Sign in with
Sign up | Sign in
Your question
Solved

P55A-UD3 FAN CONNECTIONS??

Last response: in Motherboards
Share
June 21, 2010 5:33:14 AM

Hi, I have the P55A-UD3 Motherboard and the HAF 922 case with 3 fan's that already come with it, and another fan I bought to put on the side. However, there appears to only be 3 fan headers being, SYS_FAN1, SYS_FAN2, and PWR_FAN on my motherboard. I was wondering if anyone was knowledgeable about this MOBO's fan connections and could let me know which fan headers I should use?

I heard PWR_FAN is only for the power supply fan, and SYS_FAN2 has 4 pins with the last one being a "reserve pin", so could I hook up one of my 3 pin fans to that header, leaving 1 pin unoccupied, or would that be dangerous? I would like for my fan speed to vary with the computer's load but I don't mind settling for a constant fan speed with Molex connections, but I'm not sure if that means the fan's will run at full blast all the time... which would be annoying. Could someone PLEASE provide some insight on this topic? (I'm hoping someone has the same motherboard and number of fan's, and could tell me how they managed?)

a c 177 V Motherboard
June 21, 2010 3:05:10 PM

A standard, four-pin, PWM (pulse-width modulated) works like this: pin 2 supplies an 'always on', full 12V; the 'speed control' pin, pin 4, supplies a low-current, constant frequency (20-25KHz), square wave whose 'duty-cycle' (the British have the best name for this - 'mark-to-space-ratio', essentially the ratio of the time the signal is on [high-5V] to off [low-0V]) is varied:

what then happens is the (pin 4) modulated 5V 'control' signal operates electronic switches, on the fan, to likewise vary the 'duty cycle' of the 'steady' 12V power supplying the fan, the 'average on-time' of which equates to the fan's speed...

The neat trick GB does (well, actually - the IT8720 LPCIO on your particular board [low pin count input-output], which 'does' the fan control), with the SYS_FAN2 headers, as well as with the CPU header, if the "CPU Smart FAN Mode" is set to 'Voltage' on the "PC Health Status" page of the BIOS is that they 'lock on' the 'control' (5V) signal on pin 4, by simply supplying it with a steady, unmodulated 5VDC, which also 'locks on' the switches on-board a four-pin, PWM fan - then they similarly 'modulate (like the signals shown above, only with a 12V 'peak') the 12V power supplied on pin 2 - which means the header will work with either a 4-pin PWM, or 3-pin 'standard' fan!

What I do with all client machines I build for folks is use three fans, all PWM - one CPU cooler (an usually not a particularly 'pricey' one), one case intake, and one case exhaust - and I 'daisy-chain 'em off the CPU_FAN header (right-hand image):

You can do a minimum of five fans this way, off a single header, and that way the machine is whisper-quiet when lightly loaded, but can move air like a vacuum cleaner when loads peak...

Hope this helps - any question, feel free to post 'em [:bilbat:8]
June 21, 2010 3:53:08 PM

Aha! a brilliant response for a solution. For the time being however, until I can get some appropraite cables to daisy-chain them I will settle for using just the SYS_FAN1 and PWR_FAN 3 pin fans assuming I read your response correctly and it's safe to use the PWR one for a regular case fan.

Building this computer has been a real eye opening experience lol.
Related resources
Can't find your answer ? Ask !

Best solution

a c 177 V Motherboard
June 21, 2010 4:51:26 PM
Share

Well, NE has two products to accomplish this:
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
but, so far as I am aware, neither one is wired correctly!
I usually order the Rosewill part, and a couple 12" fan extension cables to 'canibalize' for parts... The problem with the Rosewill: it's odd to begin with, as it 'acquires' its twelve volts from a MOLEX (which really doesn't matter - twelve volts is twelve volts - and this bypasses the 'five fan limit' [minimum - maximum depends on the fans themselves, and their current draw] imposed by the current available from a PWM [four-pin] header), but only one of the 'fan feedback' pins is connected!

The 'feedback pin' (pin 3 - 'speed sense'), is connected to a hall-effect switch on the fan, that is activated (usually, twice per rev) by a magnet on the rotor assembly - and is what the board uses to do 'alarming', i.e., tell you when a fan has 'stalled'. As shown on the diagram, you want to connect this pin on each fan, to a seperate header on the board, so you can 'see' each fan's speed individually. I don't mind the 'surgery', as fan pins are easy to release from the connectors, and I typically solder up every cable 'to length', and sleeve 'em anyway...



These boards have departed from GB's usual 'standard' - on the typical GB MOBO, the PWR_FAN header 'speed sense' pin is not connected, and therefore not 'seen' on the fan speed page, nor can it be 'alarmed'... It's usually just a two-pin, 12V and ground connection, with a three pin header - this board, my guess, with the exception of CPU_FAN and SYS_FAN2, doesn't care what's connected to what! I have read info from GB tech that says the SYS_FAN2 speed is 'slaved' to the 'system temp' sensor (which, in theory, should be NB temp) - but, in practice, I've not seen it!

Another route is to use Arctic PWM case fans:
http://www.newegg.com/Product/ProductList.aspx?Submit=E...
Most already have the 'splitter header' wired in for this (and correctly!) - they are where I got the idea from in the first place!
"PWM Sharing Technology"
June 28, 2010 6:38:09 PM

Best answer selected by Teadrinker_OG.
!