Whoa! Are you lapping the raised blue/gray slug on the top of the FCPGA? It's part of the silicon chip and it's engineered to minute tolerances. You may reduce its thickness but I doubt you can finish it any flatter then the precision machines that lap it at the factory with a liquid particle suspension. Besides, silicon is not forgiving like metal, it can be cracked, chipped and scratched when lapped.
-=S_Klass=- I tweak... therefore I am.
Yep that's what I'm doing. Although like I said, I'm not having much luck with it. I probably wouldn't have tried it except I heard of people doing it before with varying success. Problem is that the heat sink is not resting completely flat on there and I can't seem to adjust it any other way, this is a last resort. I don't really want to go the thermal tape route if I don't have to. Hey Klass are you using one of those copper spacers you suggested? I've got one now, but it doesnt seem to be working.
i have a copper spacer on my P3 600eb and the temp stayed the same. I have not been able to look and see if the heatsink is sitting level, i don't have a small mirror to look. What kind of temps are you getting??
Umm...just an idea...tried taping on the spacer and lapping the whole thing...if the spacer is one of those frame thingies...then you'll make sure it's not to high to make the contakt less between sink and core...and it makes the surface your sanding larger making it easier to hold level ???
I forgot my thermal grease at work. That made a HUGE diff in my temps. w/out the grease right now I'm idling at 26c. The second I do anything to stress the cpu it rockets to 50c+, sometimes hitting 60! I have a slocket adapter so I can actually put it up to the light and see the cpu to hs clearance. By the way, there's no such thing as a 600eb. eb is the code for 133mhz cpu's. Like the 667eb for example.
You live in Sandusky? You lucky sob! Cedar point is a blast! I used to live south of cleveland.
Yeah, cedar point's one hell of a drive for me now. lol That devils drop was the freakiest ride I've ever been on.
Hey tweety, thanks for you suggestion by the way. I think I might give that a try. I'll try to hammer that copper spacer flat first though I think. The center part where they punched the hole out causes one side to be pushed out toward the center.
Donovan, I'm just wondering if by chance your using a hs&f designed for a socket 370 processor. If so, look for a slight gap at one end of the processor near the thicker end of the socket. It's a very small gap, .005-.010, you might only be able to see it using a bright light.
Donovan:I have a slocket adapter so I can actually put it up to the light and see the cpu to hs clearance.
Since the newer processors are thinner, older 370 heatsinks should be carefully looked at for this slight interference. It was what kept me from O/Cing my 566 Celeron. I ordered a newer hs&f for my setup and all was good again.
Also I thought the copper spacers were for HS stability not thermal transfer. The only place you can transfer heat on a chip is the area you were talking about lapping earlier. In theory you could be transfering heat from chip face to the HS, then from HS to the copper spacer, then back off the copper spacer to the outer areas of the pin grid. I have not been a big fan of these spacers(as you can see) except in special cases. I believe that the air gap between HS and the outer areas of the processor pin grid promotes better heat transfer. But first and foremost the HS must lay absolutly flat. Even before a copper spacer is installed. And thermal compound will not take up this gap if it's uneven. Look for a HS and socket interference. Thanks for letting me ramble.......good luck !!
Honorary member of "Snipers Anonymous"
I have a copper spacer that I just lapped and it is barely bent so that it isn't totally flat on a flat surface. I can push it down to where it is flat, and that makes me wonder if I can use epoxy to glue it onto the core so it stays flat. I think the raised part of the core was a bad idea, because it offers an area with a very small surface area to contact with a very large surface area, so that it is trying to balance on it. Anyways, I think the best thing to do for an unflat surface is to go overboard with thermal paste to where it forms a bond. Is there anything wrong with this? Please tell me if I am wrong before I do anything I will regret.
Actually I got the best temps by "everdoing" the thermal paste thing. Everyone says not too, but that worked best for me. As soon as I can get my hands on some thermal tape, I'm going to try to use 4 pieces of thermal tape to form a square around the slug. Then I'm going to put a generous layer of grease on the center of the slug. If my plan works, the tape should help to contain the paste in the middle (on the slug). I'll keep you posted. I'm trying to find someone who sells thermal tape first.