Case Cooling

[Graham]

Would be interested to know which combination of fans others have found most
effective.
AMD give the follwing formula for a good
case temperature:
*T1 = Outside of case air temperature
*T2 = Ambient inside case temp (measured within 1" of CPU / video card)
*T = T2 - T1
*T should be <= 7c to ensure proper cooling
I have been experimenting with my own case cooling and so far had the
following results when not playing games with my FX5900 + 2600XP:
No exhaust or input fan: Video 53 CPU 55
Side input fan no exhaust fan: Video 48 CPU 47
Rear exhaust fan no input fan: Video 49 CPU 48
Trouble is getting reliable results with the notoriously variable temps here
in UK. I would be interested to know what combination of fans have given
others the best results?

 

[Frank Hagan]

Would be interested to know which combination of fans others have found most
effective.
AMD give the follwing formula for a good
case temperature:
*T1 = Outside of case air temperature
*T2 = Ambient inside case temp (measured within 1" of CPU / video card)
*T = T2 - T1
*T should be <= 7c to ensure proper cooling
I have been experimenting with my own case cooling and so far had the
following results when not playing games with my FX5900 + 2600XP:
No exhaust or input fan: Video 53 CPU 55
Side input fan no exhaust fan: Video 48 CPU 47
Rear exhaust fan no input fan: Video 49 CPU 48
Trouble is getting reliable results with the notoriously variable temps here
in UK. I would be interested to know what combination of fans have given
others the best results?

Good case cooling article:
http://www.short-media.com/print.php?content.111

FRH

 

[Graham]

Thanks for the link, it was just what I was looking for - someone's already
tried all combinations!
However, in the conclusion it states "A single rear exhaust fan produces the
best results overall -This flushes the theory of more is better right out
the door", but in the ten test senarios it clearly states they used two rear
exhaust fans. By coincidence that was to be my next test so it will be
interesting to see if it matches their results...

 

[tomcas]

Would be interested to know which combination of fans others have found most
effective.
AMD give the follwing formula for a good
case temperature:
*T1 = Outside of case air temperature
*T2 = Ambient inside case temp (measured within 1" of CPU / video card)
*T = T2 - T1
*T should be <= 7c to ensure proper cooling
I have been experimenting with my own case cooling and so far had the
following results when not playing games with my FX5900 + 2600XP:
No exhaust or input fan: Video 53 CPU 55
Side input fan no exhaust fan: Video 48 CPU 47
Rear exhaust fan no input fan: Video 49 CPU 48
Trouble is getting reliable results with the notoriously variable temps here
in UK. I would be interested to know what combination of fans have given
others the best results?

I ran a test once that gave very surprising results. The case was a
cheap Maxtop brand that has a fan on the top and one on the side. To
heat things up I ran the Sandra CPU burnin at 100% usage and took
temperature readings after 70 cycles. I measured the temperature of the
CPU and motherboard as recorded by the Asus Probe utility.
The readings are in degrees C and the first reading is the CPU followed
by the motherboard temp.
Both fans running= 36/30
No fans running= 40/34
Top only = 39/30
Side only =35/29
As you can see it was better to leave the top fan off. However, one
thing I would have liked to try is to cover the top blow hole and see
the results. I suspect the top blow hole works well, it just does not
need a fan. These temps were much lower than yours as they were on a
XP1700 thoroughbred B which run very cool at stock clocking.

 

[Roger M]

Would be interested to know which combination of fans others have found most
effective.
AMD give the follwing formula for a good
case temperature:
*T1 = Outside of case air temperature
*T2 = Ambient inside case temp (measured within 1" of CPU / video card)
*T = T2 - T1
*T should be <= 7c to ensure proper cooling
I have been experimenting with my own case cooling and so far had the
following results when not playing games with my FX5900 + 2600XP:
No exhaust or input fan: Video 53 CPU 55
Side input fan no exhaust fan: Video 48 CPU 47
Rear exhaust fan no input fan: Video 49 CPU 48
Trouble is getting reliable results with the notoriously variable temps here
in UK. I would be interested to know what combination of fans have given
others the best results?

I did a little experimenting of my own. I put three rear exhaust fans in my
case. Two in the factory mounting places after cutting out the grille made over
them in the case. Then since I only had a couple of pci cards I cut out the rest
of the pci slots in the case and mounted a fan there too. That made a total of 4
fans including the psu fan. Then I used clear packaging tape cut to fit and
covered all areas where the case could pull air in from the rear. Effectively
making it pull all intake air in from the front and front side case vents to the
rear of the case. This lowered my case temperature to just 2C above ambient room
temp and kept it there. It lowered my cpu temp by 8C. About a week later I
installed a Volcano 12 and my cpu temp went down another 8C to 9C. Right now my
case and cpu temp only has a 4C to 6C when doing small stuff. When running prime
95 and with the Volcano 12 fan turned all the way up the temp will got to a high
of 43C to 44C. Before these mods it would just keep getting hotter and hotter. I
had seen it go as high as 70C before I stopped the torture test. The only
drawback is the Volcano 12 is noisy as hell and all the fans make the pc fairly
noisy.

My system is an XP2500+ at actual speed of 2310mhz on a Epox 8RDA+, with 768 mb
of Kingston HyperX DDR400 and an 8X GForce video card.


I just moved all this stuff to an Antec case (430W psu and five fans controlled
by the temp sensor in the psu) it is much quieter and still cools nearly as well
with the Volcano 12 set to the slowest speed setting..



Roger

 

[Shep©]

I ran a test once that gave very surprising results. The case was a
cheap Maxtop brand that has a fan on the top and one on the side. To
heat things up I ran the Sandra CPU burnin at 100% usage and took
temperature readings after 70 cycles. I measured the temperature of the
CPU and motherboard as recorded by the Asus Probe utility.
The readings are in degrees C and the first reading is the CPU followed
by the motherboard temp.
Both fans running= 36/30
No fans running= 40/34
Top only = 39/30
Side only =35/29
As you can see it was better to leave the top fan off. However, one
thing I would have liked to try is to cover the top blow hole and see
the results. I suspect the top blow hole works well, it just does not
need a fan. These temps were much lower than yours as they were on a
XP1700 thoroughbred B which run very cool at stock clocking.

I have done similar tests and stick to one back case fan and tidy
wires
Back case fan is right above and at right angles to the CPU fan so
does a nice job.
MTP



--
Free Windows/PC help,
http://www.geocities.com/sheppola/trouble.html
email shepATpartyheld.de
Free songs download,
http://www.soundclick.com/bands/8/nomessiahsmusic.htm

 

[Graham]

Just an update with two rear exhaust fans there has been an appreciable drop
in temps -
Video : 46c
Ambient : 35
CPU : 44
Although I do beleive that blocking the grill for fan 2 might have had a
similar effect as explained by 'Roger M' to stop hot air being drawn back
in.
Still feel the ambient temp. (as reported by Nvidia temp tab for FX5900) is
too high though - any suggestions?