-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
FEATURED Building PWM Controller for 4 wires PWM fan
- Thread starter Brutal-Force
- Start date
- Joined
- May 10, 2009
Work email might work, they'd certainly have information about you on file! Could email mbently about it.
On an entirely different and somewhat more on-topic note, I'm contemplating designs for a non-PWM fan controller via the same chip I used for the PWM version.
On an entirely different and somewhat more on-topic note, I'm contemplating designs for a non-PWM fan controller via the same chip I used for the PWM version.
So I just finished reading this thread, I think it's a great read. As I went through, I had a similar thought as Bobnova did
There's already a ton of good stuff going into the PWM control, so what I was thinking was why change it? My thought is to tack on a little bit at the end of the circuit so that you can hook up either a 4 pin fan as is or a 3 pin fan with the extra bit. The circuit I found looks like this (This might have been mentioned elsewhere in this thread):
Lifted from this thread
So essentially take the PWM out from Bing's PWM circuit and feed it into this circuit. Now your 3 pin fan is as good as a 4 pin fan. (At least in theory)
I don't remember all of my circuit theory, but essentially uses the MOSFET as a solid state switch to power on and off the 12V rail to the fan and have it spin as per the duty cycle.
There's already a ton of good stuff going into the PWM control, so what I was thinking was why change it? My thought is to tack on a little bit at the end of the circuit so that you can hook up either a 4 pin fan as is or a 3 pin fan with the extra bit. The circuit I found looks like this (This might have been mentioned elsewhere in this thread):
Lifted from this thread
So essentially take the PWM out from Bing's PWM circuit and feed it into this circuit. Now your 3 pin fan is as good as a 4 pin fan. (At least in theory)
I don't remember all of my circuit theory, but essentially uses the MOSFET as a solid state switch to power on and off the 12V rail to the fan and have it spin as per the duty cycle.
Last edited:
- Joined
- Aug 1, 2003
- Location
- E600000 - E60000FF
So I just finished reading this thread, I think it's a great read. As I went through, I had a similar thought as Bobnova did
There's already a ton of good stuff going into the PWM control, so what I was thinking was why change it? My thought is to tack on a little bit at the end of the circuit so that you can hook up either a 4 pin fan as is or a 3 pin fan with the extra bit. The circuit I found looks like this (This might have been mentioned elsewhere in this thread):
Lifted from this thread
So essentially take the PWM out from Bing's PWM circuit and feed it into this circuit. Now your 3 pin fan is as good as a 4 pin fan. (At least in theory)
I don't remember all of my circuit theory, but essentially uses the MOSFET as a solid state switch to power on and off the 12V rail to the fan and have it spin as per the duty cycle.
Interesting concept..
Aren't most "high end" fans .. PWM..
So I had a thought for a way to get around this as well, although it's not nearly as clean. One could take the output of P1 (the In+ to the OP AMP) and split it so it also goes into a second OP AMP configured to amplify the signal by a factor of 12/5. This will convert it from a 0-5 volt signal to a 0-12 volt signal. This can then be fed into Bing's 3 pin fan controller instead of the R1, R2, and Pot. Although I'm not 100% sure this would give a full 0-12 range based on the voltage split.
Another option would be to take the PWN output, and throw it through a low pass filter to get an average DC voltage between 0 and 5 volts, then feed that into the OP AMP to convert from 0-5 to 0-12 and feed into the adjustible voltage regulator in Bing's 3 pin fan controller.
Not sure if anyone was able to follow my thoughts, I suppose a circuit diagram would be better. Are there any good free ones? (Sadly I don't have my college lab software anymore) I'll try to put a circuit diagram together to explain what I mean.
TK
- Joined
- May 10, 2009
An Arduino forum character by the name of Grumpy Mike (it's accurate) came up with this for PWMing non-PWM fans:
View attachment 98030
Full(er) thread here. It's only a 2a mosfet though.
View attachment 98030
Full(er) thread here. It's only a 2a mosfet though.
The article Dave quoted from Nidec is correct and makes sense.
The method of PWM-ing the power line for non pwm fan is not recommended, since that is a really crude method to control fan speed by turning "the whole fan's power" on and off pwm style at high speed.
Few reasons I could think of :
Here, repost a pic from Dave just to make it easier to imagine the whole mechanism inside a fan, actually for me its quite dizzying to think about it inside my brain, especially seeing this picture of san ace from Dave, with the stator that has so many poles, and think of how complex and accurate the firing mechanism is needed. Thanks Dave, I really love this particular shot, looks so cool.
View attachment 130458
Btw, that circuit from techpowerup is not optimal, the pwm control signal from the mobo is not beefy enough to drive or turn the mosfet on and off optimally. Usually it needs a few extra components called mosfet driver to drive the mosfet hard enough.
Again, this is not recommended method, especially if you're using good and expensive fan, undervolting a 3 or 2 wires fan is still way to go imo if you have to slow it down.
The method of PWM-ing the power line for non pwm fan is not recommended, since that is a really crude method to control fan speed by turning "the whole fan's power" on and off pwm style at high speed.
Few reasons I could think of :
- The fan speed rpm/tacho line will be useless since "normally" when then fan is powered "constantly", it will generate 2 pulses (on and off) per rotation as the standard protocol. Since the fan's power was turned on and off using this method, then the tacho signal will turned into meaningless noise.
- Even the pwm signal frequency is above human hearing range, sometimes it will make the fan generates weird noise when it was slowed down
, which defies the main purpose to make it quite from the 1st place. One of the culprit is there is a chance that the impeller rotation it self and the way that the stator was turned on and off unexpectedly will create some harmonic interference, this is somehow related to the next explanation.
- I believe we're all aware that the fan's impeller has the embedded strong permanent magnet, and the way it rotates is because of the magnetic force created by those stator poles. This firing mechanism needs a delicate and precise timing in energizing up those coils to magnetize the right poles that will create an optimal torque and at the right timing to spin and maintain the impeller rotation.
So when you turned the whole fan's power on and off unexpectedly using that method, there is a chance that in a brief moment that the stator will not firing up correctly, making the stator poles push/pull the wrong magnetic poles at the rotating impeller. This might happened even there is a hall effect sensor which it's job to sense the impeller's magnetic position, cause this sensor it self is wired to the circuit that control the firing mechanism, but alas, the control circuit is useless if they're turned on and off.
To simplify, just imagine when the fan is on and the control circuit is about to fire a stator pole since it senses the correct position of the magnet from the rotating impeller is about to arrive, then suddenly it was turned off momentarily, and then turned on again, this will confuse the control circuit and probably will be firing up the wrong poles. To make the scene more complicated, any electronic circuit, like the control circuit, needs time to be fully awake and operational after it was powered on.
And the worst part is, as the Nidec statement, this condition will create a hazardous electronic spike because the wrong combination of the rotating magnetic and the stator poles will somehow turned into an electric generator, instead of consuming/using the electrical power to push or pull the impeller's magnet in maintaining the rotation, and the unwanted generated electricity or current spike might toast the electronic part.
Here, repost a pic from Dave just to make it easier to imagine the whole mechanism inside a fan, actually for me its quite dizzying to think about it inside my brain, especially seeing this picture of san ace from Dave, with the stator that has so many poles, and think of how complex and accurate the firing mechanism is needed. Thanks Dave, I really love this particular shot, looks so cool.
View attachment 130458
Btw, that circuit from techpowerup is not optimal, the pwm control signal from the mobo is not beefy enough to drive or turn the mosfet on and off optimally. Usually it needs a few extra components called mosfet driver to drive the mosfet hard enough.
Again, this is not recommended method, especially if you're using good and expensive fan, undervolting a 3 or 2 wires fan is still way to go imo if you have to slow it down.
Last edited:
So here is version 1 of the PWM to analog voltage circuit. I'm not 100% sure it will work, I think I also may need to decrease the R6 value as re-reading the spreadsheet it looks like Vout is 1.25 volts higher than Vadj.
Let me know if you think this is crazy, or crazy enough to work.
R1-R3 and C1-C3 is the low pass filter, it's a 3rd order one to make sure that the frequency is good and attenuated.
Next is the op amp to scale the voltage from 0-5 to 0-12
Finally it feeds into Bing's fan control circuit to supply enough current to power a fan
Let me know if you think this is crazy, or crazy enough to work.
R1-R3 and C1-C3 is the low pass filter, it's a 3rd order one to make sure that the frequency is good and attenuated.
Next is the op amp to scale the voltage from 0-5 to 0-12
Finally it feeds into Bing's fan control circuit to supply enough current to power a fan
T=K, I don't think it will work as expected, honestly I'm not quite follow how it will work ?
Also why are you using a comparator LM311 instead of common op-amp ? And the non-inverting input is not connected, even its an op-amp will not work that way.
Fyi, the LM317 regulator will not provide full 12 volt at max speed if needed, that regulator will drop quite some voltage, worry the max volt for the fan probably about 9-10 or even 8 volt.
Also why are you using a comparator LM311 instead of common op-amp ? And the non-inverting input is not connected, even its an op-amp will not work that way.
Fyi, the LM317 regulator will not provide full 12 volt at max speed if needed, that regulator will drop quite some voltage, worry the max volt for the fan probably about 9-10 or even 8 volt.
You are right, I was working on that circuit while distracted a bit I'm afraid, made some stupid mistakes. I meant to connect the filter into the + instead of into the -, and looks like I also forgot a second resistor to turn it into a proper non-inverting amp (I meant to use an Op Amp picture instead of a comparator as you guessed).
Anyway, while re-designing the circuit, I had (what I hope is) a much better idea: I don't need to connect the OP AMP output voltage into an adjustable regulator, I can just use a high power OP AMP instead. The one I'm using is 3A, although there are even higher power ones for more money if you need the extra juice. The one I used in this circuit is linked. It is $1.71 from digikey. (A 5amp alternative, although some extra wiring is required).
So without further ado, hopefully this one makes much more sense. I send the output of the low pass filter into the + terminal of the OP AMP, and then the - terminal is hooked up as a non-inverting OP AMP with a gain of ~2.4, which is also 12/5.
Last edited:
T=K, without looking through the circuit design, only reading the datasheets from these high power op-amps, just fyi, the L 165 for 3 Amp or the stronger one OPA48 that is capable of 5 Amp, their capability will not reach or even close to 12 Volt if needed, this will be a problem if you need the fan to spin at max speed close to 12 volt.
Even using merely 0.5 Amp fan, I'm afraid the max voltage will be about 10.x volt or even worst around 9.x volt at larger current, are you sure this is what you want ?
Don't mean to rain on your parade, just a suggestion, I think it is better start a new thread specially to discuss this interesting topic , it will have better and wider exposure to other fellows than discussing it in here, which is mainly in controlling 4 wires pwm fan only.
Even using merely 0.5 Amp fan, I'm afraid the max voltage will be about 10.x volt or even worst around 9.x volt at larger current, are you sure this is what you want ?
Don't mean to rain on your parade, just a suggestion, I think it is better start a new thread specially to discuss this interesting topic , it will have better and wider exposure to other fellows than discussing it in here, which is mainly in controlling 4 wires pwm fan only.
Last edited:
