Ultrasonic Wake-up device (Results)

I’m getting to the point of diminishing returns on my ultrasonic wake-up device.  I have fiddled with transformers, different transducers, and filter components.  It works, but pretty much as I expected, the results are not great.  The maximum range is about 8′, and the transducers need to be perfectly aligned at that distance.  There is in bright sunlight.  Clearly this is not going to work with a device that might be up a sleeve.  
I knew I was optimistic in thinking that it might work.  I have to remind myself that it was a great learning experience, and I should recap what I achieved so I feel better.
  1. I found some circuits online and butchered them to meet my peculiar needs.  In particular I designed everything around the lowest operating current comparitor I could find (and understand)
  2. I actually built a transmitter and receiver which are clearly the most complex things I have built, and are all analog which is so much harder than digital stuff.
  3. I designed a PCB for the receiver, and only made one mistake – though it was kind of fatal – leaving off a critical wire.  I guess there is some way to check out designs to make sure they are all fully connected.
  4. I figured out how to solder, and more importantly de-solder stuff using David’s hot-air gun.  That’s a really useful thing to buy.
  5. I debugged the two devices using the scope.  I figured out how to generate an FFT so I could make guesses about the source of the unwanted noise.
  6. I figured out how to make an appropriate low-pass filter to get rid of all the MHz crap, the source of which eludes me.
  7. I wrote transmitter software for an nRF52 to send a 40kHz (sine-ish) signal to the transducer using a couple of transistors in push-pull mode.
  8. I wrote receiver software to listen out for a signal, and reset the comparitor if it heard a signal.

Range and reliability demo

The transmitter wakes up every 2 seconds and sends a sequence of 20  u/s “wakeup” pulses. 
The nRF52 receiver is normally in deep sleep consuming 400nA.  When a u/s pulse is detected my receiver raises a GPIO and the nRF52 resets(reboots).  All it does is flash an LED and go back to sleep..


The transformer is adjustable.
Useful explanation.

Leave a Reply