View Single Post
Old 08-09-2007
jcgamo88 jcgamo88 is offline
500W User
Join Date: May 2007
Posts: 50
Thanks: 0
Thanked 0 Times in 0 Posts

Sorry, took me a while to get my lazy butt to do this. The images might be a bit large =\

I have a 12cm fan mounted to the power supply just above the HS feeding it more than enough air and the proc still idles at 43C and then loads at 59C. Sometimes after a load it likes to then jump to 63C and stay there until I hit it with subzero liquid from my compressed air can.

I also don't have the mounting bracket or a proper HS :s

Here's my setup drying, after I finished:

I tried to be careful about loading the paper (having metal particles get lodged into the paper) so I'd be able to avoid scratches and use the paper at least one more time. Unfortunately I don't think I used enough water and the results were thus:

What I did was to dunk the paper into the basin of water and then place the paper on a piece of glass. the wetness on the glass help prevents the paper from slipping around. I like to call it the reverse "slip 'n slide" effect. :P

Also I wanted to use a lint free cloth (Paper towel is often lint free) that was wet to wipe off the processor each time I submerged the paper again and rinsed the basin. However I totally forgot and that mgiht be another reason for the paper loading.

I turned the paper on end so it stood tall rather than sat wide. I stroked the processor on the bottom 1/4 of the paper about 10 - 15 times with 1 finger lightly touching. Somehow I've mastered a trick whereby using 1 finger, upon each stroke the processor turns a little bit therefore eliminating my need to phsically turn the processor myself 90 degrees every now and then. After about 15 strokes I moved the processor up another 1/4 until I reached the top. Then I'd submerge the paper into water again and repeat twice more.

I was careful to also dump the water from my basin when I switched grits, wipe off the sides, rinse, and refill the basin for the next grit. Unfortunately in the 2 hours of doing this I failed to rinse every now and again and I suspect that is the reason for the paper loading.

I used many grits to also ensure I first removed the nickel, and then to finally polish the copper well. In order of course to fine I used: 180, 220, 280, 320, 400, 600, 800, 1000, 40micron, 25m, 20m, 15m, and 10m.

Afterall was said and done the loaded paper caused the processor to appear scratched up, but still reflective. The purpose of lapping is not only to remove the less conductive nickel on a processor's IHS, but to polish the copper so to remove more of the microscopic troughs and trenches which AS5 or other thermal compounds are used to fill.

And don't worry, no water entered the hole in the IHS (what the ehck is that there for anyhow?!). I was surprised by this. But some of the metal particles actually build up in there also helping to seal it off until I was done and cleaned it out.

In the end the copper was reflective and exposed. I've yet to test the heat again because unfortunately I'm not made of money and I'm running out of AS5 You can be sure thought that when I do test it I'll post here first.

never mind the bits of solder in the background here. Just remnant of my attempt to remove and AGP riser and place it onto an OEM board that has the space, but was cleverly left off to cut costs.
Reply With Quote