Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Ali Chen who wrote (109445)	5/5/2000 3:29:00 AM
From: Petz	Respond to of 1579750

Intel's "flip chips" are CRACKING due to thermal stress: (from HARDOCP) We posted earlier that OCers US had some reports of Coppermine cores being easily damaged, either through temperature variations or physical stresses. We have our take on this. Here is an explanation from Hard\|OCPer Aidan Carter, no relation to Jimmy, so get off his ass. Ive been following this up, I'm quite interested myself in what's going on here. Here's what i've come up with so far. Most of the cracked flip chips have been breaking on one corner, or on the edge. This roughly corresponds to the location of the cache on the die (according to intels spec sheet, attached). It also seems to be occuring not only on peltier cooled systems, but on passive cooling systems as well. Initially, the culprit was thought to be thermal stress, generated by peltiers. After seeing most of the reports from people who were using only large heatsinks (alphas, GWs), the suspicion shifted to mechanical stress on the slug. Latest indications seem to be a combination of the two. The real problem is the fact that the cache and the processor core generate heat at different rates. When you combine this with a very slightly mis-aligned high-stress joint, you seem to get these problems. This problem only seems to be further excerbated by the greater temperature differential generated by peltier cooling. Solutions? Well personally, I'm not going to mount any coppermines without a copper shim from now on. It's been speculated that a thermal pad might be a better solution for coppermines than thermal grease, as it allows for some flex between the two materials. Thanks to JC's page for this link to HARDOCP: hardocp.com I just can't believe the bad luck Intel is having. It is apparently related to the fact that the cache is on the corner of the die, a high stress point. I strongly suspect that these mechanical problems are what is delaying the >533 MHz Celerons and possibly >800 MHz PIII's. Petz