Sign in with
Sign up | Sign in
Your question

Fan directions, conflicting information.

Last response: in Overclocking
Share
December 28, 2012 2:14:19 PM

There appears to be some conflicting information about the direction one should install case fans. I decided to test which configuration appears to be the most effective.

Most people seem to be in agreement that the CPU heat sink fan should be drawing air over and away from the CPU, in other words the exhaust facing away from the CPU. However the disagreement appears to be over the rear case fan/s and whether or not they should draw air into/or expel air out of the case.

I have a Coolermaster HAF X with one rear case fan, curiously the manufacturer installs this fan with exhaust facing out, suggesting they believe this to be the more effective configuration.
I have experimented with the configuration of the fan and I found the exhaust facing inwards (opposite to the manufacturer's configuration) on average to decrease the CPU core temp by 2°C. I agree that further statistical analysis may be required to establish whether or not this is significant.

Data included here: https://docs.google.com/open?id=0B24QGy-OPsWXSzRxb3BmV0...

I tested by running 10 stress tests (5 of each config) using IntelBurnTestV2 and recorded data using AIDA64.
Any insight into this would be appreciated.
a c 120 à CPUs
a c 89 K Overclocking
December 28, 2012 2:36:42 PM

The accepted airflow configuration in a computer case is front to back in an attempt to keep the airflow as laminar as possible. This provides cooling to all components within the case. Reversing the rear fan may bring your CPU temperatures down but has the potential to create a stall in the airflow over other vital components (such as the south bridge which is usually passively cooled). I'm not saying your idea is bad but there are other considerations than CPU temp.
My $0.02
December 28, 2012 3:58:29 PM

This is insightful thank you.
Related resources
a c 120 à CPUs
a c 89 K Overclocking
December 28, 2012 4:20:10 PM

In further consideration also, a complete top/back to bottom/front configuration may work and provide for greater CPU cooling since the air has passed over less components by the time it goes through the CPU's HSF, which should (I think) keep the airflow laminar through the front of the case... an unconventional option I think but a possibility - could work but then again, you would have all that hotter air flowing through the case... I'm going to think on this for a bit more...
December 28, 2012 4:30:52 PM

I see your reasoning. I would be interested to know if this made a significant difference. Please keep me informed if you decide to test this.
a b à CPUs
December 28, 2012 5:53:55 PM

Air flow is measured (as I'm sure you're aware ) in CFM for ultimate efficency you should have the same amount being forced in as is being forced out Eg: a fan in the front @1500 rpm + 1 in the rear @1500 ,2@2000 f+2@2000 rear etc. a fan in the side panel blowing toward the cpu antother up top(or somewhere) pulling out as for a cpu the push pull seems to drop temps by a few degrees, but if your not over clocking it is negligible, the point being same cubic feet per minute in as out. I like youe thinking outside the box,(or case as it may be.
!