Pc Randomly Shut Down

Hello

I've been looking at a fairly old amd pc, not sure of exact specs off hand but think it's dual core 3.2ghz 2gb ram. Windows 7 32 bit.

The PC seems to be running well, quick etc. Recently it has started shutting down no bsod or anything and it seems to be totally random nothing in event viewer indicates anything funky.

Did a deep clean and removed some adware malware etc, so fairly certain not a virus

Google suggests most common reason is a faulty power supply which I can easily replace, but what is the best way to rule all other options out?

Could it be harddrive? Graphics card? Cpu etc

 

Since just about "all other options", at least in terms of hardware depend on good, clean, stable power, you should start by swapping in a known good power supply to rule that one out.

If there is anything to try first, make sure the interior and vents are clean of heat trapping dust and all fans spins properly. And while in there, inspect for leaky and bulging capacitors on the motherboard. Concentrated on the ones surrounding the CPU socket.

 

I don't know if there's an AMD version of coretemp, but I would want to see if maybe you've got unusually high temps going on.

 

Core Temp works on Intel and AMD processors.
http://www.alcpu.com/CoreTemp/

 

And it works well too. It is what I use for real-time monitoring on all my computers.

But it is always best to verify temps with a second hardware monitor when you first install any temp monitor. I use HWiNFO64 for that - though the amount of information it provides can be overwhelming at first. Speccy also provides hardware information but it frequently displays incorrect voltage information. For example, it is currently showing my 12V is only 0.048V. And my 5V is only 3.36V. If my power supply were outputting those voltages, this computer would not be running. Temps and other information, however, are in line with CoreTemp and HWiNFO64.

 

Thanks so much for all your help it's greatly appreciated. The fans are spinning but the PC was VERY dusty inside going to swap out the PSU for sure, and I'll check core temps just out of interest what do you recommend as the best way to clear dust out. I know you can get those pressure air canisters are they OK or might they do more harm than good?

 

In terms of your electronics, they are very safe, as long as you hold the can level so they blow out the decompressing dusting gas and don't spew out super cold liquid that can frostbite on skin, or even cause super heated electronic components to crack (not likely a worry here) since your computer should be powered off.

It should be noted those cans do NOT contain air. Instead, they contain difluoroethane or similar gas, typically used as refrigerants and these gases are extremely hazardous if inhaled. In extreme cases, it can damage the central nervous system, or even be fatal. So use in a well ventilated area or better yet, outside. Note in some jurisdictions, these are controlled substances and sold in a controlled manner, similar to cans of spray paint to [hopefully] dissuade “huffing”.

 

I don't know if I have fully resolved this yet time will tell, but I just wanted to say thanks for the advice and tips. I can always rely on MGs!

I checked for bulging capacitators and all ok, Swapped out the PSU, reseated the RAM

I must say the thing was caked in dust, so under the CPU fan is that called a heat sink? That was so dusty the fan likely couldn't even get through to cool the CPU. I cleaned it all out and also the fan on the GFX card, so hopefully this will resolve the issues for him!

Also used Core temp and in idle the CPU was about 34 degrees which seemed normal.

 

34° C (93.2° F) is well within normal tolerances for the CPU. What are the GPU temps like (they might go to 68.9° C [156° F] — beyond that may be too hot)?

 

Yeah, the typical configuration is a CPU fan blows onto the heatsink. The heat sink is typically a big block of aluminum with large fins that "radiate" the heat away from the CPU. The base of the heatsink is typically made of aluminum or copper. If those fins were caked with dust, then effective cooling was surely compromised.

I generally recommend inspecting case interiors once a month, then clean as necessary. This interval can be adjusted once you have done it a few times to see what your dust build up patterns are like. If you live in a dusty environment and keep your windows open most of the time, if you have a herd of cats and dogs, or the computer is located where there is a lot of foot traffic stirring up dust, you may have to clean once a month. If those variables don't apply to you, you may only need to clean once or twice a year.

FTR, I will never buy a case again that does not have removable, washable air filters. Air filters do not eliminate the need for periodic interior cleaning, but they greatly increase the time between interior cleanings.

I also now only buy cases with side panel windows. With a see-through panel, monthly inspections are as simple as shining a flashlight in there. Plus, it is easy to see if your fans are still spinning properly without having to open the case.

BTW, GPU temps can be blistering hot and still be okay. As seen here, many were hitting 85°C, 89°C and even 90°C and still chugging along just fine. And I've seen charts where some breach 100°C under load and the system is still stable. It just depends on the specific GPU and card.

 

An overheating power supply, due to a malfunctioning fan, can cause a computer to shut off unexpectedly.