ok i have a lot of knowledge about the automotive field, and fabracation field, and some in the computer field. i worked for a machine shop for car heads. when we milled them {made them flat and smooth} we had a bench with a 18" round polish/grind stone in the middle that spun about 1000rpm. it could not only shave the heads but give them a close to polish look when done right. my idea is to take a thick peace of glass about 1/2" to 3/4" thick and say 8-10" round, mount it flush in a smooth workbench type setup. using an electric motor about 1' away so there wont be a magnetic field close to object being lapped. i can make a controller to set the rpm, or even a small gearbox to reduce the rpm. then if i have like 2000 grit wet/dry mounted on the glass i could lap cpu's and HSF's in no time with a polished finish.. dose this sound plausible? or like something that could work, or dose it sound like a waste of time.
i would just use 2500-2000 sandpaper and hand lap it to be safe. do the heatsink too. make sure its all level and not sanded too much on one side otherwise the heatsink will not get direct contact.
my idea is to take a thick peace of glass about 1/2" to 3/4" thick and say 8-10" round, mount it flush in a smooth workbench type setup. using an electric motor about 1' away so there wont be a magnetic field close to object being lapped. i can make a controller to set the rpm, or even a small gearbox to reduce the rpm. then if i have like 2000 grit wet/dry mounted on the glass i could lap cpu's and HSF's in no time with a polished finish.. dose this sound plausible? or like something that could work, or dose it sound like a waste of time.
May work. Have a lens grinder than a family member gave me (her husband was an optician) but never bothered to use the device (more interested in it for polishing a mirror for a telescope though).
Lapping machines could make the process so much faster, and with the proc, so much safer (can just take off what's needed, all as flat as possible).
usually when they mill heads they make sure everything is leveled! cause you know what will happen when you slap the head on the block!
yup exactly, it works for a bit if your lucky befor it go's boom.... lol... i have seen that alot in the auto field. i have access to just about everything i need, and my room mate runs a fab shop that has an ezrouter and cnc lathe, so the only thing i would realy have to buy out of pocket would be electric motor, control box, and glass. but that shouldnt be to much. think i am going to start planning and designing this thing just to see if it is something i can build and price the materials that i will need. i love makeing things and modding them so you never know, it might just work.
usually when they mill heads they make sure everything is leveled! cause you know what will happen when you slap the head on the block!
Um, the gasket will make up the difference? It takes a LOT of stupidity to mill a head that badly... Processor? That'll be sight to see if this guy pulls it off. I'm rootin for ya, OP!
Especially since you're in Gainesville! I lived there for a few years hehe.
2000g even with a power device will not stand up to the load and be toast long before most HS will be flat.
Emery, on the other hand is designed for metals and would flaten the surface much quicker and last far longer.
Dependent on the surface quality of the HS surface...
1. 320g or fine first
2. 600g emery
3. Perhaps 1200 or 1500
4. then what ever max grit you chose.
BTW, my first sheet of 320g emery has lapped at least 7 or 8 copper HS for me and still cuts well... 2 Ultra 120, 1 TRUE, 1 IFX 14, several BT
Also, the automotuve refinishing type wet/dry cuts better and last longer with water mixed with dishsoap like you might formulate for washing dishes. Also, the likes of WD40 or LPS1 also works well to lube the wet/dry paper for longevity but is a bit more fun to clean off the HS.
Further more, like the issue with many Thermalright HS which were milled with excessive clamp pressure which causes the bowing effect. That will be a challange to the process to avoid.
It should work i did somthing like this with my friends computer and it worked pretty good. I would just finish it off by hand to enshure quality and levelnes.
Originally Posted by The Duke
2000g even with a power device will not stand up to the load and be toast long before most HS will be flat.
Emery, on the other hand is designed for metals and would flaten the surface much quicker and last far longer.
hmm, guess i will have to look into using emery then. i am thinking that if i use a small motor and a gear reduction i can reduce the rpm of the lapping wheel, at the same time i can reduce power going to the electric motor to lengthen its life span. the gearbox will basically make more torque. now one issue i have is should i make an aluminum table top, steel, or just one with a good flat rubber mat to help aginst static build up? i am thinking that even at 100-200rpm would be more than enough..
i must agree, cpu spinning at 220 and comeing unclamped would not be a good idea. not only would i have to replace someones 100+ cpu but what if i was in the path of flight.....ouch... i have been messing with coral draw x4 trying to come up wit a basic design but man this is a slow process lol. i am actuly thinking about makeing desk with a 2ft square recess glass top and the 10" lap in the center.. off set it by the thickness of the sand paper and make it adjustable so if i try a new paper i can adjust it to it. sill in the process of blueprints and figures but we shall see if maybe i can do this.
Originally Posted by The Fryer
i must agree, cpu spinning at 220 and comeing unclamped would not be a good idea. not only would i have to replace someones 100+ cpu but what if i was in the path of flight.....ouch... i have been messing with coral draw x4 trying to come up wit a basic design but man this is a slow process lol. i am actuly thinking about makeing desk with a 2ft square recess glass top and the 10" lap in the center.. off set it by the thickness of the sand paper and make it adjustable so if i try a new paper i can adjust it to it. sill in the process of blueprints and figures but we shall see if maybe i can do this.
I was thinking about this last night... Maybe it'd be a good idea to salvage a few burnt mobos and use the CPU locking mechanisms as "holders" that you can interchange as need arises. That way, if it pops loose, you'll have a larger thing flying at you! (or it'd make it more secure...)
actuly useing the ziff sockets would be good if i could make a handle so i could hold it, but it could also make the lapping unlevel if not done right becuse you only realy want to use the weight of the cpu to push it on the sand paper.
This is an older thread, you may not receive a response, and could
be reviving an old thread. Please consider creating a new thread.
Overclock.net
A forum community dedicated to overclocking enthusiasts and testing the limits of computing. Come join the discussion about computing, builds, collections, displays, models, styles, scales, specifications, reviews, accessories, classifieds, and more!