Sign in with
Sign up | Sign in
Your question
Solved

Odd air cooling situation

Last response: in Components
Share
May 27, 2010 8:31:52 PM

Ok guys, I built my first quad core system back in March, and I'm picky about airflow and cooling, so I've added some high flowing fans and opened up the case to allow all the flow I can get.

Here's the specs
Lian-Li Lancool K56
SCYTHE 120cfm rear fan - Thermaltake 78cfm front fan
Asus M4A785TD-V EVO
Athlon II X4 620
Corsair 2Gb DDR3 1333

I'm running AS 5 on the stock heatsink, and for the most part have been seeing idle temps in the 32c range for the CPU and 30c for the MB. Which doesnt seem bad to me, but the strange part of this and the reason I'm posting just happened yesterday!
Had a buddy of mine over to talk tech and hang out, and for some reason i cant remember now, needed to turn off the ceiling fan that I keep running in the room my computer is in. I consider myself to have a decent idea of how airflow works, but this has really freaked me out. About 5 minutes after I turned off the fan I glanced at my temp and volt monitor and my temps had dropped dramatically.
As I type this for your advice, my cpu is idling at 25c with my MB at 23c!! How the heck can a ceiling fan, on or off, affect my case airflow this much? I'm really baffled!
Thanks in advance for any ideas you might share!

Best solution

a b ) Power supply
May 27, 2010 8:39:29 PM

Are you sure it was the ceiling fan? Did you try again turning it on and off and measuring the temperatures? Did you make sure the CPU had the same usage the whole time?

The only thing I can think of is that the air on the ceiling will be slightly warmer than the air on the floor (hot air rises), so it blows slightly warmer air down. How this can change the temperature by 7C, does not make sense at all.
Share
May 27, 2010 9:07:55 PM

@ enzo,
very sure its the ceiling fan, we did several re-checks, turning the fan on, turning it off, even reversed the rotation (which has the same effect as turning it off).
We are thinking some sort of turbulence pattern from the air getting behind the desk, but its little more than a guess at this point. And I'm sure CPU usage was same....btw it had never before run this cool!
m
0
l
Related resources
May 27, 2010 9:47:46 PM

Must be some cooling fan to create much cooler air in your case, enough to drop the temps drastically.

can you run Prime 95 with the ceiling fan on and off to see if the temps are affected? I doubt it.
m
0
l
a b ) Power supply
May 27, 2010 10:21:50 PM

It's probably creating positive pressure outside the case, which inhibits the normal case air flow, because it's pushing against that pressure.

m
0
l
a b ) Power supply
May 27, 2010 11:39:39 PM

Was the window open? Is the air outside the house cooler than inside the room? If so then that would lower your overall system temps for sure.

Also, if your case fans are automatically controlled by the motherboard are you sure that they did not fluctuate in rpm?

m
0
l
May 28, 2010 12:16:14 AM

hundredislandsboy said:
Must be some cooling fan to create much cooler air in your case, enough to drop the temps drastically.

can you run Prime 95 with the ceiling fan on and off to see if the temps are affected? I doubt it.


I have in fact run Prime in both conditions, with the ceiling fan ON my cpu temp is 52c, however, with it off it runs at 41c.
m
0
l
May 28, 2010 12:19:14 AM

4Ryan6 said:
It's probably creating positive pressure outside the case, which inhibits the normal case air flow, because it's pushing against that pressure.



That is my working theory right now, either positive preasure, or some turbulence under my desk from the ceiling fan. It sure surprised me when I saw it.
m
0
l
May 28, 2010 12:21:48 AM

ssddx said:
Was the window open? Is the air outside the house cooler than inside the room? If so then that would lower your overall system temps for sure.

Also, if your case fans are automatically controlled by the motherboard are you sure that they did not fluctuate in rpm?



House is closed up, with AC on, ambient temp sitting at 76f, guess I could have told everybody that bit at the start! And I've got the fans in the case manualy controled, and at full speed. No fluctuation in fan speeds noted.
m
0
l
a b ) Power supply
May 28, 2010 1:06:46 AM

At least you know how to get the best cooling out of your machine now!
m
0
l
May 28, 2010 1:19:20 AM

the celing fan is pushing the hot air down from the celing. The flow of the air makes us feel cool but it just mixes the hot celing air with the cooler air near the floor. With out the fan the cool air sits low and the hot air is at the celing. In most rooms there is a trap at the door that traps a foot or two of hot air in the room even with the ac on. Just get in the top bunk of a bunk bed and you will notice the difference (ask my daughter).

Mac
m
0
l
May 28, 2010 1:36:22 AM

macusn said:
the celing fan is pushing the hot air down from the celing. The flow of the air makes us feel cool but it just mixes the hot celing air with the cooler air near the floor. With out the fan the cool air sits low and the hot air is at the celing. In most rooms there is a trap at the door that traps a foot or two of hot air in the room even with the ac on. Just get in the top bunk of a bunk bed and you will notice the difference (ask my daughter).

Mac



While I agree with the basic idea here, i need to clarify something, with the fan running reversed, as in pulling air up instead of pushing it down, the temps remain the same as having the fan off. So I'm not sure this really covers the situation completely because I'm still moving the air from top to bottom in the room, just no longer killing the cooling in the computer case.
m
0
l
May 28, 2010 1:39:53 AM

A quick thanks to everyone for all the great answers, this may just have to be put down in my book as one of those strange mysteries of airflow.
m
0
l
May 28, 2010 1:46:06 AM

The flow (up or down) does not matter it is the mixxing of the air that matters. Is the thermostat in the room with the computer? Calibrate two thermometers and place one in the room and one near the thermostat for the ac. Then move the therm. in the computer room around, on the floor out of a draft, near the celing, on the same level as your base of your compuuter (but not near it), I think you will see a difference. Then try it with the fan on. I dont disagree that the air flow may impead the pressures/airflows but I think that the temps may suprise you.
Mac
m
0
l
May 28, 2010 2:14:55 AM

-10
mac
m
0
l
May 28, 2010 2:34:11 AM

macusn said:
The flow (up or down) does not matter it is the mixxing of the air that matters. Is the thermostat in the room with the computer? Calibrate two thermometers and place one in the room and one near the thermostat for the ac. Then move the therm. in the computer room around, on the floor out of a draft, near the celing, on the same level as your base of your compuuter (but not near it), I think you will see a difference. Then try it with the fan on. I dont disagree that the air flow may impead the pressures/airflows but I think that the temps may suprise you.
Mac



Sorry Mac, gotta disagree here, with fan blowing down my cpu was idling at 32, with fan pulling up it will idle at 25, to me thats pretty significant difference. But then again I think you are on a right path with the spot checking room temps, me and my friend were just disscusing the point that the AC vent is approx 6 feet at an angle away from the computer, with both at or near floor level. We had not factored in before now whether the AC was blowing at the time of our discovery.
So our current working theory is that prior to our turning off the fan there was never an opportunity for cold air to pool in the floor (fan was always pushing it away). We just happened to turn the fan off as the AC kicked on, so the case fans had alot cooler air to suck into the system and it was effectively ACing the computer. Of course I'm still open to debating these theories, I've really enjoyed the conversations
Thanks again, Dead
m
0
l
May 29, 2010 1:52:06 AM

Best answer selected by Deadstick50.
m
0
l
May 29, 2010 1:56:09 AM

enzo matrix said:
Are you sure it was the ceiling fan? Did you try again turning it on and off and measuring the temperatures? Did you make sure the CPU had the same usage the whole time?

The only thing I can think of is that the air on the ceiling will be slightly warmer than the air on the floor (hot air rises), so it blows slightly warmer air down. How this can change the temperature by 7C, does not make sense at all.



I chose your answer as the best because it really came down to the mixing of warm air with what was coming out of the house AC unit. Because we were mixing the air, the cold from the AC didnt have the chance to "pool" around the desk and get pulled into the computer case.
So, good call enzo!

Dead
m
0
l
!