A Rant About PWM (in general) & Z97 MBs

Nuluvius

Nuluvius

Nuluvius

Associate
Joined
25 Nov 2011
Posts
49
This is a prime example of bad advertising:

http://forum.corsair.com/v3/showthread.php?p=710824

The headers on these latest motherboards are not PWM despite having four pins. They actually use voltage (as I am sure most of you are already aware) to control RPM instead. Most fans have a lower bound voltage tolerance on spin up of somewhere between 30%~40%

With default settings they certainly won't start up from a cold boot as the temps would not be high enough to warrant a 30%~40% ramp up. Only after some time operating would temps get high enough to signal an increase thus starting the fans.

This is a configuration issue at the end of the day. Rectified by setting the base tolerance to just enough to kick the fan off in both BIOS & any software that supersedes at the OS level.

I don't think there's nearly enough documentation on modern fan control methodology to be honest. Certainly the advertising glosses over the fact that MB headers are actually voltage regulated. There's only two true PWM signal headers on the boards; the main CPU fan header and the CPU OPT header.

Indeed one could well plug in a splitter and daisy chain the PWM signal around their case. However doing so only gives you PWM in the context of the CPU and NOT the individual temperature sensors located around the board. So in that respect it ends up being totally pointless and even detrimental for case fans; imagine the scenario where the GPU or the chip set gets hot ahead of the CPU. Basically your context is not globally considerate.

Yes fan speed control is argued to not matter with case fans. I disagree, we have a feature-set there, why not take advantage of it. Personally I can't be bothered to use a control unit as I'd prefer for the software to do what it's been designed for therefore saving me flicking switches and groping knobs all the time.
 
Last edited:
I have an ASUS board (PWM control on CPU headers, voltage control on other headers) and much prefer it to all round PWM as voltage control is a much better option.

PWM is fine for CPU fans where in the event of a problem you would want them to ramp up to ludicrous speed, but it's totally pointless on case fans which will never be needed to go silly fast. Voltage control is a much better option as you directly control the fan instead of using trickery to do it, PWM is just a cost saving feature at the end of the day and the drawbacks compared to voltage are well documented.

*EDIT*

If your the guy posting in the thread you linked then to make them run full speed on boot simply disable the Q-fan setting in the BIOS and just use AIsuite to control speeds while in Windows (the BIOS setting is supposed to be a set and forget curve choice, you don't need it if oyur using AIsuite too). However what's happening in that video shouldn't be able to unless the is a problem with the board/fans or something is set very wrong as it appears the fans simply aren't getting enough power to run, maybe try messing with the minimum RPM setting in the BIOS.
 
Last edited:
ubersonic said:
I have an ASUS board (PWM control on CPU headers, voltage control on other headers) and much prefer it to all round PWM as voltage control is a much better option.

PWM is fine for CPU fans where in the event of a problem you would want them to ramp up to ludicrous speed, but it's totally pointless on case fans which will never be needed to go silly fast. Voltage control is a much better option as you directly control the fan instead of using trickery to do it, PWM is just a cost saving feature at the end of the day and the drawbacks compared to voltage are well documented.

*EDIT*

If your the guy posting in the thread you linked then to make them run full speed on boot simply disable the Q-fan setting in the BIOS and just use AIsuite to control speeds while in Windows (the BIOS setting is supposed to be a set and forget curve choice, you don't need it if oyur using AIsuite too). However what's happening in that video shouldn't be able to unless the is a problem with the board/fans or something is set very wrong as it appears the fans simply aren't getting enough power to run, maybe try messing with the minimum RPM setting in the BIOS.
Click to expand...

Yeah exactly. No I'm not the guy in the thread.
 
Asus claims their Z97's are PWM/DC and will set themselves correctly depending if you plug in a 3pin fan or a 4pin fan and that is for all headers (read on Asus forum last week),

AFAIK I need to change mines in Bios (Z87) as they are not auto.

PWM is overated IMO and it ticks.
 
PWM is the superior technology as there are fans that do not even start at 7-8V, so they can never be truly silent. Moreover, changing the voltage linearly does mean the change in RPM will be linear. PWM does not turn the motor on and off as many people think. The motherboard/fan controller sends a 5V signal, that is pulse width modulated, that determines the percentage at which the motor should spin. Then there is a tiny PCB with a chip that regulates the motor's speed.
 
Last edited:
That is your opinion but I do not share it and I can hear the ticking.

Reason why until recently Noctua did not do any PWM fans and made their own in-house PWM controller but it still ticks but less than others.
 
Last edited:
I have about 10 PWM fans, not a single one ticks. And my ear is very fastidious, I can clearly hear 140mm fans at 800RPM from the other end of the room. No ticking. Actually you are the first person I hear that complains about this.
 
I have 12 in two different machines and some tick and some don't, my Akasa 140s don't but the 120s do!
 
jh30uk said:
Again PWM all tick just Noctua are far less ticking, and its not at full speed and not all peeps will hear it.
Click to expand...

No, they don't. I have over 30 PWM fans and none of them tick. Not sure where you are getting this from. Some people on OCN have more PWM fans and I have never heard of a global ticking problem that affects all brands and models.
 
Read the reviews and read what Noctua say on matter as you cannot have done much reading on PWM in that case.

If you cannot hear its due to your hearing and again its only at certain lower speeds due to the PWM freq.

None of this is to do with OP which I did answer BTW, just stating they are not the "be and end all".
 
Last edited:
I already did in post #9.

QUOTE:


Noctua Announces New PWM Fans

Press Release by
btarunr
Wednesday, November 30th 2011 12:21 Discuss (21 Comments)
Enlarging its portfolio of premium quality quiet fans, Noctua today announced PWM versions of its renowned NF-P12, NF-B9 and NF-R8 fans. Having received more than hundred awards and recommendations from the international press, the classic NF-P12, NF-B9 and NF-R8 fans helped to found Noctua's reputation as a top-tier manufacturer of premium grade quiet cooling components. The new PWM versions build on the proven designs and feature Noctua's novel NE-FD1 IC for fully automatic control and ultra-low power consumption.

"Our customers have been asking us for PWM versions of our fans for a long time, but the available PWM fan driver ICs simply weren't up to our quality standards", explains Mag. Roland Mossig, Noctua CEO. "With our NE-FD1, we've now got our custom-designed IC that allows us to build PWM fans that consume less power and remain quieter at low speeds."


Most conventional PWM fan driver ICs put out square signals that cause sudden torque impulses. These impulses can cause minute deformations of the entire fan structure which lead to audible "clicking" noises that are especially noticeable with many common PWM fans running at slower speeds. Adopting Noctua's Smooth Commutation Drive (SCD) technology, the custom designed NE-FD1 PWM IC slows down the slew rate of the output signal in order to give a smoother, less sudden torque impulse. This way, PWM switching noises can be effectively suppressed and the fan runs even quieter, especially at low speeds.

Noctua's NE-FD1 IC was not only fine-tuned for superior quietness, it has also been carefully optimised to allow for ultra-low power consumption. As a result, the new PWM models consume 25-40% less power than their non-PWM counterparts.

All models come with Low-Noise Adaptors, 30cm cable extensions and 4-pin y-cables that allow to connect a second PWM fan to the same fan header for automatic control. Equipped with Noctua's signature SSO bearing and topped off with a 6 years manufacturer's warranty, the NF-P12 PWM, NF-B9 PWM and NF-R8 PWM are a proven premium choice for the highest demands.


http://www.techpowerup.com/155944/noctua-announces-new-pwm-fans.html
 
Last edited:
I found this out when i bought like six apache blacks in my Z77 build and i could hear them tick. I was going mental about that when they cost like £14 each off OCUK. I told them they should be removed from the store or the description changed because i only fired them up outside of DSR period.


£60+ worth of ticking plastic that i felt conned on considering they are silent. Maybe the buzzing is meant to simulate an actual AH-64??
 
I will be honest here, I have taken down PWM fans to 500 rpm or so but not lower. I would like to think my hearing is pretty good so it may be that the tick happens at even lower speeds but so far I haven't had any such issues.
 
You must log in or register to reply here.
Back
Top Bottom