How do I lap a P4 heat spreader?

[ytoledano]

Hi, I've just received some 500 grit paper and I want to lap my P4C's heat spreader because I kinda really scratched it and I also heard that you get better thermal conductivity between the heat sink and the die. I was thinking of putting using a piece of Styrofoam and insert the CPU pins to it and then moving the HS over the paper. Will that work? Where can I find a tutorial on how to do something like this?

Abit IS7 @ 275 FSB, VAGP @ 1.65, OCZ PC4200 RAM @ 550Mhz, P4 2.4 @ 3.3Ghz Vcore @ 1.625, GeXCube Radeon 9600P @ 450/626.
Thermalright SLK-947U, P3 HS @ NB.

 

[jclw]

Styrofoam is a good idea. Cut a piece of Styrofoam about 3"x3" and set the chip in it.

You'll also need a perfectly flat surface - I find glass/mirrors work well. Make sure it doesn't have any scratches in it.

Put the sandpaper on the flat circle and start doing circles or figure eights with the processor. Rotate the processor 90 degrees every few turns to make sure you lap it evenly.

*Dual PIII-800 @900 i440BX and Tualeron 1.2 @1.7 i815*

 

[pIII_Man]

just pull it off!

Intel uses pretty crappy phase change wax under the heatspreader, lapping it probably wont change too much.

If it isn't a P6 then it isn't a processor
110% BX fanboy

 

[ytoledano]

OK, I just booted after lapping and I think there is something like a 3C change, maybe a bit less. Probably wasn't worth it... I'm too scared to pule the IHS off, I heard you really have to fight the processor to get the IHS off.

Abit IS7 @ 275 FSB, VAGP @ 1.65, OCZ PC4200 RAM @ 550Mhz, P4 2.4 @ 3.3Ghz Vcore @ 1.625, GeXCube Radeon 9600P @ 450/626.
Thermalright SLK-947U, P3 HS @ NB.

 

[pIII_Man]

thats true...on the tualatins i have delt with removing the IHS can be tricky. To remove the IHS, i put the cpu upsidedown in a vice...so the IHS is the only thing being gripped. Then i get a heat gun and heat up the IHS so the phase change material underneat melts...and then i gently twist the cpu in both directions until eventually the RTV sealent breaks away. If you neglect to heat up the IHS then you crack the core in the process of twisting off the IHS (yes i have done it).

I havnt seen this method documented but it works the best for me...normally i can get an IHS off in 10 minutes this way...

If it isn't a P6 then it isn't a processor
110% BX fanboy<P ID="edit"><FONT SIZE=-1><EM>Edited by piii_Man on 03/27/04 07:54 PM.</EM></FONT></P>

 

[ytoledano]

I thought there was a reason for P4s to shut down at 90C how much do you heat the processor? Why doesn't it die?
Oh, and the whole Stryrofoam thing didn't help - I put the CPU on the thing, pins down, and pushed really hard but it just wouldn't stick. So I used some cloth glued to some foam that I found lying around in the house. I put the CPU, IHS down, on the paper and worked in circles.

Abit IS7 @ 275 FSB, VAGP @ 1.65, OCZ PC4200 RAM @ 550Mhz, P4 2.4 @ 3.3Ghz Vcore @ 1.625, GeXCube Radeon 9600P @ 450/626.
Thermalright SLK-947U, P3 HS @ NB.

 

[Lando95]

Just how thick is the IHS? Also, will your method of removing it work with a 2.8C?

PIV 2.8 @ 3.37, 2X256 Geil Platinum PC4000 @ 1:1, 240fsb, SP-94 with 92mm Tornado, IC7, (2)Raptors in raid 0, BFG5900nu flashed to 5950ultra bios, 500 watt psu.

 

[pIII_Man]

it was about 2-3mm thick, but this was with tualatin core celerons, however i beleive the dimmensions of the IHS on the tullies and NW's is the same (or at least similar). I cant think of why this method wouldnt work for a NW core cpu. Although you have to remember if you are using a heatsink without a copper base/copper slug insert you probably would be better off leaving the IHS on.

If it isn't a P6 then it isn't a processor
110% BX fanboy

 

[Lando95]

Thanks, I still have my 2.53ghz P4, maybe I will try it on that first. And yes, I do have a copper HS, an SP-94.

PIV 2.8 @ 3.37, 2X256 Geil Platinum PC4000 @ 1:1, 240fsb, SP-94 with 92mm Tornado, IC7, (2)Raptors in raid 0, BFG5900nu flashed to 5950ultra bios, 500 watt psu.

 

[jammydodger]

Correct me if im wrong but isnt the whole point of the heat spreader to increase the surface area of the CPU so that heat can be dissapated more easily? Surely by removing it temps will increase?

P4 2.6@3.38
512Mb PC4000
2x120Gb 7200.7 in RAID0
Waterchill KT12-L30
Abit AI7
Ge-Force4 Ti4200

 

[jihiggs]

im convinced the heat spreader is there to keep bumbling clods from crushing the core.

wpdclan.com cs game server - 64.246.52.144:27015
now featuring (optional) cheating death!

 

[Lando95]

Well, according to what i have read, there are two purposes of the ihs, 1) protect the core, 2) dissipate heat evenly across a large area. However, the actual core is only a small portion of the size of the ihs. {I think the arctic silver website that shows a picture} The effects of the ihs on heat dissipation are very minimal. The heat transfer is centered directly above the actual core. Theoretically, it would be better to remove the ihs, since you could use high heat conductive copper, such as a waterblock or etc. I have not yet decided if i want to hack away on my great oc'ing 2.53, it goes to 3.4.

PIV 2.8 @ 3.37, 2X256 Geil Platinum PC4000 @ 1:1, 240fsb, SP-94 with 92mm Tornado, IC7, (2)Raptors in raid 0, BFG5900nu flashed to 5950ultra bios, 500 watt psu.

 

[jammydodger]

I would be too scared of crushing the core to remove the IHS I think.

P4 2.6@3.38
512Mb PC4000
2x120Gb 7200.7 in RAID0
Waterchill KT12-L30
Abit AI7
Ge-Force4 Ti4200

 

[tombance]

THe IHS simply adds another heat transfer boundary into the eequation. Even if the heatspreder has a large SA, the core itself still has to make contact with the IHS, and the core doesnt hava a very large SA. It would be like sticking an extra lump of metal between an Athlon core and the HS, even if the SA is very large, it will only increase temps.

I personally would just take it off, but then I have enough experience with AMD CPU's not to crush a core (I think...).

<A HREF="http://service.futuremark.com/compare?2k1=7454540" target="_new">Yay, I Finally broke the 12k barrier!!</A>

 

[pIII_Man]

well on a aluminum based heatsink...the heatspreader does help because the copper heatspreader allows heat to be dissipated a little more evenly across the heatsink. However if you have a copper/copper based/or HS with copper slug then the IHS really serves (as said earlier) as just another junction that heat must pass through.

If it isn't a P6 then it isn't a processor
110% BX fanboy

 

[tombance]

Yeah, I can understand that. On a aluminium HS, it would be like having a copper insert in the base, with the only difference to a standard hybrid being that the interface between aluminum and copper wont be quite as good a standard HS.

<A HREF="http://service.futuremark.com/compare?2k1=7454540" target="_new">Yay, I Finally broke the 12k barrier!!</A>

 

[jclw]

A word of caution:

There have been many people that have tried to remove their IHSs (on Northwoods) and have ended up with a broken processor. For those that did manage to remove the IHS there was not that much of a difference in temps.

The general consensus on most overclocking forums is that it is best to leave the IHS on.

*Dual PIII-800 @900 i440BX and Tualeron 1.2 @1.7 i815*