DIY sonar scanner (practical experiments)

so this video is supposed to be about the sonar that i made it's actually working really great and i will show this however while conducting my experiments during the last month i just realized how many modern technologies are based on the same physical principles so if you're curious to find out what starlink medical ultrasound and even stone skipping have in common this video is for you this video is sponsored by isla a few years ago i tried to use a car proximity sensor to perform a scan of my lap it was a super simple project but i quickly found out that it doesn't work because the beam is too wide to get any detail a few months back i wanted to revisit this idea with a new approach the way these ultrasonic distance sensors work is that they send out a short 40 kilohertz pulse and measure the time until the echo returns since we know the speed of sound we can calculate the distance to the obstacle the frequency is not audible to humans unless you're a batman however this 40 kilohertz pulse is similar to a wave on the water surface just like the highs and lows on the surface of the water the sonic wave creates high and lows in air pressure you can even create a simple experiment levitating stuff on these pressure waves but that's a topic for another day the problem is that the wave is spreading in all directions and we can't tell where the echo is coming from these transducers seem directional but they are still made to detect a wide range to not hit anything with your car that's slightly off axis of the sensor so since working with these wide angles won't work like that we need to use some tricks so what is the solution if one transducer is not good enough take more transducers a few years ago i stumbled upon a video on youtube about phased arrays even though the name didn't sound as catchy the tech behind this was really fascinating instead of only using one pile source we can take several sources creating a narrow directional beam the math looks scary at first but there are really simple ways to demonstrate what happens there so back to our example using one source we get a perfect circular wave however using two the waves overlap they are basically adding up so low and high cancel each other out while high and high or low and low amplify each other this is also called interference if we look closely we might be able to see regions with stronger waves now the interesting part is if we change the timing of the pulses the direction of those is changing [Music] maybe we can get a clearer view in the simulation so what we got here is a constant wave with one source red are highs and blue are lows this is basically like our water surface now let's add a second source since it is a continuous wave now we are able to see the actual shape of these what are called lobes the two sources are in sync they are basically creating highs and lows simultaneously they have the same phase what happens if we shift the face of one source slightly the angle of the lobes is changing [Music] having multiple lobes isn't very helpful if we bring the sources closer together we get fewer but wider lobes at half the wavelength we only have two lobes left but in my case the ultrasonic transducers are directional enough to block the other side if i transfer this curve to the simulation this is what we get changing the face now it moves but it still doesn't look like a beam which we can use to scan a room however compared to a single source it looks like the beam is more narrow let's take a look at just the power of the highs and lows one source two sources how about more [Music] and this is basically a faced array the beam quality improves with each added source it's able to concentrate the power to one strong beam the main lobe and some weak side lobes changing the phase shift between the sources to shape and steer the beam is called beam forming this deck is used everywhere like radar wi-fi 4 and 5g antennas a prominent consumer electronics example is the stalling antenna kankita sacrificed his antenna and exposed a faced array inside unlike dish antennas which are physically built to point in one direction this dish adjusts its position at startup and then uses the phase shift of the array to communicate with the closest satellite without moving it ensures a short pink time to the local ground station even though the constellation has many satellites already there are only a few above europe or germany where i live at the same time so using a faced array to concentrate the sending power and track a satellite is crucial in this case another example closer to my project is medical ultrasound the ultrasound probe is basically using an array of tiny transducers to create a scanning beam and convert the returning echoes into an image this topic touches so many interesting fields i could go on for hours but let's get out of the rabbit hole and back to the basics can we recreate that with our water experiment there's only one way to find out [Music] when i press the button the microcontroller will let the solenoids tap the water with different phase shifts [Music] if you look closely you can actually see the directional wave isn't that cool [Music] the higher the delay between the tabs that's our phase shift the steeper the angle [Applause] so if you are good at stone skipping you could actually create a wave with a specific direction it's even better visible if i only filter changing parts of the image [Music] [Applause] [Music] crazy isn't it [Music] let's see if we can also make it work with ultrasonic transducers ok we know it will i spoiled everything in the intro but anyways i started with an array of 8 transducers these are working best at 40 kilohertz and we found out the closer the sources the cleaner the beam so i picked smaller transducers i used kaiket to design a board with a few amplifiers to drive them with firepower just like last time i ordered the boards at isla which is today's sponsor this time however i also ordered a stencil for the surface mount components the service is local fast and reliable since you are uploading your kiket file they can assure great manufacturing precision and you can easily match your parts with their database to order even an assembled version it makes it also easy to share your project with your colleagues customers and friends you can create a shared link and they will be able to order your creation directly from isla check out my link to this project below since i used the express service i received my boats within a week from the local factory cool let's assemble it [Music] doesn't it look cool the board is only populated with transmitters i will use a single receiver to detect the echoes it can be driven by 15 volts but it still needs the 40 kilohertz signal to each transducer to control the face individually i decided to use an esp32 microcontroller to do that activated [Music] a little bit of code later you can see the eight waveforms and a nice phase shift i can control it from the browser which also shows a simulation which i also coded [Music] but how can we make sure that our beam forming is working okay we could take a receiver and measure the signal strength everywhere in front of that but that can be automated i decided to exploit my pick and place machine for that from all my cnc-like machines it has the biggest working space and it also can be controlled by g-code to drive around for me [Music] so i attached a receiver to an arduino nano that calculates the signal strength using dft [Music] sends it to my browser control room attached the array to the frame and implemented the scan procedure let's go [Music] this is so cool [Music] the voltage is a little bit too high but still we are seeing a beam compared to a single transmitter i scanned before let's try a lower voltage [Music] it's really creating a directional beam you can see some side lobes but it's way better than i expected [Music] the only thing left is to compile it to a working sonar with a higher frame rate higher than one frame for five minutes i soldered a small adapter board to attach the microcontroller directly to the array and printed a small frame to attach the receiver since the echoes are way weaker i made a small amplifier board this improves the signal a lot code again and here we are this is the first official prototype of my diy sonar scanner it sweeps a field of 60 degrees in front of it since the pulses have a hard edge you can still hear them it's really awkward in person this is the camera the speed is limited by the depth i decide to scan and the angular resolution of the scan it's basically a physical limitation it's not very portable but i'm working on that [Music] these are a few tests how well we can see individual objects [Music] let me display the sonar as an overlay cool [Music] i'm really happy how it turned out i hope i was able to tickle your curiosity if so consider subscribing to my channel for more of these projects i put some links to the parts tools and product files in the description big thanks to isla for sponsoring this video please check them out also great thanks to all my supporters and helpers around the lab that really helps when i put more effort in bigger projects i see you next time bye

Channel: bitluni

Views: 890,961

Keywords: maker, diy, tutorial, esp32, esp8266, electronics, tinkering, project, coding, arduino, atmel, microcontroller, espressif, phased array, ultrasound, ultrasonic, radar, starlink, technology, physics, teaching, waves, interference, sonar, scanner, tinker, beamforming, pattern

Id: z4uxC7ISd-c

Length: 14min 29sec (869 seconds)

Published: Sat Feb 19 2022