Question about overclocking a CPU and catching fire...please read!

Pck21

Pck21

CAGiversary!
Feedback
9 (100%)
So a little back story to the question before I post it. My girlfriend's sister's husband is a pathological liar. He has been caught cheating on his wife (they were dating at the time and it was in her bed, in her parent's house!) red handed, states he owns a lot of property but can't prove it, can't hold a job down for more than 6 months (no joke, that's the longest so far), and he's basically a scum bag that everyone hates. So here's what he told my gf's parent's 2 nights ago...

Apparently, he and his "class" from the local community college took a trip down to the University to compete in a "computer competition." This is a lie because my gf's brother goes to the community college and I go to the University and both of us didn't hear of any competition and we're both tech guys.

Second, he claims that his team won "money" for winning the competition and he "overclocked" a CPU "so much that it caught on fire." :face palm:

I wasn't there to refute such a claim, unfortunately, but my question is this: Is it physically possible for a CPU to catch on fire instantaneously in one sitting and one afternoon of overclocking? I don't know the exact computers they were using during this "competition" but I suspect they would be Dell pre-made computers.

I'm calling BS because I don't think it's physically possible for a CPU to "catch on fire" in the first place with the thermal paste around it and with a cooler on top. Wouldn't a too overclocked CPU just shut down?

Thanks for the replies!
 
Try turning on your PC without a CPU cooler. It'll destroy itself. You'll probably not get flames, but I'd say smoke is a possibility.
 
If there is a heat sink connected to it? I really doubt it. But, if the thermal contact is really poor, it might happen.

There was an article on Tom's hardware a loooooong time ago which tested what would happen if you suddenly removed the heatsink from the CPU while it was running a game. This was back in the days of P4s and the famous heat machine Athlon Thunderbird 1.4GHz. They posted videos of each Intel and AMD test run.

They removed the heatsink and immediately began measuring the surface temp of the CPU using an infared thermometer. In the background it showed the monitor running a looping Quake 3 demo.

The Intel chip, with its thermal protection circuitry, immediately throttled the frequency down to save itself. You could see the Quake 3 running at about 0.2fps in the background. The surface temp of the CPU was like 100C (could be wrong, as I don't remember clearly anymore). Now I am really speculating here but I think in the end of the video they put the heatsink back on and once the temp dropped down some the CPU upped its frequency and you could see Quake 3 speed up in the background again.

The AMD Athlon, with no thermal protection, once the heatsink was removed from the CPU the computer immediately locked up as the CPU died. Within a few seconds smoke staretd to rise from the top of the CPU (probably the thermal paste beginning to decompose). The infared thermometer showed a surface die temp of like 600F.

The thing is though, I think all modern CPUs have thermal protection integrated into them now such that it would shut itself off before ever reaching such dangerously high temps. So, maybe if the thermal protection either was defeated, or broke/died in the overclocking, and the chip suddenly suffered from catastrophic cooling failure (tantamount to removing the heatsink completely), you might see some smoke.

I'm curious if that Tom's hardware video is still around it was pretty cool to watch hehe.

Ruahrc
 
Setting fire to a CPU would require temperatures fatal to any nearby observers. It was the thermal conducting paste that ignited in those videos. About the only thing on a CPU that would ignite at relatively low temperatures (high temperatures being stellar fusion) is the ink on the labeling.

You know the old joke about PCs running on blue smoke and if you release the blue smoke the computer stops working? The smoke is from electrical insulation burning when a low voltage connection, say a front LED light, is connected to far higher voltage than intended. I had that happen once almost twenty years ago when I building a system with a really badly designed motherboard. There were pins in close proximity to each that any modern board would handle differently.
 
Like said before it prolly couldn't happen ... and the other thing is why the hell would you win for setting a cpu on fire? Wouldn't that be a bad thing?

Anyone can go to the mother board settings and crank everything u as high as it goes and melt shit, the real trick is making it faster then anyones and be stable right?
 
[quote name='strikeratt'] ... and the other thing is why the hell would you win for setting a cpu on fire? [/quote]

Well he was going for the trifecta after crushing the most beercans on his forehead, and turning the most doughnuts in the JC parking lot with his 78 Trans Am.
 
[quote name='strikeratt'] ... and the other thing is why the hell would you win for setting a cpu on fire? Wouldn't that be a bad thing?
[/quote]

I came in here to say this exact same thing.
 
You must log in or register to reply here.
bread's done
Back
Top