PWM fans on voltage controller?

[termonator28]

So I read this: PWM is more efficient. With a voltage regulator if you need 6V and have 12V, the other 50% of your power is completely wasted. (dissipated as heat) PWM systems are much more efficient and wast very little power.
So if I get a PWM fan, but hook it up to a voltage regulator, like the fan controller on my case, does it defeat the purpose of PWM? I'm asking because I'm looking to replace the fans in my case, they're pretty loud. Thanks.

 

[chegalhavara]

The PWM fans / connection is indeed more efficient as for power waste and management of your fan speeds allowing them to work also at less than 50 of their maximum speed and managing a more efficient fan curve through BIOS or Software or also a fan controller.
The fan works always at the same voltage although the power is turned off/on (pulsing) at regular intervals to manage their speed, thus achieving a very precise control of the fan´s speed using less power. That means the Fan will allways work at the voltage it was designed for, probably also increasing it´s lifespan.
With a 3pin connector that is achieved through a voltage increment from 50% upwards.
It is usually advised to give preference to 4pin connectors over 3 pin connectors.
BUT 4pin Fans also work on 3 pin connectors and the reverse is also true.

 

[JackNaylorPE]

PWM fans have the advantage of being able to lower the speed threshold to a lower rpm. Their difference is cost and low speed hum / clicking. It has no real tie in to efficiency and if it did, the difference is immeasurable. It's PWM that actually applies 12v power constantly, voltage control uses less power when running at 6v than it does at 12v. PWM tho is not constantly on ... its a power on / power off kinda thing. So using DCV all the time at 6 V isn't much different than using PWM at 12v half the time.

PWM would have an advantage if it was able to run fans at say < 50% speeds but does the amount of fans provided allow that ?... if fans are temp controlled, can they move enough air to keep the CPU / GPU / case interior at desired setpoints ?.... maybe / maybe not. So depending on your case cooling and load, you may not be able to take advantage of this "advantage"

PWM would have an advantage if it was able to run fans at < 50% speeds and DCV could not. But the fact is today's better designed DCV fans have no problem matching PWMs low speed controls. I am using DCV control and the fans have no trouble doing 25% of full speed getting down below 300 rpm. I just have them shut off when temps are below my established setpoints in which case they are at 0 rpm.

But as you have the option of using PWM control to move the fan at the exact same speeds, there is no difference in power used, so therefore there can be no greater efficiency.

A typical 140mm fan costs about 2 cents a month if running at full load... mine are off most of the time ... typically rad fans at 350 - 500 rpm while working in CAD (case fans off), 650 rpm when gaming max, and near 800 when stress testing. So each fan costs me about a penny a month ... when it's on at 1250 rpm... at 600 rpm about 1/4 of that.... at 300 rpm, immeasurable.

Controlling PWM devices with variable voltage is a bad idea.
http://www.swiftech.com/pwmcontrollers.aspx

Nidec, maker of PWM fans, writes:

Pulse-width modulation of DC operating voltage to modify fan speed [edit: in PWM devices] is not recommended. Transients generated by that approach can irreversibly damage motor commutation and control electronics and dramatically shorten the life of a fan

There is a best of both worlds approach. Using a PWM fed Fan control PCB to control the speeds of DCV fans.

Advantages:
-Lower fan cost
-Same low speed control
-No low speed PWM humming or clicking
-No sloppy fan splitter cables
-Can set ramp up and ramp down speeds ... for example on rad fans, have fans ramp down slowley ... CPU may not be as hot with load removed but the 2 liters of water in the loop is
-Can automatically turn some fans off when not needed (idle) and turn on again when you apply some load.

It would be nice if I knew the fan size and what we have to cool but let's assume you have up to (6) 140mm mounts:

(1) Phanteks fan Control PCB ($19.99)
http://www.newegg.com/Product/Product.aspx?Item=9SIA4RE5032993

(2 to 6) Phanteks 140mm 1250 rpm fans ($11.99 each)

a) Mount the PCB at convenient location
b) Connect a PWM or DCV header to the fan hub with provided fan cable.
c) Connect up to 6 of the above fans to the hub
d) use your Fan Control Utility provided with the MoBo to control fan speeds.
e) You do not have to connect the SATA Power cable to the hub ... 1) each fan draw a a max of 0.14 amps and the headers are rated for 1 amp.... 6 x 0.14 amps = 0.84 amps < 1.00 amps so OK and 2), if you supply 12V power tot he hub and then if you are using a DCV header asa feed, the 12v signal from SATA connector will override the MoBo header 3 - 12v signal and you will have no speed control.

If you want more than 6 fans, then you need to use a PWM MoBo Header ONLY and with 7 ore more fans you will need the SATA Power cable so as not to overdraw more than 1 amp from the MoBo.

I am using 3 of these hubs as follows:

CPU and CPU _OPT=> two water pumps

CHA-1 DCV header => Hub No. 1 => (6) 140mm rad fans om 420mm radiator
CHA-2 DCV header => Hub No. 2 => (4) 140mm rad fans om 280mm radiator
CHA-3 DCV header => Hub No. 3 => (6) Case Fans

No low speed PWM Hum / Clicking
PWM low speed fan control
Low cost $12 3-pin fans ... 316 rpm on 1250 rpm fan (25%)
Best performance / noise ratio fan on market

In essence all of the advantages of PWM and none of the disadvantages.

 

[TJ Hooker]

It has no real tie in to efficiency and if it did, the difference is immeasurable. It's PWM that actually applies 12v power constantly, voltage control uses less power when running at 6v than it does at 12v. PWM tho is not constantly on ... its a power on / power off kinda thing. So using DCV all the time at 6 V isn't much different than using PWM at 12v half the time.
[...]
But as you have the option of using PWM control to move the fan at the exact same speeds, there is no difference in power used, so therefore there can be no greater efficiency.

Yes, the power delivered to the fan is the same. The inefficiency of non PWM fans comes from the fact that they are controlled with a linear voltage regulator. If the linear regulator is outputting anything less than the 12V (full speed) input, it will dissipate the excess power as waste heat. That being said, I think the power draw of a typical case fan is fairly low, don't think I'd be overly concerned about it.