Big 220mm Fan Experiments/Results. More tests...

Control: management of fans, temp/rpm monitoring via soft/hardware

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Big 220mm Fan Experiments/Results. More tests...

Post by Bluefront » Sat Mar 03, 2007 5:17 am

Here are the results from my first big fan experiment.......helped along by Felger Carbon. I was trying to run this whole computer on the one big fan...intending to also use a fanless Zen PSU when complete. The result....I could use a fanless PSU, but the CPU will overheat even with the fan at 12V. From what I found, there is plenty of airflow through the case at any voltage, but this airflow is not concentrated enough to cool the CPU.

Specs....

ASRock board

P4-2.6 NW....about 70watts max.

Antec NSK 4400, stock 380w PSU.

Xclio 220mm fan mounted to 45mm from the XP-120 heatsink.

Initial testing with the big fan at 12v (very noisy due to high airflow).....the CPU idles about 37C, at 50% CPU usage reaches 53C, and at 100% goes off the chart......had to stop the test.

At 12v, at max CPU usage, everything in the case is running very cool.....except the CPU. Obviously this setup won't work with the P4-2.6. Here's the fix. Put a CPU fan directly on the XP-120. This fan is within 20mm of the big fan, but doesn't seem to change the noise factor. Before I installed the CPU fan, I tried various ducts, and air deflectors....nothing worked very well. I need another fan on the CPU to fix the noise problem, and the cpu heat problem.

With a generic 120mm on the XP-120 running on 5V (about 800 rpm), and the big fan at 5V (very quiet, only air noise) here's the results. The CPU idles at 31C, and maxes about 45C (CPUBurn). Much better......I could lower the voltage below 5V on both fans with a different fan controller, and get a much lower sound level.

Amazingly.....when the computer is maxed out, the 120mm fan at 5V, turning the big fan to 12v does not affect the CPU temp....same at an idle. The rest of the board components, and HD, give better temps, but not the CPU.

Perhaps a different CPU heatsink would give better temp results running fanless in this setup, but I doubt it. Conclusion....this big fan makes a good case fan, but not a CPU fan. I think this experimental setup can be tweaked into a really quiet computer.....running very cool. But it will take careful management of fan-speeds, and airflow. YMMV.

Image

Image

Image
Last edited by Bluefront on Sat Mar 03, 2007 9:28 am, edited 1 time in total.

Iateronmly
Posts: 17
Joined: Wed Feb 28, 2007 7:51 am
Location: Ninilchik Alaska
Contact:

Re: Big 220mm Fan Experiments/Results.

Post by Iateronmly » Sat Mar 03, 2007 6:50 am

Bluefront wrote:
Perhaps a different CPU heatsink would give better temp results running fanless in this setup, but I doubt it. Conclusion....this big fan makes a good case fan, but not a CPU fan. I think this experimental setup can be tweaked into a really quiet computer.....running very cool. But it will take careful management of fan-speeds, and airflow. YMMV.
Yah - I love big fans, but, they're not that great at spot cooling. The increase in airflow through your system will help you run your spot cooling fans a lot lower though and get better temps - but there isn't enough velocity/momentum to the exiting air to get it through your CPU heatsink, it'll just flow around it as you experienced.

So far in my own experiences:
- they're great for reducing the system board's tempatures (MOSFETs/PLLs etc which may help overall system stability)
- They substantially reduce tempatures vs heatsinks along with a 120mm intake/exhaust at lower noise.
- In the above scenario you can reduce spot cooling specific fans (GPU/CPU) further than with a single 120mm intake or exhaust.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Sat Mar 03, 2007 9:24 am

Well maybe I spoke too soon.....I made another, bigger duct, using an old fan frame, with some foam extender strips attached. This ducts the airflow all the way over to the big fan frame. I estimate the square duct area covers about 35% of the total area of the big fan......this works.

With the right fan control (haven't started that yet), this new duct allows an idle temp of about 34C @5v........and 46/47C @12V maxed out. The maxed out temp at about 7V is about 54/55C. It looks like all it took was a longer/wider duct. Right now this duct is temporary....till I make a better one.

And I measured a TT Big Typhoon.......without any ducting, and the removal of the TT fan, the heatsink would be within 1/8" of the big fan. It's the same size as the XP-120, so it might work better since the heatsink would be closer to the fan. More testing is in order. This looks very promising. :D

Image

edit....Ambient during testing today is 23C.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Sat Mar 03, 2007 3:33 pm

Next test and probably the final heatsink.....a Big Typhoon. It is almost 45mm taller (without the 120mm stock fan) than the XP-120. This places the heatsink within 1/8" of the big fan.....so I don't need a duct. I did add two deflectors to the big fan, because the Typhoon doesn't center on the big fan quite as well as the XP-120.

Temperatures......very slightly better than the XP-120 at maximum CPU usage. Could go either way. I'm going to use the Big Typhoon. It is a more secure mount, easier to install, and I don't have to remove the heatsink to change the ram....like the XP-120.

My opinions have changed......If you get the right heatsink, and duct the thing properly, you can run your computer on the single big fan. When this project is completed, it'll be quiet, plenty quiet with one fan. I have yet to tweak the exhaust airflow, and add dampening. There's a thread on the entire project in the gallery section.

Image

Image

Post Reply