Mad lapping action! (Lapping Tips) Pics too

[GiZzO]

Spent the last 3 hours lapping a HS for a friend. Tried a little something different this time using body repair rubbing compound, and wow did it step it up much further then i was able too with 2000 grit sand paper. Leaves an amazing finish, with 2000 grit your finish doesn't really come out so mirrored its kinda cloudy, and you can see very small scratches if you look at it at a certain angle, this stuff will totally remove those scratches, and totally come out like a freakin mirror.

Pic of HS base finished off with body repair rubbing compound

Pic of orginal base
The OZC Gladiator aka PowerCooler PCH137, base is so nasty. I bought quite a few of these, and man the base finish feels like someone tied this heatsink to the back of there car and dragged it 50 miles, its worse then old Global Win HS's, Volcano II which are known to be poor also. Its pretty bad pretty jagged deep grooves, also noticed when i lapped a few of em took alota excess time cause of its unevenness.

Another Pic

 

[McCarthy]

My FOP32 was at least that rough if not worse. Amazing really, the cost to do a better job can't be much if anything.

When I lapped mine, I decided to lap the itty bitty retail heatsink from my Celeron 300A. Don't know what alloy they used there, but it was a lot harder than the aluminum Globalwin uses! FOP took 1 sheet, no biggie. Wore out 4 sheets of wet/dry (wet) sandpaper on the little retail before I gave up. Wasn't that rough anyway, just hard.

Forget if I had a point. Nice finish

 

[Swanny]

Hey, can I trade you my mom's antique miror for one of those?:Q

 

[Mikendi]

Nice finish there anyone remember the flame wars over lapping Celerons? There were some heated battles then

 

[Swanny]

Lapping Celerons? How would you do that? Just lap the metal thing on top?

 

[mothman]

You might want to check this article ....

Why heatsink polishing might be a bad idea

 

[MrThompson]

mothman has a point. Hand lapping will always create a convex surface compared to machine lapping. The question is, how bad is your HS. Sometimes hand lapping is an immprovement.

Polishing compounds are a bad idea. They stay in the microscopic surface imperfections where the thermal compound should be. A better solution is to do your final lap with a fresh sheet of 2000 grit and Arctic Silver as the lube. This way you have Arctic Silver mixed in with the abrasive particles and fine metal particles from your HS. Take a look here for a visual representation of the thermal interface.

 

[GiZzO]

It won't come out convex, if you lay it on a thick peice of glass. From all my experances so far lapping has had its positive results. Thats an interesting arcticle though i remember kinda reading something about that before. You could kidna promote the paste of ozzing off the sides by rubbing the hs around on the CPU core kinda pushing and wiggling getting the excess to squeeze out, you shouldn't be putting on that much in the first place . The CPU core is somewhat corse too, both surfaces don't need to be. Alpha HS the finish on there bases are pretty smooth, GW probably takes in the factor you might be using there supplied silicone paste which is pretty thick stuff compared to AS II which is very fine, and probably for the ppl that think they have to use the whole packet for 1 cpu .

Back in the days with my 366 i was pretty much maxed out at 560mhz, did the same lapping/polishing on both the HS base and the CPU core down too the copper, and was able to push it into 616 perfectly stable, it would always choke at the windows screen before. I went through alot trying to push it further resited it many times pumping the voltage too 2.8v Lapping the cpu core down the the copper, not untill i lapped the HS i was able to hit 616 no problems.
Id think as long as you don't have too much excess paste you should see great improvemnts.

 

[MrThompson]

I have always had positive results lapping too. Both cores and heat sinks may not be flat from the manufacture. Ask a qualified machinist about hand lapping and they will laugh. Tell me you have not felt at HS or CPU rocking a bit while you lap it. Yeah, that?s right it happens. Anyone who is honest and has lapped a HS has felt this. My statement on hand verses machine lapping stands.

Here is a pic of an AMD CPU that is convex. Notice the distribution of thermal compound at the edges of the die. A Danger Den Maze 2 with a four bolt hold down was used on this chip so the possibility of a HS rocking on the core to produce these results is minimized. Lapping would help this core. I usually lap cores until the etched codes are gone. If you see copper, you loose.

 

[GiZzO]

True but on a HS a little rocking won't hurt cause if the edges are rounded a little doesn't really matter, the center of the HS will be pretty flat though. Never went to the trouble of lapping the new generation CPU cores with the smaller cap, i would imagine it would pretty tough to keep that small area flat against the sand paper, last cpu i lapped was that 366 ya know with that big fat nickle plate, which back then you did wanna go down to the copper, it was pretty easy to keep those level cause you got a big area to balence on.

 

[Mday]

if you lap either the hs or processor cap (of the celeron originals) use artic silver 2.