Computer shutting down and other oddities

KopeAcetic

Reputable
Mar 31, 2015
16
0
4,510
Hello,

My computer has been randomly restarting for between 2 to 4 weeks now. There are no errors in the event logs other than the kernel-power 41's basically just saying that the computer shut down unexpectedly.

I had originally thought it was my ssd going bad because when I booted back up every once in a while there were a ton of orphans and fixes chkdsk found (it used to run chkdsk every 3rd shutdown or so).

Another oddity is that if I turn it back on immediately after it shuts down, it turns off as it's initially trying to boot up, but if I wait a few minutes it usually boots back up.

There was one time where it kept shutting down a bit after it got into windows (had like 6 to 8 shutdowns in a row) which then it did a system restore to a previous safe point, and it became more stable, but still was restarting...This might have just been luck, or a conflict?

Someone said there were 2 windows 7 updates that are known for issues, so I uninstalled them, and it still was shutting down so I reinstalled them.

I tried to run a memtest but it crashed after almost completing a first pass.

I'm thinking PSU or Mobo?

Possibly an unrelated issue, I bought a new power cord and surge protector to see if this was the issue (i noticed my power cord was being crushed by my desk...just wanted to try it) but it was still shutting down. Since then a couple times my monitor wasn't detecting my GPU, I took my comp apart and cleaned out what dust I could from the heatsinks of the vid card and reset it, including taking out and putting the plugs back in.

Any thoughts?

*edit #1* One thing I forgot to add, when my computer boots up my chrome seems to be forgetting passwords and login information.

*edit #2* So, I noticed my cpu's temp peaking at 70's while opening multiple chrome tabs, I cleaned out the dust from the heatsinks and the temp seems to be more manageable. When I tried to turn the computer back on after that it didn't detect my gpu again, I took it out, and reset it and it detected it again..but when I got to the windows login screen my screen went dark....then came back 1 or 2 minutes later, this is now happening each time I boot up, and some wonky stuff is going on with the graphics of some games. I'm thinking I either destroyed my cpu or my gpu.
 
Solution

What decides when you computer can power off or on? Its power controller. This controller will power off a machine if a bad or maginal hardware condition exists. To say more (to see a defect) means using a meter with requested instructions. Otherwise the fewer who really know this stuff cannot provide assistance.

You have two choices. Find a defect before fixing or changing anything (as recommended). Or just start replacing good parts until the defect stops. Replace any part in any order since most any part (including...

DeaDSouL

Reputable
Apr 1, 2015
10
0
4,510
first think I would think of is power-supply
if it's ok

then make sure the cpu temperature is normal not so high
is the cpu-cooler doing its job ?

good luck
 

westom

Distinguished
BANNED
Mar 30, 2009
931
0
19,160

What decides when you computer can power off or on? Its power controller. This controller will power off a machine if a bad or maginal hardware condition exists. To say more (to see a defect) means using a meter with requested instructions. Otherwise the fewer who really know this stuff cannot provide assistance.

You have two choices. Find a defect before fixing or changing anything (as recommended). Or just start replacing good parts until the defect stops. Replace any part in any order since most any part (including and not limited to GPU and CPU) might cause that problem irregardless of what others suggest. Those are your only two choices.


 
Solution

KopeAcetic

Reputable
Mar 31, 2015
16
0
4,510
Thank you for your answers :).

Unfortunately I was incorrect in my CPU temp, and saw that it was coming in at 70+ degrees C when opening chrome tabs and nothing else open.

I opened the case and saw my cpu heatsink and graphics card heatsinks caked in dust so I blew them clean with condensed air...but did not hold the fans in place and I think my CPU got friend because of that, or the constant overheating.

When I tried to log into a game after that the graphics were super accelerated (the characters were twitching and the ambient effects were crazy fast) and I couldn't even select an option to go into the game (the buttons activated as if I was clicking on them, but nothing happened).

I'm purchasing a new CPU and a GPU (GPU was older anyway and I wanted to upgrade, CPU I'm pretty sure I blew up) and I'm also getting a coolermaster D92 to help with my cpu's heating (better safe than sorry).

Thanks again for your time and your replies.
 

westom

Distinguished
BANNED
Mar 30, 2009
931
0
19,160

If a CPU blew up, then the system BIOS did not even execute let alone boot the Operating System. You have no reason to even suspect CPU failure.

Near zero heat has been confused with hardware destructive heat. Confused because you are ignoring what must exist in every recommendation and conclusion - numbers. Temperatures are ideal at below 90 degrees C so that signal timings do not vary too much; cause a software crash. That is temperature for reliable software execution; not for hardware damage. Lower that temperature and all signal timings go back to normal.

Hardware damage occurs at temperatures exceeding 200 degrees C. You do not have that.

Furthermore, if a CPU is from Intel, then even removing its heatsink does not cause hardware damage. A design that even existed in the 80486 processors so that a CPU cannot destroy itself.

Do not confuse slightly high temperatures that upset signal timing with much higher temperatures necessary for hardware damage.

Previously posted were symptoms explained by something completely different and that you do not understand. You have become fixated only with heat. Then worry about temperatures that are so low as to not cause hardware problems. That only cause hardware damage when hearsay forgets to mention what temperatures are necessary for hardware damage.

See an earlier post for problems consistent with your symptoms. Not one posted symptom even implies CPU failure. That conclusion is a classic example of wild speculation.
 

KopeAcetic

Reputable
Mar 31, 2015
16
0
4,510
I'm sorry I honestly don't understand what you are saying.

I should have posted my specs at the beginning, and I doubt this will even have any difference in the answer, but the CPU is an AMD.

I do agree, that testing and figuring out the problem would be absolutely best but to be honest I have absolutely no idea how to do that at this point, I'm way way over my head here and assuming information is correct.

The reason I was thinking the heat is the issue is that with just opening chrome browsers brought it to above 70 degrees.

I could still play games, and everything was ok but when I played and taxed my system for longer periods of time (sometimes it seemed randomly) it would shut down.

After cleaning the fans in my cpu and gpu my system hasn't shut down once, I haven't been taxing anything but it hasn't shut down. When I try to open a specific game the graphics look hyper accelerated and I can't actually click to start the game, the clicks won't register on that one screen. Because this started right after I cleaned the heatsinks, I assume I damaged either the CPU or the gpu when I blew condensed air into it to clean it off, and I didn't hold the fans in place, so dynamo effect maybe? I'm not thinking my cpu blew up, I just am assuming I damaged it.

Again I'm sorry I honestly don't know what to do to figure out what the issue is, and I don't want to be a complete bother on these forums (I've read things on these forums many times in the past however many years but this is my first time creating an account and posting...I'd rather not anger the community because it seems pretty great).

Do you have any suggestions on how to figure out what the actual issue is? I'm assuming the psu is out now because of how long I've kept the comp on without it resetting at all (unless it's a problem with the psu being overtaxed...but I doubt that's an isssue, it's a 700 watt corsair bronze psu). You also mentioned software, what software could be the issue? I noticed I installed the samsung software magician's new update right before this happened as well, but i didn't change my ssd to IDE (previous updates forced me to switch from AHCI to IDE, but I just let this one go through). I've since uninstalled the software, rebooted my comp and set my SSD to IDE and reinstalled it, but the computer was still resetting at this point after I did this.

I guess specs might be a good idea if you're still interested in helping:

CPU - AMD FX 8350
MOBO - ASUS Sabertooth 990fx r2.0
Vid Card - Nvidia GTX460 (this is changing probably today to a gtx 970)
Audio - ASUS Xonar Essence STX
Ram- G.SKILL Ripjaws X Series 8GB X2
Seagate 2tb HDD
Samsung 840 pro 256 GB SSD
Case - Antec 900

Again thank you for your time.
 

KopeAcetic

Reputable
Mar 31, 2015
16
0
4,510
When I ran 3dmark 11 Demo it ran pretty well. My cpu got up to 74 degrees, my gpu handled the heat and stress well, it lagged and frame rates went down to 9 on a couple of stressful points but it didn't crash.

I then ran Prime 95 again and when my cpu got up to 73 degrees, it froze again.

Does this help at all?

I should probably also say the temp increased by about 1 degree a second. This was monitored by HWM.
 

KopeAcetic

Reputable
Mar 31, 2015
16
0
4,510
HWM_snapshot_4_4_15.jpg


Here is an image of HWM a bit after running 3dMark11.

I was looking at my CPU's Vcore voltages, not sure if they are correct?
 

westom

Distinguished
BANNED
Mar 30, 2009
931
0
19,160


That HWM implies at least one voltage (not a CPU voltage) is defective. However it is using a meter (on motherboard) that is not yet calibrated. So integrity of those numbers is unknown until calibrated with a meter. The meter, request some instructions, and minutes of labor mean three digit numbers can be posted. The next answer then says what is or is not defective - without all that speculation that is currently dominating every post.

Every part you have suspected is only wild speculation. For every assumption, I could include another 10. You are shotgunning. Suspecting parts only on wild speculation. For example, if a PSU is suspect, then you also have a number that says why it is suspect.

Shotgunning is replacing parts based upon wild speculation (ie suspect heat). Since most any part can create your symptoms, then anyone will suspect anything. That is shotgunning. In a TV show CSI, described is how problems are solved. Follow the evidence. Currently no facts exist. No hard evidence exists to follow. Provided in the first paragrasph is how to make progress.
 

KopeAcetic

Reputable
Mar 31, 2015
16
0
4,510
Thank you again for the response Westom.

I have looked up a few guides online for testing voltages of the mobo. I'm a little confused on what specifically I would be testing: The Capacitors? To do this a guide I found: http://www.wikihow.com/Troubleshoot-for-Motherboard-Failures Looks like it's grounding to the case, and then attaching the red side of the meter to the capacitor itself, is this correct?

Another youtube guide I found was checking the cpu voltages I think by testing holes in the mobo as opposed to the capacitors. You said it's not calibrated so I need to check with a meter, do i just test everything or is there something specific I'm looking to test.

I actually don't own a meter so I"m going out and purchasing one right now to test.
 

KopeAcetic

Reputable
Mar 31, 2015
16
0
4,510
After getting the meter i unplugged everything and tried to see if i could figure out what to test. After a half an hout of looking through the motherboard manual and my computer i decided to put it together again to check the forums. after putting it back together my video card wasn't being detected and so my monitor didn't turn on again like what happened before. I tried resetting it and plugging the cables back in, nothing happened. I tried a couple more times to no avail so I tried to set it in another PCIE slot which still didnt work. My computer turns on, and the graphics card is getting power (lights signal power but not large loads) but no signal to the monitor. I'm typing this on my phone.

*edit* Woo got it to recognize my video card. I tried switching everything back to the slots they were originally in, and tried to re-seat and hook up the wires again and got it to go after a few tries.