Overclock.net banner

1 - 5 of 5 Posts

·
Registered
Joined
·
664 Posts
Discussion Starter #1
I'm in a sticky situation here. My current case is a Corsair 800D where there are plentiful 5.25" bays. I have a Lamptron fan controller there powering my pump and around 11 system fans. Unfortunately, my new case, a Lian Li PC-08, does not have 5.25". Consequently, I cannot fit my fan controller resulting me to revert back to my motherboard's fan headers.

My P9X79 WS has 4 Chassis fan headers, 1 CPU fan header, and an OPT CPU fan header (just another fan header that supports the CPU).

Now while researching I stumbled across this thread: http://webcache.googleusercontent.com/search?q=cache:xyvKiBfceOwJ:https://rog.asus.com/forum/showthread.php%3F23025-Y-Splitter-for-3pin-Fan-on-P9x79-Deluxe+&cd=1&hl=en&ct=clnk&gl=us (Using google's cache as the website is down at the moment).

In this thread, users found from the motherboard's manual that a P9X79 Deluxe has a CPU fan header current rating of 1A. As well as this, a user allegedly confirmed that this current rating applied to all the fan headers in the system. Checking my P9X79 WS' manual, I found that my CPU fan header has a current rating of 2A. The question is: does this 2A rating apply for the chassis fan headers as well?

I have a 4 to 1 three pin fan splitter. After reading around, I've deduced that drawing too much power from a fan header can damage it. Using Prolimatech 120mm fans I found these specifications:
  • Rated Current: 0.18 Max.Amp
  • Rated Power: 2.16 Max.Watt
Next question: while 4 x 0.18A is only 0.72A, would this be the max current draw? Should I be worried here connecting 4 fans to 1 header?

Any other suggestions regarding my lack power sources for fans? I don't think case modding is an option with a 5.25" bay in my Lian Li PC-08 haha. Thanks all, really appreciate the time for responding and reading!
biggrin.gif
 

·
Overclock Failed...
Joined
·
13,565 Posts
Power the fans off of a 12 volt molex and control those that you want speed control for from a 4 pin header.

Not available, but do you get the idea?
 
  • Rep+
Reactions: xNiNELiVES

·
Registered
Joined
·
664 Posts
Discussion Starter #3
Quote:
Originally Posted by billbartuska View Post

Power the fans off of a 12 volt molex and control those that you want speed control for from a 4 pin header.

Not available, but do you get the idea?
Simultaneously somebody else gave me a similar idea. Located in this thread I created: http://www.overclock.net/t/1585816/5-25-bay-dliemma/0_30.
There I found that I needed at least 12.96w of power per header/channel. Motherboard headers cannot provide this at 1A (12w). I've been looking at fan controllers, but I've been close minded to PWM controllers. I think though that one such as this could do the job as it provides 30w per channel!

Read up the other thread if you want. Thanks for the help!
 

·
Overclock Failed...
Joined
·
13,565 Posts
Quote:
Originally Posted by xNiNELiVES View Post

Read up the other thread I guess for more of my thoughts I guess. Thanks for the help!
The next point to be made is that controlling rad fan speed based on CPU temp is just wrong. It works, but not correctly.
When load is added to a CPU it heats up- very fast - so the fans speed up. But the water heats up much more slowly and additional fan speed isn't needed for a few minutes. The inverse happens when the load is removed, the fans slow down, but the water is still warm.

A better, more accurate, and more efficient way to do it is to base fan speed on water temperature.

Another, even better way to do it is to base the fan speed on the temperature of the air entering the rad on which the fan(s) are mounted. You set the fan speed to deliver the amount of air the rad needs to keep the water at the temp that will give you the full load CPU temp you want. Then as the ambient air temp changes (and the loop's Delta T* changes) the fans speed up/slow down to maintain the water temperature (and Delta T) you need.

* Delta T is the difference in temperature between the water in the loop and the air entering the rad(s) and is a measure of the cooling efficiency of a loop.
 
  • Rep+
Reactions: xNiNELiVES

·
Registered
Joined
·
664 Posts
Discussion Starter #5
Quote:
Originally Posted by billbartuska View Post

The next point to be made is that controlling rad fan speed based on CPU temp is just wrong. It works, but not correctly.
When load is added to a CPU it heats up- very fast - so the fans speed up. But the water heats up much more slowly and additional fan speed isn't needed for a few minutes. The inverse happens when the load is removed, the fans slow down, but the water is still warm.

A better, more accurate, and more efficient way to do it is to base fan speed on water temperature.

Another, even better way to do it is to base the fan speed on the temperature of the air entering the rad on which the fan(s) are mounted. You set the fan speed to deliver the amount of air the rad needs to keep the water at the temp that will give you the full load CPU temp you want. Then as the ambient air temp changes (and the loop's Delta T* changes) the fans speed up/slow down to maintain the water temperature (and Delta T) you need.

* Delta T is the difference in temperature between the water in the loop and the air entering the rad(s) and is a measure of the cooling efficiency of a loop.
Oh I didn't mean to actually base the fan speeds off of my CPU temp haha. I was thinking of using my motherboard's temperature sensor. Thanks for enlightening me with regards to those other ideas. Only problem is establishing a system to actually ramp the fan speeds off of the temperature of the air entering the radiator. I could mount temperature sensors and everything and then use some fan software... It just seems like it is going to get too complicated haha.
 
1 - 5 of 5 Posts
Top