What would be a feasible approach/circuit-design to make a self-powered wireless pushbutton, assuming it is even realistic?

This is what I mean by each of the three terms:

  • Self-powered: Power derived only from the mechanical action of pressing the pushbutton

  • Wireless: Pressing the pushbutton makes an RF transmission (let us consider a case with peak current draw of 40 mA during the transmission)

  • Pushbutton: Any type of pushbutton I can get at a hobbyist store or even build myself, but not a pushbutton where I have to turn a crank to activate the switch ;)

I would like to set up a little "network" project in my home (indoors), with these pushbuttons located at various points, but would like to do away with any battery power source, hence I'm experimenting with the self-powered idea. After all, the pressing of the pushbutton does bring in mechanical energy that could be used, and moreover, the resulting RF transmission event will be the only time the circuit will be alive or need to draw current (of around 40 mA).

My broken thoughts so far:

  • I am considering using a capacitor/supercap that charges up during the mechanical event.
  • Perhaps I could use some method of harvesting (piezoelectric, gear-based, etc.) of the mechanical energy from the push.
  • I've noticed that there is this interesting chip that might be of use here: the LTC3588
  • 1
    \$\begingroup\$ Someone already did it... ecogeek.org/component/content/article/2170 \$\endgroup\$ – Axeman Aug 24 '12 at 19:12
  • \$\begingroup\$ I'm not looking to be the first or anything; this is a home experiment for fun. Or, if you mentioned that as an actual option for my experiment, I can't (don't want to) afford it -- their product costs 100+ USD! \$\endgroup\$ – boardbite Aug 24 '12 at 19:21
  • \$\begingroup\$ I wrote it to answer to your "assuming it is even realistic" ... Yes, it can be done. Now I too would like to read some answer about how can it be done "at home" :-) \$\endgroup\$ – Axeman Aug 24 '12 at 19:26

For anyone who is interested in a fairly straightfoward piezoelectric approach, I discovered one successful report (using a piezoelectric pushbutton from a lighter) in the following paper from 2001 by two researchers at the MIT Media Lab (incidentally the paper is titled very similar to my Question!):

The following extract from the paper summarizes their method well:

A 4.4 μF tank capacitor integrates the charge transferred from a button strike. This, in-turn, powers a MAX666 low-dropout linear regulator, which provides a stable (although very inefficient) +3 volts supply until the tank capacitor's charge is drained. When the MAX666 is activated, the HT12E digital ID encoder is enabled, producing a repeating 12-bit serial ID broadcast via the On- Off-Keyed (OOK) transmitter module.

A drawing of their circuit: enter image description here

Images of their board/components: enter image description here

Here are further specific details I picked up from their implementation:

  • For the piezoelectric/button section, they took the core of a Scripto “Aim 'N Flame” lighter and modified the spring action to make the strike softer.

  • The piezo element generates peaks around a few thousand volts -- this is passed through a transformer, with a 90:1 turns ratio, outputting 30V at the tank capacitor.

  • The conversion efficiency (mechanical to electrical) for the piezo and transformer together is 7%.

  • At the point after the linear regulator, they measured 0.5 mJ to have been delivered (at 3V).

  • For the wireless transmission, they used a Holtek HT-12E encoder which generates 8 bits of ID and 4 bits of data, which are in turn transmitted by an RFM HX1003 (418MHz, 7.5 mW consumption, 50 ft range).

| improve this answer | |

I think you could probably do it in a semi-reasonable fashion if you did something like take a lightswitch and connect it to a gear. Have that gear drive some kind of generator, and harvest the electricity (you would probably need some gearing and a spring). You could also do this with a custom-made push-button: push-button gear drive

Now, this obviously requires you to build some sort of button yourself in order to make the contraption go, and you'll have to follow it up with a very-low loss rectification scheme, or have a reaaaaaaaalllllly long stroke on the button presses. You can do it, but I don't know that it's practical. The RF transmission part seems less the issue and more the mechanism for gaining enough charge. You can run current-day electronics on very little power, but how to generate any power at all seems like the issue.

Woo doodling!

| improve this answer | |
  • \$\begingroup\$ How efficient or lossy would the conversion to electrical energy with this gear-based method be, versus some method using a piezoelectric? (Also, cool doodle - did you draw it on a tablet? Reminds me of certain textbooks from the 1980s!) \$\endgroup\$ – boardbite Aug 25 '12 at 4:02
  • \$\begingroup\$ The piezoelectric has the problem that you may generate the amount of electricity you need, but in the timeframe you're looking at (a button press) to get it you'll take an extremely high voltage (say 100+ volts) at some really small current and have to convert it over without much loss because the pulse will be very short. The nice thing about the gear is that you can control the output voltage and the stroke of the button press can extend the duration. (Thanks! The doodle was just done in GNU paint on my laptop) \$\endgroup\$ – Kit Scuzz Aug 25 '12 at 8:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.