Holy !!! - Proof CPU's can handle HEAT

[AlabamaCajun]

Console, The plastic would not melt until past 140 or so, It must have softened temporarily.

T, For images of sink off chips, check youtube, they have a clip on an exploding Athlon (not for the squeamish).

PS2, I don't think it would be enough with just the spreader but you could probably go fan less and a regular stock sink.

 

[Rinne]

Well if you look at it from a strictly hypothetical situation it is without a doubt possible. Because it's simply a matter of moving enough air across the core I would think. In practicality it's beyond me as to whether or not standard PC fans would be up to the job. If it's worth anything, tom's used to have a video (it's still there but you have to hunt for it now) with a willamette P4 at 1.5GHz running without the heatsink. The proc throttled but otherwise remained functional the entire test with nothing but the IHS. I guess the ambient temp in the room, the TDP and the thermal threshold just happened to be balanced well enough to cause that. That chip had a 55W TDP so I'm sure it could be recreated with one of the cooler running chips of today.

It is OK, as long as the actual wattage doesn't exceed 7watts, probably a value a throttling willamette will run at.

But what I ask myself, why didn't the computer shut itself down?

 

[ps2cho]

It is OK, as long as the actual wattage doesn't exceed 7watts, probably a value a throttling willamette will run at.

But what I ask myself, why didn't the computer shut itself down?

Some stock BIOS' have the autoshutdown disabled. I know 3/4 IIRC of my motherboards owned had it disabled at stock. My current Biostar TForce7050 had it disabled.

 

[Moto7451]

Some CPUs have no problem running that high at all. My old G3 was rated at 105°C at 800MHz (stock voltage for that chip) with an expected lifetime of 20,000 hours MTBF.

 

[bing]

If you've read electronic chip datasheets, you'll notice most of them the max operating temperature at the silicon die (tjunction) is 105 C and some even up to 150 C.

 

[Moto7451]

Usually not modern CPUs though.

 

[AzAzEl656]

Sheesh. I got my 4600+ to 81C and I thought it was going to explode.

 

[Goonda]

60 C was my best. I have not blown up any CPUs till date.

 

[UBERCOOL]

I burned up an Athlon XP 2600+ once because my dad thought it would be alright to boot it up to see if it posted without a heatsink. This just enforces how fast that thing got past its temperature threshold because it never even got to post.

 

[Gautam]

Running with no heatsink has been done with mobile AXP's and Core Solos.

 

[ps2cho]

60 C was my best. I have not blown up any CPUs till date.

We all keep our personal temperature thresholds well above what a "real" limit should be.

Most people agree for a desktop chip, 50c-60c for AMD is the limit.

Well some desktop chips they put in laptops idle at 55c and load up to 75c and work fine!

I still will keep mine cool though

 

[Douken]

There was this thick solid plastic cover still attacked to the bottom of the heatsink! Apparently, I went a little too quickly while assembling the parts and didn't notice the plastic covering.

ROFLMAO!!!111

 

[AlabamaCajun]

Wait, SOI means "Silicon on Insulator", not "Sink on Insulator"

 

[ps2cho]

ROFLMAO!!!111

LOL That is what I said when I read it. Well I guess if you have never built a computer before...you could kinda miss it....maybe?

I was always a little more comfortable running higher temperatures and voltages than most people recommend. Heck I ran my AXP 2000+ at 1.95v @ 55c full load for a good 6months. Most people said that 1.95v on a Socket A processor was reaching the limits of air cooling

Ah the heck with it!