Sign in with
Sign up | Sign in
Your question

CPU Lapping - Wet/Dry? And When Is It Enough?

Tags:
Last response: in CPUs
Share
April 5, 2008 4:42:16 AM

Hi,

so I've been reading up on lapping guides and decided i'd give it a test run on an old Athlon 64 3000+.

I started with 320 grit sand paper, then 600 grit, eliminating the aluminum layer entirely to reveal the copper beneath it all across the IHS. Initially, i moistened the sand paper with a little water, but this became a big hassle as residue started to clump up and make the lapping process akward, so i stopped the wet method. Is there a difference? Should I just stick to dry?

Also, I'm wondering what the criteria is for a successful lapp. I've read that you can check to see how flat your CPU's IHS is by putting a razor blade across it and holding against light, and i've done this on my test run, but a little light still peeks through, although nowhere near as much as before I started lapping.

Are successful laps those that have no light peeking through between the razor blade and the IHS? Should I go back to 320 grit or 600 grit to make it entirely flat? I plan on getting 1000 or 1500 grit paper for the final lapp on this test CPU.

And what about motion? I wad doing straight lines back and forth, trying not to apply pressure onto the CPU. I've read that circular and figure 8 lapp patterns might work too?

also, I was doing this test run because I plan on getting a Q6600 which I might possibly lapp if the need arose.



Thanks

edit: btw, this is what a lapped CPU looks like:

More about : cpu lapping wet dry

April 5, 2008 6:25:04 AM

AH, WFT kind of zombie coin is that? haha JK, that guy on the coin just freaked me out a bit.

I haven't lapped before but I feel like I should say this. I do safety stuff in laboratories so I know little about metal toxicity. Metals are bad to put in your body so dry sanding metal isn't a good idea on a health standpoint. Metals are quite reactive on a biochemistry standpoint and really mess with DNA and proteins. Putting copper particles into your lungs isn't exactly good for the health.

Wet sanding will help prevent the particles from going airborne and finding their way into your lungs. That is just something else to consider in your sanding adventures. Sorry to be the party killer here but maybe you weren't aware of this stuff.

"Health Effects

Copper is necessary for good health. However, very large single or daily intakes of copper can harm your health. Long-term exposure to copper dust can irritate your nose, mouth, and eyes, and cause headaches, dizziness, nausea, and diarrhea. If you drink water that contains higher than normal levels of copper, you may experience vomiting, diarrhea, stomach cramps, and nausea. Intentionally high intakes of copper can cause liver and kidney damage and even death. Very young children are sensitive to copper, and long-term exposure to high levels of copper in food or water may cause liver damage and death. Copper is not known to cause cancer. We do not know if copper can cause birth defects in humans. The seriousness of the effects of copper can be expected to increase with both level and length of exposure."
April 5, 2008 6:36:37 AM

I assumed it wasn't the healthiest thing to inhale... So water it is, maybe i'll just use more sheets if they get soiled too easily/fast or have some good ventilation / face mask.
Related resources
April 5, 2008 6:41:25 AM

Water? I thought people used oil.

Edit: Oh I see graysky used soap water. Nevermind!
April 5, 2008 7:11:43 AM

SpinachEater, Do you go outside without a mask? He is not doing this everyday so take a deep breath, with a mask of course.

I think he will live.

All jokes aside, I am interested in what you do for a living. It sounds like it might be interesting.
April 5, 2008 7:45:51 AM

I've always used vegetable oil and a circular motion. Just the same, your results are impresive. That with 600 grit? Great!!
If you get the hsf that good, your biggest delema may be getting a small enough quantity if TIM to have be effective.
Beautiful work.
April 5, 2008 7:53:54 AM

I actually lapped BOTH my HSF and my CPU. I read many guides online also. You are supposed to use water and a small drop of dishwashing soap. Like actual dishwashing soap, NOT handsoap. Also, since your doing your cpu, why not do the HSF? It wont cost you anything, other than time of course.

You can PM me if you would like and i can lead you through the whole process. :) 
April 5, 2008 9:02:21 AM

Gotta keep in mind lapping is not a solution for new guys or mild overclockers. The purpose is not to fix an overheating problem, but still to improve cooling efficiency for maximum overclocking performance. Many quality heatsinks come from the factory with mirror finishes and need not be touched. Anyone looking to solve their overheating problem should consider a better cooler and reading instructions on how to apply thermal compound, before attempting to think about lapping the processor. Just don't want to put this idea into n00b heads, Thank you.

As to when enough is enough - once the plate is smooth and looks like its polished to a mirror finish, hit it with some rubbing alcohol to ensure uniformity. If you did not use a straight edge to do the sanding, like a piece of glass - use a straight edge now to check for any significant low spots that may have been ground into a little too much. Unlevel surfaces can be worse than just leaving the stock surfaces alone.
April 5, 2008 10:17:18 AM

endyen said:
I've always used vegetable oil and a circular motion. Just the same, your results are impresive. That with 600 grit? Great!!
If you get the hsf that good, your biggest delema may be getting a small enough quantity if TIM to have be effective.
Beautiful work.


heh, that's not my pic, that was just a reference pic i found on legitreviews.com. The test CPU I used 600 grit with has decent reflection, but the coarsness of the paper really shows. I think i'm going to do what rockbyter suggested and scrounge for a piece of glass to lapp on further.

I assume anything beyond 1500 grit is overkill? I also read that it's not so much the mirror finish or shine that's important but the flatness. And SuicideSilence I might take you up on that offer :) 

Finally, what kind of temprature improvements have you guys noticed after lapping?

I've read everything from 10 degrees drops to some rare increases in tempratures... :ouch: 
April 5, 2008 10:42:47 AM

For me, the difference was about 5-7 degrees in C.
April 5, 2008 12:55:43 PM

And it's just not enough to be worth the effort. Oh, I'm lazy.
April 5, 2008 1:27:07 PM

Zorg said:
And it's just not enough to be worth the effort. Oh, I'm lazy.


:lol:  . o O (I feel the same way)

Also... I know I'd be pissed, after doing all that work to get a mirror like surface, and later on find a scratch on the bottom surface.

I've already seen a dot (hole on the P4 IHS) on the bottom of my old Zalman 7700 copper that was imprinted on the bottom of the HS. In the past I've seen scratches appear on the bottom of aluminum HS.

So I'll be damned if I spend 1-2 hours making something look rather shiny and flat to find out later it has a bunch of scratches from either putting it on or taking it off, which basically defeats the purpose of lapping both parts.

And besides that... that is what thermal grease is for.. to fill in the gaps. :sweat: 
!