Overclock.net › Forums › Cooling › Air Cooling › OCN Dave's Air Cooling Guide
New Posts  All Forums:Forum Nav:

OCN Dave's Air Cooling Guide - Page 188

post #1871 of 2323
yes please do dave..you can go crazy..from the smallest to the tallest..
post #1872 of 2323
I have a few high rpm PWM fans im trying to repurpose for a htpc, and id like them to run at the lowest possible PWM %. Is there a mod I can do to the pwm signal line that would trick the fan into sitting at its minimum speed all of the time?
post #1873 of 2323
post #1874 of 2323
Okay I have a flow question and want to see what everyone here things.

In a SFF if you use a cooler like the one pictured....



Are you better off using the fan to blow down or pull up.

Follow the logic for a moment, by blowing down you are often pulling air against the exhaust fan at the rear of the case and are blowing warm air from the CPU onto the motherboard. If you make the fan a pull up then you are pulling the cases intake air, over components, up through the heat sink and then exhaust right into the intake path of the rear exhaust fan thus moving the heat quickly out of the case.

From a pure logic point of view this would seem a better solution. I am curious if anyone has tested this? You would need to test not just CPU temp, also motherboard and internal case air as well.
post #1875 of 2323
Quote:
Originally Posted by Mopar63 View Post

Okay I have a flow question and want to see what everyone here things.

In a SFF if you use a cooler like the one pictured.... Warning: Spoiler! (Click to show)

Are you better off using the fan to blow down or pull up.

Follow the logic for a moment, by blowing down you are often pulling air against the exhaust fan at the rear of the case and are blowing warm air from the CPU onto the motherboard. If you make the fan a pull up then you are pulling the cases intake air, over components, up through the heat sink and then exhaust right into the intake path of the rear exhaust fan thus moving the heat quickly out of the case.

From a pure logic point of view this would seem a better solution. I am curious if anyone has tested this? You would need to test not just CPU temp, also motherboard and internal case air as well.
Your logic is sound. thumb.gif

I've found most of the time pulling air away from motherboard results in lower temps, both for motherboard and CPU. Air pushing into cooler hits mobo, turn out, hits GPU, RAM, I/O housings, heatsinks and turns up beside cooler and fan, and when it gets above fan is sucked back into cooler .. around and around it goes!!

When I tested the AXP-100 on open bench with a 140mm fan flowing air across test station the CPU ran 5-8c cooler with cooler fan pulling rather than pushing.
Quote:
AXP-100 w/ TY-100 pushing in
Tested with i7 920 stock (130w CPU)
Handbrake @ realtime

Room ambient . Idle CPU; rpm; Cooler Exhaust. . 100% CPU; rpm; Cooler Exhaust . mobo; NB; Cooler intake*
21.5c . . . . . . . . . 33-34-35-31c 2400rpm 28.0c . . 67-68-67-66c 2400rpm . 39.5c . . 36c . . 51c
22.0c . . . . . . . . . 33-34-35-31c 2250rpm 28.0c . . 68-68-69-67c 2250rpm . 39.5c . . 36c . . 51c . . 30.0c
22.5c . . . . . . . . . 36-35-39-33c 2000rpm 30.0c . . 69-70-70-69c 2000rpm . 40.0c . . 39c . . 51c
22.5c . . . . . . . . . 35-34-38-33c 1750rpm 29.5c . . 72-73-73-71c 1750rpm . 43.0c . . 40c . . 54c
22.5c . . . . . . . . ; 35-35-39-34c 1500rpm 30.0c . . 76-78-77-75c 1500rpm . 47.0c . . 40c . . 57c
23.0c . . . . . . . . . 37-36-41-36c 1200rpm 31.0c; 39c & 46c mobo & NB
2400rpm is maximum rpm mounted pushing in
*Temperature of air 30mm above TY-100

Intake air is much warmer than room ambient.. like 8c warmer!.

AXP-100 w/ TY-100 pulling out
Tested with i7 920 stock (130w CPU)
Handbrake @ realtime

Room ambient . Idle CPU; rpm; Cooler Exhaust. . 100% CPU; rpm; Cooler Exhaust . mobo; NB; Cooler intake*
21.5c . . . . . . . . 32-31-35-31c 2450rpm 26.0c . . 62-62-63-61c 2450rpm . 33.5c . . 34c . . 46c
21.5c . . . . . . . . 33-31-35-31c 2250rpm 25.5c . . 63-64-65-63c 2250rpm . 34.5c . . 37c . . 45c
21.5c . . . . . . . . 33-32-37-33c 2000rpm 30.0c . . 64-65-65-64c 2000rpm . 35.5c . . 34c . . 48c
21.5c . . . . . . . . 35-34-39-34c 1750rpm 28.5c . . 66-67-66-65c 1750rpm . 37.5c . . 35c . . 50c
21.5c . . . . . . . . 34-33-38-32c 1500rpm 27.0c . . 69-70-70-69c 1500rpm . 39.5c . . 34c . . 51c
21.0c . . . . . . . . 35-34-38-33c 1200rpm 28.0c; 34c & 47c mobo & NB
*I neglected to put a probe under cooler, sorry.
. 2450rpm is maximum rpm mounted pulling out


Running TY-100 pushing in and find the heat difference is because of air temperature over cooler / fan intake. Running at 2400rpm the air 30mm above AXP-100 is 28.5c with 22c ambient. That's with a TY-140 40cm away and 15cm above work top blowing 22c ambient over test setup. It also shows in the exhaust temp readings with TY-100 pushing into cooler.

Now when testing or setting up a new system I always monitor the cooler intake air temp and adjust / change fans (case & cooler) to supply cool air instead of preheated air.
post #1876 of 2323

so theres no like hard mod I could do like throw on a diode/ or capacitor to the pwm line?
post #1877 of 2323
Quote:
Originally Posted by nick779 View Post

so theres no like hard mod I could do like throw on a diode/ or capacitor to the pwm line?
the 4th pin is a signal lead. Trying to change the signal to be 0% up to a certain temp and then ramp up would not be easy.
Changing the resistance only weakens the signal.
You could try lowering the 12v power and see if fan will run with less voltage. I've never tried it but 8-9v instead of 12v pulses would make fan run slower .. if it even works.
Edited by doyll - 1/12/15 at 6:52am
post #1878 of 2323
If you want full control just get a top notch controller that can actualy turn the fan off..

If any ones interested drop by and check my current creation in the making..did a lot of catching up today..
post #1879 of 2323
Quote:
Originally Posted by doyll View Post

the 4th pin is a signal lead. Trying to change the signal to be 0% up to a certain temp and then ramp up would not be easy.
Changing the resistance only weakens the signal.
You could try lowering the 12v power and see if fan will run with less voltage. I've never tried it but 8-9v instead of 12v pulses would make fan run slower .. if it even works.

Honestly, I dont want any control with them. I want them to sit at the minimum PWM speed constantly regardless of what the pwm signal from the mobo is telling the fan to run.
post #1880 of 2323
Quote:
Originally Posted by nick779 View Post

Honestly, I dont want any control with them. I want them to sit at the minimum PWM speed constantly regardless of what the pwm signal from the mobo is telling the fan to run.
My first suggestiion should allow you to set them at the lowest possible speed they will run.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Air Cooling
Overclock.net › Forums › Cooling › Air Cooling › OCN Dave's Air Cooling Guide