Gigabyte Z97X UD5H 4 pin fan headers are not all PWM???

beekermartin

Distinguished
Jan 22, 2008
580
0
19,010
I am in the process of building my first custom water cooled loop. A few months ago I knew I wanted a fan splitter that was powered by the power supply and not the fan header. I ended up getting this: https://www.amazon.com/Swiftech-8W-PWM-SPL-ST-Way-PWM-Splitter-Sata/dp/B00IF6R4C8
I have that installed on the CPU header and five radiator fans connected to it. It works perfectly when connected to the CPU header.
The issue I am having is the EK D5 pump I am using is also PWM and powered by the power supply. Essentially the fan block and the pump are wired the same. The 4 pin connector only uses the third and fourth wire. The third is the rpm sensor and the fourth is the PWM signal. The ground and power wires go directly to the power supply. My issue is the 4 pin headers on this motherboard are not all the same. Only the CPU header works as a traditional PWM header. The remaining 4 pin headers all control speed by adjusting the voltage not by using a PWM signal. Gigabyte calls it VCC for the fourth pin on those headers. So with the pump, or the fan block, connected to any 4 pin header besides the CPU header the pump or fans all run at 100%. That is because there is no PWM signal being sent they run at 100% all the time.

WTF! Why did Gigabyte do this??? Is there a work around like a setting in the bios that will make the headers work like a traditional PWM 4 pin header? What is the advantage or reason that Gigabyte did this?
 
have you tried Gigabyte's fan control software?
it let's you set individual curves. fan speed is controlled by power regulation, works like a charm and eliminates the PWM sound at low rpm.
at least with my Z170-Gaming 7 which I think has the same configuration
 

beekermartin

Distinguished
Jan 22, 2008
580
0
19,010
The problem is the Swiftech splitter and D5 pump both only have a 3 and 4 pin wire. Since, it appears, all of the four pin headers except for the CPU header don't regulate the speed using the 4th pin PWM signal I don't see how it could work. It can control fans using the 12v 2nd pin connector by reducing the voltage from 12v. Which is how traditional 3 pin non PWM fans are controlled.

Essentially Gigabyte drops PWM on all but the CPU header and I am beyond pissed about it. I don't understand why they would do this. PWM is basically the standard and has been for years. Why would Gigabyte not have it on all their 4 pin headers???

I found this rant on Overclock.net:
Awesome rant. (Click to hide)
You can completely judge a company on the tricks, lies, and lack of support that they offer their customers.

With todays motherboards and the technology available there is NO reason at all that we should be forced to buy multiple hundred dollar fan controllers. Don't fool yourself $50-100 fan controllers suck... and as for $300 fan controllers? Well there are motherboards that as long as you use PWM fans can do almost everything that those expensive fan controllers can do.

The thing is that Gigabyte STARTS OFF on a bad foot by lying through their TEETH!!! On the cover of the boxes they say they have great fan control RIGTH ON THE BOX. Yet I have what I think is the most expensive Gigabyte Z87 board The OC FORCE and it has what 1 single PWM header? Even the secondary head is NOT PWM. Yet in the BIOS it lists ALL headers by PWM%. Don't call it PWM % since it isn't PWM.

For gigabyte if the header is BLACK it is 3 pin voltage control with a 4th pin that according to gigabyte is there because some PWM fans don't run at 100% when hooked up to 12v and no PWM signal. Well why don't you just spend the extra and give us real PWM control.


THE BOTTOM LINE!!!

There are companies that offer multiple channels of PWM control AND external thermal probes that you can use to control those PWM channels!!!

What does this mean???

It means that you have complete control AND if you have water-cooling you can connect a probe to an inline sensor so that it can handle all your radiator fans.

YOU CANT BEAT THAT!!!

Yet Gigabyte can't even give us 1 spare PWM header on a $400 motherboard!

GARBAGE

Then when a company decided to give a BASIC feature that EVERYONE has and then when it breaks they don't replace it...

WELL that is poor service!


I have to decided what to do next. Either get a fan controller to control all my radiator fans or connect them all to the mother board 4 pin but really 3 pin headers directly using splitters and control their speed using voltage control instead of true PWM. I am so pissed off at Gigabyte for this. I will think hard before buying another product for them.

 
I honestly don't see what all the fuss is about.
I stopped using PWM fans except for my CPU some time ago.

currently I'm using 1xPWM CPU cooler fan and 3xtraditional fans
in Gigabyte's SIV a whole menu is dedicated to fan control
I can set my fan curves for the CPU fan just like I can for my other non-pwm fans
and it works perfectly.

sometimes I think people just rant without even bothering to look at solutions that are right in front of them.
I can't give you the technical explanation as to how the fan speed is controlled by the board, I guess it's due to power management. but I can control all of my 3-Pin fans as my PWM-4-Pin fan the same way with the board's utility, setting curves ranging from 0 / 40% up to a 100% in 1% steps. and I can test this, it's really working and not just regulating in 3.3V / 5V / 12V steps.

so in theory it should work for you as well that way, depending on the splitter.
also usually the CPU AND the CPU_2 header are "real" PWM headers while the SYS-headers aren't, but I don't know that exact board too well I admit.
 

beekermartin

Distinguished
Jan 22, 2008
580
0
19,010


If the CPU OPT was PWM I would be all set. It isn't. Only the main CPU header is. Since I just bought the splitter I linked to and 5 PWM fans I am pissed off. The splitter is essentially useless because I can't use it with this mobo since it only uses the third and fourth pins. Essentially the fourth pin is not used on this POS mobo except for the CPU header. I have no choice but to connect the D5 PWM pump to the CPU header and I will need to find another way to connect the 5 radiator fans to the motherboard. I know I can use traditional splitters that will draw power from the mother board header. I can then control the fans using voltage control. I didn't want to do that because I didn't want all 5 fans drawing that much power from the mother board headers. I don't have a choice at this point. All because Gigabyte saved a few pennies by not using PWM on all 4 pin headers. I understand if they wanted to save a few bucks. I don't have a problem with that. Just use 3 pin headers instead so there is no confusion.

Is it ok to use voltage control with a PWM fan?


 

beekermartin

Distinguished
Jan 22, 2008
580
0
19,010
I already have a two way 4 pin splitter. Should I just get this for the three remaining fans? https://www.amazon.com/gp/product/B004CLFOK4?tag=lpastore-20

That is only if it is ok to control PWM fans using voltage control instead of an actual PWM signal.

The fans I am using are these: https://www.amazon.com/Cougar-Vortex-Cooling-CF-V12HPB-Black/dp/B008N323U6

From what I have found they use a maximum of .18 amps at 12 volts. So 3 will be well below the recommended maximum of 1.0 amps on a single header.
 

beekermartin

Distinguished
Jan 22, 2008
580
0
19,010
I ordered the 3 fan splitter. I will run two fans on the 2 fan splitter I already have and 3 fans on the new splitter. This is the easiest solution without getting a digital fan controller. I like this solution better because I can set custom curves for the pump and fans using the same Gigabyte SIV software.

I am still pissed at Gigabyte for putting 5 four pin headers on this mobo but only one of them is a real PWM 4 pin header. The others are really three pin headers with an extra useless pin. This shouldn't be allowed and is complete BS in my opinion.