Powering a 3 V motor with 2xAA batteries for several hours

I have a 3 V toy motor that I want to run with some kind of stability using 2x AA batteries. In terms of load, imagine a little fan. I'm not sure how much torque that would add.

The motor is a brushed DC motor that needs about 400 mA to start. I measured the resistance across the leads, and I got about 0.8 ohms.

I don't know how to measure its inductance, but I bet it's significant, because with just the batteries + motor, there was visible sparking when I broke contact. So I added a ceramic capacitor in parallel to the motor to prevent this.

This is the current state of my circuit. I estimated the resistance of the batteries by measuring their current in a short circuit.

Suppose the batteries have 2000 mAh of capacity. And that I'm happy with the motor's behavior operating at 500 mA. So an approximation of its theoretical upper limit on runtime is 2000 mAh / 500 mA = 4h. But I have no idea what I can shoot for in real life.

But the AA batteries supply way too much current to be connected in this way for very long. At present, the batteries start out by supplying over 1A. After a few minutes, the batteries are HOT. I have decided to interpret this as a warning from the universe. If I'm not mistaken, it's because my circuit is operating with too little resistance (almost operating in a state of short-circuit), and the chemicals inside the battery can't do their thing at that rate without producing significant heat. With my multimeter, I kept an eye on the current, and it drops slowly over time, and since I have little fan blades connected to the shaft, the noise produced by the fan evolves in such a way that it's like, imagine you see a nice flower and it just starts drooping slowly. It is unpleasant to hear.

My question is:

If we assume that my diagram is correct enough, then by adding a resistor of 3.2 ohms in series with the motor, I should expect about 500 mA thru the motor (at least for a while). I'm sure it would run with more stability than now, but is this too naïve an approach for what I want?

I did a bit of research, and it sounds like this is problem of limiting the current. I wonder if a circuit like the one mentioned in the following question might be useful?

Help dimensioning resistors in a two NPN transistor current limiter circuit

I also looked up current limiting and according to Wikipedia (and on this site) there is a technique called pulsed width modulation which sounded like using a 555 timer (since I've used one for blinking LEDs before), picking the resistors to control the length of and between the blinks, and then using a capacitor to smooth it out.

• If you want to target long run time, you must use PWM instead of series resistance to reduce speed and current draw. You use the motor inductance itself to smooth the current out. You'll need a FET with very low Vgsth to turn on properly at 3 V (and lower as your cells discharge) Commented Sep 5, 2023 at 16:55
• I will look into this. Plus it seems like a technique I can lean into for other projects as well.
– rcot
Commented Sep 6, 2023 at 4:02
• Using a 555 timer for PWM is very unlikely to improve run time--the 555 consumes a ton of power on its own, even when it's not doing anything. There are lower-power PWM chips out there--consider an LTC6992, perhaps. Commented Sep 6, 2023 at 13:07

If the motors are specified for 3 volt operation, they should work with a 3 volt power source with no need for any current limiting.

You may need to use two D cells rather than two AA cells to get enough current capacity and to avoid battery heating.

• The 2000mAh capacity of a fresh Energizer AA alkaline battery cell is listed at 250mA and when its voltage has run down to 0.8V. A battery does not FORCE all its current, instead the load DRAWS as much current as it needs and as it can. You are overloading the little AA cells and need larger cells. Commented Sep 5, 2023 at 18:43
• I've got some other batteries I can use so I'll test it with those too. Ultimately, I don't want to use 2xD cells, but I will test it. If I can't use 2xAA batteries in some reasonable configuration (i'm sure someone can invent some wacky arrangement that would work), then my aim is to at least figure out what the challenges are and why it's not such a good idea. The overloading element is clearly a good reason why my current configuration is not a good idea.
– rcot
Commented Sep 6, 2023 at 15:55
• You've shown that two AAs won't work. Four or six AAs in series/parallel might work, but you do definitely need a battery pack capable of much more current than two AAs can supply. Commented Sep 6, 2023 at 17:00

Simply adding a series resistor to limit your motor current is a very inefficient method to use. While the motor will have it's current reduced the series resistor will be dissipating (wasting) a good amount of the battery power.

Your example of estimating the motor current is only valid at the very brief moment of power application or in a motor stall condition. When a motor is actually running there is also a condition known as Back-EMF. The generated Back-EMF of the motor will actually decrease the motor current significantly as the motor speeds up. (More on Back-EMF.)

The current limiting two-transistor circuit you referenced also has a drawback, it drops about 1.6V ahead of the motor. This is more than half your battery voltage. So with that arrangement guess where most of the power dissipation will be going to.

The PWM method using a 555 chip is a reasonably simple way to minimize the average motor current and provide basic speed control. However running at 3V or lower is still a challenge, it would be best to use the CMOS 555 version (such as LMC555 or TLC555). The LMC555 actually specifies operation down to 1.5V. As in the comments you would also need to use a driver MOSFET/transistor capable of switching efficiently at 3V or lower. The datasheet for the TI LMC555 contains application circuits including some with PWM outputs. In addition to the TI datasheet a google search brings up a huge number of similar application circuits that use a 555 for PWM motor control.

There are also dozens of small commercially made speed control modules that utilize 555 chips, some claim operation down to 3v or lower, however most seem to be using LM555 chips that are only specified down to 4.5V operation.

• Ah! I'm having an insight. The emf bit is very interesting bc now I know I can't think of the dc motor like an inductor. After the voltage across an inductor drops to 0 and it gets a steady current, I can use ohms law, as long as the voltage doesn't fluctuate. That's what I did and it's wrong. Bc in the motor, the coils are made to spin bc the current passing thru them interacts with the magnet. But now that they are spinning, the magnetic field from the magnets is changing (relative to the coils), inducing an e-field. So even tho it's a motor, it's still a generator.
– rcot
Commented Sep 6, 2023 at 15:46
• I'm going to test out the motor a bit more to map its specific behavior better and probably test it out with a 555 (maybe with an alternate power source) and i'll look into other chips in the meantime. Maybe my goal of running the whole thing on 2x AA is not reasonable. If it is not, I'd like to know why, ideally, thru experimentation.
– rcot
Commented Sep 6, 2023 at 15:51

But the AA batteries supply way too much current to be connected in this way for very long. At present, the batteries start out by supplying over 1A.

• If this happens without mechanical load on the motor, the motor is broken. The current the motor draws should be commensurate with the mechanical load.

• If this happens with mechanical load, the load is too high. 1A @ 3V is 3W - not out of line for a fan, depending on the fan size. You're mechanically overloading the motor if that's the case. The fan usually has to be chosen/designed for a particular motor, so that it runs at the highest efficiency point of the fan-motor system. If you mix-and-match fans and motors, most of them will perform poorly, as you've experienced.

I did a bit of research, and it sounds like this is problem of limiting the current.

If the motor is not broken, it is a problem of mechanically overloading the motor, given the available power supply. It's a transitive relation: if the batteries are overloaded, the motor is overloaded for the application it's in.