lurch63
10-21-2001, 08:12 PM
Well just now I sort of got a bee in my bonnet as to the true effects of the height that a computer is placed in a room and the ambient temperature in relation to system temps, now at times I have heard people referring to ambient temp when they are actually referring to their system temp, and what I have found is that this assumption can not be made. So to keep it readable, for this brief moment I’ll try to be organized,
Subject #1)
1.0ghz tb avia @ 1.33ghz currently w/ mc462
Case: full tower w/ 2 120mm deltas on side, front and rear 80mm fans, vent holes cut 5.25 bay, 3.5 bay and enlarged holes in lower front. Vertical location is approximately 1meter or 2.5ft.
Subject #2)
1.0ghz tb pre-axia stepping w/ all copper gladiator
Case: mid tower w/ front and rear 80mm case fans, 4x2 inch hole cut above CPU in case with grill over hole for ventilation. Vertical height 0 meter or 0ft ;)
At the time both units where running UD which is just about as hot as I can get them. And using a Springfield precise indoor/outdoor digital thermometer for measurements. Now while my basement my be cool there is nothing that is particularly special about it kept in perspective of another room being warmer and dryer, cooler moister air will still go nearer to the floor while hot and dry air will rise. Another assumption to be made is that the measurements taken by these devices is accurate, so the exact data must be taken in stride. Although from great differences conclusions can be reached neglecting these variances.
Data:
Ambient temp at : 0m =64.4F/14.4C
at : 1m = 73.1F/18.3
Humidity
at: 0m = 61%
at:1m = 59%
CPU temp
subject #1 = 95F/34C
subject #2 = 104F/40C
System temp
subject #1 = 78F/26C
subject #2 = 80F/27
Results: So what does it all mean? Well first looking at the height that there is a 3.9 degree Celsius drop from 1 meter to the floor, so if your looking for that last bit of cool, keep it to the ground. Secondly on that note there was a 2% difference in humidity, so if your running a peltier rig that you feel is getting really close to condensation, raise it up. It just might be the edge you need to stop your board from shorting out. (but don't quote me on that :) )
Secondly we get to look at the difference in ambient and system temperatures. First it is important to note that subject #1 has much better cpu and case cooling than subject #2, in relation to system and ambient temperature this can now be measured. So seeing that the difference in ambient temperature and system temperature was 7.7C for subject #1 and 12.4C for subject #2 this idea of equal system and ambient temps must be thrown out the window.
So first two points having been made. First the height in a given room does have an important effect on temperature and humidity. Secondly system temp in no way implies ambient temp. lurch
Subject #1)
1.0ghz tb avia @ 1.33ghz currently w/ mc462
Case: full tower w/ 2 120mm deltas on side, front and rear 80mm fans, vent holes cut 5.25 bay, 3.5 bay and enlarged holes in lower front. Vertical location is approximately 1meter or 2.5ft.
Subject #2)
1.0ghz tb pre-axia stepping w/ all copper gladiator
Case: mid tower w/ front and rear 80mm case fans, 4x2 inch hole cut above CPU in case with grill over hole for ventilation. Vertical height 0 meter or 0ft ;)
At the time both units where running UD which is just about as hot as I can get them. And using a Springfield precise indoor/outdoor digital thermometer for measurements. Now while my basement my be cool there is nothing that is particularly special about it kept in perspective of another room being warmer and dryer, cooler moister air will still go nearer to the floor while hot and dry air will rise. Another assumption to be made is that the measurements taken by these devices is accurate, so the exact data must be taken in stride. Although from great differences conclusions can be reached neglecting these variances.
Data:
Ambient temp at : 0m =64.4F/14.4C
at : 1m = 73.1F/18.3
Humidity
at: 0m = 61%
at:1m = 59%
CPU temp
subject #1 = 95F/34C
subject #2 = 104F/40C
System temp
subject #1 = 78F/26C
subject #2 = 80F/27
Results: So what does it all mean? Well first looking at the height that there is a 3.9 degree Celsius drop from 1 meter to the floor, so if your looking for that last bit of cool, keep it to the ground. Secondly on that note there was a 2% difference in humidity, so if your running a peltier rig that you feel is getting really close to condensation, raise it up. It just might be the edge you need to stop your board from shorting out. (but don't quote me on that :) )
Secondly we get to look at the difference in ambient and system temperatures. First it is important to note that subject #1 has much better cpu and case cooling than subject #2, in relation to system and ambient temperature this can now be measured. So seeing that the difference in ambient temperature and system temperature was 7.7C for subject #1 and 12.4C for subject #2 this idea of equal system and ambient temps must be thrown out the window.
So first two points having been made. First the height in a given room does have an important effect on temperature and humidity. Secondly system temp in no way implies ambient temp. lurch