Phanteks Enthoo Pro fan controller??

[GamingPC 4]

I am going to build a gaming pc and am planning to use the Phanteks Enthoo Pro case. I have looked at the case and it seems to have a fan connector that connects 3-pin fans to the motherboard so they can be used as 4-pin ones. Because they will work as 4-pin fans, how will I control them? Will I need an additional fan controller or do I have to control them on a program on the computer?

Thanks in advance to everyone who answers.

 

[Shneiky]

3 pin Fans are controlled by their input voltage. Usually 5V is the lowest speed and 12V is their highest rated RPM.

4 pin fans are controlled by PWM. That stands for pulse width modulation.

You can't control 4 pin fans when plugged into 3 pin motherboard fan header and vice versa. It will send the fan at maximum RPM immediately.

You can plug 3 pin fans into 3pin fan headers and 4 pin fans into 4 pin fan headers and control them in the BIOS. You can set their idle speed and let them ramp up if your temperature increases significantly.

Or you can also pick up a software that you have to deal with withing Windows.

I am not certain how the Ethnoo fan controller actually is, but what they describe is that you can plug fans into their hub and then plug the hub to the motherboard. If it so, than both the BIOS and the software solution works.

I use NZXT Sentry Mix 2 fan controller. You can plug in 3 or 4 pin fans and control them by sliders. I prefer the analog approach.

 

[Amencerment]

I have the Ethoos Primo, and the fan controller that came with mine is a 3 pin x 6 fan setup. It does come with a 4 pin to MoBo cable but I did not set that up as I connected it by the Molex as all 6 of the fans I use are non-variable set speed.

For the Variable fans and I have 5 I use the Bitfenix Recon fan controller to set the fan levels at certain temperatures. Now I use speedfan for my monitoring.

 

[hwc1954]

The Phanteks fan hub is supposed to generate a variable voltage signal proportional to the 4-pin PWM signal from the MB. This is supposed to control the 3-pin fans connected to it.

It's really a brilliant product, on paper. Problem is that it doesn't work. It only controls the speed of the 3-pin fans over a very tiny range of a couple hundred RPM. For all intents and purposes, they just run full speed.

It would probably be a pretty simple thing to adjust some resistor values and make the thing work. There's no reason somebody can't build one that does work.

 

[synphul]

Something else to be careful of with the enthoo fan controller since it splits up to 6 fans off one mobo fan header, be aware of what size fans they are and how much power they draw. It's not a good idea to overload the fan headers and a 6 to 1 splitter may allow that to happen depending how many fans and what their max draw and starting voltages are. Not saying you're guaranteed to burn out your mobo's fan header, but it's a possibility if that fan hub is full.

Another reason for it not having as wide a range of control may be due to trying to create support for a wider variety of fans. There are a number of 140mm fans that won't even start spinning until around 6-7v. Between 5-6v and they just twitch a little without enough juice to actually get them going, too low of voltage range may make for quite a few fans that appear not to work with it and would cause even more headaches for users. It would create a headache scenario of then trying to determine of all the fans out there, which are compatible and which aren't. If you prefer not to mess with fan profiles in the bios or through software there are hardware fan controllers (both digital and rheobus style) that work extremely well for fine control.

The tradeoff is the looks. I have a 4 way fan controller mounted in a 5.25" bay and even though it's possible to set them via software or through the system I've found it's much easier to control them manually. No software compatibility issues, no fiddling with settings. Too loud? Twist the knob until the sound matches my preference. Too warm? Turn it up some. Plus it allows me to slow the fan just enough to deal with the acoustics. For instance many controllers whether switches or noise adapters are locked at 5v, 7v and 12v. The sound levels of a particular fan may improve significantly at 8v or 9v and provide better airflow than 7v so I'm not stuck with a particular voltage and get better sound control without unnecessarily sacrificing airflow.

 

[hwc1954]

The Phanteks fan hub gets around the header limit issue by taking a PWM signal from the 4-pin header and the +12V signal from a PSU molex. It would be a fantastic product if it worked.

A floor of 6 or 7 volts would be fine. In fact, 6 or 7 volts would be perfect for case fans which you typically want to keep moving a little air even when the CPU is not loaded.

The problem is that the Phanteks only goes down to 10.5 V, and only then with a rock bottom 20% PWM input signal. So the operating range is from 10.5 V to 12 V and it only hits 10.5V when CPU temps are in the 20C range. For all intents and purposes, it runs the fans 100%, something that can be easily accomplished without it.

Really a shame. Fantastic product that is simply not tweaked properly. Many MBs implement excellent voltage control of 3-pin headers driven by the PWM signal. There's no reason an aftermarket hub, like the Phanteks, couldn't do the same thing.

 

[synphul]

No it doesn't. Maybe that's why your fans are running so close to full speed, the use of a sata power cable with the enthoo fan hub causes them to run at full speed. From the user manual, page 31:

"The 12v sata power cable can not be used to power the PWM hub if connecting to these types of 4-pin connectors, due to the interference with the RPM regulation by voltage (resulting in the fans running at full RPM). The PWM hub will draw its power from the 4-pin connector, which is limited to a total device consuming 30w in total."

Also noted in the owner's manual that it can only read one PWM signal, so if connected to the cpu fan header on the mobo it will relay that rpm signal to the rest of the fans connected to the hub and make them all run at whatever speed is set to the particular fan header used on the mobo.

 

[hwc1954]

No it doesn't. Maybe that's why your fans are running so close to full speed, the use of a sata power cable with the enthoo fan hub causes them to run at full speed. From the user manual, page 31:

"The 12v sata power cable can not be used to power the PWM hub if connecting to these types of 4-pin connectors, due to the interference with the RPM regulation by voltage (resulting in the fans running at full RPM). The PWM hub will draw its power from the 4-pin connector, which is limited to a total device consuming 30w in total."
.

That's section of the manual is referring to fake 4-pin headers that are really variable voltage controlled, not PWM controlled. A true PWM controller has a fixed 12V wire (doesn't matter where it comes from) and a pulsed PWM signal that modulates the 12 volts. You can test a PWM header by removing the 4-wire. The fans will accelerate to full speed. This is what happens on both my Dell and Asus motherboards -- not surprising since both have PWM controlled 4-pin CPU headers that work with PWM fans according to the Intel spec.

With a fake PWM header, you can only use the Phanteks as a dumb splitter. Fans connected to it would go up and down in speed with the variable voltage, just as they would without the Phanteks.

I tried the Phanteks without a SATA connection, just to test the possibility that my Dell and my ASUS PWM fan headers were not actually PWM. Didn't make any difference. Same misniscule range.

The design flaw of the Phanteks fan header is highlighted in this review:

http://forums.vr-zone.com/hardware-depot/3167542-phanteks-pwm-fan-controller-hub-review.html

He measured a 10.6 volt output with a 20% PWM input signal (which is idle speed for the CPU fan). So even at idle, fans connected to the Phanteks are going to be running at 10.6 volts (90% of full speed).

 

[hwc1954]

Also noted in the owner's manual that it can only read one PWM signal, so if connected to the cpu fan header on the mobo it will relay that rpm signal to the rest of the fans connected to the hub and make them all run at whatever speed is set to the particular fan header used on the mobo.

That's what it is supposed to do. Read the PWM signal from a 4-pin PWM header on the MB and adjust the speed of voltage controlled 3-pin fans connected to the Phantek hub. It would be fantastic product if it actually did that. Unfortunately it dosn't. Well, at least not in any useful way. As the PWM signal based on CPU temp varies from 20% speed to 100% speed, the Phanteks voltage going to the 3-pin fans varies from 10.6V to 12V. Useless. Might as well just run them at full speed.

 

[synphul]

All I'm pointing out is the fact that the manual clearly states it simply as - if you use the external power, the fans will run at full speed. Regardless of delving into pwm specifics and how other standard pwm's operate and trying to overthink it. If not following the directions it's hard to expect different results. I'd say the fan hub is working properly and chalk it up to user error. Try running your fans directly off the mobo without the sata aux power to the fan hub and I'll bet you get different results.

Take any 4 pin fan and using an adapter plug it directly to either sata or (more commonly) molex and it too will run at full speed. That's how fans have worked for the longest time now. Using direct power from the power supply negates pwm and runs fans full out. Plug that same 4 pin fan into the mobo and it will work properly - aka, leave out the external power.

The supply cable for the fan hub doesn't have to be plugged into the cpu fan header, it can be plugged into another mobo fan header. If in fact for whatever reason it's running your cpu fan at too fast of a speed/too much power for your other fans, plug the cpu cooler fan into the cpu fan header and run the hub off a different one and it won't have them cranked so high.

 

[hwc1954]

No. The manual does NOT state that simply. Just the opposite, in fact. It says to connect the SATA power cable when using PWM CPU header and to NOT connect the SATA cable when using a voltage controlled header. You quoted the manual out of context:

Connecting the 4-pin to CPU_FAN

For full PWM functionality, Phanteks’ PWM hub requires users to connect
the 4-pin connector to the “CPU_Fan” connector of the motherboard,
because all motherboard manufacturers implements a PWM
signal modulation on this connector. Connect the SATA 12V power to
power the PWM hub. Not all motherboards have their CPU_Fan connector
set on PWM signal modulation by default. Please consult your
motherboard documentation for this matter.

Connecting the 4-pin to other 4-pin header (besides the CPU_Fan)

Other 4-pin connectors can be found on modern motherboards besides
the “CPU_Fan” connector (e.g. “CPU_Fan2”, “CHA_Fan”, “OPT_Fan”),
however not all motherboard manufacturers implement a true PWM
signal modulation onto these connectors. These type of 4-pin connectors
modulate the RPM by voltage, which has a smaller control range
compared to modulation by true PWM signal.
The 12V SATA power cable can not be used to power the PWM hub if
connecting to these types of 4-pin connectors, due to the interference
with the RPM regulation by voltage (resulting in the fans running on
full RPM). The PWM hub will draw its power from the 4-pin connector,
which is limited to a total device consuming 30W in total.

It's all academic. The Phantek fan hub doesn't work as intended no matter how you hook it up. It is only useful as dumb splitter when connected to a variable voltage header, limited to the ~1A current capability of the header. Might as well just buy a $5 splitter.

It's a shame. A fan hub that actually does what the Phantek is supposed to do would be a very useful product, allowing you to control a large number of 3-pin fans with the thermal controlled PWM signal of the CPU header.

Try running your fans directly off the mobo without the sata aux power to the fan hub and I'll bet you get different results.

I did try that. only one connection to a perfectly functioning PWM header. No SATA connection. Same results. Fans ran at or near 100% speed. No useful speed control whatsoever. Trust me, I wanted the thing to work. I bought it because I thought it was a really useful product. What can I say? It just doesn't work as it supposed to work.

 

[synphul]

What I'm wondering, if say you have the fan hub set to run off the cpu fan header (for example since it'll be most likely to be true pwm) - is your cpu fan being controlled across a broader range by itself, without the hub used? I mean are you getting a range of 20% fan to 100% fan speed control using just the cpu fan on the cpu header (mobo) using pwm? This will tell how well the root of the pwm is coming off the motherboard for a baseline. Since this pwm signal is what's being sent to the fan hub, if the pwm control is a very narrow range or if for whatever reason your cpu fan is having to run faster to keep the cpu cool, it will force the rest of the fans on the shared connection to run faster as well.

I'd find it difficult to believe that a pwm controlled cpu fan (say on a cooler, single fan to make it more simple) is being controlled with a range of 60% speed variation, allowing the cpu fan to drop down low and for some unknown reason forcing the other fans to run full out when using the hub. If your cooler fan is running at 80%+ speed you can't expect the other fans on the shared hub to run any slower than that. If in fact your cpu fan is running at say 40%, then I can see expecting the fan hub to drop the others to 40% speed as well.

 

[hwc1954]

Yes, the PWM CPU header on the Dell and on the ASUS motherboards work perfectly. I've had the CPU fans down to 20%. They'll go lower, but that's the low cut off I set in both SpeedFan on the Dell and in BIOS on the ASUS, because I really don't want the CPU running with fan speeds below that. I actually had the ASUS set for a few minutes to have 0% fan speed at 20C CPU temp. It dropped the CPU fan speed down to around 300 rpm (on a 2000 rpm fan). So, yes, there is a full range of control on PWM headers (depending on the temp/fan curve you set up or manual adjustment in SpeedFan.

The Dell has properly implemented variable voltage 3-pin headers that track the PWM temp curves and fan speeds. The only difference is that the 3-pin headers will only drop the fan speed down to 40%, but otherwise they track the same, clearly using the same PWM signal as the underlying input.

I fully expected the Phanteks hub to work similarly and was quite surprised when it didn't. It's a pretty simple IC chip voltage regulator circuit to produce a variable voltage DC output based on a PWM input signal.

 

[Amencerment]

Again let me explain this... Now this should be the same from the Primo and lower in the Ethoos line...

I have all my "Static Fans" no speed control on the Ethoos fan controller that means it will all run at the same speed of the 6 fans I have in it's place are 140mm roughly 80ish CFM so they kick up about 22+DB, and the viable fans I have on Bitfenix Rouge fan controller with the speeds adjusted for the temp around the sensors.

The 3 pin fans will run at 1 continuous rate with no variation in speed, now the fans attached to the controller are 3 pin, but they have a viable adjustment for the speeds. These fans I have accompanied with heat sensors so they adjust to the levels desired.

For the Ethoos controller I would not consider it a true controller as all 6 fans are static with no variation.

 

[synphul]

True, I wouldn't call it a true 'controller' either since it doesn't really control anything. At most I'd consider it a fan hub (aka splitter) with what's supposed to be 'smart control' in relaying the pwm signal. Far as I know it's supposed to work if plugged into the mobo without aux power and aux power overrides this and runs all at max speed same as plugging a variable speed fan directly to a molex. The only time I've seen a variable speed fan work on a molex is when it's equipped with an actual switch which is entirely different. I'll have to try it sometime and see what I can come up with in regards to using the fan hub. It's an ok perk I guess, though in my particular situation isn't needed since I have a rheostat hardware fan controller I plan to use. Still waiting for my replacement side panel to come in. I've never been fond of software fan controls either bios or speedfan since they have been known to be glitchy. Once I get my system transferred into the pro case it'll be easy enough to test the fans with and without the fan hub vs the hardware controller.

I don't trust software sensor reports 100% especially when it comes to fans. In either speedfan or hwinfo64 it appears like my cpu fan is running around 800rpm but no way to confirm it. I tried running prime95 and intel burn test to see if I could get the fan speed to ramp up and it barely changed (so it may not be reporting the cpu fan rpm's). Although I have a difficult time telling since my cpu cooler fans are pretty quiet whether my cpu is oc'd or not, even under stress testing I haven't heard them 'rev up'. The only fan for sure I can tell is affected is the gpu fan because it whines so loud at higher rpm's it's easy to identify over everything else. Maybe I'm not hearing the cpu fan over the rear exhaust (scythe s-flex) which admittedly isn't the quietest. It's getting replaced with the new build so maybe I'll be able to discern differences in software/bios fan controls better once the whole system is quieter.

 

[Amencerment]

Yeah I agree 100% with the fan "hub" on the Ethoos, that is why I just use the Molex instead of plugging it into my MoBo. I know all 6 fans are going to run at the same speed so I prefer not to tax my MoBo as I use a pretty beefy PSU.

With my controller and the heat senors they are 1 unit with no software, The unit (Bitfenix Recon) has a grouping (5) temperature sensors placed near the controlled fans.

I use the speed fan to watch my, well fans on my 3rd screen but not to control the fans, it is purely a monitoring software.

 

[hwc1954]

The only time I've seen a variable speed fan work on a molex is when it's equipped with an actual switch which is entirely different. I'll have to try it sometime and see what I can come up with in regards to using the fan hub.

I have four Noctua PWM fans connected with a molex to the PSU and controlled by the PWM CPU fan header. Simple splitter that takes the PWM (4th wire) and TACH (3rd wire) connection from the CPU fan header. Takes the +12V and Ground from a PSU molex. Sends them to four 4-pin connectors, one for each fan. Works perfectly with PWM fans. Plug a 3pin fan (no PWM signal) into the splitter and it runs at full speed. PWM fans vary in speed depending on the PWM signal.

I've had no trouble getting SpeedFan (once I figured it out) or the ASUS bios to report accurate speeds and control the fans. I prefer the BIOS, but only because SpeedFan reverts to 100% fan speed if you close it or log out of Window. I have my BIOS profile to run those fans at 10% at 20C and 100% at 50C (AMD CPU) with a smooth curve in between. That's pretty aggressive cooling. Setting them to run 100% at 60C also works fine and keeps them a little slower most of the time. Right now the CPU temp is idling at 30C and the PWM fans are running at about 45% speed (900 rpm). The BIOS uses CPU socket temp as the input. Speedfan can use either socket temp or core temp. The main difference is that the core temp falls immediately when the load stops. The CPU socket temp (and fan speed) falls a little slower.

These are two fans on the CPU cooler, one rear exhaust fan right behind the cooler and one front bottom intake fan.

 

[hwc1954]

If I were building a system from scratch, I would just get all PWM fans and control them all from CPU PWM fan headers using the simple splitter described above. Easy and effective, especially if your BIOS lets you customize the fan temp/speed curve.

 

[Marc Sayer]

Synphul, hwc1954 is right, you are misreading the manual. The comment, "The 12V SATA power cable can not be used to power the PWM hub if connecting to these types of 4-pin connectors," is not referring to a CPU 4 pin connector, or any other true PWM 4 pin connector. It is referring to the type of 4 pin connector in the preceding paragraph in the manual, that describes a type of 4 pin connector that while it has 4 pins, works like a 3 pin connector and relies on voltage to adjust speed, rather than a PWM signal. When hooking up to one of these non-pwm type 4 pin connectors, you will override the voltage control feature if you hook up the sata connector as well. BUT with a PWM 4 pin connector this is not the case. You are supposed to hook up the sata connector with the PWM type 4 pin connector. This will ensure there is sufficient power for all the fans, but the voltage will still be regulated by the hub interpreting the PWM signal and using that to control voltage to the 3 pin connectors in the hub.

I recently built a computer using this case and hub. All my case fans (6 of them) are controlled via the hub. The hub is plugged into the CPU connector on the mobo (asus p6x58d premium). And the hub is also connected to one of my sata power cables. I use the asus ai suite fan control program to control the speed of the cpu fan connector which controls all my case fans through the phanteks hub. My h80i is plugged into an uncontrolled 3 pin connector thereby supplying a full 12v to the h80i and leaving the real cpu fan speeds to the built in h80i fan control.

I did not experience the inability to slow the fans down that hwc1954 is describing. In fairness I did not measure minimum voltage, or actual fan rpms, I simply relied on the sound of the fans (since that is the thing I want most to control, to keep the computer quiet unless it is running hard and getting hot). But using the hub I was able to reduce the speed of the case fans so that they are virtually inaudible when set to minimum. And they still hit full speed when needed. For all intents and purposes, when plugged into the hub they appear to behave exactly as they do when plugged into a controlled 3 pin connector on the mobo, slow to the point of being almost silent when the computer is idling or running cool, and full speed when the computer is under a significant load or running hot. The asus fan control program gives me several preset profiles as well as a fully customizable profile. And it certainly seems to be able to fully control the 3 pin fans connected to the hub. They will not drop down to the point where they stall, but they do get very, very slow. So, assuming that the hub really can't drop the voltage as low as a 3 pin connector could, it still gives a pretty wide range of speed control, at least in my setup.

 

[CamT15]

Okay just built a PC using the Enthoo pro I was having the same problem and I read most of this conversation and sort of caught on to what you guys were saying.

If I connect all my fans to the "controller" in the back then take the CPU cord from the "controller" and plug it into the CPU fan header on my mother board without plugging in the SATA power cable then the fans will work correct?
I see that the "controller" in the back is really just a way to connect all the fans so that's what I'll use it for, but will what I said work or is that to much power being drown from the CPU fan header on my mother board.
Also I have the corsair hydro 105 does this play into effect any way? I just plug the pump directly into the mother board and it works.

 

[Danny_66]

this is a old thread...i know
i had a look at what everyone is saying and i though i should just try and disconnect the molex from the fan hub
IT WORKS!
and i have it running of the cpu opt

 

[weaselworx]

No it doesn't. Maybe that's why your fans are running so close to full speed, the use of a sata power cable with the enthoo fan hub causes them to run at full speed. From the user manual, page 31:

"The 12v sata power cable can not be used to power the PWM hub if connecting to these types of 4-pin connectors, due to the interference with the RPM regulation by voltage (resulting in the fans running at full RPM). The PWM hub will draw its power from the 4-pin connector, which is limited to a total device consuming 30w in total."
.

That's section of the manual is referring to fake 4-pin headers that are really variable voltage controlled, not PWM controlled. A true PWM controller has a fixed 12V wire (doesn't matter where it comes from) and a pulsed PWM signal that modulates the 12 volts. You can test a PWM header by removing the 4-wire. The fans will accelerate to full speed. This is what happens on both my Dell and Asus motherboards -- not surprising since both have PWM controlled 4-pin CPU headers that work with PWM fans according to the Intel spec.

With a fake PWM header, you can only use the Phanteks as a dumb splitter. Fans connected to it would go up and down in speed with the variable voltage, just as they would without the Phanteks.

I tried the Phanteks without a SATA connection, just to test the possibility that my Dell and my ASUS PWM fan headers were not actually PWM. Didn't make any difference. Same misniscule range.

The design flaw of the Phanteks fan header is highlighted in this review:

http://forums.vr-zone.com/hardware-depot/3167542-phanteks-pwm-fan-controller-hub-review.html

He measured a 10.6 volt output with a 20% PWM input signal (which is idle speed for the CPU fan). So even at idle, fans connected to the Phanteks are going to be running at 10.6 volts (90% of full speed).

WRONG. This Fan Controller will work when used properly! Its running in my case with a exact voltage rage limited ONLY by the fans themselves. The reason I came to this page is because I'm tracking down a voltage short that I believe is coming from this fan controller. WATCH OUT WITH PHANTEKS! My cases RGB headers built into the case is mislabeled/wired incorrectly. In my case they crossed the Green and Blue. I'm suspecting their electronics design or quality control is flawed to be kind. However, this fan controller is usable for FULL motherboard control when used right. The one in the study may have been flawed or broken like a great number of Phanteks products are.

 

[laurenspeek]

No it doesn't. Maybe that's why your fans are running so close to full speed, the use of a sata power cable with the enthoo fan hub causes them to run at full speed. From the user manual, page 31:

"The 12v sata power cable can not be used to power the PWM hub if connecting to these types of 4-pin connectors, due to the interference with the RPM regulation by voltage (resulting in the fans running at full RPM). The PWM hub will draw its power from the 4-pin connector, which is limited to a total device consuming 30w in total."
.

That's section of the manual is referring to fake 4-pin headers that are really variable voltage controlled, not PWM controlled. A true PWM controller has a fixed 12V wire (doesn't matter where it comes from) and a pulsed PWM signal that modulates the 12 volts. You can test a PWM header by removing the 4-wire. The fans will accelerate to full speed. This is what happens on both my Dell and Asus motherboards -- not surprising since both have PWM controlled 4-pin CPU headers that work with PWM fans according to the Intel spec.

With a fake PWM header, you can only use the Phanteks as a dumb splitter. Fans connected to it would go up and down in speed with the variable voltage, just as they would without the Phanteks.

I tried the Phanteks without a SATA connection, just to test the possibility that my Dell and my ASUS PWM fan headers were not actually PWM. Didn't make any difference. Same misniscule range.

The design flaw of the Phanteks fan header is highlighted in this review:

http://forums.vr-zone.com/hardware-depot/3167542-phanteks-pwm-fan-controller-hub-review.html

He measured a 10.6 volt output with a 20% PWM input signal (which is idle speed for the CPU fan). So even at idle, fans connected to the Phanteks are going to be running at 10.6 volts (90% of full speed).

WRONG. This Fan Controller will work when used properly! Its running in my case with a exact voltage rage limited ONLY by the fans themselves. The reason I came to this page is because I'm tracking down a voltage short that I believe is coming from this fan controller. WATCH OUT WITH PHANTEKS! My cases RGB headers built into the case is mislabeled/wired incorrectly. In my case they crossed the Green and Blue. I'm suspecting their electronics design or quality control is flawed to be kind. However, this fan controller is usable for FULL motherboard control when used right. The one in the study may have been flawed or broken like a great number of Phanteks products are.

And what is the correct way to set it up then? True PWM + SATA Power? Or 3-pin(or 3-pin like) connectors with SATA. Or True PWM without SATA?

 

[laurenspeek]

So I´ve disconnected the SATA power cable from the fan-HUB. It was doing 700-780 RPM before constantly (running only the standard fans 1 up front, 1 at the back). I've only set it up to the same true PWM 4-pin connector on my Z370 (System 4) and the RPM seems variable, dependent on temperature. So: if you have true PWM 4-pin connectors on your mobo, don't hook the fan-HUB to the SATA-power connector. Although I do feel that the controller is a little hesitant to go to full speed even when there is a high load on the GPU, CPU. Result is that I manually put it on 100% when I feel it is needed.