- May 30, 2014
- 6
- 0
- 4,520
Hello,
I am in a process of designing a custom water cooling loop for a brand new pc, and I am quite a bit uncertain when it comes to the issue of fan control.
What I am trying to do, is establish 3 basic zones of control, in which a single PWM signal is replicated. For example, I want to control the speed of 8 fans on a quad 120mm radiator with a single motherboard PWM header, control a different set of 6 fans on a different radiator with another MB PWM header, and finally a set of 5 fans of various sizes, acting as case fans, all PWM enabled, with a third MB PWM header.
What I have now, are effectively 3 zones, all of which I plan on controlling via adjusted temperature curves (first zone for CPU temp, second zone for GPU temp, third zone for overall temp).
I have found very little information concerning this subject on the internet, yet it would seem that all the builders on youtube, etc. have this problem figured out to the point, they feel like it is not necessary to mention it.
The way I see it, is I have several options:
1. Buy a fan controller - software controlled can be expensive
2. Buy lots of splitting cables - motherboard headers might not be able to handle the load.
I decided to investigate the second option. The problem of overloading the motherboard is very real, given the fact that two of the headers will service 8 fans. However, I can buy special splitting cables that draw power from the PSU, rather than the motherboard. This can get quite expensive also, with a significant number of splitting cables (3x power splitting PWM, 4x 4-1 PWM, 1x 3-1 PWM, 5x 2-1 PWM. 6x PWM extender cables) present to connect all the fans throughout the entire case (a 900D).
A savings measure presented itself, in the form of a Swiftech 8-1 PWM splitter box, that has its own PSU power connector. That would reduce the number of needed cables greatly. However, it has come to my attention, that this box has some issues when coupled with Corsair fans, namely, the replicated PWM signal is too weak to provide effective duty cycle regulation when used with 4+ fans.
Since most of my fans are in fact SP120 Performance fans, this presents a real problem. The reality of the PWM signal being too weak might even affect the solution of splitting the PWM signal via various splitting cables, as mention earlier, rather than the box, and that the box is not entirely to blame.
My question is, what are the possible solutions to the problem of controlling as many as 19 fans in 3 separate zones of control, and how to compensate the strength of replicated PWM signal issue with the Corsair fans, without going as far as soldering a custom PWM amplifier board? Cost effectiveness is a factor as well, since I am already planning to spend little over 4000 euros on my computer, and I cannot go overboard anymore.
I am in a process of designing a custom water cooling loop for a brand new pc, and I am quite a bit uncertain when it comes to the issue of fan control.
What I am trying to do, is establish 3 basic zones of control, in which a single PWM signal is replicated. For example, I want to control the speed of 8 fans on a quad 120mm radiator with a single motherboard PWM header, control a different set of 6 fans on a different radiator with another MB PWM header, and finally a set of 5 fans of various sizes, acting as case fans, all PWM enabled, with a third MB PWM header.
What I have now, are effectively 3 zones, all of which I plan on controlling via adjusted temperature curves (first zone for CPU temp, second zone for GPU temp, third zone for overall temp).
I have found very little information concerning this subject on the internet, yet it would seem that all the builders on youtube, etc. have this problem figured out to the point, they feel like it is not necessary to mention it.
The way I see it, is I have several options:
1. Buy a fan controller - software controlled can be expensive
2. Buy lots of splitting cables - motherboard headers might not be able to handle the load.
I decided to investigate the second option. The problem of overloading the motherboard is very real, given the fact that two of the headers will service 8 fans. However, I can buy special splitting cables that draw power from the PSU, rather than the motherboard. This can get quite expensive also, with a significant number of splitting cables (3x power splitting PWM, 4x 4-1 PWM, 1x 3-1 PWM, 5x 2-1 PWM. 6x PWM extender cables) present to connect all the fans throughout the entire case (a 900D).
A savings measure presented itself, in the form of a Swiftech 8-1 PWM splitter box, that has its own PSU power connector. That would reduce the number of needed cables greatly. However, it has come to my attention, that this box has some issues when coupled with Corsair fans, namely, the replicated PWM signal is too weak to provide effective duty cycle regulation when used with 4+ fans.
Since most of my fans are in fact SP120 Performance fans, this presents a real problem. The reality of the PWM signal being too weak might even affect the solution of splitting the PWM signal via various splitting cables, as mention earlier, rather than the box, and that the box is not entirely to blame.
My question is, what are the possible solutions to the problem of controlling as many as 19 fans in 3 separate zones of control, and how to compensate the strength of replicated PWM signal issue with the Corsair fans, without going as far as soldering a custom PWM amplifier board? Cost effectiveness is a factor as well, since I am already planning to spend little over 4000 euros on my computer, and I cannot go overboard anymore.
Facebook
Twitter
Instagram