I have a confession to make, though I know I?m not the only computer user guilty of this particular sin:  I?m terrible at remembering to regularly clean the dust out of my computer.  I run Ad Aware, update McAfee, and install Windows critical updates to keep my installations running smoothly, but when it comes to actually blowing out the case, I habitually forget.  As much as I?d like to present this article as the epic conclusion to a months-long effort to carefully track and monitor how (and to what degree) dust buildup affects system temperatures, the reality is more mundane. 

This article didn?t start as an in-depth investigation into dust; it began as a personal investigation into why my CPU temperatures were running so high after a random temp check produced alarming results.  The end results, however, were interesting enough to share. 

The DustTestbed:

System Specs:
Athlon 64 FX-53 (Socket 940, AMD OEM cooler)
ASUS SK8V
2 Gig Mushkin ECC DDR400
2xWD800JB PATA drives in RAID 0 Configuration
ATI Radeon 9800 Pro.

My system obviously isn?t the coolest or quietest rig in existence, but I?d carefully checked temperatures (and dusted) back in July when I added additional RAM and upgraded to the faster CPU.  I?ve had the case open a few times between then and now, but hadn?t dusted or checked temperatures until I randomly decided to take a look during some routine maintenance. Since I run Folding@Home constantly as a background service, my CPU usage is always pegged at 100%.  Sandra?s reported CPU temperature, however, was rather alarming: I expected a full load temperature of between 52-55?C, but not the blistering 65?C Sandra reported.  For a modern high-end Prescott that?s acceptable (even good), but its quite warm for an Athlon 64. 

Here?s what temperatures looked like with the case off and with the case on; all temperatures are at full CPU load.  In order to ensure I wasn?t picking up any confounding variables, I did a full set of BIOS and driver updates, as well as checking to ensure all my fans were rotating properly. None of it made any difference; my case-closed temperature remained at 65?C, my case-open, 59?C. 

So how much of a difference does a good dusting make?  See for yourself.

The drop in temperature simply from blasting the dust out of my CPU cooler and case is huge?far larger then I would?ve predicted.  My case closed, full load temperature is 9?C cooler than my case open temperature, even though the ambient air is the same 72?F. 

next >

---

• Dust: From Nuisance to Major Problem (1 of 2)
• Conclusion:  Dust Early, Dust Often. (2 of 2)