• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Lapping question

KingGheedora

Diamond Member
Just want the summary says. What's the cheapest piece of glass or other near perfectly flat surface I can use to lap a heatsink and CPU?

 
Go to your local hardware store & ask for a piece of scrap glass. I think of the 4+ times I've bought glass for this purpose I've been charged MAYBE once and it was like $1-2. Most places will just give it to you since they were just going to toss it anyway.

Viper GTS
 
Scrap glass huh, wish I had asked for that. I just bought a plate of window glass. I left it in the cardboard frame so I wouldn't cut myself handling it.
 
Sandpaper is really freakin hard to tape down with regular desk tape... does masking or duct tape work better?
 
Soak sandpaper in water for 1/2 hour. Put wet sandpaper of lapping surface and start lapping. Soon, the sandpaper will flatten and stick to the lapping block. Let the weight of the CPU cooler do the work. DO NOT apply too much pressure. Clean lapping block and sandpaper after 50 strokes. DO NOT allow the soft Al or Cu clog the sandpaper.
 
I saw a guide about lapping a few days ago for the first time. What is the point of lapping? Lapping guids talk of heatsink being concav or convex, is that the only reason? That the heatsink does not fit the CPU perfect and therefore cant remove as much heat?

If thats it, why arent the manufacturers just making HSs that fit the CPU?
 
Creating a very smooth and flat surface cost more $. If you want to optimize heat transfer between two mating surface, then you need to lap temp to maximize surface contact area.
 
You should lap both the HSF AND CPU to ensure both surfaces are flat. Don't go beyond about 800 grit because after a point smoother is NOT better.

You want metal to metal contact on the microscopic ridges left by the lapping process with compound filling the voids. If the surfaces are too smooth then you don't get any metal to metal contact because the surfaces "float" on the compound itself.

 
Originally posted by: Henny
You should lap both the HSF AND CPU to ensure both surfaces are flat. Don't go beyond about 800 grit because after a point smoother is NOT better.

You want metal to metal contact on the microscopic ridges left by the lapping process with compound filling the voids. If the surfaces are too smooth then you don't get any metal to metal contact because the surfaces "float" on the compound itself.

QFT

a perfectly flat unpolished surface is what your aiming for.
 
I'm about to lap my Q6600 today and I need to ask, do I really need to wet the sandpaper? Can't I just tape it down and get to work?

and if I do...how? Just leave a little on the sandpaper? what does it do?
 
Originally posted by: Aflac
Sandpaper is really freakin hard to tape down with regular desk tape... does masking or duct tape work better?

I went with silver duct tape and it worked well.
 
Originally posted by: Henny
You should lap both the HSF AND CPU to ensure both surfaces are flat. Don't go beyond about 800 grit because after a point smoother is NOT better.

You want metal to metal contact on the microscopic ridges left by the lapping process with compound filling the voids. If the surfaces are too smooth then you don't get any metal to metal contact because the surfaces "float" on the compound itself.

Hmm, I'm curious. What do you mean by "float" and what causes it? Like does the thermal compound become more adhesive with a flatter surface? Or do the metals not like each other? Or is there like some electromagnetic force thing going on?

Just wondering because I always read these threads and people claim to get whatever they're sanding to be 'perfectly flat', but at the grit sizes they use I'm always like "That's not completely flat..." since microscopic scratches would be visible at microscopic levels; but if there's a point where flatter is worse, then there must be some sort of 'optimum' grit level to be used, but I've read people using grits ranging from 300-2500 - was there science behind your 800 grit maximum? Also, what if both the cpu and heatsink were completely flat, would there be a need for thermal compound anyway?

And then there's the case of the possibility lapping at an angle, where you'll probably get good contact on one side but then on the opposite side, there wouldn't be any contact at all, which is one of the reasons why I haven't tried lapping yet.

I think I confused myself.
 
He means something like this....

A) Floating

....Heatsink....
___________
TTTTTTTTTTT
........CPU......

B) Metal to metal

.....Heatsink.......
\/T\/T\/T\/T\/T\/
/\T/\T/\T/\T/\T/
........CPU..........

Where T= thermal paste
Flat lines in A represent perfectly lapped components
\/ in B represent visibly flat heatsink

Most people tend to use too much thermal paste - so there would be thermal paste between the ridges on B which would otherwise be touching. Hope that helps visually. Its late, good night 😀
 
Has anyone seen a link to where they have done any real dimentsional measurements on this stuff? I played with lapping years ago and there was a slight improvment, but I think it was more from removing material between the heat sink and the core. A chunk of wet sandpaper on glass or granite is a very crude method for lapping and you could end up with less surface contact than you started out with. You're always going to have some amount of roll-off at the edges that way, not a plane surface.
 
Back
Top