Making CU -and AL shiny. How?!
 

I've tried to make a mirror-like finnish on a aluminium-block i reasently made, but without results. I lapped it steps, 120,320,400,600,1000.
But I can't get that shiny finnish! Same with copper!

How do you guys do to make them transform into a mirror!?

Jewelers rouge...

Also called polishing compound. You can get it at ace, along with the biggest softest buffing wheel you can find. What you have done is lapped the surface very flat and smooth, now you can take the next step and work your way up the polishing compound grits. They arent really rated in grit, but I would advise working with white compound first, then up to the highest color grade (red)? Stay away from emry as it will tend to eat more than you want to...

What you do is slap the buffing wheel in a drill, or drill press, or get a bench grinder buffing wheel. Then put some rouge in that spinning beast and buff till your heart is content.

...then clean off the residue with acetone and/or rubbing alcohol, as this compound will affect the TIM, if you leave any.

Is 'AutoSol'(car shops) a universaly available metal polish?, or 'Brasso'?...

I read somewhere that ArticSilver can pit metal if it's not left 24hours after lapping/polishing for oxidisation to build up a protective layer, is that right?...

Getting it shiny and keeping it shiny are two different things :D . does anyone know if BladeRunner laquers the CPU side of his blocks?. it would bugger up the TI would'nt it?...

no fellas, thats not the way its done

dry lap with the finest grit
an informative thread is here
and another here (the posts by N8)

do not ever use a buffing wheel or any such 'tool'

I think BladeRunner knows better than that!

The copper will oxidize if left exposed to ambient air. I don't believe that AS would have a detrimental effect on the copper, in fact, I would use it as soon as possible, after lapping. Once polished, and cleaned, I'd heat up the surface with a hair dryer, or something, to make sure that the cleaning fluids evaporate completely.

The rest of the bottom will oxidize, but who cares?

Nice. I've seen that (h) link before. (had to hack my way back in!)

I agree, the buffing wheel is a very bad idea.

I think we need (another!) lapping article. N8 has lots of good info, but we need something that can be reproduced at home (or at least nearby!).

Kind of weird that this came up today. I have an old one page article on copper cold plate lapping that I put up exactly 3 years ago today. One the cold plates I was working on was for dual pelts for a slot 1 processor. It is somewhat dated considering what people are doing today and the pictures are not that clear since they are screen-caps from a camcorder. The info from that page was related to 'at home' techniques, since I only did that stuff at home back then. In fact, my whole site is pretty out of date. That's what happens when you have to work full time, have a family, and constantly have side projects (computer+household.) The web stuff isn't top priority anymore.

I have been thinking about a current article using both 'at home' methods and more advanced materials and equipment, but I haven't had much time with a 3 week old newborn at home.:dome:

I'll have to see what I can throw together.

Here is that old page for now: http://www.wsu.edu:8080/~i6735189/cpu/coldplates.html

Congrats N8!!!

I had to ask! :D, he uses 'T~Cut' car paint restorer cause it can be sprayed with laquer without any special cleaning. Can you get T~Cut in US and Europe?...

I'd like to see an article on the best way to lapp different things, how tight a circular motion is best, how to avoid the edges rounding, ect. Also an article on lapping chipsets(NB & GPU), I could'nt get a hit anywhere searching for this!!. How many VIA or NVidia chips you seen that are'nt dipped badly into the middle?, even the Cu/nickle~plated covered Ti cores are!.

I usualy us a mirror for lapping HS's, an oldish one(not antique but not the paper thin crud you get nowadays) with thick glass to lay the paper on. when I lapped my NorthBridge I got some scraps of 6mm glass cut into 1.5 and 2inch(1.5 covered the chip nicely & did'nt hit the Caps & resistors[to much!]) squares then covered them with W&D. I'd do my GPU like this to before I H20 it.
One thing I've learnt is not to be scared of using a course paper to start with if there's a very uneven surface(chipsetHS's) to start with, you can loose patience and start to loose the 'nice circle' motion if you see too little progress.

BillA!, why not use a buffer/tool?, do you mean just on the TI/CPU~side surface?. the body of the blocks OK though is'nt it?...

I dont agree with not using a Buffing Wheel . As long as you dont use it on the CPU contact side it will have no detremental effects whatsoever. In fact, it will further the polishing you have already done by lapping the other sides as long as you dont force the metal into the wheel.

Look at my avatar if you want to see what a buffing wheel can do. Thats my home-brew block.

You'll get a shiny finish, but it'll affect the flatness, that's the problem.

polishing with the AS3 though, seem like a smart move. Dunno. Will have to wait for N8 to come up with something, unless one of us writers can do it first.

Whats the problem? Im not talking about the interface side of anything here... I'm talking about the top, and the sides of the block. Who gives a crap if the top and sides are microscopically concave or convex? Point is, you cant tell by looking at it and it achieves a better finish than lapping alone.

Polishing with arctic silver is pointless. Its not an abrasive, its a thermal interface material. The only thing I can see this helping is if you rub some into your lapped side (CPU side) to kind of "prime" the copper for its actual interface.

Sorry, I WAS talking about lapping the block.

A buffing wheel would indeed work quite nicely, as you demonstrate every day! I've used mine to buff glass, and it works great. It's trickier to do on plastics...

Yea, we don't give a crap about the sides and top flatness. At least I don't. The idea is to not use a wheel on the mating surfaces. The finish on the sides and top are 'usually' just for aesthetics anyway, which some people are concerned about and some aren't.

You can polish with just about any material as long as it contains particles that are as hard or harder than the surface you are polishing, and the size of the particles are fine enough for the surface finish you desire. Copper is harder than silver, so you would not be able to polish copper with Arctic Silver, 'unless' there are other particles in Arctic Silver that are harder than copper. I think the idea of using AS while polishing is to saturate any surface pores with the AS, as you said. I haven't thought hard enough about the viabililty of doing that to give an opinion at this time.

You would'nt be saturating the pores with just AS though would you?, there'd be all that caribundum(?) crap from the paper as well as Cu particles and plain ole dirt. seeing's as you'd have to clean it all off with iso~alcohol it's a waste of AS3...

How about thining the AS3 with Iso~Alcohol to get it into the pores better(if getting the pores full is a problem). I rub some in then wipe it off gently when I mount a block usualy

Good point about the AS3, N8!

What I see is that it is very important to clear whatever was used to shine the block, so that the TIM joint consists of AS3, and only AS3, as much as possible. Any other substance, even in the least (microscopic) amount, will reduce performance.

The acetone wash, followed by a hairdryer/heatgun drying seems to be best, as far as I can see, right now (but hey, it's still early!).

I thought about mentioning the contaminants issue, but like I said, "I haven't thought hard enough about the viabililty of doing that to give an opinion at this time." ;)

Now that I have thought about it some more, here is a long-winded response.

It would depend on what suface you were doing the final polishing with. If you were using SiC sandpaper, 600/800/1000/1500/2000 grit, you would certainly be contaminating the AS3 with copper dust, which might not be a bad thing, except you may be increasing the electrical conductivity of the AS3. You would be taking a chance of some of the SiC grit detaching from the paper and getting mixed in as well - not a good thing unless it is one of the very fine grit specs. 1000 grit and finer 'may' not cause any interface problems. The SiC grit doesn't come off very easily unless the paper is very used or old, in which case the bonding adhesive is old and worn out.

If you are just doing the AS3 'polish' on a clean cloth, I don't think that would do anything unless there are some other particles in the AS3 that are harder than copper. I suppose it wouldn't hurt anything if this were done AFTER final cleaning, if you wanted to charge any pores and surface imperfections with AS3. However, you stand the likely chance of getting cloth fuzz and particles on the surface and mixed in with the AS3, which is not good.

If you are finished lapping/polishing your surface, I would clean it with soap and water first, then water, then isopropanol (you can buy 91% stuff pretty easily at the grocery/drug store). Around here you can buy 50%, 70% and 91% isopropanol, the other percentage is usually just distilled water. The higher the concentration of actual isopropanol, the better. If you have acetone, then you can use that as well, BUT I would still do the final cleaning with the isopropanol as it leaves the least amount of surface residue. Acetone evaporates so quickly, it usually leaves behind some spotty residue of what it has dissolved - unless you use a lot of it to ensure removal of all contaminants. It will also leach all the oils and dirt out of your skin and redeposit that onto the surface of what you are holding, so nitrile gloves are useful for copious acetone washing.

You can buy nitrile gloves at almost any drug store/pharmacy these days. They are thin and strong, and pretty damn chemical resistant (you can also use them for cleaning around the bathroom, using bleach, changing the oil in the car, etc.) Acetone will eat through them over time, depends on the quality and thickness of the glove.

You can blow the surface dry with canned air, or a blow dryer as BB2K mentioned. I would not try to blow dry it with your mouth, as you will just be breathing moisture onto the surface.

I usually apply the AS3 with my finger (inside a nitrile glove). I like to put on the nitrile glove(s) and then wash my hands like normal with water and soap (with the gloves on), then dump a bunch of isopropanol on them to clean them off some more, and then let them air dry for a minute or use canned air. They are then very clean and you will not contaminate the AS3 application.

*An important consideration - I know my pieces are very flat, so I only put on a translucent smear of the AS3, so I don't follow the 'AS3 application guide' that they provide. Unless you know for sure, you might not want to follow my AS3 application technique (using the glove/finger.)

I have put things together without any TIM (AS3), and find that since the metal surfaces are so flat and polished, they like to surface weld together. The big problem in doing this that I have found is that I must get the surfaces in the correct and final orientation the very first time I put them together, because if I need to shift them a bit, the surfaces stick together and gall, and I need to repolish or sometimes relap them. If they do stick and gall, they are not flat together across the whole surface anymore, and heat transfer is not as good. I have found that a very very thin layer of TIM (AS3) prevents this. It is somewhat difficult to line every mounting hole up properly and never move/shift the heatsink/waterblock/coldplate around, and a very thin layer of AS3 prevents the galling of the metals during slight alignment changes.

One of these days when I have time to do it, I would like to provide some results of direct interface, super polished and flattened surfaces with TIM vs no TIM. It will take some time.....

:rolleyes:

hope you find the time N8
sounds quite interesting

What I have found to work best for cleaning is a spray can of degreaser cleaner for my Paslode Impulse cordless framing nail gun. It's purpose is to clean out the aluminum combustion chamber of the gun without hand rubbing. It evaporates almost instantly. You can get a can at any lumber yard that deals in Paslode merchandise for about $6.

Thank you N8!

We're all looking forward to your experiment results, whenever that is!

Very interesting writing in this thread! N8 and BillA, I have greatly enjoyed all of your respective posts and articles on the TI issue. Perhaps in the future one of you, or both perhaps could put together a comprehensive article on "lapping for dummies" (for lack of a better terminology) in which you could outline basic, intermediate, and advanced lapping techniques and show the differences in degree of flatness, interface quality, and physical effect on performance. I dunno, if I had a better knowledge base on the subject and $$ I might try to tackle something like it but I don't. I really think it would open a lot of people's eyes on this often overlooked subject.

N8 could you explain what galling is?,and how it would ruin flatness?, I understand surface welding,(or sucking :p).

Also what is your take on the silver writing that AMD prints onto the core?, I've had some of it be far from flat,(some little silver 'spikes'), and it can cover a fair percentage of the core area(10%?).I've noticed that it leaves an imprint on the block everytime you mount it!, this can't be good can it?. Is there a way of removing this with a solvant rather than an abrasive?. or is there a need to?. Any idea what it actualy consists of?...

Also what's the best AS aplication procedure?. put it on with finger in glove(sounds best for ease) I've heard of a Cello/Scotch~tape method before but am not 100% sure on procedure. but I think you apply the stiky side to some AS thats spread on glass or something flat to pick up a thin layer, then press this layer onto the core or block(better than core I think, more pores to fill)...

I'd like to see you do an article as well ;) ...

Thanx!...:) ...

PS. what is SiC paper?, I use the wet'n'dry that is indiginous to auto shops :D. I always seem to get a 'black dirt' (like when you lapp ALu) when I lapp Cu with this, this cause it's cheap?...

Galling: like blistering.

SiC: Silicone carbide (water resistant)

I see!, so you need a lubricant in there to hold a vacum kindof. Would Iso~Alcohol work?, or would it leave residue or create airpockets when evaporating?...

Any ideas on the AMD silver writing?...