I've tried looking around for as much information as I could gather to build really small wearable electronics that could power themselves by generating energy with movement.

I know that the technology already exists in watches. There are watches that produce their own energy using a small generator that spins when the wearer moves its arm around. The energy is stored in a capacitor to make the watch work quasi-infinitely as long as the wearer moves enough throughout the day so that enough energy is generated. (some info here : http://en.wikipedia.org/wiki/Automatic_quartz )

I have trouble finding information about just how much current can this technique output for practical uses and if this technique can be adapted to work with other electronics.

In spirit of actually having real questions and not just vague stuff, here's a couple of questions I hope can get answered

  • How much energy can these devices produce / store?
  • Is there any other devices aside from watches that are powered using only kinetic energy?
  • Is there any schematics on how to create a kinetic generator for low-power devices?

Edit : Some people have said that it is hard to determine if it's possible since I didn't mention exactly what I would be doing. I don't have any project right now but I want to play around with wearable electronics that are motion-powered instead of battery-powered that you need to replace or recharge.

I'm thinking of using the lowest-power MCU I can find (something like the IT MSP430 microcontrollers) and reduce the current usage to the maximum by using standy mode as much as possible and doing very little active work, maybe just for logging multiple sensor data every couple seconds.

  • 6
    \$\begingroup\$ Kind of depends on how vigorous you want to be - see for example shake flashlights. \$\endgroup\$ Commented Aug 14, 2012 at 15:14
  • 1
    \$\begingroup\$ I had totally forgotten about shake flashlights! Thanks for reminding me. For clarifications I'm thinking more about something that works relatively well while being passive like a watch, you don't need to make a conscious effort to charge these watches, it's done with your normal body movements. \$\endgroup\$
    – user11519
    Commented Aug 14, 2012 at 15:18
  • 4
    \$\begingroup\$ Don't expect too much from this. For the watch it works well because that uses only microwatts. The flashlight has to be shaken vigorously for 15 seconds for just a couple of minutes of light. For low-power microcontrollers, in the micro-amperes, you'll be more successful. \$\endgroup\$
    – stevenvh
    Commented Aug 14, 2012 at 15:22
  • 1
    \$\begingroup\$ One technology to consider is piezo piezo.com/prodeh0nav.html \$\endgroup\$
    – kenny
    Commented Aug 14, 2012 at 15:24
  • 2
    \$\begingroup\$ These energy harvesting boots may be of interest. They generate a continuous 6 to 9 watts from walking, which is quite impressive. \$\endgroup\$
    – Oli Glaser
    Commented Aug 14, 2012 at 16:08

2 Answers 2


I saw this answer and glanced at my watch, it just so happens to function in the way you describe. There's a large half-disc pendulum inside which presumably is used to collect the kinetic energy. It can run for about let's say one day If I walk about 2-4 miles a day. If I move throughout, indefinitely, but then, watches require very little power to begin with.

Energy Storage How much energy you can store only depends on your capacitors and batteries, etc, it's not inherent to the the energy generation method you want to use.

Other Devices I don't see why you would look at it this way as mentioned before. Decide on what you want to use kinetic energy to power and then do a few basic calculations based solely on first principles, assuming 100% efficiency of kinetic energy -> chemical / electrical energy and you'll quickly see what's doable and what's not. As a general guide, microWatts, up to a maximum of say a few milliWatt is probably all you can do (this very much depends on the type of physical activity you're using, more strenuous activites could very much reach into Watts)

Schematics I can't give you one, but I can tell you the design is very simple. Take the pendulum in my watch mentioned earlier(here's a picture of one from your article):

enter image description here

While I can't specify exactly how thing's work internally, what I do know is this:

"Electromagnetic induction is the production of an electric current across a conductor moving through a magnetic field"

After reading the provided article, the pendulum is used to spin a pinion(visible in the picture) - just a small gear, which in turn drives an electric generator(which has been documented plenty). Anyway, you're likely to get an AC voltage from the electric generator. You can use perhaps a bridge rectifier, maybe a voltage regulator as well and store this in the energy storage device of your choice.

Finally Like I said before, decide on what you want to power first, that's the only way you're going to be able to do this properly. Make an upper maximum required power for the circuit you want to power and proceed from there. Feel free to ask for any more help but based on the information so far, this is all I can do.


~ People are making vibration coupling backpack type units that produce in the few Watts range.

~ A typical Chinese squeeze light when operated frenetically will give about 1 Watt. After 5 minutes of use at that level your hand drops off or feels like it has.

~ A properly build hand squeeze power unit could make more like 5 Watts at the same level of effort.

~ A properly built hand crank unit with reasonable throw - say handle at 150 mm radius will make 10 Watts at reasonable effort. and 20 Watts with rather busy effort. At the 10 Watt level you could do it for an hour +

~ A properly designed pedal unit (many aren't) can make 50 Watts at a level where an only somewhat fit person could do it for an hour, 100 Watts at a level where a moderately fit person wishes they were fitter and more if you are a sports enthusiast. People with sports gear usually make claims to higher power levels than I usually measure. I can do 500 Watts pedalling for about 10 seconds but then need to die for 10 minutes or so.

~ Shaker torches are probably somewhat below hand squeeze lights in power. A properly mounted hand crank should be able to be mounted to operate as you walk.

~Piezeo is OK but is usually VERY low power. Newer higher Wattage versions can be bought for money.

~If you wear a windturbine on your cap it can be made to spin as you walk and will make a few milliwatts. If you mount a 1 metre diameter WT on your cap they will certify you - but power out at a fast walk can be a Watt or few.

~Foot fall and leg bend offer much power.

~ Work = Force x distance. = F x d Power = work per time = F x d / t

~ You can work out heel strike force, travel etc for an in shoe alternator.

~ Anything using hot and cold will produce very very very close to zero power.
Carnot efficiency = (Thot-Tcold) / Thot
Real efficiency = Carnot by say 25% if lucky.