Ultrasonic Sound Gun (Parametric Speaker)

For this project I built a gun that shoots out a narrow beam of ultrasonic audio. The sound can only be heard by people inside the narrow beam, or through a nearby source when the audio gets demodulated.

I was inspired to build this project after watching CodeParades awesome video Turning Sound Into a Laser. I definitely recommend watching his video before you watch mine.

Ive seen parametric speakers like this for sale on some sites but I wanted to make on for my own. It works by taking an audio source, in this case a Bluetooth speaker circuit, which then gets modulated to 40khz through a 555 timer circuit. From the 555 timer the output gets amplified and then is then sent to an array of ultrasonic transducers.

40khz is outside of the audible human range which means we cant hear it, however once the sound waves strike an object, the 40khz sound gets demodulated and you can hear the sound that's playing over the bluetooth speaker module. Because you cant hear the ultrasonic sound, It makes it seem like the sound is coming from the object. If you stand in the narrow beam of sound waves the sound can be heard, but standing just outside the beam the audio is silent.

Supplies

Here's what you need to build the Ultrasonic Sound Gun

1. Bluetooth Module Link

2. 555 Timer Link to 2,3,4,7 and 8

3. 100k Resistor, 2k Resistor, 1k resistor Link to 2,3,4,7 and 8

4. .1uF Capacitor, and small capacitors to tune apx 100nF Link to 2,3,4,7 and 8

5. An Amplifier ( I used an L298n H Bridge module, any Mosfet will do) Link

6. Ultrasonic Transducers (the more, the better) Link

7. Various Switches and LED's as needed Link to 2,3,4,7 and 8

8. Perfboard Link to 2,3,4,7 and 8

9. A 3d printer is optional

Step 1: Assemble Circuit

The 555 circuit is a tricky one to get right. The slightest variance in capacitance will cause the entire project to modulate improperly and the sound will be inaudible. I recommend using a piezo buzzer instead of a transducer at first as its easier to hear if its working properly. The circuit says 160pF but this could be off slightly, just add or remove very small capacitors until the circuit is modulating at 40khz.

Once the circuit was functioning properly I added all the components to a perfboard and soldered them up. FYI the perfboard circuits capacitance was different then the one I used in the breadboard so be aware you may have to do some more tuning.

To assemble the transducer array I just pushed the leads through the holes in the perfboard with gentle pressure (I broke a couple trying too hard). I put all my transducers in order and marked the first one with tape to make sure they get wired in parallel.

Step 2: Assemble Components

Next I designed a 3d printed case in Fusion 360. This definitely wasn't my best work and I cut a lot of corners, but nevertheless I uploaded the STL's 1,2 and 3 above. Because I cut corners I had to use a soldering iron to poke holes for the various switches and LED's. Everything fit quite nicely though, so I still like it.

Due to the frequency of the circuit, you might have problems with induction if the components are too close, It never came up with the one I built but its definitely a possibility.

If you don't have a 3d Printer, that's fine you can just get creative and put the components into any other case

Step 3: Done

Now you're done!

If you followed my circuit, the left switch boots everything up and the LED lets you know the bluetooth module is ready to connect. Next you can connect your phone and start playing some sound. I like to either use the microphone or just play a song. Pressing the trigger will cause the ultrasonic audio to play when you have your finger on the trigger and flipping the toggle switch will cause the music to play constantly.

I really liked building this project even though I had many problems building it. Its definitely not a project for beginners as there's a lot of troubleshooting to be done.

Thanks for reading this Instructable, hope you enjoyed it.

Comments

[unturnedguyalex]

How much range does this have?

[TeDv8]

depends who/what is your receiver, what detail do you want it to be, what volume, etc. it is possible to get prety decent range similar to a normal small speaker, however you have to use higher voltages(up to 110V at the transducer(output) side to get the most power, the parts stated here can't handle that, but 4 powermosfets rated for higher voltage can, use a voltage pump or a step up transformer or such to increase the voltage.
at around 5V it will work but will be soft, at around 18V you will have prety loud sound having a prety far range. closer to the 110V(for example 100V) would give you something loud enough to make people have trouble hearing anything else.

[rafaelalfrota]

Hello, I am curious about how the appropriate spacing/design of the transducer array system is determined? How do you determine where to place each transducer for optimal beam focusing?

[TeDv8]

it needs to be around 0,4288 centimeters for 40khz transducers, since the wavelength is 0,8575cm(8,6mm)(european units)
you need to use half the wavelength to get a good beam, smaller distances also work especially when looking to factors of the wavelength, but half the wavelenght tends to be the most effective compared with how doable it is. if you can not get that distance(above all the transducers themselves are 1,6cm which is almost 2 times the wavelength.
so you can use 2 times the wavelength in this case which would be 1,71cm or 17,1mm. but know that sizes over half the wavelength, will make it less directional, however due to hardware limitations you can't really do so without smaller transducers.
so in this project 17,1mm difference would be close to optimal.
however the transducers aren't perfect and also aren't spot transducers but area transducers where the horn/membrame area is larger than the wavelength, due to this you can see it as many different spots and so in reality no spot will be really good and no spot will be really bad. on average it will be better to stay close to the 17,1mm(1mm between the transmitters). try to keep the error within around 1mm for optimal results so either place them so close that they all touch eachother, at most 2mm from eachother, or in between, 1mm difference would likely be the best you could do by hand.

TLDR : keep 1mm+-1mm(0 to 2mm) between all transducers. try to keep the spacing between the transucers around equal 1,1mm would be the optimal spacing for 40khz transducers which can be reached with these modules, the actual optimal spacing would be 4,3mm, but the transducers are 16mm big so they can't reach half the wavelength of distance. due to the transducers being imperfect and being area transmitters instead of spot transmitters you can't really optimize it to perfection, so spacing doesn't come to accurate, the stated numbers should give you the best results

[Constructed]

The beam could definitely be tuned based on the wavelength and location of each transducer, but for this project I wanted to keep it simple and accessible to just about anyone. The transducers are just put neatly into the perfboard and soldered with the correct polarity to make them in phase with each other.

[Coolerman]

Great looking project,,! But how far?? what is the range??

[en2oh]

Great project. Thanks for sharing it.
I thought the underlaying principle at work here was the interaction between a modulated and an unmodulated signal.
not really sure how modulating a single ultrasonic signal will lead to clean audio.
(Kind of like the experiment of sending a 1,000hz tone to one ear and a 1500hz tone to the other. You end up ‘hearing’ a 500hz tone, apparently coming from the middle of your head. )

[GokudoG]

Thanks for sharing ,very cool . will try it after the pandemic .

[KushagraK7]

That's a cool project. Kep it up!