Will this work?

[Mighty Burger]

I'm a starter. I'll try to keep it short.


It's my first circuit I plan to actually use for a while. It's going to be a little desktop fan that can be powered by either a wall adapter or batteries. The fan is 12V 550mA (Noctua's NF-A14 3000RPM specifically. Yes, I know it's crazy overkill, and yes, I know it's suppose to be used in a PC).

When the wall adapter is plugged in, the P-channel MOSFET's gate to source voltage will be 0, so it will cut off the batteries and the fan will only be powered by the wall adapter. When the wall adapter is unplugged, the gate is at 0 volts, so the gate to source is -12V, and so the P-channel MOSFET will turn on and let the batteries power it.

R2 will be a potentiometer located on the front of the device. Unfortunately, the fan speed will be non-linear, but that's not too much of an issue.

So, will this work? If not, what's causing the issue? (I'd really like to avoid using a switched barrel jack, but I can use one if it's 100% completely absolutely totally necessary.)

Thanks in advance! 

[rx8pilot]

Your wall adapter looks like it will backfeed the converter.

Sent from my horrible mobile....

[rx8pilot]

I take it back, the body diode will block that. Your converter wouldn't be switched off since it can conduct through the same body diode.


Sent from my horrible mobile....

[michaeliv]

Seems like it would work, though having a diode in series when the wall adapter is powering the fan is not ideal as you will drop voltage over it.
Also, the buck converter will always be working, draining the battery, even when the adapter is powering the system.
Also, the adapter will be feeding into the buck converter(through D2 + P-CH body diode), which might not be good as well.

[Zbig]

PWMing the power it not the proper way to control "PC-type" brushless DC fan's RPM. You're likely to experience one or more of those:

• Narrow RPM range. Including not starting at all unless you dial the duty cycle up to almost 100% and then keeping more or less constant RPM up to 100%
• Audible noises
• Reduced fan's lifespan

You see, the BLDC fan itself contains a driving/timing circuitry that won't like getting fed PWM train as its power source.
Either get a so-called PWM fan that takes DC voltage for power and PWM as the control signal or redesign your circuit so the fan is getting proper variable DC voltage.

EDIT:
Or use one of the many dedicated fan controller ICs available.

[bobcat]

Ignoring the loss from the buck converter running all the time, D2 should be in series with the drain on Q2, with D2 anode connected to Q2 drain.

[Cerebus]

PWMing the power it not the proper way to control "PC-type" brushless DC fan's RPM. You're likely to experience one or more of those:

• Narrow RPM range. Including not starting at all unless you dial the duty cycle up to almost 100% and then keeping more or less constant RPM up to 100%
• Audible noises
• Reduced fan's lifespan

You see, the BLDC fan itself contains a driving/timing circuitry that won't like getting fed PWM train as its power source.
Either get a so-called PWM fan that takes DC voltage for power and PWM as the control signal or redesign your circuit so the fan is getting proper variable DC voltage.

EDIT:
Or use one of the many dedicated fan controller ICs available.

A quick search shows that the fan the OP is planning to use is the type commonly referred to as a "4 wire PWM fan". So, it has:
Pin 1 - Ground
Pin 2 - 12V Supply
Pin 3 - Sense - Tachometer output, open-drain or open collector, two pulses per revolution.
Pin 4 - Control - a 25 kHz PWM speed control signal, 5V logic levels

Thus a radical redesign of the OPs circuit is on the cards. If it was me, I wouldn't pass up the availability of the sense output to make a closed loop control circuit.

The 'class' specification for these fans can be found at http://formfactors.org/developer%5Cspecs%5C4_Wire_PWM_Spec.pdf

[Zbig]

A quick search shows that the fan the OP is planning to use is the type commonly referred to as a "4 wire PWM fan". So, it has:
Pin 1 - Ground
Pin 2 - 12V Supply
Pin 3 - Sense - Tachometer output, open-drain or open collector, two pulses per revolution.
Pin 4 - Control - a 25 kHz PWM speed control signal, 5V logic levels

Thus a radical redesign of the OPs circuit is on the cards. If it was me, I wouldn't pass up the availability of the sense output to make a closed loop control circuit.

The 'class' specification for these fans can be found at http://formfactors.org/developer%5Cspecs%5C4_Wire_PWM_Spec.pdf

Pretty much all but the most bare-bones ones have a tachometer output. PWM control - only some of them and it's usually mentioned explicitly as an extra feature. My very quick search for "Noctua NF-A14" seemed to imply there are actually two versions and as the OP haven't mentioned that, I assumed he's got the basic one. But I could very well be wrong, i.e. it could be that each NF-A14 has PWM control.

[Delta]

PWM the control line directly from the 555.

If you insist on not using a switch barrel jack, use a relay to disconnect the battery pack, its coil powered from the wall wart...