Slowing down a DC motor

[Mutant Funk Drink]

I have this project that I'm doing where I have a solar panel charge some AA NiMH batteries, and those batteries power a small 0.5 volt DC motor.

So far, I've found that 2 batteries can power the motor for about 8 hours. That's great, but I would like it to last even longer than that.

What I would like to know is how to slow down the motor so I could get it to run off the batteries for as long as possible. I know that sounds stupid, but the motor runs at around 1000 RPM, and in reality, I only need it to run at about 120 RPM. It really doesn't need to be so fast, and if the motor was doing fewer RPMs, that would mean less battery power being used at once, right?

I'm looking for a simple way to do this. Apparently there's some way to do it that involves an IC, but I'm looking for something that's easily replicable that uses simple components.

[Rourke]

Voltage divider.
http://en.wikipedia.org/wiki/Voltage_divider

[Pee Vee Proots, M.D.]

I was going to suggest capacitors or resistors, but it's been a while since I made a circuit.

[vovka351]

pulse width modulation, either use a fucking IC (555) or make a two transistor oscillator.

Nothing else will effectively reduce the current draw of the motor... I'll assume you've got the intellect to have wired the batteries in parallel already, yes?

[Mutant Funk Drink]

Yes, the batteries are going to be wired in parallel(the batteries have to be charged in parallel anyway).

What advantage does PWM have over using resistors? I'm not that experienced with electronics, so I wanted to go for something simple, but using a two transistor oscillator sounds like an attractive idea.

[Spatula Tzar]

Resistors burn off the extra power as heat. PWM is far more efficient, and gives greater torque.

[Mutant Funk Drink]

That's good.

So can I assume that using PWM will make it so that the circuit will take longer to drain the batteries?

Also, what about acoustic noise? I imagine sending a frequency of square/sawtooth waves to the motor would create some irritating noise.

EDIT: Looking at PWM circuits, it seems like they are relatively complicated. The circuits aren't really complicated at all, but I would think that creating an oscillation wouldn't need complex circuitry or ICs. After all, a joule thief supposedly oscillates, but it consists of nothing more than an NPN transistor, resistor, and an inductor. Is there a reason why a PWM circuit couldn't be that simple?

EDIT 2: What about a circuit like this?

http://wild-bohemian.com/electronics/flasher.html

Would I be able to modify this so that it drives my DC motor? Seems like all I'd have to do is just replace the LEDs with the DC motor on one side, and change the resistance to adjust the flash rate. But again, the only other thing I'm worried about is noise that might be generated.

[Mutant Funk Drink]

Bump. Anyone else have any input?

I tested out that flasher circuit in the falstad circuit simulator, and it seems to work. I reduced it to a single LED, and changing the two center resistors to 1.4k makes the circuit flash at a frequency of around 60hz. I don't know what effect of speed that will have on a DC motor, but if I made this circuit, I'd probably use a pot to control the speed.

I also noticed that the circuit creates sawtooth-like waves, rather than square waves. Does that matter in controlling a motor?

[Spatula Tzar]

The circuit has a 50% duty cycle if both the resistors are equal, so the motor will receive about half power. This doesn't quite equate to half speed, but it will be significantly reduced. The motor will be fine with a sawtooth.

[Mutant Funk Drink]

Thanks, Tzar!

I have built the circuit, and it works. But my problem is that now the motor has trouble starting up. If I'm using 3 AA batteries, this isn't an issue, but with the 2 AAs, I have to turn it by hand just a little to get it started, and it now has virtually no torque. I'll probably have to decrease the resistance and settle for a bit more speed.

[Spatula Tzar]

Did you keep the 470 ohm resistor in series with the motor? If so, get rid of it. This is only to limit the current of the former LED. You could also replace the bipolar junction transistor (BJT) with a field effect transistor (FET) for better performance.

[crazzyass]

Originally Posted by vovka351

pulse width modulation, either use a fucking IC (555) or make a two transistor oscillator.

Nothing else will effectively reduce the current draw of the motor... I'll assume you've got the intellect to have wired the batteries in parallel already, yes?

Oh shit, we have Tony Stark up in here. Not everyone is an electrical engineer, stop being a dick. Some people have less experience than others, and you gain experience through either mistakes or by asking someone. OP here is taking the wise choice and asking for help. So get the fuck out.

OP: Spatula's advice seems spot on. What's the project for, if you don't mind me asking?

[Mutant Funk Drink]

Originally Posted by Spatula Tzar

Did you keep the 470 ohm resistor in series with the motor? If so, get rid of it. This is only to limit the current of the former LED. You could also replace the bipolar junction transistor (BJT) with a field effect transistor (FET) for better performance.

I did not get rid of the resistor entirely, but I reduced both of those 470 ohm resistors to 100. In the circuit simulator, removing the resistor you are talking about seems to work just fine, but when I remove the other resistor, the circuit stops working properly. I guess I'll try moving the one for the motor tomorrow, since it seems like there's no reason to have it.

Originally Posted by crazzyass

OP: Spatula's advice seems spot on. What's the project for, if you don't mind me asking?

It's for a small hydroponics project(for lifting water).