today's video takes a closer look at a componentthat you might have used before, or you might have seen on other peoples' projects. it's called an ultrasonic transducer. it's designed to be a microphone or a speakerfor frequencies that are too high for humans or most other animals to hear.most people probably haven't had a reason
3d printed lower receiver for sale, to use one of these transducers all by itselfbut you might have used it inside one of these ultrasonic rangefinder units like the parallaxping, or one of these generic models from ebay.these two rangefinders use separate transmitter and receiver units. they're even labeled onthis board. the way these work is the transmitter
sends a little ultrasonic click that we can'treally hear, and then it times how long that click takes to propagate toward some obstacleand bounce back, and then land at this receiver. you'll also see some rangefinders like thisone from maxbotix that uses a single transducer to transmit and to receive. this results ina design that's more compact and possibly even more accurate at the cost of somewhatmore complex electronics. this is an older project of mine that usesone of these rangefinders as kind of a user interface. but you also see these a lot inrobots, to detect obstacles. nowadays, lidar is starting to become a viable alternative,but these are still super useful and it's nice to know how they work and how they mightbe modified.
even though the actual principle these operateon is identical, they really do have very different capabilities. the parallax sensoris really nice because it can give you a digital readout immediately when the echo is received,and then you can just time it yourself with arbitrary resolution, and the components areall well chosen so you get a fairly stable output. and these little ebay modules... youcan find them i think if you just search for this code, hc-sr04, you'll find these. andthey operate on a similar principle to the parallax ping, they're just made more cheaplyall around, so the output isn't as stable. and then these sensors from maxbotix, i reallydon't like them. they try to be clever and they have a microcontroller that can be configuredto send back results in various formats, including
a voltage and a time and a serial message.but this is all after it's been quantized to one inch increments, so it's really nogood at all for any application where you want to sense something with some nuance toit. so here i've got some test equipment set upthat will approximately replicate the distance measurement technique that these little ultrasonicrangefinders use. and we can experiment with different frequencies and different speakersto see how it works. so the bottom trace is coming directly fromthe signal generator. i have it set up to generate a burst of sine wave pulses. in thiscase, four cycles, and then it stops for 100 milliseconds. then the top trace is comingfrom this little microphone here. so, let's
plug in the speaker and see what happens. so the amplitude on the bottom trace goesway down because the speaker has 8 ohm impedance and the signal generator output is 50 ohmimpedance, so the two act like a voltage divider and the speaker doesn't get that much voltage.but that's fine. then on the top trace we can see that the 1 khz pulse is getting mixedin with all the environmental audio but it still stands out pretty well. we can pullthis away and the amplitude decreases, but more predictably the phase also shifts accordingto the speed of sound. so this would work, but the waves don't propagate very far andit's super annoying. so let's see if we can get the frequency up out of the human rangeof hearing.
so as i start changing the frequency here,you can see the wave shape is distorting like there's a resonance nearby. and that soundslike it. you can even hear it getting louder. so, right around there we're hitting somekind of mechanical resonance in the speaker that's boosting our signal. kind of like atuning fork. and after i get past around 9 khz it's muchquieter. the wave continues to shift in shape as it distorts a little bit. right around30 khz now. i'm still hearing a click from the speaker, but if you zoom in you can seewhat's going on there. so i'm trying to drive this speaker here at 30 khz, and the speakerseems to be responding to the dc component in this pulse where it actually switches fromoff to a sine wave, so i think that switch
where it goes from dc to sine wave is turninginto the click we're hearing. and some of the 30 khz is making it through, but reallynot very much, and there's so much distortion from that initial click that it would makeit difficult to actually measure the 30 khz portion reliably. 48 khz... 55... yeah, none of the ultrasonicis getting through at that point. so this is a short sine wave burst at 40 khz,the same as one of these ultrasonic rangefinders would use. and we can see a little bit ofthat 40 khz showing up in the microphone, but mostly it's just this terrible attemptthat the speaker makes to respond to the pulse train showing up and then disappearing. ifi drive the speaker with 40 khz continuously
some of it makes it through, but it's veryinefficient. it's finally inaudible because now the speaker is no longer responding tothe lower frequency components of the pulsed signal. so let's compare this speaker withan ultrasonic transducer. right away we can see it's much more efficient at generatingthat 40 khz pulse and it doesn't have that terrible low frequency distortion that madethe speaker annoying. so that's pretty good. got a nice strong fieldright in front of it. seems to fall off after around 30 degrees, which is what most of theseare specified for. let's try changing the frequency to see how this one responds. you can see it looks like the transmitterreally wants to ring at 40 khz. if i give
it a much higher frequency, then it filtersout most of it except that 40 khz energy that's left. hopefully the scope display isn't gettingtoo crowded. i added this new purple trace which is an fft, so the frequency domain representationof this received audio signal up at top. here the peak is right at our 40 khz frequencythat we're transmitting. let's go up. let's try transmitting a higher ultrasonic frequencythan 40 and see what happens. so that peak is shifting down, and some of the energy ismoving over into the higher frequencies, but not very much of it. a lot of it is just thatpeak moving to a different harmonic. if you're familiar with this behavior this will rightaway look like a resonator. this is basically
acting like a tuning fork. it's designed toresonate at 40 khz, and so if we help it out by sending it exactly a 40 khz input signalthen a lot of that energy will make it through. but if we give it a different input signal,very little of that will make it through and it really just tries to convert that energyback into 40 khz vibration because that's what it's really designed to mechanicallyoscillate at. here i'm down at 19 khz... we're getting a little of that 40 khz ring still,almost nothing at 19. so this would be a very bad general purposespeaker. oh, now i can hear it clicking. that's interesting. yeah, this would be a very badgeneral purpose speaker. it's really more like an electrically actuated tuning forkthat vibrates at this one particular frequency.
that said, people have made a particular kindof audible speaker using these transducers. if you search for "parametric speaker" you'llfind some interesting examples of that. this is another type of ultrasonic transduceri'm interested in. this one comes from one of those alarms you can put in your car thatbeeps if you're about to back into something. and it was really cheap; i think the entirekit of dashboard mounted thing and four of these transceivers was something like $15on ebay. i'm interested in these things because they serve a similar function, but ratherthan being kind of fragile and open to the elements like the traditional little metalcans, this one seems pretty rugged and waterproof, and the actual front surface of this thingis the ultrasonic emitter, which is pretty
interesting. this is yet another different type of ultrasonictransducer. you might have seen these in an ultrasonic levitation video from mike of mikeselectricstuff.they're pretty interesting. they're usually used for ultrasonic cleaning devices, andso they're actually designed to operate at a fairly high power, but let's just see howthey would compare to the little ultrasonic transmitter in this same circuit. so this whole contraption is actually designedto resonate at this very high frequency. the energy we're putting into this doesn't getus nearly as far because we have so much mass to move around. it does work as a transmitter;if i get it really close to the microphone
you can get a weak signal. after this initialenergy has gone into getting it started vibrating, you can see actually some back-emf from thatvibration going back into the transmitter after that pulse has ended. in fact, thatringing lasts for quite a long time. not a lot of actual energy gets through this systemjust because this signal generator can't move so much mass without a lot more output power,but we do see such a long reverb tail as the resonator slowly dies down. if i squeeze thetransceiver, i can damp that much faster. that also changes the frequency very slightly. so far we've seen these two transceivers thatwere clearly designed both to transmit and receive. but we also have these much morecommon devices where one transducer is clearly
marked as a transmitter and one is clearlymarked as a receiver. normally i would have thought that this would just be a differencein the circuit layout and not in the transducers themselves, but when i tried buying thesetransducers individually, these really common inexpensive ones that are used in all theultrasonic rangefinders... they're only sold as transmitters or receivers, usually in pairs.so i'm curious what the difference is between these devices. from the outside they lookpretty much identical except for the stamped t and r. alright, here's the one marked receiver. nowif i switch to the one marked transmitter... that's very similar. it's actually slightlyless. i've tested a handful of t's and r's
from the same vendor and presumably the samebatch, and they're very similar to each other but in most cases, i think in all the cases,the r performs slightly better as a transmitter than the t, so maybe if you buy these testthem yourself and don't go by the markings? so the t.. that one was an r and it seemedto have about a 30 degree angle. about the same here. so this is the same setup, but i've movedthe microphone down here and turned up the gain somewhat. and now we can use this setupto measure distance. but you can see there's quite a lot going on, on the scope now. thattop trace has ambient audio from the room, and it has that early transmission path directlyfrom the transducer to the microphone. then
finally it has the signal we're actually tryingto measure right there. now we have pretty much the same setup exceptthat instead of receiving using a normal microphone, the receive side is just this ultrasonic transducerconnected directly to the o-scope. and now you can see the signal to noise ratiois way better. this is because both the transmitter and the receiver now are acting like selectivefilters that only really want to pass that 40 khz signal. so at the mechanical levelactually, the receiver is ignoring the room noise, ignoring my voice. that gives thiskind of a receiver much more ability to pick out signals in a noisy environment. you canfilter out a lot of the acoustic noise mechanically before it even reaches the electronic preamplifier.
if i rotate my hand, the signal pretty reliablyreflects. there's always some part of my hand that's pointed roughly toward the sensor.but because there are so many different surfaces the signal can reflect off of, all of thosedifferent reflection paths merge together in the air to create an interference patternthat changes as i move my hand. so that's why the shape of the wave and the amplitudechange so much, even if the reliable pulse at the same time can still let us measurethe distance accurately. this piece of cardboard reflects the wavemuch more efficiently than my hand. it's big, flat... in fact, if i manage to hold thisat just the right angle parallel to the table you can see multiple reflections as the wavesbounce again and again between the table and
the cardboard. this technique of bouncingsound back and forth between parallel plates might remind you of a laser. this kind ofresonant cavity is also how a microwave oven's magnetron tube is tuned. the nice reflectionwe get with the cardboard gives us a hint that maybe we can guide these waves... usingsome kind of waveguide. this piece of vinyl hose works a lot like an optic fiber for thisultrasonic sound. so it leaves me wondering, how much couldthis technique be extended? just like it's popular to put your surface mount led on thecircuit board and then use a light pipe to throw that light directly where you want iton your project's enclosure, i wonder if you're building a robot how practical would it beto put the ultrasonic transducer on the circuit
board, and then have features in maybe a 3dprinted robot body that direct the beams all in different angles. now we have the transmitter at the left aimeddirectly at the receiver on the right. the top trace's scale is exactly a thousand timesgreater than the bottom trace. here we're showing the transmitted signal with 10 voltsper division and the received signal with 10 millivolts per division. as you might expectfrom how efficiently the cardboard seems to reflect the ultrasonic waves, not a lot ofthe ultrasonic energy would get through this piece of cardboard. on the other hand, ifi completely cover up the transmitter with my hoodie the amplitude only goes down byabout half. the waves make it pretty well
through cloth. this thin plastic cling filmblocks about as much as the cloth. if i leave the film somewhat loose then it transmitsthe energy pretty efficiently then as i pull it tighter you can see it doesn't transmitas much ultrasound, then if i pull it even tighter now it it actually starts to oscillateat a different frequency on its own. so i have the transmitter on a turntable andit's aimed for maximum signal amplitude which unsurprisingly is directly at the receiver.i continue to get some signal even way out here at 90 degrees. if you're lucky enoughto find a data sheet for the transceivers you have, they might have a graph in polarcoordinates showing you the power at every angle. and usually they drop off pretty sharplyaround 30 degrees. these seem similar. it
almost looks like they radiate energy justa little bit from the sides of the cylinder. so that got me thinking about making 3d printedobjects that have space for a transducer and a built-in waveguide. this one is just a simpleelbow joint. there's a 90 degree bend. the tube is about the same diameter as the endof the transducer. so if this design works correctly i would expect this porthole tobehave a lot like the original opening on these transducers would. so i mounted this with the transducer pointingthat way, just as a control to see whether this was successfully redirecting the flowof acoustic waves upwards. and i was still getting some kind of a signal, so i thought,oh, maybe it's radiating off this flat surface.
but if i cover the porthole then i get nothing.so it looks like now the transmission amplitude is pretty respectable, and it definitely doesseem to be aimed in the direction we expected it, coming straight out of that porthole.so it does look like this 3d printed elbow joint seems to be working. but i did noticethat the off-axis radiation appears to be more spread out than with the original unmodifiedtransmitter. it doesn't fall off quite as fast and there's this kind of strange loberight here, which... actually, that seems to be lined up with the corner. if i turnthe turntable to maximize the wonkiness of that wave, so right about there... the cornerof this box is aimed directly at the receiver. i think if i just rouned all the edges inthis thing it would probably perform pretty
well. so this is the next experiment i wanted totry. i know i can just take these ultrasonic waves going down and redirect them out thatway 90 degrees. but this is a little bit more ambitious. this has a little cone in the center,so the idea is you aim the ultrasonic waves straight in and the center of that beam hitsthe cone, and that disperses the waves 360 degrees. and hopefully it comes out thesevents in a usable form. so i've tried to design this not really knowing anything about properacoustic waveguide design. i kind of went on intuition and i tried to design this soeach of these little compartments, each little vane, maintains the same area as it reachesthe inside of this cavity. so they change
aspect ratio and become taller and more narrow. i'm hoping that means the pressure waves don'texperience an environmental difference that causes them to reflect. so, if we're luckythe ultrasonic waves efficiently bounce off the cone then just spread out and exit throughall of these vents. this is a tight pressure fit just to make sure there's a nice airtightseal and it doesn't wobble it's way out. that seats tightly against a lip in the interiorthat should set it at the right depth. well i'm pretty happy with that too. the outputisn't as strong but it seems pretty evenly spread all 360 degrees. there are some specificangles that are basically dead spots, but they're quite small. this has managed to spreadthe energy out over quite a large range of
angles. well that's pretty neat. this mightbe pretty close to an omnidirectional ultrasonic transmitter. sort of like an antenna. so as we've seen, all of these items are ultrasonictransducers. mechanically they're different, but electrically they all operate on the principleof piezoelectricity, the same as these little discs. you might have seen these little piezoelectricdiscs in greeting cards or in other cheap electronic devices, but they're actually prettymarvelous. this common piezoelectric buzzer is a sandwich of materials. on the bottom,any kind of flexible metal, in this case brass. on the top, a very thin layer of silver orany other conductive material. but in the
middle, that's the piezoelectric material. you might have seen the initials pzt. somepeople will tell you that stands for "piezo transducer", but really those are the initialsfor the chemical compound that makes up most piezo transducers, lead zirconium titanate. piezoelectricity is a relationship betweenmechanical strain and an electric field potential. so in this case, by applying a voltage acrossthe piezo crystal it will actually grow slightly. because it's bonded to this piece of metalthat won't grow, it causes the whole thing to flex like the head of a drum. these piezo discs also use mechanical resonanceto amplify their sound, but they tend to be
tuned for the audible range. this electronic whiteboard marker we lookedat in an earlier video uses a thin film polymer transmitter, but these other ultrasonic transducersprobably use the same pzt crystals that this simple buzzer uses. because this transducer is so large it's easierto see the components. we can see here too layers that have probably some elastic propertiesto them. but most importantly there's this layer of piezo ceramic material. probablyseveral stacked layers, actually. and when you apply voltage to those layers, they'llexpand and drive the mass a little bit in one direction. then the elastic will pullit back, and it will set up a mechanical oscillation.
then by adding energy into the system at thesame frequency as the mechanical oscillation, you can drive a continuous wave (a continuousvibration) actually really efficiently. this isn't designed to transmit energy into theair. it's designed to bolt onto something and vibrate that thing. i mean, it will transmitenergy through the air, as we've seen, but it doesn't really do that especially efficiently. so that leaves us with these two. and theseare both designed to transmit 40 khz waves through the air. so a transmitter like this would really beno fun to use outdoors. anything gets in there and it probably just stops working completely.so the automotive transmitters are designed
somewhat differently. this whole front plateis... you can feel it actually ticking as it transmits, so this must actually be directlyabove the piezo element. but then it's also somewhat mechanically isolated from the exteriorof the housing here. and it looks like there's some kind of rubber piece in between thatis probably actually providing a weather seal and giving the interior bit a little roomto wiggle. i'm curious if we can see any internal differencesbetween the t and the r here. from the outside they sure look the same. now we can brute force this... stick it ina vise, get the dremel, just cut through the side here. but i'm interesting in seeing ifwe can take this apart the way it was put
together. so i wonder if we can get this outeasily just by un-bending the edge of this tube. this might be easier than i thought,the metal is pretty soft. this looks like it's answering some questions but maybe askingsome new ones too. oops... the screens were just held into thefront of the tube by this little metal ring. so the transmitter and receiver both workfine with the case off, though we can see they have a somewhat wider radiation pattern.here's a much closer look at the transducer marked t for transmit. you can see the plasticbase holds the two sturdy pins in place, but then two very fine bond wires carry the electricalsignals up to the actual piezo crystal. so the piezo crystal's going to be somewhereinside that larger disc area. but that little
metal hat on top seems to be sort of likea horn that pushes the air as the transducer moves up and down. the vibrations in the transmitter are muchtoo subtle to see, and they seem to be too subtle to move around something like thisplastic bead. but if i touch the piezo transducer or the vibration-spreading cone with my tweezers,you can hear it screech as the two things vibrate against each other. this is the one marked r for receiver. sofar i've noticed two differences. one: one of those leads has been pushed up a bit, andit's almost touching the piezo transducer. but that might just be something i did toit during testing or disassembly. it might
not be a manufacturing defect. the other differenceis in the shape of this little resonating cone at the top. over here on the transmitteragain, that little resonator on the top is just a little stamped metal cone that lookslike a perfect little horn to push the sound waves up into the air. but over here on theone marked receiver, it looks like they filled that cone in with just a drop of glue, justa drop of clear epoxy or something. so i wonder if this serves to change the shape of thesensitivity pattern, or whether it kind of damps out extra vibrations after the pulsehas been received. it's kind of strange that this one seemed to be just a little bit betterat transmitting than the unit without the drop of glue.
probably because this is one of the cheapertransducers, instead of a metal plate with a feed-through gasket for one of the wires,this is really just a plastic plate with this tiny thin layer of metal on top that isn'teven electrically connected to either of these legs reliably. so if we put that aside, thenwe're just left with this plastic assembly with the actual ultrasonic transducer mountedon it. so this disc where the horn is mounted seemsto be the actual metal backing plate for the piezo transducer. then on this side you cansee a little hint of the electrodes and the yellowish piezo ceramic material. alright, i'm not worried about breaking thebond wires or this plastic piece, but i would
like to do this without breaking the piezocrystal if i can. it looks like this is stuck onto the plastic base with kind of a siliconeputty. so here's what's left of the bottom plate.now we can get a much clearer idea of how this was put together. the piezo crystal wasattached to this plastic baseplate using a flexible silicone adhesive. the baseplateholds it only at the edge, leaving the middle free to swing up and down. here's the tiny ultrasonic transmitter coneagain. now that we've removed it from the supports, on the opposite side we can seeour familiar little piezo disc. so again we have a metal piece on the outside. in thiscase it looks like it's steel. then the kind
of yellow pzt crystal, then the silver electrodeand tiny wires connecting them to the pins. so that was the transmitter. let's take alook at the receiver and see if we can notice any other differences. well, so far it looksthe same. there's the drop of glue inside the cone like we saw before. and on the otherside it looks like they might be using a little bit more silicone goop, but that's probablyjust unit to unit variation. it doesn't really look like this is substantially differentfrom the transmitter. alright, i had one last thing to tear intoin this video: these inexpensive automotive ultrasonic transducers. i'm not sure why theybother telling you which side is up. maybe it's a weatherproofing issue? in any case,it looks like this cylindrical piece fit inside
an outer plastic shell, probably press-fitwith some snaps. i'm going to see if i can wedge it out through these little rectangularslots. there's a plastic cover that seems to come off pretty easily if you wedge theseplastic slots open. alright. that reveals a whole load of siliconegoop. looks like that center hole in the back panel was an injection port to fill the wholeback side with silicone. i'm just going to cut this to make it easierto deal with. so this is actually pretty close to the end of the wire. the silicone sealantjust barely covers the gap between the cable and the insulation. actually it looks likethis is coax, which is nice. i think at this stage we might be better off just peelingoff the outside of this casing. i can already
feel the inside and outside of this cylinderseparating. so that outside piece is just all coming off together. so this is just kindof an outer shell to hold everything and wedge it into the bumper of your vehicle. then thissilicone acts as a shock mount for the interior ultrasonic transducer, then it's all beensealed in the back with silicone goop. oh yeah, it seems like it peels open on thefront. just like a little silicone boot that this metal can fits into. so it looks likethere's a tiny circuit board back here, which might be... i kind of doubt there are anyelectronics on there, that's probably just for terminating the coax. but it looks likethe coax terminates to a tiny circuit board, then there's a little wire for the centerconnector running to the actual transducer
body. there's a metal spring clip connectingthe ground on the coax to the exterior of this can. just so i can separate these two without toomuch damage to either i'm going to cut this little wire in the middle. so now we havethis transducer body with just a metal can for one electrode and this little wire. well,this tiny circuit board doesn't have any components, it was just part of the coax termination.all the way in here we have the same kind of silicone we saw on that very outer seal.a little bit crumbly, but it feels like basic weather sealant. right under that white layerof silicone sealant, there's a layer of this brownish material that might be some kindof glue or some kind of other elastic. the
brown material is... that's very strange,it almost reminds me of cork. could they have just filled this with sawdust as an elasticor damping medium? maybe this is cork for moisture absorption, like in case any watergets all the way back here. oh yeah! this is a cork. hah. so there's a small pad ofcork material. it might have been for moisture purposes, or for springiness? yeah! now wecan see where that wire is actually soldered to the piezo crystal. between the cork andthe piezo crystal is this soft donut of felt. this might have supported the crystal a littlebit more gently along the edges while still allowing it to vibrate in the middle. so now we can see our old friend again. it'sthe same kind of little pzt transducer, but
this time much tinier. so this whole piecewe're left with after pulling out all the layers of spacers and fillers, it seems tobe just a single piece of turned aluminum, painted black on the front, then on the insidewe have the tiny pzt disc bonded to it. this is a pretty interesting design. it seems prettyrugged against the environment, and also it seems to produce a pretty strong acousticvibration. wow, that thing is pretty powerful. i'm gettingwaves all the way over here. well thanks so much for watching! i hope youenjoyed taking a closer look at the electronic components that make these common distancesensors work. and i hope it might have given you some inspiration to use ultrasonic transducersin some new ways, and maybe come up with some
new ways of using sound that we can't hearto measure things or otherwise make people's lives better or more fun. so thanks for watching!and if you enjoyed this, maybe check out some of my other videos or subscribe, and you'llbe the first to know when i post something new. thanks!
