PIII Coppermine Core ? Lapping it flat?

[cautery]

Related Threads:
Thermal Tape Selection and Thermal Interface Qualities
http://forums.anandtech.com/messageview.cfm?catid=37&threadid=204693
PIII Coppermine and Heatsinks - quality of the thermal interface...
http://forums.anandtech.com/messageview.cfm?catid=37&threadid=203730
PIII Coppermine Core ? Lapping it flat?
http://forums.anandtech.com/messageview.cfm?catid=37&threadid=203731
FC-PGA Heatsink ?Feet? ? how to get a custom fit
http://forums.anandtech.com/messageview.cfm?catid=37&threadid=203734
Heatsinks: Are they Flat? Answer here?.
http://forums.anandtech.com/messageview.cfm?catid=37&threadid=203739
Thermal Interface Materials: Required Thickness?
http://forums.anandtech.com/messageview.cfm?catid=37&threadid=204688

In my ?PIII Coppermine and Heatsinks - quality of the thermal interface...? thread, I told you that the core was not even close to flat. Now, it IS possible to perform a little remedial work to improve the flatness of the core. I must caution you that should you decide to follow any of the procedures in this post, you may very well damage or destroy your processor. Please follow all proper precautions to avoid ESD damage?. Grounding yourself, etc.

This procedure applies ONLY to the PIII FC-PGA Coppermine processor, NOT the Slot-1 version. I highly discourage anyone without the proper equipment and skills to perform this procedure. Performing this procedure on the Slot-1 version is MUCH more difficult and risky and should be avoided. Now with all the disclaimers out of the way, this is how you do it?.. simply really.

Materials:
1 each PIII FC-PGA Coppermine processor.
1 each NEW sheet of 3M brand 1500 grit wet/dry sandpaper.
1 each 8? x 10? piece of quality double thickness (or better) glass.
CLEAN, preferably distilled water. (actually, a high quality, very light grade of lapping oil is preferable)
Several lint free towels?. I use lens cleaning towelettes.

This is the genaeral procedure. There are tons of additional ?anal retentive? measuring tools and techniques that one can use, but for the purpose of this article, you can ?get by? with the above.

Place the glass pane on an absolutely clean and flat surface?. Please make sure you clean off your work surface and remove all extraneous materials. Lapping a $200 plus processor should not be done on a computer workstation cluttered with McDonald?s trash, pop cans, diskettes, Cds, and paper?.

Dampen the reverse side of the sandpaper with a sponge and clean water?.. Make SURE there is no label/price tag on the reverse side of the paper. The damp paper will adhere to the clean glass better, thus reducing the tendency for it to slide around. Do NOT wrap the paper around the glass, or try to tape the paper down, as this will tend to cause the paper to ?bow? up in areas which will radically increase your chances of NOT having a flat lapping surface and causing ?rounding? of the edges of the core?. This would be BAD!!!

Now, place two to three drops of clean/distilled water (lapping oil is preferable since it will reduce the tendency of the core to ?suck down? to the wet paper as the surface gets flatter?. This ?sucking down? may cause a ?hang? which in turn MIGHT cause you to ?skip? the edge over and across the paper? possibly rounding the edge) in a single ?puddle? in the CENTER of the NEW sheet of paper now located on the glass. It is critical that you use a NEW sheet of paper in PRISTINE condition (no scuffs, or creases?.). Any imperfections in the paper will likely affect the quality of the lapping results and greatly increase the chances of you damaging or destroying the CPU.

Place the CPU, core side down, in the center of the water ?puddle?. Place your INDEX finger LIGHTLY in the EXACT center of the exposed side of the processor. (You are trying to locate the force perpendicular to and in the center of the core) As the core is NOT flat (it is rounded), it is critical that you maintain as light of pressure as possible, and keep the core face parallel to the paper. Now begin making small circular patterns with the processor on the paper. Keep them small, and maintain LIGHT pressure. You are essentially ONLY trying to remove the blue coating, and this does NOT take much pressure, or much lapping.

After every 5-6 circles, check your work. Clean the core off, and repeat the process until you have a flat ?silverish? surface having removed the blue protective coating. Do NOT remove too much material, as you may well get far enough into the core that you begin sanding away circuits?. also a very bad thing, unless your intent is to create a $200+ keychain.

It MAY not be possible to achieve a perfectly flat core, as the core itself may well NOT be flat itself, but you should be able to greatly improve the surface. I do not know how much material can be removed before destruction of the core, as I don?t want to sacrifice a processor to find out.

After you have completed this procedure, remember?. 1)The core is no longer protected by the coating?. The edges will be more susceptible to fracture if you apply too much pressure on them, and 2) I would suspect that the core is more susceptible to electro-static discharge (ESD), so mount the processor to its heatsink (with non-conductive thermal compound) immediately, or cover the core with non-conductive thermal compound, or dielectric grease for storage until you are ready to mount it. (Storage in an ESD safe bag should also suffice).

Hope this helps?. Have fun.

 

[PliotronX]

Wow, your indexed threads are incredibly informative! This is gonna help a lot of people for a long long time

I just have one question.. you mention above that lapping the protective coating makes the core more susceptible to static. I was wondering, I'm going to be using some Arctic Silver on this 700E, and although the AS compound isn't electrically conductive in it's normal form, it has been forwarned that it can be electrically conductive in high compression. Such as between the heatsink and core. My question is, will lapping the protective coating potentially pose a problem using Arctic Silver compound to mate the surfaces?

Thanks in advance!


Edited for various spelling errata

 

[Mikewarrior2]

My Lapped P3 works great with Artic silver!



Mike

 

[PliotronX]

Oh oh! I have another question

When applying pressure to the sandpaper, is it okay to be touching those little thingies (are they resistors?) under the core on the reverse side? Also, are the pins really sensitive to static?

Edit: oh man am I out of it right now I lapped my 300A, and I wrapped it up in electric tape so I could apply pressure where it was needed... nevermind. Got all the info I need. Here I go baby! Thanks again!

 

[PliotronX]

Right on! Thanks Mike

 

[TimberWolf]

Clay:

One minor exception to your excellent and concise guidence (from a "former" automotive machinist and gunsmith) -

All hand lapping (of any object) should be performed using a "figure-8" pattern, rather than a purely circular motion. This will minimize the subtle and unconcious tendency to cut more heavily on a fore- or back-stroke. It also spreads the task over a larger area of the abrasive surface.

 

[cautery]

Jeez... I forgot to subscribe to this thread.... Doh, sorry guys...!

TimberWolf - GENERALLY, you are correct.... But, I believe there are exceptions to every rule.... and that the small P3 core is one of them. There is SOOOO little flat area to work with, I personally believe that a super-light "touch", in small, circular path, with finger-tip directly in the center of the reverse of the core chip, yields the best results. It takes so little pressure, and such a little material removal, I believe MOST folks would end up with rounded edges (or worse), if they tried to use the figure-8 pattern over a larger area. With proper care, attention to detail, and a steady hand, either method should be successful however.

One additional reason I recommend "small" circles is also to minimize the chance that you will "stray" into the area near the edges that MAY not be flat to the glass, but curling away from it. It's VERY difficult to get snad paper mounted to a flat surface AND insure that it is uniform... it almost always tends to "bow" upwards somewhere... even a few thousandths makes a hufe diff at these tolerances. That's another reason that I recommend a NEW sheet for EACH core lapped, and that the lapping be done in the very center of the sheet.

Finally, after thinking about it.... I definitely agree that IF you are as anal as I generally am , and choose to use lapping compounds (not generally necessary), you probably SHOULD use a figure-8 pattern, though still small, and directly on the glass.... however, this will also "lap" the glass "machine surface" which would tend to wear the machin surface over time necessitating the replacement of the glass.

Thanks TW....

 

[TimberWolf]

Clay:

I gave the matter some further thought, and decided to devote two 566 Celeron FC-PGA's for "research purposes"; I'd have to agree with your reply points. With LIGHT pressure, circular motion works just fine, with no discernable rounding or beveling of the edges (at 10x magnification, anyway ). Since I have it on hand, I used light cutting oil, rather than water.

BTW, if you're inclined to use ultra-fine lapping compound, just lay down a sheet of 2400 paper, and work on that. It will retain the compound, and protect the plate from abrasion. I have a small granite surface plate, and a vivid memory of the crusty old machinist who would probably come back from the grave to break my fingers if I ever did any lapping directly on it.

To All:

Only PLATE glass is manufactured with dimensional control for thickness and flatness (it's made using a roller process that "squeezes" it to it's final size). The typical window pane laying around your garage or basement is usually FLOAT glass, which varies in thickness, with irregular "waves" in it's surface. A readily available source for plate glass (aside from a "custom cut" at a glass shop) is a quality bevel-edge mirror block. A 9x9" block is ideal, with the added benefit that you're not likely to cut yourself on the finished edges.

 

[PliotronX]

I'm in awe of your knowledge guys

 

[cautery]

TimberWolf - Cool! ...and yes, a light cutting oil is better than water, though the sand paper seems to cleanup/last longer using water.

Good idea on using the 2400 paper... never thought of that. Yep... I used to go ape-sh!t on my airframe guys when they'd put stuff on the machine tables intended for layout work.... same-same with the lapping and machining tables...

Good tip on the glass as well.... I have been using double-paned "float" glass I suppose, but checking mine with a dial indicator shows it to be less than 0.0005" runout.... acceptable.... though this is high quality glass... I need to get some good 1/2" plate glass though.... Put some rubber feet on that stuff, bevel the edges (slightly) and you have a very good surface that won't move around on you.....

PliotronX -

 

[PliotronX]

I just lapped it! I went waaaay slow. Took about 45 minutes. I know, but I just don't have another $200 to get another one Thanks again you all!! It went painlessly actually. Just took a lot of time. I put a small strip of electric tape on the back of the chip (pin side) where I applied very little pressure.. actually I was just moving it along the paper like it was a Ouija board I got a custom fat piece of glass, perfectly flat, followed all the instructions, used some Qtips and isopropyl alcohol to clean it all off, and it's just a beautiful job no sign of beveled edges, razor-sharp it looks like. /me is proud of himself! Thanks again!!!!

 

[TimberWolf]

PliotronX:



<< . . . actually I was just moving it along the paper like it was a Ouija board . . . >>



Great analogy - That's exactly the idea!

 

[shadow]

<< After every 5-6 circles, check your work. Clean the core off >>



if you are worried about esd you should also suggest some static-safe cloths or paper to use when cleaning the core, you sure those lens wipes qualify as ESD safe materials?

 

[CADbad]

Great thread-Can anyone give comparitve temp values before and after lapping?-I would like try this procedure, but only relative to risk/reward decision.
thx

 

[PliotronX]

What makes me sad is this BH6 came out just before the whole thermal-sensing diode craze came about, so I can't compare temperatures. But I can tell you that the mobo sensor is at 26ºC