[SOLVED] Low RPM Case Fans

[jdfnnl]

I am currently working on a ventilation project involving two 140mm fans mounted in parallel to a 12V battery. Is it possible to find such fans capable of running well at low RPMs?

Would it be possible to connect these to a simple 12V rotary switch and have them work well at low RPMs. Many of the fan specs I've been have the RPM value listed from a number around half its top speed. I actually have a Noctua NA-FC1 for 4-pin fans but the knob is so small.

I already tried connecting two 120mm fans individually, but somehow one no longer works and the other spins only at half its rated amperage. Though I wonder if any of the components in 4-pin fans would be damaged if subjected to 12V current since admittedly I connected the wires at random in my attempt to identify the two motor leads.

Would a 3-pin fan impart better luck?

 

[Darkbreeze]

If you are trying to run on a straight 12v signal with no PWM controller, then you need to use a 3 pin DC controlled fan. On a straight 12v signal, a PWM fan with no controller to cycle the 12v signal will only run at full speed and PWM fans are not designed to run on lower voltages. They are designed to run on a pulse width modulated 12v signal only, unless they are a PWM fan designed for a different voltage, which do also exist, but which would still need an appropriately designed for the specific voltage PWM controller.

Only a 3 pin DC controlled fan, with the benefit of the built in controls on most modern motherboards that allow for the use of both DC and PWM fans, or a similar hardware control device, will be capable of safely (For the fan) running at a lower RPM by applying a reduced voltage.

If you want to use PWM fans, then that Noctua controller is probably your best bet and I believe it comes with it's own power adapter so it does not require a motherboard connection to work.

 

[Paperdoc]

I think you are headed in the right direction, but need to grasp a couple details. 4-pin computer fans in the 140 mm size are well suited to low-speed operations for two reasons. One is that, as larger-diameter units, they are designed for lower speeds than small fans but still good air flow. The second is that the new 4-pin PWM tye of computer fan can be run at lower minimum speeds than 3-pin DC fans that may stall at low voltages. That is because the PWM system gives the motor a small "kick" (well, a tiny nudge) with each pulse of the PWM control signal.

By the way, PWM is widely used with industrial DC motors for speed control, using PWM power supply units. HOWEVER, the PWM fans used in computers are quite different and cannot be used that way. They need the type of PWM fan power system built into computer motherboards. That Noctua Noctua NA-FC1 unit is a stand-alone version of the power supply / speed controller you need for them.

Using 4-pin computer fans with an appropriate PWM fan controller will give you better speed control and slower minimum fan speed than you can achieve with older DC fans and a variable voltage supply. Do NOT try to use that Noctua controller with 3-pin fans - you would get NO speed control!

So, a combination of standard 4-pin computer fans and that Noctua controller will do the job for you. Note that the controller comes with a fan Splitter to allow you to connect up to three fans in parallel to its output. The controller is rated for a max load of 3A at full speed, and most such fans will draw about 0.15 A max each, so no problem. Note also a useful feature of that controller: it has a button you can push to ensure that the lowest fan speed it will allow is 300 RPM (so they don't stall and fail to re-start) , which is pretty slow and probably matches the minimum speed spec of many 4-pin fans you might consider buying.

Here's how the pinout connections work. Look closely at the female fan connector with 4 holes, and the mating male socket with 4 pins on the controller. On the female connector you will see two ridges running down one side, and the male socket has a little "fence" with a slot in it so you can only plug in one way. Looking at that male socket, the "fence" is under the first three pins, and the fourth is outside that. (The female connector ridges fit around the fence.) The pin functions are:
Pin #1 - Ground (- of DC power)
Pin #2 - +12 VDC at all times
Pin #3 - speed signal (see below)
Pin #4 - PWM signal.

The speed signal is a series of two pulses per evolution generated by the fan motor and sent back to the controller on Pin #3.

The PWM signal operates at about 20 kHz and is like a "square wave", except that a true square wave in on exactly 50% of the time. The PWM signal also switches from fully on to fully off in one cycle, but the "% ON" value varies from 0 to 100 %, depending on what speed is being required. Inside the motor there is a small chip that uses this signal to modify the flow of current from the fixed 12 VDC power supplied on Pin #2 through the motor windings to alter the motor speed.

The Noctua unit comes with a few cables designed to plug into common connectors in a computer system, which you are not using, and supply power to the Noctua control module. At least one of them ends in a female standard fan connector, just as your fans do. With the info above on the pinout details, you can rig a custom connection using one of those cables to your 12 VDC battery. It needs only the connections from Pin #1 (well, hole #1) to battery (-), and hole #2 to battery (+). You need no connection to the other two pins.

As I said, the fans generate a speed signal, and ONE of your fans' signals will be relayed back to the Noctua unit. Although it does not display that speed in any way, apparently it does use that in limiting the minimum speed to 300 rpm IF you push the button.

 

[jdfnnl]

As I said, the fans generate a speed signal, and ONE of your fans' signals will be relayed back to the Noctua unit. Although it does not display that speed in any way, apparently it does use that in limiting the minimum speed to 300 rpm IF you push the button.

After reading this entire posting, I am not sure if you advise connecting 4-pin fans directly to a 12V motor via pin 1 and 2, though I think it is clear I cannot adjust the speed by reducing the voltage. For this I would need a 3-pin fan and could possibly employ a simple rotary switch, though the minimum speed would be higher than that of a 4-pin fan with a PWM controller. Does this sound correctly distilled?

Also, are there any components in 4-pin fans that would be damaged if subjected to 12V current?

 

[Darkbreeze]

After reading this entire posting, I am not sure if you advise connecting 4-pin fans directly to a 12V motor via pin 1 and 2, though I think it is clear I cannot adjust the speed by reducing the voltage. For this I would need a 3-pin fan and could possibly employ a simple rotary switch, though the minimum speed would be higher than that of a 4-pin fan with a PWM controller. Does this sound correctly distilled?

Also, are there any components in 4-pin fans that would be damaged if subjected to 12V current?

4 pin PWM fans ARE subjected to 12v voltage. That's the whole point. They are ONLY (If they are 12v PWM fans) used with 12v, and the 12v is cut off, then on, then off, MANY times per second/minute, but what the fan sees from the controller is ALWAYS 12v. It does not ever see "10v" or "7v", in order to reduce speed. When the speed needs to be reduced it simply reduces the duty cycle to a lower figure, say for example, a 40% on and 60% off duty cycle rather than a 100% duty cycle for full speed operation.

Much might depend on the specific PWM controller or motherboard as well, because my Noctua NF-A14 PWM fans will NOT run below about 33% duty cycle with one of my Z170 motherboards which is a Gigabyte Z170x-Gaming 5 without random fan locations invariably stalling out intermittently, but when those same fans are used with my Hero VIII motherboard, I can run them all the way down to about 23% duty cycle which makes for a somewhat significant different in RPM when it comes to trying to get them as quiet as possible for normal no load or low load operations.

As to the other part of your reply, yes, I believe you understood the gist of what was said exactly right. Not sure what you meant by "connecting 4 pin fans to a 12v MOTOR via pin 1 and 2", but again, 12v is fine for four pin PWM fans, but you need a PWM controller. You CANNOT run PWM fans on low voltages. You MUST use a controller that sends ONLY a 12v signal, and pulses that signal for speed controls.

If you MUST use variable voltage to control speed, then you want a 3 pin DC controlled fan but as Paperdoc stated, you might not be able to get them to as low of speed as with a PWM fan. I have some Aerocool Deep silence 140mm 3 pin DC controlled fans that run at pretty low RPM without stalling though. I think I had them running at around 550rpm running off my NZXT Grid+ v2 controller and they were for all intents and purposes mostly inaudible.

 

[madmatt30]

4 pin PWM fans ARE subjected to 12v voltage. That's the whole point. They are ONLY (If they are 12v PWM fans) used with 12v, and the 12v is cut off, then on, then off, MANY times per second/minute, but what the fan sees from the controller is ALWAYS 12v. It does not ever see "10v" or "7v", in order to reduce speed. When the speed needs to be reduced it simply reduces the duty cycle to a lower figure, say for example, a 40% on and 60% off duty cycle rather than a 100% duty cycle for full speed operation.

Much might depend on the specific PWM controller or motherboard as well, because my Noctua NF-A14 PWM fans will NOT run below about 33% duty cycle with one of my Z170 motherboards which is a Gigabyte Z170x-Gaming 5 without random fan locations invariably stalling out intermittently, but when those same fans are used with my Hero VIII motherboard, I can run them all the way down to about 23% duty cycle which makes for a somewhat significant different in RPM when it comes to trying to get them as quiet as possible for normal no load or low load operations.

As to the other part of your reply, yes, I believe you understood the gist of what was said exactly right. Not sure what you meant by "connecting 4 pin fans to a 12v MOTOR via pin 1 and 2", but again, 12v is fine for four pin PWM fans, but you need a PWM controller. You CANNOT run PWM fans on low voltages. You MUST use a controller that sends ONLY a 12v signal, and pulses that signal for speed controls.

If you MUST use variable voltage to control speed, then you want a 3 pin DC controlled fan but as Paperdoc stated, you might not be able to get them to as low of speed as with a PWM fan. I have some Aerocool Deep silence 140mm 3 pin DC controlled fans that run at pretty low RPM without stalling though. I think I had them running at around 550rpm running off my NZXT Grid+ v2 controller and they were for all intents and purposes mostly inaudible.

In my experience its only been the initial kicking up of the fan from standstill that is a real issue with voltages..
Most fans will run right down to 3v no issue (I have arctic cooling p12's that will do that) but won't actually start below 4v.

 

[Darkbreeze]

Mostly, that's probably true, to a point. I've seen DC fans stall out though at low voltages. Or not necessarily "stall", but jerk and buck, like they want to run but won't. I imagine that's pretty bad for the motor.

For PWM fans, it's definitely not. They WILL stall out randomly if the duty cycle falls too low. Even if they are already going. That's exactly why that Noctua controller HAS the feature to stop fans from running below 300rpm, so they don't stall. I'm not sure what that duty cycle would be, but I'm guessing it would have to be like 20% or less, for a 1500rpm fan.

 

[jdfnnl]

There is a PWM fan controlled with variable voltage in this video:

Are there any components in 4-pin fans that would be damaged if subjected to 12V current?

 

[madmatt30]

There is a PWM fan controlled with variable voltage in this video:

Are there any components in 4-pin fans that would be damaged if subjected to 12V current?

There's no physical reason they would be.

4 pin pwm and 3 pin voltage controlled fans are both 12v.

Straight 12v to either would just run them at their maximum stated speed.

I'm not sure there are any physical differences really.

What you will find though is that 3 pin fans tend to have a lower max rpm, I think this is set lower simply so that people who don't have voltage control don't have to run them full pelt all the time.

For example

Corsair sp120 3 pin - 1400rpm
Corsair sp120 4 pin pwm - 2350rpm

Arctic cooling p14 3 pin - 1000rpm
Arcric Cooling p14 4 pin pwm - 1400rpm

Minimum required voltage on all those fans is specced at 7.2v

As I stated before I know the 3 pin p14 can go down to 3v without issue as can the Fractal Design Dynamic x2 gp-14.

I have no idea about the rest, this is from actual personal experience of using them.

 

[Zerk2012]

Actually you could use any fan and a potentiometer to vary the voltage.

I had some old Enermax TB Vegas fans that had them included with the fan came with just a 2 pin molex adapter.

 

[Karadjgne]

It's all fan dependent. Some DC fans won't start below 60% (7v), most will go to 40% (5v) but from there down its hit or miss as to exact starting/running requirements. Much of that has to do with the efficiency of the motor, starting a fan at 5v takes considerably higher amperage than starting a fan at 12v.

Most PWM fans fail to start below 10v (ish) and very few will run below 9v.

Just need to make a decision, run at a variable voltage, needing 3pin DC fans, or run at a constant voltage that's pulse modulated and requires 4pin PWM fans. Either will have Amperage requirements, but pwm is limited to @ 10 fans by the pwm signal strength, DC fans are limited by the control amperage rating (1A for a mobo header, or 2-3 DC fans normally.)

There's a big difference between starting voltage and running voltage. Those Arctic fans require 7v to start, but might run fine at 4v. Amperage capped. Starting power is always far higher than running power.

 

[madmatt30]

It's all fan dependent. Some DC fans won't start below 60% (7v), most will go to 40% (5v) but from there down its hit or miss as to exact starting/running requirements. Much of that has to do with the efficiency of the motor, starting a fan at 5v takes considerably higher amperage than starting a fan at 12v.

Most PWM fans fail to start below 10v (ish) and very few will run below 9v.

Just need to make a decision, run at a variable voltage, needing 3pin DC fans, or run at a constant voltage that's pulse modulated and requires 4pin PWM fans. Either will have Amperage requirements, but pwm is limited to @ 10 fans by the pwm signal strength, DC fans are limited by the control amperage rating (1A for a mobo header, or 2-3 DC fans normally.)

It would be helpful to know exactly what this 'ventilation project' actually is in all honesty.

We could then probably recommend an actual fan model that would be suitable between us from personal experience.

 

[Karadjgne]

Lol, yep. Definitely curious about that.... 🤔

 

[Zerk2012]

That is true about fan dependent, had a fan controller with different fans connected and some would cut completely off when some were running slow.

 

[Darkbreeze]

That is true about fan dependent, had a fan controller with different fans connected and some would cut completely off when some were running slow.

I have seen this happen, consistently, even when several of the SAME fan are in use, at very low RPMs.

And for what it's worth, while this isn't the common wisdom or word on the subject, I think PWM fans can be damaged by consistently feeding them a too low voltage. Not the controller or IC itself, but the motor, which is technically not designed to be used with very low voltages. Possibly the hall effect sensor as well, but I'm not positive on that. Certainly they CAN run using lower voltage as well, but it is not wise, nor is it intended for them to be run on anything other than the voltage for which they were designed to be run on.

 

[Paperdoc]

The design of 4-pin fans includes a backwards compatibility feature in case they are connected to an older 3-pin fan header. Such a header does not provide any PWM signal - it does not even have a Pin #4. But it does it job solely by altering the DC voltage supplied on Pin #2. If a PWM fan is connected to such a header, it WILL run with speed controlled, because it gets NO PWM signal to modify the power supplied, and so its motor just runs according to the voltage available. HOWEVER, many people suggest this is NOT the ideal way to power such a fan. Moreover, if your plan is to supply reduced voltage to reduce speed and NOT to provide a PWM control signal, then it is surely wiser (and slightly cheaper) to use older 3-pin fans.

If you are set on using some means of voltage reduction for fan speed control, you will need to experiment to customize, AND you will not be able to change components easily thereafter. Why? You plan, for example, to simply use a muti-position switch to insert one of several resistors into the power supplied from the 12 VDC battery source. For any chosen speed, the value of that resistor will depend entirely on what is the CURRENT flowing through the total load, since the voltage drop across the resistor depends on the current. (And, of course, the current depends on the voltage drop across the motors, etc.) You plan to have two fans in parallel drawing current from this source. If ONE of the fans is removed or fails, then suddenly the current is altered and the voltage drop across the resistor is reduced, making the voltage supplied to the remaining functioning fan higher. Or, if you replace one fan with a different model with different electrical characteristics, that also changes the resistor value needed.

Some above have suggested a potentiometer, rather than switched fixed resistors. Technically, although it is the same component, that device will be used as a series variable resistor, and not as a potentiometer, but that's a minor point. Such a system would make it easier to adjust for changes in the fan load, because you would simply re-adjust the control knob to get what you think looks right.

You should be aware of what mobos do for the issue of starting and stalling a fan. This goes back to the fact that, to START a stopped fan, you need a certain minimum voltage or it just won't start. AFTER it is running, you can reduce the voltage a certain amount until it stalls. But once it has stalled, it will never re-start (and may overheat with a low voltage constantly supplied) unless you raise the voltage to the minimum starting value. So you can NOT set a low voltage (for minimum speed) with a resistor, then shut down, and then turn on the power again. It will NOT re-start! A mobo will always start its fans at full 12 VDC, wait a couple seconds, then reduce the voltage to what is needed for the desired speed. In normal operation it will never reduce the voltage below some minimum value (typically around 5 VDC) to avoid causing the fan to stall. It also monitors the fan speed signal from Pin #3 (which your simple proposed system cannot do) and, if that signal indicates no speed, will automatically raise the voltage to full 12 VDC for a short time to re-start the fan, then reduce it to the desired setting once again.

By the way, there is an aging and wear effect to consider. As a fan wears its bearings in normal use, the bearing friction slowly increases and the fan runs slightly slower. More important;ly, the minimum starting voltage must be increased to offset this effect.

You have not told us WHY you want to run slow fan speed in your project, and that might help to clarify advice. Are you concerned about air flow rate, or fan noise, or mechanical vibration, or what???

 

[jdfnnl]

This is for a situation of living in a vehicle -- having the option of minimizing noise would be a bonus.

A would potentiometer decrease the current draw on the battery, or would the variable resistor draw as much as the fan's motor does not?

So I am thinking of going with a couple Phenteks PH-F140 because they seem robust and quiet. Maybe I ought try out both the PWM and non-PWM version and see which seems to perform best ..

 

[Darkbreeze]

If you are looking for quiet as the biggest priority, my suggestion would be to look at Noctua and Thermalright, maybe even Be Quiet. Be Quiet tends to achieve some part of their low noise levels by limiting the maximum RPM several hundred RPM below what competing models are at, which has the additional effect of lowering maximum performance, but in this case that's clearly probably not an issue.

In general, Noctua and Thermalright at the the top of any shootout, comparison or shortlist when it comes to having the lowest decibel level at any preassigned RPM. Phanteks makes decent quality fans and very good cases, but they are not, in my opinion, on the same level as Noctua and Thermalright when it comes to fan development or performance.

 

[Paperdoc]

I fully agree with Darkbreeze above. Noctua fans have reputations as quieter for the same performance, and long-lived. IF you are getting the Noctua PWM fan speed controller box you cited, go ahead with one or two of their PWM fan types. But if you are going to do this with a variable resistor in series with the fans, I'd go with the 3-pin older models.

When you connect a variable resistor in series with a fan and turn it to reduce the fan speed by increasing the resistance and reducing the voltage to the fan motors, it also reduces the current draw. So in terms of Watts of power, a slower speed means using less power and having a smaller impact on battery life. Two typical 140mm 3-pin Noctua fans connected in parallel with a series variable resistor on the supply side would pull a max of 0.2 A at full speed, and less at reduce speeds. I'd suggest you look at their Redux line (grey colour) which are slightly less expensive than their newer regular line (2-tone browns). Do not go for their iPPC line - they are much more powerful fans that blow a lot of air, but also generate more noise.

 

[Karadjgne]

Phanteks fields some of the best value stuff out there. From cases to fans to accessories, anything they stick their name on is good quality for the price. Doesn't make them the best quality, just best value.

For fans, Noctua, Noiseblocker and Thermalright (NOT Thermaltake, don't confuse them, not in the same class or even close).

As far as battery power goes, moot point. A couple of decent fans won't draw enough to make a difference. You'll get much larger swings in draw just from the cpu and Windows 'idle' spikes/dips. You are looking at 2w-3w ish worth of draw at best with 2x fans at 50% speeds. The Noctua 3pin even come with low and ultra-low voltage adapters (9v and 7v I believe) so can be capped easily at 60-75% max rpm.

A manual, variable rate switch is a disaster in a pc. Loads constantly change, and you really have no idea of the actual load at any given time, so you'd end up trying to guess the needed rpm with every change. Play a little game, forget to change fan speeds up, everything overheats. Get done playing a little game, forget to turn speeds back down, drains the battery quicker than expected. Far better to allow automatic control by the motherboard within ranges you set.

 

[Paperdoc]

Now you've raised a question in my mind. I did not think OP was looking for a fan cooler system for inside a COMPUTER. I though the post replying to my why? query indicated he / she is looking for small air circulation fans inside a vehicle.

 

[Darkbreeze]

Methinks some clarification from the OP is definitely called for here. I understand if the circumstances might seem moderately uncomfortable, but we are not here to judge anybody, only to help find answers to the questions that are posed.

 

[Karadjgne]

Op said vehicle. I took it to be something like an RV with a desktop that's used for gaming, TV, media etc. Which would explain the battery requirements if the camper wasn't hooked to power at a campsite.

 

[Darkbreeze]

If this was for a desktop system in a vehicle, there wouldn't be any NEED for any of what the OP is looking to do because it could easily be done through the BIOS fan profile configuration. He wouldn't need an external controller or potentiometer, or any of that.