concur with above; PSU 'testers' are just short of useless....

 

They will identify a dead voltage, one that wont turn on, etc...but are usually incapable of applying any sort of meaningful load...


Asus Z270A Prime/7700K/32 GB DDR4-3200/GTX1060


#12 BugsBunny527

BugsBunny527
  • Topic Starter

  • Members
  • 13 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:11:25 PM

Posted 02 July 2017 - 10:04 PM

I took a healthy PC and swapped the GPUs. I ran glmark2 on both PCs with a 1080p video running in the background for about an hour and everything was fine, except the suspect GPU had crappy scores on both PCs (285 on the healthy one and 817 on the misbehaving one, vs ~6000 for the healthy GPU in the bad PC), but I'm gonna chalk that up to nvidia's fubar power saving feature. Swapped them back, stresstested with glmark2 again, then BOINC+GTA5 @ clean 59fps, no crashes. ¯\_(ツ)_/¯

#13 pctestcard-com

pctestcard-com

  • Members
  • 1 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:05:25 AM

Posted 02 July 2017 - 10:57 PM

I would like to give a couple suggestions here:

1. give the PC a quick discharge first by unplugging the power cord, holding down the power button over 20 seconds. then connect the power cord back.

2. power on the PC, get to the Bios, reset the Bios default.

3. check on the PC power switch and cable and see if that is faulty an causing the poor connection to trigger the improper power ON and OFF command.

4. reseat the power connectors, CPU and also check on the CPU temp, previous freeze could be caused by the higher temp.

5. I wonder if the PC been experience the power shortage or strike before, 1.5 year PC should not have the damaged capacitors unless due to some reasons.

 

Hope these help.

Bill

Tech support manager WPTinc.



#14 BugsBunny527

BugsBunny527
  • Topic Starter

  • Members
  • 13 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:11:25 PM

Posted 03 July 2017 - 07:11 PM

Thanks for the info Bill, just regarding #5 the capacitors are fine, forgot to mention that above.

 

Anyhow, the PC is still working fine. I memtested it for 6 hours, and I'm gonna leave it on overnight, so far no RAM errors (it ran much slower after 1st pass but I read that's normal). Reconnected and reseated most everything by now, except mobo and CPU, because they're both a pain to remove and I'd need new thermal paste. If it was a loose connector I guess it got fixed when I first pulled out the GPU.

 

I'm not worried by CPU & GPU temps, I've checked while idling and during the stress tests and they're OK. I used to have an Intel CPU that idled in the low 80s °C for a long time before I replaced the thermal paste and got rid of the stock cooler, and that one still runs the PC I'm typing on here. I do wonder if it could've affected the PSU. Room temp went up to 30°C last week (around 20-23°C now). I'm not betting on this tho, I have a much older PC with some failing parts in the same room, and it ran perfectly fine all the time.

 

 

(edit: fixed newlines)


Edited by BugsBunny527, 03 July 2017 - 07:27 PM.


#15 buddy215

buddy215

  • Moderator
  • 13,261 posts
  • ONLINE
  •  
  • Gender:Male
  • Location:West Tennessee
  • Local time:04:25 PM

Posted 03 July 2017 - 07:26 PM

You have a Linux distro installed so you likely have a Linux DVD or Flash Drive that you can use as LIVE. Try booting from it and

using it for an hour or so. If the comp doesn't lock up during that time then you can eliminate the RAM and power supply as being the problem.


“Every atom in your body came from a star that exploded and the atoms in your left hand probably came from a different star than your right hand. It really is the most poetic thing I know about physics...you are all stardust.”Lawrence M. Krauss
A 1792 U.S. penny, designed in part by Thomas Jefferson and George Washington, reads “Liberty Parent of Science & Industry.”




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users