Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home	.	.
Prices: Mainboards ABIT ASUS Chaintech Shuttle Soyo Tyan CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ Search Prices:

Despite all necessary precautions on both the user and the manufacturer's sides, we have been able to inadvertently fry the AMD XP2200 Thoroughbred processor. It is still not entirely clear what happened but we have a few pointers of what to avoid if you don't like the smell of fried horse flesh.

Montage of the CPU and the TAISOL HSF pictures, the latter mirrored to provide coincidence for the burn-marks on the HSF bottom and the CPU surface components. At first glance, the CPU does not look too bad, even though the HSF was too hot to touch it.

Let's roll it up backwards, the result is shown above, that is, a burned out AMD XP2200+ "Thoroughbred" CPU next to the Taisol CGK760172, the latter mirrored to make the identification of the hot spots on the CPU causing the burn-marks on the copper bottom of the HSF easier.

The top view still looks sort of ok, however, a short look at the bottom of the CPU shows the blisters in the OPGA packaging that are an almost certain tell-tale sign of a dead CPU.

The bottom of the CPU reveals the full damage in form of blisters and discoloration of the OPGA.

OK, so what happened? The story is either very easy or very complex. Let's start with the easy version. The CPU was installed correctly, the heatsink was lapped to an almost mirror finish using 400 grid water-proof sandpaper and all of it was working just fine. The heatsink compound used was Arctic Silver3. The system was turned on, the phone rang next door and 15 min later, the only thing left was a dead processor. The more complicated story goes as follows.

• Because of the inability of the 1006 BIOS to change the multiplier on the Thoroughbred, the BIOS was flashed to ver. 1007.
• Changing the multiplier still resulted in a no-boot until the next reboot to come up at default settings.
• The CPU HSF was hooked up to the PWR fan header on the board instead of the CPU fan. The two headers are located directly adjacent to each other.
• Because of the low noise level of the Taisol fan, its shutting down went completely unnoticed in the overall noise level of two other PCs running.
• Powering the system Off and On again caused the CPU fan to run for about 30 seconds before stopping again.
• Moving the fan connector to the CPU fan header kept the fan running
• According to sources close to ASUS, one of the changes of the 1007 BIOS has been to accommodate requests that if the system goes into idle mode the PWR fan will shut down to cut down on the noise level.

next page:    => The Final Analysis =>