SI
SI
discoversearch

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.
Pastimes : Dream Machine ( Build your own PC ) -- Ignore unavailable to you. Want to Upgrade?


To: Dirk Hente who wrote (3324)10/31/1998 9:55:00 PM
From: pae  Read Replies (1) | Respond to of 14778
 
Dirk,

I hear you on no favorite. Sounds like you are going for whatever works. My concern at the moment is my Gateway ATX Tower P5-166. This morning it had just one fan blowing out of the power supply into the case - but very close to the P5 which has a huge heatsink but no fan. I flipped the PS fan to blow out (what a pain in the butt!) and added a low front fan to suck in, but with the case open that P5 is uncomfortable to touch. As in quite warm. I begin to understand that due to the PS fans proximity to the P5, it could give more benefit to blowing in: more volume, higher velocity right at the P5; versus the likely OVERALL more efficient "cool-in-the-bottom-hot-out-the-top" "assisted convection" approach that I have switched to. I slid the cover back on to focus the flow past the pentium - but I've got no means to monitor other than malfunction itself. (I'm on that machine right now.) And frankly the flow from those two fans FEELS pathetic.

I may spring for that drive bay front temp monitor/alarm/probe/fan thing that was in one of JW's links. While I would prefer not to sink any more dollars in the P5-166 MB/CPU/memory in anticipation of coming upgrade - I would like it to NOT catch fire in the meantime. <g>

Perhaps I'll put one of those Radio Shack squirrel cage fans on the back wall with superlock fasteners and point it right at the P5 heatsink. Just wish I had a way to measure other than taking the cover off and touching it.

Well, the thing hasn't caught fire in 2 years - it'll probably last another couple of months. <g>

Paul



To: Dirk Hente who wrote (3324)11/3/1998 12:24:00 AM
From: Sean W. Smith  Respond to of 14778
 
pae,more or less I'm experimenting around with no clear favourite for one method. What I'm doing is:
1. adding a second fan at the front side
2. changing the flow direction of the fan(s,) always one out and one in
3. adding/removing vents
Thats all. I never leave my case open. When you leave your case open, the air velocity is reduced very much so that the heat dissipation of some parts is done only by natural convection and radiation. This may be not enough and there is some danger of overheating. Usually the CPU is not affected by this because you have a fan attached to it so that you get forced convection. Leaving the case open may even have a positive effect here because the cpu fan now gets fresh cold air. If you ask me I think its too risky to leave the case open.
The nice thing about the PII/Celeron CPU's is the fact that they have a thermal protection build in. You can never thermally destroy the cpu by overclocking because the CPU just disables the clock when it is getting too hot. Your PC just stops working, thats all.
As long as you dont have high performance parts (450MHZ cpu, 10.000rpm hd's and 3d accellerator cards) thermal management shouldn't be a big prob.




Most of the boxes we design at work are very dependent on this and many will overheat and malfunction with the cover removed. Often times we have large fans blowing on them when we have them open and probed up for debugging. Up until 1-2 years ago it wasn't as big a deal but its becoming worse.

Sean