Fan Problems

[Mighty Burger]

I'm creating my first permanent project. It's going to be a fan that can be powered by batteries or a wall adapter. It also has a speed control, which is the problem. I'm using PWM, and I'm fairly sure it works, but I don't have a sillyscope oscilloscope to test it. However, the fan's reaction to it doesn't. I'm testing it with a fan from an old computer before I place my very fast and slightly expensive fan into the circuit. Normally the fan (now talking about the old fan) spins perfectly, with 12 volts, and it spins even down to 3 volts. I've been told controlling the voltage is a poor way to control the fan speed, so I've went with PWM. Here's the circuit:

(Note: in testing I eliminated the barrel jack and just used the batteries as a power source.) However, when I turn it on it makes a silly noise:
https://www.youtube.com/watch?v=UbJRvniuXgg&feature=youtu.be
The moments when it is silent is when no power is connected. It wiggles and makes the strange noise when I connect power to it. The noise also changes pitch when I change the PWM duty cycle.

I hope the solution to this issue is to simply replace one or two components. I really don't want to redo the whole circuit, but if I have to, I will. I hope reducing C1's capacitance would probably fix it?

One last thing to note. The fan in the video is the old fan, and it doesn't have a 4th wire. However, the new fan does have a 4th wire. This extra wire is there for the computer to control the fan speed, but no google search is giving me any answers to how to take advantage of that extra wire. The only information I've collected is that doing so would be very difficult and would require a lot more components. I'm just going to PWM the power.

[Mr.B]

PWM the power is not the best way to do it. (And sometimes not the way to do it at all.)
That 4th wire you are talking about is for sending a PWM signal to the fans own internal speed controller.

Perhaps have a read of this article: http://www.overclockers.com/pwm-fan-controller/

[Monkeh]

One last thing to note. The fan in the video is the old fan, and it doesn't have a 4th wire. However, the new fan does have a 4th wire. This extra wire is there for the computer to control the fan speed, but no google search is giving me any answers to how to take advantage of that extra wire. The only information I've collected is that doing so would be very difficult and would require a lot more components.

http://www.formfactors.org/developer%5Cspecs%5C4_Wire_PWM_Spec.pdf

Not difficult, not remotely. Ponder why they'd introduce a method of controlling fan speed which is more difficult and expensive..

I'm just going to PWM the power.

Don't.

[michaeliv]

What MOSFET are you using ?
Try spinning up the fan by hand and then connecting the power, does it work then ?

[Marco]

Not difficult, not remotely. Ponder why they'd introduce a method of controlling fan speed which is more difficult and expensive.

Cause they didn't want a high dI/dt signal across the cable presumably. The driver/tach circuitry could have used a rectifier and smoothing capacitor to get consistent power with direct PWM control, so that doesn't matter.