Sign in with
Sign up | Sign in
Your question

AAAAAAAARGHHHH!!!!!!!!!!!!!

Tags:
Last response: in CPUs
Share
August 14, 2002 8:56:18 AM

I just destroyed my beloved, famous and highly overclockable Tualatin Celeron 1400MHz @ 1764MHz during an amateurish attempt to remove the heat-spreader. Sigh :-((

It's a boxed retail, but I guess Intel won't take this one back ...

So now I have to get myself a replacement. Which one should I go for ? A 1200MHz or another 1400 MHz ? The 1400MHz didn't reach the magical 133 MHz FSB, but maybe a 1200MHz one will do. The only thing that bothers me is, that I read somewhere that the 1400MHz version is a new stepping and I'm not sure the 1200MHz is as overclockable.

Please don't give me any suggestions about dumping the mobo and get myself a Athlon XP1600+ on a KT333 board. It's tempting but I also have a stick of 256MB PC133 SDRAM lying around.

Still wiping the tears off my face... this one really hurt.

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> [sej]

More about : aaaaaaaarghhhh

August 14, 2002 9:20:14 AM

try rmaing it.



:wink: The Cash Left In My Pocket,The BEST Benchmark :wink:
August 14, 2002 9:45:31 AM

The heat-spreader is still on, but if somebody takes a closer look on it, they will discover the scratches on the side of the heat-spreader. It will be difficult for me to explain how they came there.

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> [sej]
August 14, 2002 9:47:53 AM

Wait, the Celeron tualatins don't come with the integrated heatspreader right? So that means all you were doing is taking off your own. I think Intel's warranty would take that, won't hurt to try.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
August 14, 2002 10:11:24 AM

Quote:
Wait, the Celeron tualatins don't come with the integrated heatspreader right?

Oh, yes they do, that is, the 1400MHz version does, I don't know about the other lower speed versions.

By the way, imgod2u, have to put the benchmarks of my late Tualatin Celeron on your homepage as you promised ?

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> [sej]
August 14, 2002 11:47:24 AM

LOL... When I first saw this thread... I guessed you killed your CPU, but I would've thought you would've done it tryin to push it as far as possible though... :wink:

<i>Past mistakes may make you look stupid, but avoiding future ones will make you look smart!</i>
August 14, 2002 4:24:23 PM

Quote:
The heat-spreader is still on, but if somebody takes a closer look on it, they will discover the scratches on the side of the heat-spreader. It will be difficult for me to explain how they came there.

Heat sink installation and/or removal? It was a really stubborn heat sink you know... ;) 

<A HREF="http://www.nuklearpower.com/comic/171.htm" target="_new">The corpse you find may be your own.</A> - Black Mage
August 14, 2002 6:27:49 PM

Why would you try to remove the heat spreader? I don't understand the reasoning. Do you think you can make a better thermal interface between the die and your heatsink than a bonded heat spreader + heatsink? Isn't the heat spreader there for the purpose of increased surface area?

Why not just lap the surface of the heat spreader and heatsink?

<b>I have so many cookies I now have a FAT problem!</b>
a b à CPUs
August 14, 2002 8:29:32 PM

Dude, all the latest Intel chips come with heat spreaders, even my PIII 1000EB Coppermine came with a heat spreader.

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
a b à CPUs
August 14, 2002 8:32:08 PM

According to its name it's there to increase surface area, but in actuallity it's there to protect the core. I took the heat spreader off my Celeron 1.2, but did not see any improvements in cooling, so I put it back on.

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
August 14, 2002 8:46:27 PM

Quote:
Why would you try to remove the heat spreader?

Because... I'm a fool.

The idea was that any additional layers between the core and the very smooth surface of my heatsink would result in less efficient cooling. The heat-spreader makes the CPU a lot more robust, both in handling and prevents it from burning in the absence of a heatsink, but I think lower temperatures can be obtained without. I've read about people removing the heat-spreaders on their P4's, and they got more efficient cooling.

Initially I thought the heat-spreader was made of pure ALU, but actually my scratches revealed that it's made of copper with an ALU surface, so it may not be as bad as I thought.

Well I've learned a lesson, let this be a warning to the rest of you. Don't mess with it, unless you have the right tools, I didn't...

<i>/Copenhagen - Clockspeed will make the difference... in the end</i> :cool:
August 14, 2002 8:53:21 PM

Hmmm, so you can remove it. I thought the bond would be stronger and thus a good thermal interface (OK, so it depends on how it was bonded).

<b>I have so many cookies I now have a FAT problem!</b>
August 14, 2002 8:57:00 PM

Well dude....if Intel says anything when u RMA it can't u tell them the scratches got there when u tried to take all the TIM Residue off the Processor??? i dunno how big the tim is, so its just a suggestion....

ALSO....Crashman will tell u the 1200 Tully celeron is not a good overclocker.......so id recomend a P-3 1400Mhz Tully...not the celeron Version.....

<A HREF="http://www.anandtech.com/mysystemrig.html?id=13597" target="_new">-MeTaL RoCkEr</A>
August 14, 2002 9:12:52 PM

OK, now I'm not so sure. According to Crashman the heat spreader is easy to remove so now I'm wondering how good a job the interface really does.

This is a little different but I once tried wedge a piece of copper between a Duron and the heatsink. I thought the increase surface area would help but cooling got about 4-5 degrees worse. From a little reading I learned that it is hard to overcome the fact that I now had two interface points. One of those interface points is still the size of die, so nothing gained.

I thought heat spreaders are "bonded" in a more effective way. So which would be more effective, heat spreader or no heat spreader?

<b>I have so many cookies I now have a FAT problem!</b>
a b à CPUs
August 14, 2002 9:24:08 PM

It has some kind of grey putty on the inside and is bonded at the edges with some kind of black rubbery stuff. You can slide a razor under the edge to cut the black stuff. I replaced the black stuff with black RVT automotive sealant when I reattached it, and used a little ASII on the core to fill in the missing grey stuff. The spreader itself is made of Copper and is plated with a soft silvery metal which I believe to be Tin.

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
August 14, 2002 11:09:06 PM

Was there any difference in temps after you replaced the spreader?

<b>I have so many cookies I now have a FAT problem!</b>
a b à CPUs
August 15, 2002 12:36:01 AM

I think I reduced the core temp by 1C at idle without the heat spreader, and 0C at full load. Putting it back on didn't make any difference than before it was removed.

<font color=blue>You're posting in a forum with class. It may be third class, but it's still class!</font color=blue>
August 16, 2002 6:35:56 AM

Quote:
Why would you try to remove the heat spreader? I don't understand the reasoning. Do you think you can make a better thermal interface between the die and your heatsink than a bonded heat spreader + heatsink? Isn't the heat spreader there for the purpose of increased surface area?

Why not just lap the surface of the heat spreader and heatsink?


for the last time, fatburger proved heatspreaders RAISE TEMPS, even bonded its just ANOTHER useless thermal layer to impede transfer.

its for protection of the core ONLY.

:wink: The Cash Left In My Pocket,The BEST Benchmark :wink:
August 16, 2002 6:57:16 AM

I would lie my ass off and say I didn't do a thing to it. Try to get the freebie. Then dispute the charges on my credit card unless 60 days have passed.

My 2 cents.

<P ID="edit"><FONT SIZE=-1><EM>Edited by killian2 on 08/16/02 02:57 AM.</EM></FONT></P>
August 16, 2002 7:11:25 AM

well i didn't read what every one here has posted, but sand paper does wonders if the scratches aren't very deep.

<b><font color=orange>sing to prolong HDD life; spin right round like a record baby Right round round round
August 16, 2002 3:21:12 PM

If you read the other posts you'd see that I am already aware of the problem of multiple thermal interfaces.

I was pointing out that lapping can get some cooling improvement and you don't have to risk damaging the core by removing the heat spreader.

OK, so I don't know how much cooling improvement can be gained by removing the spreader. I'll accept that it does improve it. I would think the losses are relatively small compared to the gains that Copenhagen has obtained from water cooling.

Further, if as you say, the sole purpose of the heat spreader is to protect the core, it's ironic that Copenhagen destroyed his core by removing that which is supposed to protect it. Don't you think?

<b>I have so many cookies I now have a FAT problem!</b>
August 16, 2002 3:32:22 PM

Quote:
Further, if as you say, the sole purpose of the heat spreader is to protect the core, it's ironic that Copenhagen destroyed his core by removing that which is supposed to protect it. Don't you think?

To us, ironic. To Intel, poetic justice. :( 

<A HREF="http://www.nuklearpower.com/comic/171.htm" target="_new">The corpse you find may be your own.</A> - Black Mage
August 16, 2002 4:16:25 PM

Quote:
To us, ironic. To Intel, poetic justice

Nah! If Coppenhagen successfully RMA's it, now that's poetic! Not ethical but poetic.

<b>I have so many cookies I now have a FAT problem!</b>
August 16, 2002 4:23:43 PM

Quote:
Nah! If Coppenhagen successfully RMA's it, now that's poetic! Not ethical but poetic.

True. :) 

Hmm ... I wonder if anywhere in the waranty statement it explicitely says: Waranty void if heat spreader removed.

Who knows. Maybe it isn't even unethical?

Of course, then we could also get into the whole debate of morality and ethics are just individual points-of-view and we could say that Intel selling the chip for so much more than it costs them to make is unethical. Or that hindering the Celeron so badly by halving the cache and stripping the FSB is unethical. And therefore sticking it to Intel with a successful RMA would simply be the ethical thing to do.

But then, that's the joy of being human. We can rationalize a justification for anything if we try. ;) 

<A HREF="http://www.nuklearpower.com/comic/171.htm" target="_new">The corpse you find may be your own.</A> - Black Mage
August 16, 2002 10:53:05 PM

The 1400s are available in both the tA1 and tB1 steppings. I believe the whole line from 1000 to 1400 will be available in the tB1 stepping. I'd get the 1200. At 133 it'll be running at 1600.

- JW
!