Does PWM work independently of Q-fan?

[setter]

Thanks for the clarification on this wja96, ive used the arctic 12025 pwm fans in the past (albeit only 2) i then moved over to sharkoon silent eagle se's, theese cant be connected together like the arctics, hence my use of the splitter cable, gave me a bit of peace of mind as well as no power was being drawn from the cpu header.

My pwm setup.

2 cpu fans and rear exhaust are on pwm.

 

[twist3d0n3]

i've currently got 6 apaches running off the cpu header, via 2 akasa pwm splitters sliced together.

bit of solder to link the pwm line to both splitter cables with a bit of shrink wrap around and you're sorted

i would post a pic but it's a bit messy in my case at the mo, and i'm a tad embarrassed

 

[Alfie]

But do I need the molex power?

 

[WJA96]

So i'm safe to run 4 fans off one PWM header on the P6TD with the Gelid splitters that don't have additional molex power? If so, why does the Akasa have have the additional molex power? Will I be ok with 4 Apaches? Are they daisy chainable do you know?

I'm guessing that Akasa supply additional power because they don't know what fan header you're pulling power from. The CPU PWM fan header is in the Intel chipset specification, but companies like Gigabyte often supply one or more PWM fan headers for other locations too. These could be powered at almost any level from 3W-almost anything. If you were using one of those you might not be able to daisy-chain the fans.

I'm also interested to know what happens if 1 of the 4 PWM fans in the chain is rated for different RPM?

PWM works by rapidly turning the full power on and off so the fan almost pulses. As it pulses faster or slower the fan runs faster or slower. If you have a 1200RPM fan on the same splitter as 4 other 1800RPM fans they will all run in proportion, so at 50% the 1200RPM fan will run roughly 600rpm and the 1800RPM fans will run at roughly 900RPM. PWM is usually fairly linear between 40% and 100% power, below that you get odd effects with some fans. I like the AC 12025PWMs and I've long been a supporter of PWM as it's the only technology that lets the fan stop completely and restart under automatic control.

 

[WJA96]

But do I need the molex power?

No, but it won't hurt.

 

[setter]

I'm guessing that Akasa supply additional power because they don't know what fan header you're pulling power from.

I was under the impression wja96 that the akasa splitter draws no power from the cpu header? merely the pwm signal with the power coming direct from the psu, i have mine connected to the cpu header on a p6t deluxe, used the same setup on a p5q deluxe as well.

 

[Alfie]

i've currently got 6 apaches running off the cpu header, via 2 akasa pwm splitters sliced together.

bit of solder to link the pwm line to both splitter cables with a bit of shrink wrap around and you're sorted

i would post a pic but it's a bit messy in my case at the mo, and i'm a tad embarrassed

You must post a picture. no one has posted a pic of apaches installed yet. I will of course when I've finished my rig.

 

[TheMrWibble]

1st post so go easy on me
5 arctic cooling pwm fans running of 1 header nps here. (4x 120 , 1x80 , ep45-ud3lr)
Now only if i could tie in one of the front fans to ramp up with GPU temp id be 110% satisfied.

 

[WJA96]

1st post so go easy on me
5 arctic cooling pwm fans running of 1 header nps here. (4x 120 , 1x80 , ep45-ud3lr)
Now only if i could tie in one of the front fans to ramp up with GPU temp id be 110% satisfied.

In theory, you'd just need to splice the fan into the PWM header on the graphics card, but the only graphics card spec I have seen (8800GTX) only allowed 5W on the fan header so it might struggle to power two fans. I've certainly bodged a 120mm PWM fan onto the graphics card fan header with a Thermalright HR-03 heatsink in the past and it worked brilliantly.

 

[TheMrWibble]

Could i take the pwm signal only from the gfx card and power the extra fan off a mobo header?
Tbh i didnt think the gfx card would be a pwn type fan and more just a couple of voltage jumps but thinking about it it must be as it steps up nice and linear with the fan profile i bunged on the nvidia control panel hehe staring me right in the face so to speak
Big thankee m8ty ill go dig the old weller out.

 

[twist3d0n3]

You must post a picture. no one has posted a pic of apaches installed yet. I will of course when I've finished my rig.

i should have some screws arriving today to properly attach the 3 newer apaches on the underside of the rad, so once i've got them i'll take a few snaps and upload them

 

[Alfie]

I have set up 4 PWM fans (3x Akasa Apache + 1x Enermax Cluster) on my CPU 4 pin fan header. I can confirm that by daisy chaining witht the Gelid splitters you do not need additional molex power so I don't know what Akasa are up to with their cable.

I have to say I am so glad I went for it in the end, it's a work of art to have a silent computer and then watch all your fans whoosh to full speed when you run prime 95.

I highly reccommend this setup to everyone, PWM is the way forward!

 

[Big.Wayne]

Glad your glad Alfie!

Fact is the apaches wont spin up properly with any Q-fan enabled [on the P6TD Deluxe]

I wonder if that is your your board or all boards? . . . seems like the fans don't like running with whatever low voltage your board is giving them with Q-Fan . . . *or* do Apaches just not run with reduced voltage period?

PWM is the way forward!

Seems to be a nice technology as you have found out some fans motors don't work properly if not given the full 12v, thats been a problem for years but most fans do work ok with as low as 7v . . .

I guess we will start too see motherboards featuring only 4-pin fan connectors soon!

 

[WJA96]

I wonder if that is your your board or all boards? . . . seems like the fans don't like running with whatever low voltage your board is giving them with Q-Fan . . . *or* do Apaches just not run with reduced voltage period?

Well, if they're fed power from the additional 4-pin molex then they'll always have 12V on them. That could be part of the problem. If they're only using 3 of the 4 pins on the connector then they should revert to 'normal' mode and if they are using motors with a high stall voltage I would think they would just stop and not restart.

Seems to be a nice technology as you have found out some fans motors don't work properly if not given the full 12v, thats been a problem for years but most fans do work ok with as low as 7v . . .

From personal experience the stall voltage on all 12V fans is about 4.5V which may be why most rheostat fan controllers only go down to 5V. I love PWM fan control - the mCubed Tbalancer I've been using for the last few years has developed into probably the ultimate noise/temperature control system and I agree, it surely can't be long before we see cases or motherboards with that sort of system built in.

 

[Alfie]

to be honest, it's totally crazy that we've had the 4 pin PWM for so long and it's not the standard. especially as it's backward compatible with 3 pin fans.

 

[setter]

My pwm fans usually start of very slow as well, i have to hit F1 to continue the boot sequence as the, no cpu fan warning pops up. Once in windows they run properly, i use asus ai suite's fan xpert to configure fan speeds, qfan enabled in bios on the turbo profile.