Lapping

I know that ive seen a bunch of different dirrections in the posts here for lapping a heatsink, could some one point me in the dirredtion of the "best" way to do it. Thank You.

All articles are the same IMHO.

Get:

Waterproof Sandpaper (You'll be using it wet, to prevent scratches in the heatsink, or heavy scratches I should say).

Grits Needed: (2) 200, 400, 600, 800, 1000, 1500, and 2000 (Optional for a very clean mirror finish).

A piece of glass. You want to lap it to be flat as possible and glass is as flat as it can get!

Instructions:

Get a square container for the water and sandpaper. Soak the sandpaper for 15mins, before starting the lapping.

Since the heatsink has clips, you don't want to be sanding them. If you have a small piece of string (I used dental floss), tie the clips so they are held up, at the end you can cut them.

Get a towel that you wouldn't need to probably throw away, lay it on the floor. Put glass ontop of the towel. The towel is meant to support the glass, and the sandpaper color will go into the water and stain the towel so make sure you don't use something you like

If your fan has a fan which is detachable, I would recommend you taking it off, if it isn't then becareful lapping and don't get the fan wet.

15mins Later

Lapping Starts

Take the 200 grit out of the water, then get the heatsink and skim the base of it into the water, or if you don't have a fan on it just dip it in.


You must now put the heatsink on the sandpaper and move the heatsink back and fourth. The start is very critical, the reason I asked you to have two, 200grits is because the start may take a little long. Every 30secs - 1min you should wet the heatsink and the sandpaper. The purpose of wetting is to prevent heavy scratches, and to stick the sandpaper onto the glass.

Remember you shouldn't be applying pressure to the heatsink or you might be pushing into different areas of the heatsink which then grinds on the sandpaper and the other parts of the heatsink don't. You should be gently pushing the heatsink back and fourth by its side.

Keep doing that until you see the base is 100% flat. YOU MUST NOT MOVE ONTO A new grit until you finish the start, if you do the rest is a waste and you'll end up with an uneven layer.

When the start is done you should start seeing a flat layer with a minor mirror.

The start is very long, so don't think you maybe doing it wrong. Once you get it flat, move onto the next grits. The next grits, I recommend you should spend 10-15mins with each. At 1000 grit you should start spending like 30mins with the lapping. Although I recommend you finish it with 2000, if not at least 1500.

Threw the process of lapping you should see the sides of the heatsink coming in slowly as they are being sanded down.

At the end when your done, make sure before you use the heatsink you clean both the core and the heatsink (Especially), with isopropyl. Reason? It's because the metal is still under the heatsink so you need to take it off, use a magnifying glass to determine whether its gone or not too. While your at it, clean the fins of your heatsink fan if you can.

Snip the floss holding the clips, reattach the fan.

Your done.

Hoot's Corner

Here is a piece of yesterdays goodies

Originally written by Hoot
Still, some people take a simple process and elevate it to some form of overclocking pilgrimage. You don't need 5 grades of sandpaper and several polishes. You do need some common sense technique though:

Start with some 600 grit Wet-or-Dry paper. Keep your paper wet with slightly soapy water always. Rinse the paper every few minutes with that water to flush out the suspended particles. Depending upon the severity of the finish, you want to adjust your downward pressure and speed of stroke.

At the start, use firm downward pressure and a slow circular motion. After you get the surface flat, use less downward pressure and faster circular motion. At the end of the process, you should be using almost no downward pressure and very fast motion.

When you're done, rinse the paper with clean warm water and sit it somewhere to dry - yes, you can reuse it many times. I generally use a fresh sheet at the start and, when I get to the final phase, I shift to an older used sheet. As the sheets age with use, they become less abrasive and an excellent polishing surface.

Speaking of surfaces:

You are not grinding optics for a Laser Based Defense Weapon here. No need to find a sheet of glass or other calibrated flat surface. All I have ever used is a flat, clean Formica counter top. The waste water/slurry cleans off very easily and you get points for cleaning up your mess. If this process takes more than 15 minutes, then you really need the practice anyway.

Click to expand...

lmao You are not grinding optics for a Laser Based Defense Weapon here

Yodums said:

Grits Needed: (2) 200, 400, 600, 800, 1000, 1500, and 2000 (Optional for a very clean mirror finish).

Click to expand...

Very optional. There was a really good article that went somewhat deep into lapping for a mirror finish, or just lapping for a smooth/flat surface. The test showed that anything over 800 grit is basically a waste, and even the extreame 1500grit and beyond actually started to raise temps.
They figured that a mirror finish removed all the microscopic "mountains" and "valleys", so any heat sink compound applied would just be a layer between the cpu and h/s.
With a flat smooth surface (not mirror finish), you've basically made the microscopic "mountains" into platues, so you still have valleys in between. And if you used a nice, flat, even surface, then all the platues sould be of equal hieght. So when you place your h/s on you cpu, the platuaes will make contact with the cpu surface, while the h/s compound fills in the valleys.
So theres no need to have a mirror finish, just a smooth flat finish. Flat is the key.

If you want to improve you lap job, take a fresh sheet of paper and use your favorite thermal compound for the lube on the final lap. This will put some thermal grease in the microscopic surface imperfections along with the abrasive and heatsink residue.

Regarding a flat verses mirrored surface, the higher the polish of the surface, the less thermal compound will be required. Even the best theraml compounds have thermal resistance. Reducing the amount of thermal compound will result in lower temps.

I guess I'll throw my two cents in here too. I definitely have to agree with the comment about not grinding a mirror for a laser defense system. Try to always keep the true objective in mind -- creating a flat, smooth surface on the heatsink. If you can see yourself in the heatsink when you're done, good for you, but that is NOT the goal. (And you have way too much time on your hands.) Please do not, I repeat, NOT use any kind of polishing compound to achieve a higher shine. The compound will get into all those microscopic nooks and crannies and it won't come out. You want thermal compound in there, not polishing compound.

If your heatsink is in very bad shape to start, you might want to begin with something rougher than 200 grit. I'd say start with 200. If, after a few minutes, you can see the high spots wearing down but it doesn't look like you're making a lot of progress, drop back to 100 grit or maybe even 60 grit. If you want to go beyond 200 grit to get a smoother surface, I don't think it will hurt anything. I went up to 1500 grit when doing mine. Can't say if the 1500 did much of a constructive nature, but it did make things shine real nice. (Which, of course, is NOT the objective.)

Use of glass as a base is not a bad idea, but not required. Just be sure to use a flat, smooth surface. Any irregularities on the surface of the base will be transferred up through the sandpaper and could make the process a lot more difficult. I speak from experience here.

Do your best to hold the heatsink flat against the sandpaper when sanding. Due to the motion of sanding there is a tendency for the heatsink to rock a bit. Rocking will wear down the edges faster than the middle and result in an uneven surface. The lower (closer to the sanding surface) your grip on the heatsink, the less the tendency to rock.

Finally, go to the Artic Silver web site and find their instructions for applying thermal compound. Even if you don't use Artic Silver, the instructions are still worth reading.

WOW, there are way too many different ideas on what is right with this whol process. I think that Ill try it with the glass and start with 400 grit and end up with 1000 grit. Anyhting beyond there seems like a waste to me, and besides my heat sink is not that bad.

The one thing that I am still trying to figure out is what is the idea behind the pressure. Seems to me that you would not want to apply any, and just allow the sand paper to do its job. But hoot says that pressure is important. What does everyone think of this. Ill try it later this week and post my results. Thanks for the help guys.

I think you're seeing different instructions because of different circumstances and sometimes different objectives. Everybody does what they think will work best in their situation. Just be sure you understand what you're trying to accomplish and then work out a strategy. If you're lapping your heatsink because it's what all true overclockers do, then you don't really understand your goal and you would be well advised to abandon the whole thing. If you're lapping your heatsink because it isn't flat or there are scratches on the bottom or there are significant machining swirls on the bottom or the manufacturer stupidly put some oil on the bottom so it would look pretty when you took it out of the package (thank you, Alpha), then you have a clear understanding of your objective and should be able to wade your way through all the advice and find a path that will work for you.

The whole thing with the multiple sandpaper grits is for faster removal of the surface. The first things I lapped were my Celeron CPUs. They are the old style with the integral one inch heat spreader. The corners on these were way higher than the middle and I had to remove quite a bit of material before I got it close to flat. I wasn't making enough progress with the 200 grit so I dropped back to a coarser sandpaper to speed things up. The 200 would have done the job, but it would have taken hours. The coarser sandpaper left a lot of scratches, so I moved up through a series of different grits until the surface was pretty smooth.

As far as pressure goes, the amount of material removed depends on the grit of the sandpaper and the pressure applied. You could use 40 grit and very little pressure and all you'd end up doing is scratching the surface rather than removing a lot of material. You can remove more or less material by varying the pressure. Start with light pressure until you get a feel for it and see how much material you're going to have to remove. If you need to remove a lot of material, the job will go faster if you apply a bit more pressure.

Well, the method I just told you, I ended up with no scratches but very small minor ones which you couldn't see without a maginifying glass.

and waht does a flat surface improve? whn i buy a HSF,
the HS isn't already flat?

Starfox said:

and waht does a flat surface improve? whn i buy a HSF,
the HS isn't already flat?

Click to expand...

I think it's flat fox, question is. Is it flat enough.