First post, by Kahenraz
- Rank
- l33t
It looked good from a distance but then I inspected it from the side.
Super gross!
It looked good from a distance but then I inspected it from the side.
Super gross!
I don't know if this is a result of liquid metal or what. But there is some kind of textured surface adhered to the top of the die. I've never seen this before.
It's also very chipped around the sides.
I tried to scrape it off with a nylon spudger but whatever it is it's very solid. It also will not dissolve with a citrus cleaner which further suggests that it's some kind of liquid metal that has solidified.
Now that's a nasty way of mounting a cooler...
Stuck at 10MHz...
Wow, 🤣. Does the CPU work?
I cleaned everything up and reseated the heatsink and it works fine. It only looked scary. There is some real damage but the affects appear harmless.
They put the HS on 180 degrees off from what it was supposed to be. Could have fried the cpu but sounds like you got lucky !
Long ago had a friend do similar and it fried the cpu, think it was a AMD soc 754 iirc.
Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun
Horun wrote on 2021-12-11, 02:53:Long ago had a friend do similar and it fried the cpu, think it was a AMD soc 754 iirc.
This couldn't happen with socket 754. With Athlon 64, AMD started using an integrated heat spreader, which brought the total height of the CPU package above that of any part of the socket. As a result, heatsinks for socket 754 and onward could be completely flat on the bottom and oriented in whichever direction, they didn't have that one recessed edge like socket 7/370/462 heatsinks had. Also Athlon 64 had integrated thermal protection, they could throttle down or even shut down if they began overheating. Even if you managed to mount the cooler incorrectly, it shouldn't have fried the CPU.
If it's an AMD CPU that your friend fried, it would have probably been a socket 462 one, so a vanilla Athlon or Athlon XP. Those still lacked thermal protection and didn't come with an integrated heat spreader, so you had to orient their cooler correctly.
Nothing wrong with the direction that heat sink was mounted at all.
There's a glitch in the matrix.
A founding member of the 286 appreciation society.
Apparently 32-bit is dead and nobody likes P4s.
Of course, as always, I'm open to correction...😉
frudi wrote on 2021-12-11, 03:53:If it's an AMD CPU that your friend fried, it would have probably been a socket 462 one, so a vanilla Athlon or Athlon XP. Those still lacked thermal protection and didn't come with an integrated heat spreader, so you had to orient their cooler correctly.
It could also be with s.754 - with Turion or Sempron mobile, which the heatspreader doesn't have and the core is drowned lower, so the stock cooler doesn't reach it. It sits on a frame around the socket. But the burn should not occur - the overheat protection should shut down the processor before it burns, as you wrote.
AFAIK Pentium IIIs were WAY more tougher than Duron/Athlon when it came to mounting a HSF in reverse.
Though, shouldn't the HSF itself show some resistance when trying to mount it backwards?
"Enter at your own peril, past the bolted door..."
Main PC: i5 3470, GB B75M-D3H, 16GB RAM, 2x1TB
98SE : P3 650, Soyo SY-6BA+IV, 384MB RAM, 80GB
frudi wrote on 2021-12-11, 03:53:Horun wrote on 2021-12-11, 02:53:Long ago had a friend do similar and it fried the cpu, think it was a AMD soc 754 iirc.
If it's an AMD CPU that your friend fried, it would have probably been a socket 462 one, so a vanilla Athlon or Athlon XP. Those still lacked thermal protection and didn't come with an integrated heat spreader, so you had to orient their cooler correctly.
Yes now am thinking it was an Athlon soc 462 cpu.
Caluser2000 wrote on 2021-12-11, 07:26:Nothing wrong with the direction that heat sink was mounted at all.
Heheee yeah that is a perfectly mounted heat sink 😀
Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun
Badly chipped edges do not always result in dysfunctional CPU. My PIII is also badly chipped to the point I would never have bought it myself, but I got it as a replacement and it works. It's best to test it using Prime95 in all modes for a few hours.
I usually use alcohol, soft paper tissues and nails to clean CPU dies. You could also use plastic to scratch off burned in thermal paste. It's only necessary to flatten it, no need to remove it fully.
Pentium III 900E, ECS P6BXT-A+, 384MB RAM, NVIDIA GeForce FX 5600 128MB, Voodoo 2 12MB, 80GB HDD, Yamaha SM718 ISA, 19" AOC 9GlrA
Athlon 64 3400+, MSI K8T Neo V, 1GB RAM, NVIDIA GeForce 7600GT 512MB, 250GB HDD, Sound Blaster Audigy 2 ZS
Caluser2000 wrote on 2021-12-11, 07:26:Nothing wrong with the direction that heat sink was mounted at all.
Wat?
appiah4 wrote on 2021-12-11, 19:14:Caluser2000 wrote on 2021-12-11, 07:26:Nothing wrong with the direction that heat sink was mounted at all.
Wat?
I'd say it was missing a smiley to indicate sarcasm, but it's really hard to tell due to Poe's Law.
I thought IBM was born with the world
The HS is indeed mounted 180 degrees rotated. It's really quite facepalm worthy.
The TIM looks like AS5 or something similar to that and it can be a real pain to clean up the mess.
When TIM residue is cooked to the CPU die, you could try and scrape it off with a plastic card like an old credit card or an old bank card.
If that had been a sA CPU, it would likely have fried right away if the system had been turned on with a CPU HSF mounted like that.
I even had an Athlon fry itself within seconds with a correctly mounted HSF but having forgotten to apply TIM.
When underclocked (like 1200MHz to 900MHz) I've had the sA CPU survive several times when having forgotten to apply any TIM but only if I noticed my error within seconds after which I switched off the power.
s370 CPUs have always seemed much more resilient to abuse from clueless folks.
On a similar note, I have tested motherboards with socket 370 CPUs without a heatsink and heatsinks without any thermal paste for quick testing such as whether the system turns on or reaches POST. I can't recommend this practice but I have never damaged a CPU in the process. Not irreparably, at least.
I used to use a basic heatsink with thermal tape for this use case but that ended up being a bad idea.
Tetrium wrote on 2021-12-14, 03:03:The HS is indeed mounted 180 degrees rotated. It's really quite facepalm worthy. The TIM looks like AS5 or something similar to […]
The HS is indeed mounted 180 degrees rotated. It's really quite facepalm worthy.
The TIM looks like AS5 or something similar to that and it can be a real pain to clean up the mess.
When TIM residue is cooked to the CPU die, you could try and scrape it off with a plastic card like an old credit card or an old bank card.If that had been a sA CPU, it would likely have fried right away if the system had been turned on with a CPU HSF mounted like that.
I even had an Athlon fry itself within seconds with a correctly mounted HSF but having forgotten to apply TIM.
When underclocked (like 1200MHz to 900MHz) I've had the sA CPU survive several times when having forgotten to apply any TIM but only if I noticed my error within seconds after which I switched off the power.
s370 CPUs have always seemed much more resilient to abuse from clueless folks.
When testing Socket 370 CPUs I usually don't even clamp the PSUs down, just laying it down on the CPU is enough foor testing it to POST. So the wiseass that is me at some point thought he could get away with just placing the HSF on top of a Socket A CPU to test it as well. It took 1 second for magic smoke. The CPU died, and killed the motherboard as well, somehow.
I've had such good success misbehaving with Intel processors I would have done the same with an AMD chip. I don't have any Socket A boards but if I ever get one this will be a good thing to know.
If what you say is accurate then the cheap thermal tape I was using would probably have fried the chip as well.
Kahenraz wrote on 2021-12-14, 09:12:I've had such good success misbehaving with Intel processors I would have done the same with an AMD chip. I don't have any Socket A boards but if I ever get one this will be a good thing to know.
If what you say is accurate then the cheap thermal tape I was using would probably have fried the chip as well.
I don't know about that, no thermal compound seems to be fine for testing Socket A CPUs for me. Maybe I am just living dangerously..