• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Lapping Guides

Correct me if I'm wrong, but don't you just have to sand the thing with a high grit sandpaper. It isn't rocket science.
 
Use very fine sandpaper on a flat surface and sand until it's smooth and shiny. When it gets shiny, use an even finer grain sandpaper until it's shinier! And well, you get the idea. Probably not a good idea to buff the surface (the buff medium might interfere with the cooling abilities of the TIM) though.
 
It's not about "shiny" it's about flat, "shiny" never hurts, but the main point it to make the base flat. Depending on how "off" your base is you might want to start with something like 280-320 and work your way up (however 400 makes it most of the time, just might take a bit longer), most people stop at 800-1000 grit, I went up to 10,000 because I thought "why not". All you need is a piece of glass and your sand paper, keep the base flat, some people will say you need to do back and forth movements, some will say circles and others figure eights, I say use what ever works, so long as you keep it flat.

Easy PC Kits has all the stuff you need at pretty good prices.
 
It's not about "shiny" it's about flat
Sanding smooths the surface of the material, making it uniformly flat. We know that light reflects off of smooth surfaces (as opposed to rough where it is scattered in all directions), so more light would reflect right back at you if it was highly polished. What does that mean? A very well lapped base should act as a mirror. Shiny = flat. It's worth noting if it is highly polished but not done so evenly across the entire surface, you'll be able to see distortions in your 'mirror' similar those fun houses at the fair. A warped reflection is bad.
 
You can indeed have a "flat" surface without having a shine, and being shiny does not mean flat, the angle can be very small and still effect the contact area. I have seen (and done) many tests with lapping where you stop at 800 (not shiny at all) or go up to 10,000 grit (mirror finish) with no further effects on temps, the biggest impact is by getting the surface flat, after that, you can polish it if you want.
 
I got a kit of wet/dry sand paper at AutoZone consisting of 220, 400, 800, and 1000, and a piece of glass from home depot and some blue masking tape.

That should work right? I'm supposed to wet the sand paper and use a drop of soap to make it lubricated as well yes?
 
You can, I always found wetting on the higher grit papers to make it harder, as it wants to stick to the paper, so I always do mine dry or use rubbing alcohol to keep the shavings from stopping up the grit, however, like I said, whatever works for you.
 
If your lapping a heatsink with pointed edges you may wish to round them off so they dont get stuck on the paper....

Also wear rubber gloves or you will get blisters from the paper + its probably safer.... and just be careful

I didnt wetsand my cpu but whatever you want...

and make sure to clean with 90+% alcohol or the purpose made cpu cleaner...
 
You can indeed have a "flat" surface without having a shine, and being shiny does not mean flat, the angle can be very small and still effect the contact area. I have seen (and done) many tests with lapping where you stop at 800 (not shiny at all) or go up to 10,000 grit (mirror finish) with no further effects on temps, the biggest impact is by getting the surface flat, after that, you can polish it if you want.

To touch on this point a little bit more...

What is trying to be controlled is the thickness of the TMI required. TMI has a thermal conductivity of around 8 w/m*K (or degrees C, same thing in this case). Copper is ~400, while aluminium is ~200. It should pop out immediatly that 0.010" of your thermal paste is like adding a 1/2 inch block of copper, in terms of linear resistance. Flatness in the GDT sense is the deviation from two parrellel planes to the nominal surface. Surface finish is a measure of the defect or groove depth on the surface. If you've ever lapped something by sanding in one direction, turning it 90 and moving to a finer sand paper, the grooves your leaving is the surface finish. While it looks nicer, those grooves are well under 0.001" well before you pick up the "mirror like" shine. Changing the thickness from 0.020" to 0.010" or even 0.005" will make a difference in temps, of course it will never do more than the theroetical limit from the thickness of your TMI. A much worse problem is when people attempt to spread TMI over the chip or heatsink instead of using some form of the "dot" method. There were some GREAT youtube videos some guy posted in [H]ard's TMI review comment thread that showed the airbubbles that got trapped when you used that method.
 
Back
Top