Computer shutting off without warning

[Pazzaar]

Hi, I have been recently having a problem with my Sony Vaio computer

My computer has been shutting down without warning, but I'm not sure why. It's shutting down in a similar fashion to when my computer has overheated in the past - I.E the screen just goes black and the power light fades away almost instantly.

However, I have recently replaced the heatsink on my laptop, and I can tell you for a fact that the computer is NOT overheating.

Here is the incredibly weird thing though. The computer does not appear to shut down ever when I don't have any games open. However, when I start up a game - including clients for some games, like League of Legends, the computer will shut off in usually 5-10 minutes.

Even stranger still, sometimes I can play games for 3 hours, and the computer will not turn off. The problem only appears to happen when i play some certain games. To see whether the computer was overheating, I alt-tabbed out of a 'problem' game, and watched my temps on CPUID HW monitor, and sure enough, no overheating occured - temps stayed at aroun 60 degrees C, and the computer after about 10 minutes just shut off without warning.

I have zero idea why this is happening, but it is of course annoying. I have tried updating graphics drivers, but it changes nothing.

If anyone could help me it would be greatly appreciated. Thanks!

 

[smarteyeball]

However, I have recently replaced the heatsink on my laptop, and I can tell you for a fact that the computer is NOT overheating.

Here is the incredibly weird thing though. The computer does not appear to shut down ever when I don't have any games open. However, when I start up a game - including clients for some games, like League of Legends, the computer will shut off in usually 5-10 minutes.

Whats the Gpu temp though? You mentioned checking the temp - was that the CPU or the GPU temp?

Certain games will utilise the GPU more than others. The CPU may be 60c but the GPU may be exceeding 100C +

If the GPU temp is exceeding the threshold, then it will shut down like described. It doesn't mater what the CPU temp is.

otherwise, are you running on battery or mains? Is the power adapter the correct one for your system?

 

[Pazzaar]

However, I have recently replaced the heatsink on my laptop, and I can tell you for a fact that the computer is NOT overheating.

Here is the incredibly weird thing though. The computer does not appear to shut down ever when I don't have any games open. However, when I start up a game - including clients for some games, like League of Legends, the computer will shut off in usually 5-10 minutes.

Whats the Gpu temp though? You mentioned checking the temp - was that the CPU or the GPU temp?

Certain games will utilise the GPU more than others. The CPU may be 60c but the GPU may be exceeding 100C +

If the GPU temp is exceeding the threshold, then it will shut down like described. It doesn't mater what the CPU temp is.

otherwise, are you running on battery or mains? Is the power adapter the correct one for your system?

The GPU temps stay at around 50C, so its not that. Sometimes, for example last night, I've been able to play a game for 3 hours with no problem. However, if I opened up a different (less demanding) game, the computer would shut down in 5 minutes.

The power lead is the same lead I have been using for 3 years, so I would be hard pressed to say it was that. I currently run without a battery in the laptop, directly from the mains, however if I were to disconnect the lead from the mains, the laptop's power light would instantly go off. However, in this instance, it fades away after the screen goes black from the laptop shutting off

 

[ShamrockRig]

Could it be that you might be accidentally knocking the lead out of the input port? therefore causing it to lose power and instantly shut down?

 

[Pazzaar]

Could it be that you might be accidentally knocking the lead out of the input port? therefore causing it to lose power and instantly shut down?

Definitely not that

 

[Pazzaar]

Okay, I have a development. The problem has begun to appear OUTSIDE of games, so it would appear it is not the games which are the problem. It seems to occur when I am doing something that has some demand in the process, like working on Photoshop for example.

I tried booting in safe mode, and sure enough, the computer still shut down after a bit when I was working in Photoshop. This would lead me to believe that the problem does not lie with the GPU.

That leaves the motherboard, CPU, HDD, PSU, or something else that is a bit random. If anyone has any ideas what else could be causing these random shut downs that would be great!

BTW, the computer IS NOT overheating! The fans are spinning at normal rates, and the temps of the CPU are at arounr 45-50C at idle temps.

 

[linnemeyerhere]

I wouldn't trust the temps but would highly suggest you check first hand by opening up and blowing out all fans and coolers and while there have a good look around for cold solder joints, loose wires, or evidence of discolored mobo.

 

[matts6887]

Also; while in there; look for any possible capacitors that look like they may be about ready to burst open; as any signs of that can be a major problem. I know linne said not to trust the temps; and i do but i dont agree with that. What i would do is try and compare temps of the hardware monitor with those of the bios(if your bios has a temp monitor in it) but if not then try and compare it with another hardware monitor to see if the temps are the same. If they are different; then ya gotta try and figure out which one is right. And yes you should blow the dust and dirt out of the fans as part of maintenance. Also; check for broken or loose wires or cables/connectors etc.

 

[carwiz]

Run this Diagnostic Tool so we can see the logs and system setup.

 

[Pazzaar]

Ok, I am less than enthusiastic to start prodding around inside my computer straight away, but I guess I'll do it later, when I have more free time

Run this Diagnostic Tool so we can see the logs and system setup.

I used the diagnostics tool and then created the zip file as requested, and have set it as an attachment to this post. I hope it can help solve this issue I am having!

I'll update this forum later with whatever I find when I open up my laptop!

 

[linnemeyerhere]

We'll if you're sure it's not high temps then have you looked for updated drivers for your GPU and other mobo drivers.

 

[carwiz]

Is this the same Alistair that frequents a certain machinist forum? "CCWKen" over there.

Your system event log ends in December of last year so it wasn't much help but I noticed several problems in the application log. Windows Update is failing as well as a couple of events that could be caused by malware or errant drivers. The two below are particularly suspect.

Event[29118]:
  Log Name: Application
  Source: Windows Error Reporting
  Date: 2014-01-31T22:59:40.000
  Event ID: 1001
  Task: N/A
  Level: Information
  Opcode: Info
  Keyword: Classic
  User: N/A
  User Name: N/A
  Computer: Alistair-PC
  Description: 
Fault bucket 3979255468, type 5
Event Name: RADAR_PRE_LEAK_64
Response: Not available
Cab Id: 0
Problem signature:
P1: MsMpEng.exe
P2: 4.4.304.0
P3: 6.1.7601.2.1.0
P4: 
P5: 
P6: 
P7: 
P8: 
P9: 
P10:

Event[29231]:
  Log Name: Application
  Source: SideBySide
  Date: 2014-02-01T04:53:39.000
  Event ID: 33
  Task: N/A
  Level: Error
  Opcode: N/A
  Keyword: Classic
  User: N/A
  User Name: N/A
  Computer: Alistair-PC
  Description: 
Activation context generation failed for "C:\Program Files\Autodesk\Composite 2013\python\lib\distutils\command\wininst-8_d.exe". Dependent Assembly Microsoft.VC80.DebugCRT,processorArchitecture="x86",publicKeyToken="1fc8b3b9a1e18e3b",type="win32",version="8.0.50608.0" could not be found. Please use sxstrace.exe for detailed diagnosis.

For test purposes, uninstall Akamai NetSession Interface and Pando Media Booster.

Check your hard drive for unnecessary files or off-load photos, videos, etc. to an external disk. It appears that you may be running out of disk space.

Run a complete virus and malware scan after clearing some disk space.
Run SFC /scannow from an elevated command prompt. (Use Option Two.) Report back any console messages. Note that if errors are found, you should reboot and run it again. It can take several cycles of this procedure for the scan to repair files and links.

 

[Pazzaar]

Okay, I can now safely assume that the computer believes that the CPU is overheating, and hence shutting the computer down. Putting the CPU under any stress that will take it into the 60C region causes the computer to shut down. I can confirm this is not a windows problem or HDD issue because it does this when I am in the BIOS too. Great.

I can only assume that the sensors which I have access to are not telling me accurately what the temperature is. I find it odd that the computer fans do not increase in RPM at all though, to attempt to cool down the CPU. The CPU can detect when it is too hot that it shuts down the computer, but it can't detect when it is too hot that it needs to up the RPM on the fan? Just a bit strange

Now, coincidentally, about 20 days ago I was forced to change the Heatsink on my laptop (linn might remember this), and I bought a new part. The heatsink is connected to the fan on my laptop, so I ended up getting a brand new heatsink and fan. I cleaned off nearly all of the old thermal paste from the CPU and GPU, and then placed the new heatsink in position. The new heatsink came with a standard thermal tape, which was already applied.

After installing the heatsink, my laptop began to work well again - better than it had before. The temps stayed in the low 70s to 60s, and no overheating ever occured. Its only recently (in the past 3 or 4 days) that this 'overheating' issue has come about.

This seems too convenient that the CPU is having overheating issues after I replaced a device specifically created to prevent overheating, however I have no clue how this would happen. The laptop was working absolutely fine 10 days ago. Then suddenly - bam, overheating under the slightest of strain.

Is this a thermal paste/tape issue? It seems hard to believe, as why would it suddenly stop working, however I can't think of any other explanation. And also, why is the computer telling me that my CPU is at a fine temperature? The computer uses the temps it tells me to tell if the CPU is overheating, how can it possibly tell it is overheating if the temperatures are clearly in the low 60s.

If anyone could give some insight into this problem that would be swell. Sorry for the long post!

 

[linnemeyerhere]

It would seen doubtful the TEM is at fault. More likely would be the cooler itself or the mount needs to be tighten as the TEm has settle in after several shutdown cycles. I would remove the cooler and clean all parts again and then remount using a premium TEM like Arctic Silver or the like. Remember less is more and many techniques could be used. I like the five drops or you could use a spreader and cover the whole cooler with an ultra thin smear. Then tighten and let rest and tighten ever so slightly again. Power on and off watching temps for several days then try to snug the cooler again but never over tight just little snugs if it will at all.

 

[Pazzaar]

Okay, I'll do something later today

What I don't get is why the computer is telling me the temps are at 60C, but it still knows to shut down? How can it know it still needs to shut down, if the sensors which it uses are telling it the CPU is currently at 60C, an absolutely safe temperature?

 

[linnemeyerhere]

At this point you're just in the process of elimination, it's likely a good mount on the cooler but a clean and re-install can't hurt and maybe as you snug it real slow and evenly you'll get a more flush and better contact. Did you do the previous replacement?