I have two case fans. One is in the back blowing aie in just over the processor and the other is in the lowerfront just next to the cased speaker. I have both blowing air into the case from outside. In what direction should the air be blowing for more effective cooling, The only air comming out is via the powersupply.
Air flow of fans?
- Thread starter Sigma
- Start date
-
- Tags
- Fan Controller Heatsinks
I'm in the middle of some testing at the momet. My previous setup was two 80mm exhaust fans in the front, 2 80mm intakes in the back and my heatsink fan blowing down. I read somewhere on the net the my particular heatsink (Alpha Pal-6035) was desinged specifically for the fan to suck air from the heatsink. My idle temps with my 'mistaken' setup was 28-31C and about 41-42C under load. Now that I've corrected myself and reversed everything my idle temps are 33C and 46C under load. numbers speak louder than words.
okay here is how it went:
Setup 1:
two rear 80mm Antec exhaust fans
two front 80mm Antec intake fans
Black label Delta 60mm heatsink fan with airflow away from heatsink (Alpha Pal-6035)
Idle temps: 33-35C
Load temps: 47-51C
setup 2:
Two rear intake fans
two front exhaust fans
delta blowing down into heatsink
idle temps 28-32C
load temps 39-43C
as far as I can tell Setup 2 is much better in my system. don't think it has as much to do with the case fans as it does with the heatsink fan.
Setup 1:
two rear 80mm Antec exhaust fans
two front 80mm Antec intake fans
Black label Delta 60mm heatsink fan with airflow away from heatsink (Alpha Pal-6035)
Idle temps: 33-35C
Load temps: 47-51C
setup 2:
Two rear intake fans
two front exhaust fans
delta blowing down into heatsink
idle temps 28-32C
load temps 39-43C
as far as I can tell Setup 2 is much better in my system. don't think it has as much to do with the case fans as it does with the heatsink fan.
The general practice (not to say it is right in all cases) is that you intake at the front and exhaust at the back.
I'd be interested in your results if you take option 2 and reverse the fornt/rear intake/exhaust setup.
The temp differences could be totally due to the reversal of the HSF and not the case fans. You may (or may not) gett better solutions with option 3....
Also try to factor your motherboard temperature which can be just as important to stability as CPU temps...
-* This Space For Rent *-
email for application details
I'd be interested in your results if you take option 2 and reverse the fornt/rear intake/exhaust setup.
The temp differences could be totally due to the reversal of the HSF and not the case fans. You may (or may not) gett better solutions with option 3....
Also try to factor your motherboard temperature which can be just as important to stability as CPU temps...
-* This Space For Rent *-
email for application details
btvillarin
Distinguished
- Apr 10, 2001
- 2,370
- 0
- 19,780
Yeah, I was just about to say the same thing. So, you should end up with four configurations: 1) 2 front intake, 2 rear exhaust, HSF blowing towards heatsink, 2) 2 front intake, 2 rear exhaust, HSF blowing away from heatsink, 3) 2 front exhaust, 2 rear intake, HSF blowing towards heatsink, and 4) 2 front exhaust, 2 rear intake, HSF blowing away from heatsink.
Maybe put your core voltage, fsb, and multiplier settings as well?
Plus, what are your measurements based on? (i.e. Asus Probe, Motherboard Monitor, Compunurse, SiSoft Sandra) And, yeah...can you put your ambient temp as well? I'd also like to see the results of your testing.
Great job...
<i>OC...unless your computer's cheezy (is that a good rhyme?)</i>
Maybe put your core voltage, fsb, and multiplier settings as well?
Plus, what are your measurements based on? (i.e. Asus Probe, Motherboard Monitor, Compunurse, SiSoft Sandra) And, yeah...can you put your ambient temp as well? I'd also like to see the results of your testing.
Great job...
<i>OC...unless your computer's cheezy (is that a good rhyme?)</i>
Ok fellas here it is. Short and sweet, allowed 5 minutes after bootup to cool down, then did 10 minutes under Seti@home. Used Sisoft Sandra 2001 motherboard information.
Room temp was 75F
133x7.5 = 997mhz 1.75v Changed to stock for the test.
Setup 1: 2 intakes rear, 2 exhaust front, heatsink down
Idle temps Load Temps
C F
Board Temp : 25.4C/77.7F 25.4/77.7
CPU Temp : 27.2C/81.0F 35.3/95.5
Power/Aux Temp: 29.0C/84.2F 29.9/85.8
Setup 2: 2 intakes front, 2 exhaust rear, heatsink down
25.0/77.0 25.0/77.0
29.4/84.9 37.1/98.8
27.6/81.7 26.7/80.1
Setup 3: 2 intake front, 2 exhaust rear, heatsink up
25.0/77.0 25.0/77.0
33.9/93.0 38.0/100.4
26.3/79.3 26.7/80.1
Setup 4: 2 intake rear, 2 exhaust front, heatsink up
25.4/77.7 25.4/77.7
33.4/92.1 39.0/102.2
28.1/82.6 29.4/84.9
As you guys can see that the 2 intake front, 2 ehaust rear produced better board temps and aux temps, but the intake rear produced the better cpu temps. Personally I think the extra .2C board temps is well worth the 2C cooler CPU temps. I believe this is a direct reflection of the cooler air from the rear intakes being closer to the sucking of the heatsink fan. When you intake from the front, the heatsink is only getting ambient board temp air versus wind chilled air directly from the intakes above it. At least thats my take.
"It takes a big man to cry, it takes an even bigger man to laugh at him."
Room temp was 75F
133x7.5 = 997mhz 1.75v Changed to stock for the test.
Setup 1: 2 intakes rear, 2 exhaust front, heatsink down
Idle temps Load Temps
C F
Board Temp : 25.4C/77.7F 25.4/77.7
CPU Temp : 27.2C/81.0F 35.3/95.5
Power/Aux Temp: 29.0C/84.2F 29.9/85.8
Setup 2: 2 intakes front, 2 exhaust rear, heatsink down
25.0/77.0 25.0/77.0
29.4/84.9 37.1/98.8
27.6/81.7 26.7/80.1
Setup 3: 2 intake front, 2 exhaust rear, heatsink up
25.0/77.0 25.0/77.0
33.9/93.0 38.0/100.4
26.3/79.3 26.7/80.1
Setup 4: 2 intake rear, 2 exhaust front, heatsink up
25.4/77.7 25.4/77.7
33.4/92.1 39.0/102.2
28.1/82.6 29.4/84.9
As you guys can see that the 2 intake front, 2 ehaust rear produced better board temps and aux temps, but the intake rear produced the better cpu temps. Personally I think the extra .2C board temps is well worth the 2C cooler CPU temps. I believe this is a direct reflection of the cooler air from the rear intakes being closer to the sucking of the heatsink fan. When you intake from the front, the heatsink is only getting ambient board temp air versus wind chilled air directly from the intakes above it. At least thats my take.
"It takes a big man to cry, it takes an even bigger man to laugh at him."
Your results are interesting. Before I saw your numbers I would have thought that having the exhaust in the rear would be better for the cpu since the hot air rises. That would put the exhaust fan closer to the hot air since the rear fan is higher.
Perhaps having the intake fan in the rear blows the ambient air directly in front of the cpu cooler fan. Then the cpu cooler fan blows the ambient air directly onto the heatsink before the air has any chance to warm up from being in the case(in the "heatsink down" configuration).
That might also explain why the results are reversed for the "heatsink up" configuration. Your results for the cpu temp were better with the intake in front. In the "heatsink up" configuration the air for the cpu cooler fan is drawn from the area around the heatsink which is influenced by the temp of the motherboard and other components nearby.
Let me know if one of your rear fans, when used as an intake blows air directly over the fan for the cpu cooler.
If so perhaps the following would be another test:
-rear fan that blows directly over heatsink fan: intake
-other rear fan: exhaust
-front fans: 1 intake / 1 exhaust
Maybe both front fans exhaust. That would tend to create a vacuum in the box and pull more air from the rear intake that blows at the cpu cooler fan. It may mean less overall cfm for the case but more air directly at the cpu cooler fan (and a coresponding cooler cpu). Just a thought.
Perhaps having the intake fan in the rear blows the ambient air directly in front of the cpu cooler fan. Then the cpu cooler fan blows the ambient air directly onto the heatsink before the air has any chance to warm up from being in the case(in the "heatsink down" configuration).
That might also explain why the results are reversed for the "heatsink up" configuration. Your results for the cpu temp were better with the intake in front. In the "heatsink up" configuration the air for the cpu cooler fan is drawn from the area around the heatsink which is influenced by the temp of the motherboard and other components nearby.
Let me know if one of your rear fans, when used as an intake blows air directly over the fan for the cpu cooler.
If so perhaps the following would be another test:
-rear fan that blows directly over heatsink fan: intake
-other rear fan: exhaust
-front fans: 1 intake / 1 exhaust
Maybe both front fans exhaust. That would tend to create a vacuum in the box and pull more air from the rear intake that blows at the cpu cooler fan. It may mean less overall cfm for the case but more air directly at the cpu cooler fan (and a coresponding cooler cpu). Just a thought.
I use an Antec SX830 Workstation case. I don't think the 1 exhaust 1 intake in the back would work. My heatsink seems to be directly inbetween the two fans, so I believe splitting them up wouldn't have the same airflow over the cpu. I really never did get the idea of pulling hot air into your cpu fan.
Facebook
Twitter
Instagram