Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
Music Science

Researchers Develop a Paper-Thin Loudspeaker (mit.edu) 66

MIT engineers have developed a paper-thin loudspeaker that can turn any surface into an active audio source. MIT News reports: This thin-film loudspeaker produces sound with minimal distortion while using a fraction of the energy required by a traditional loudspeaker. The hand-sized loudspeaker the team demonstrated, which weighs about as much as a dime, can generate high-quality sound no matter what surface the film is bonded to. To achieve these properties, the researchers pioneered a deceptively simple fabrication technique, which requires only three basic steps and can be scaled up to produce ultrathin loudspeakers large enough to cover the inside of an automobile or to wallpaper a room.

A typical loudspeaker found in headphones or an audio system uses electric current inputs that pass through a coil of wire that generates a magnetic field, which moves a speaker membrane, that moves the air above it, that makes the sound we hear. By contrast, the new loudspeaker simplifies the speaker design by using a thin film of a shaped piezoelectric material that moves when voltage is applied over it, which moves the air above it and generates sound. [...]

They tested their thin-film loudspeaker by mounting it to a wall 30 centimeters from a microphone to measure the sound pressure level, recorded in decibels. When 25 volts of electricity were passed through the device at 1 kilohertz (a rate of 1,000 cycles per second), the speaker produced high-quality sound at conversational levels of 66 decibels. At 10 kilohertz, the sound pressure level increased to 86 decibels, about the same volume level as city traffic. The energy-efficient device only requires about 100 milliwatts of power per square meter of speaker area. By contrast, an average home speaker might consume more than 1 watt of power to generate similar sound pressure at a comparable distance.
The researchers showed the speaker in action, playing "We Are the Champions" by Queen. You can listen to it here.
This discussion has been archived. No new comments can be posted.

Researchers Develop a Paper-Thin Loudspeaker

Comments Filter:
  • It sounds much worse than I was expecting, and my expectations were not high. Interesting to see where the technology goes however.
  • by koko ( 66015 ) on Wednesday April 27, 2022 @09:16PM (#62485766)

    Much louder than that thing. 66dB is pathetic. And 20dB louder (4x as loud) between 1kHz and 10kHz is one screaming mimi. Horns can be even much more efficient than a regular cone speaker, at 100dB/W at 1 meter.

    • Why was this downvoted? He is correct. Typical speakers are 86-92dB/W at 1 meter. And as he says, horns can be higher. The article is just an expansion of piezo drivers, which have been around forever. Moto had a super-tweeter using a piezo driver in the early 80's. And unsurprising the article says one application will be ultra-sonic for detection of bodies. Same thing they use in car bumper ultra-sonic systems. Distortion numbers were absent from the article too.
      • Because this uses 100mW for roughly the same volume - 1/10th the energy. Battery powered speakers could benefit from that.

        • Because this uses 100mW for roughly the same volume - 1/10th the energy. Battery powered speakers could benefit from that.

          Exactly. Put it in a proper enclosure and it might be quite surprising.

          • At the very least a lot of remotes can now suddenly benefit from a remote finder without the dimensions to support both a speaker and AA batteries. Maybe one day they could affix to windows, remain transparent, and filter road noise while letting bird sounds pass through.
            • At the very least a lot of remotes can now suddenly benefit from a remote finder without the dimensions to support both a speaker and AA batteries. Maybe one day they could affix to windows, remain transparent, and filter road noise while letting bird sounds pass through.

              Interesting applications.

      • They never mention how large the speaker is they were testing, so there's no real way to get a SPL/W out of the numbers they are giving.

        You could argue how coy they are with numbers and measuring at 30cm suggests it ain't all that good, but that's just about it.

      • Why was this downvoted? He is correct. Typical speakers are 86-92dB/W at 1 meter. And as he says, horns can be higher. The article is just an expansion of piezo drivers, which have been around forever. Moto had a super-tweeter using a piezo driver in the early 80's. And unsurprising the article says one application will be ultra-sonic for detection of bodies. Same thing they use in car bumper ultra-sonic systems. Distortion numbers were absent from the article too.

        OMG!

        The infamous (and ear-splitting) 50 W(!) Motorola Piezo Tweeter (crossover optional!) and the JBL 075 "Bullet" Ring Radiator 27 kHz Traffic Sensor Special; I remember them well. . .

      • by jbengt ( 874751 )

        Why was this downvoted?

        Maybe because he said that 20dB louder is 4 times as loud, when it is actually 100 times the sound pressure. (though "loudness" is subjective)

    • by Bodie1 ( 1347679 )

      Almost. 6 dB is 4 times louder.

    • The dB scale is logarithmic. 10dB is 10 times as large, so 20dB is 100 times as large. https://soundear.com/decibel-s... [soundear.com]

    • by Zappy ( 7013 )

      20dB is 100 times louder.
      ~ 6dB will give you 4 times louder.

    • I agree about the SPLs. Paper coned speakers can go over 100dB if they have neodymium magnets, otherwise horns do have it for efficiency, because the match the impedance of the driver with the air.
  • by sedenion ( 8059516 ) on Wednesday April 27, 2022 @09:31PM (#62485782)

    Interesting, but this seems to be just a piezoelectric film speaker, something that has existed for several decades... Am I missing something?

    • Re: (Score:3, Interesting)

      by sedenion ( 8059516 )

      Oh, and although I like Queen, the music in this 2010ish demo is much nicer: https://www.youtube.com/watch?... [youtube.com]

    • by omnichad ( 1198475 ) on Wednesday April 27, 2022 @10:14PM (#62485842) Homepage

      It's designed in a way that the film can attach to a surface and make the surface the speaker. It has holes via photolithography to let the air move through so that the surface doesn't muffle or distort the sound.

      • It's not holes to let air move through. They make the holes and then use a vacuforming process to suck the piezoelectric film through the holes, effectively creating thousands of separate dome-shaped speakers, each the diameter of a few hairs.
    • by Pinky's Brain ( 1158667 ) on Thursday April 28, 2022 @12:24AM (#62486010)

      They pre-stress it into lots of little holes to make small domes.

      If you just try to apply a voltage to a big sheet of piezo electric material it will try to spread out as a whole, which the material resists. That resistance scales superlinearly with the size of the sheet, it only really works for very small sheets. In their case each little dome does not affect the others, so output scales linearly with size.

  • Not new (Score:5, Interesting)

    by xevioso ( 598654 ) on Wednesday April 27, 2022 @10:24PM (#62485864)

    From the first Youtube comment:
    "Do people not realize that this already exists and has for 50 ish years? It’s a piezoelectric film element and they’ve existed for high SPL use on the market since the 70’s. Headphones, full range speakers, and tweeters using multiple mechanical layouts exist using this tech, I’ve even built some. I sure hope this didn’t receive some big grant, all they’ve come up with here is a marketing application for an existing technology. Efficiency with these type of transducers is exceedingly high for only reproducing band signal at higher frequencies with very problematic dispersion and phase but full range types require huge power and have a lot of engineering challenges that no one ever overcame."

    • I agree. It reminds me of the ads for those beautifully thin 'Quad' electrostatic speakers that were advertised in UK's Practical Electronics and other magazines in the 60s/70s.
      • I agree. It reminds me of the ads for those beautifully thin 'Quad' electrostatic speakers that were advertised in UK's Practical Electronics and other magazines in the 60s/70s.

        Quad still makes their electrostatic speakers. Martin Logan makes some very nice ones as well. I have a pair of Magneplanars which are also thin film speakers using a different technology (planar magnetic). One thing they all have in common is they don't do bass well. Low frequencies require moving a lot of air, and while electrostatic and planar magnetic panels are large, their excursion is very small. I have a pair of subwoofers to fill in the ~20-50Hz region. Martin Logan builds a traditional cone

    • I used to build PA speakers (as a hobby & because I couldn't afford to buy good quality commercially made ones) & I simply call these "tweeters" because they're only useful at higher frequencies. They're no good for mid-range or bass frequencies. Tweeters come in larger, bulkier housings because the sound needs to be directional in order to give better sound separation (left & right speakers) for stereo effects. Just sticking a piezo driver to a wall will give you a tinny, omnidirectional sound.
    • I agree that piezos are not a new thing.
    • But this is wrong. It's not just a piezoelectric film speaker. They took it and pulled it through tiny holes, so it is no longer the entire surface vibrating, but thousands of separate domes vibrating. Now you can do things like flex it, or butt it up against a solid surface without the same losses. And because it isn't moving the entire sheet with a massive amplitude, it resolves many of the power issues.
  • Tinny as anything, but it's a seed from which a mighty oak grows, I'm certain.

    If nothing else, kids can now understand why the bridge of Raspberries' "Overnite Sensation" sounds so weird.

  • by Tony Isaac ( 1301187 ) on Wednesday April 27, 2022 @10:31PM (#62485872) Homepage

    This type of speaker would have basically no bass at all, just as is apparent listening to the YouTube video linked in this story. For real bass, you've got to move the air in a significant way. This is why you can literally feel the sound when using a good subwoofer. There's no way this film will be able to generate those kinds of low-frequency sound waves.

    Why is that important for those who don't care about bass? Some of the proposed uses rely on the ability to significantly move air. For example, noise cancellation in an airplane cockpit would be ineffective without the ability to emulate low frequencies as well as high.

    These speakers are interesting in some ways, but don't mistake them for high fidelity speakers.

    • by 93 Escort Wagon ( 326346 ) on Wednesday April 27, 2022 @10:48PM (#62485888)

      This type of speaker would have basically no bass at all, just as is apparent listening to the YouTube video linked in this story.

      Executives from Bose immediately declared an interest in providing further funding for this research.

    • but don't mistake them for high fidelity speakers.

      Don't confuse high-fidelity and full range. Fidelity is about the ability to properly reproduce what you can, not about how wide of a frequency range you can cover. Otherwise there'd be no such thing as a high-fidelity speaker.

      That said, colour me skeptical about the "fidelity" here too. The underlying material and it's dimensions have a huge impact on fidelity, making it effectively impossible for a device like this to consistently have any decent performance metrics.

      • When applied to audio speakers, most people would interpret "high fidelity" to apply to the ranges of sound that a human ear can hear. "Is it live, or is it Memorex?" (You had to live during the cassette tape era to know that reference!) No one would call a tweeter, by itself, "high fidelity." To get high fidelity, you usually need several different sizes of speakers working together to reproduce the full range of sound that accurately reproduce what we would hear if the sounds were "live."

        I agree with your

        • When applied to audio speakers, most people would interpret "high fidelity" to apply to the ranges of sound that a human ear can hear.

          Err no. Never heard anyone talk about high fidelity to mean the range. God knows no one has ever gone to a white goods store and seen the kids play with $30 15" subwoofers in the back and declare that to be "high fidelity". Quite the opposite, people who use the term often use it to describe sound systems that have a significantly narrower frequncy range than a cheaparse budget soundbar + sick subwoofer.

          To get high fidelity, you usually need several different sizes of speakers working together to reproduce the full range of sound that accurately reproduce what we would hear if the sounds were "live."

          But yes this was my point. Just because a technology can't apply to a large frequency range doesn't mean

    • Music is not the only use for speakers. Noise cancelling, locating your keys or remote, echo location, etc. doing it at low power increases the potential applications. I dont need Barry White vocal ranges to tell me where my remote is. Chip or Mickey ranges work fine for that goal.
      • I'm not asserting that there are no applications for this technology. I doubt noise cancellation will be effective, because this would only be able to "cancel" high frequencies at low dBs. Echo location or key location could work, but don't require this "new" technology, any standard piezo-electric speaker can generate the necessary tones.

        What I am saying is that the label "high fidelity" doesn't really apply.

  • Sounds like a maintenance nightmare. Surfaces are exposed to a lot of wear and tear, and replacing such a film sounds non-trivial. You do *not* want to cover any surface you are going to touch regularly with anything expensive or important.

    I wonder about it's performance at human voice frequencies. They use quite high frequencies as their example. "When 25 volts of electricity were passed through the device at 1 kilohertz (a rate of 1,000 cycles per second), the speaker produced high-quality sound at con

    • It suggests it's almost entirely capacitive ... they just need to drive it harder at lower frequencies.

      Can't really drive this with any ordinary amplifier, needs to be class D and specially designed for it at that, so it can recover the power stored in the capacitance.

    • "Surfaces are exposed to a lot of wear and tear...

      That's not quite so true of ceilings.

    • If I understand it right, the article suggests that the entire surface can be covered with a layer of PET film. I can't imagine getting too much better than that while still allowing sound transduction...Maybe add an elastomer if you want some scratch resistance...
  • by WierdUncle ( 6807634 ) on Thursday April 28, 2022 @04:49AM (#62486198)

    If I understand the article correctly, what they have done is produce an array of point sources, which means the net result is a diaphragm that moves like a flat stiff panel. This is quite different to existing piezo sounder technology, where a disc is made to flex by application of a voltage.

    The article says that the strong mechanical resonance that occurs in a single piezo disc is eliminated, or at least pushed into supersonic frequencies. This means that a reasonably flat frequency response might be possible, so in that sense it could produce high quality sound.

    The main disadvantage I see is that the diaphragm deflection is very small. The article mentions half a micron. Typical cone loudspeakers have deflections of several millimetres. In loudspeaker design, the sound pressure level from a diaphragm is proportional to the product of the diaphragm velocity and area. What that means is that, in order to achieve a given diaphragm velocity at low frequencies, you need more diaphragm deflection. Anybody whose has observed small bass speakers belted with bass and drums will know what I am talking about. The velocity * area formula is the basic reason why bass speakers tend to be big. In the case of this new piezo technology, I don't see any possibility of bass output, unless they can find some way of increasing the air volume movement. It is interesting to note that their acoustic output showed 66 dB at 1 kHz, rising to 86 dB at 10 kHz, which bears out what I just said. I am not sure you could get even basic comms speech out of this at a usable volume, which would cover the range 400 Hz to 4 kHz, typically.

    Where I do see an advantage is for high frequency applications. The article mentions this. It might be worth looking at it for hi-fi tweeters, above about 3 kHz. The lack of resonances in the audio frequency range would eliminate one of the problems with tweeters based on cones, domes, etc. Another application the article mentioned was broadband ultrasonics. I presume this transducer can receive as well transmit, like other piezo transducers do. The range resolution of existing (resonant) systems is limited, because pulses are splurged over many cycles, due to the narrowband signal. If you have a broadband signal, you can use things like chirp pulse compression, to produce seriously good resolution, as in some RADAR systems.

    With high frequency applications, the size of the whole array is large compared to the wavelength. This tends to produce a narrow central beam, with many sidelobes. For hi-fi applications, you definitely want to avoid this. The ideal is to have a fairly uniform dispersion to cover a reasonable listening area. This motivates tweeter cones/domes that are as small as possible. I think what could be done with this new technology is to shape the array into a curved shape, to broaden the central lobe. In the case of ultrasonic ranging and scanning, elimination of sidelobes would be a good thing. One problem you get with ultrasonic ranging is spurious echos from sidelobes.

    • In the totally weird divergent category; we could use it to generate frequency dogs hate to keep these pompous rich bitches from taking their purse dogs everywhere falsely claiming emotional support. If only we could separate them from actual service animals in some way. The most extreme example was a woman who took one on a cruise ship. Had a baby stroller and everything for it.
      • I don't know about targeting the little dogs. Is there some frequency that targets pompous rich bitches?

  • MIT marketing is operating in the dark corners of bulshitology.

    Seriously, there is so much wrong with what they state in their advertisement it hurts my brain.
    Their device can reproduce practically zero bass frequencies. It is not in any way viable as a normal loudspeaker. But they go on and compare it to one anyway. Remember, these are MIT people.

    The sheet in their video reproduces only the very top end of the hearing spectrum.
    Claiming it can be used for active noise cancellation is baffling due to the lim

    • I think there's a perfectly good article hiding in here, but, yeah, when it gets into the foreseeable applications, it just flies off the rails into nonsense-land. I guess they're trying to grab the attention of as many industries as possible to get funding from wherever they can, but I feel like stretching like this just ends up making you look bad. Maybe I'm wrong, I'm not the target audience.

      They should've selected a sound sample that's more suited for what they have at the moment. They sort of torped

      • sure, just like you're trying to get attention, what's wrong with their development?
        • I mean, if you consider participating in discourse to be "trying to get attention", then I guess you're not wrong. I don't know why you think I've got any problems with their development though. Sounds like they're doing just fine.
  • For me this is rather old news. I heard about it about a decade ago, at least. Anyway, an interesting thing is that speakers can be microphones, too. A microphone is actually a speaker in reverse and vice-versa. Imagine the possibilities... your window, or any hard surface for that matter, could be spying on you right now. I'm not being paranoid here, just saying. :)

    I'm into audio production and this interests me greatly. I'm sick and tired of these big boxes [monitor-speakers], and the thing is: the bigger

    • NSA raids your house and replaces some pictures with listening devices instead. Ones so low power that a bug sweep probably wont detect it? That would mean no EM transmissions either. Probably would have to setup an ultrasonic beam to a relay that could re-enable the window trick even with the curtains pulled. Im sure the NSA is working on it.
    • I suspect that what you've heard of is piezoelectric film speakers. This is related, but it is specifically an advancement on that technology. The article says up front that the film speakers go back many decades. Dividing the film up so it isn't a single vibrating surface is the novelty described here.
      • by dddux ( 3656447 )

        Thank you. You've nailed it. I read about piezoelectric film speakers and this novelty [dividing the film] is making it possible to make them useful for audio studio applications maybe, because they can obviously produce the high frequencies easily, but lower than 1kHz they start to sound more and more distorted. That's not unusual for the "boxes" with the common speaker drivers, too.

        Hopefully they can find a way to make the low frequencies sound good. It definitely sounds promising. What I'm also worried a

  • Imagine a street musician whose clothes are all speakers.
  • You have probably seen piezo transducers at concerts. You hear them in your car, and in a lot of electronic devices. Oddly, they are in Epson printers to squirt ink.
    • That's why the article isn't about piezo transducers. It's about vacuforming piezoelectric film through a matrix of tiny holes to create a completely different style of driver.
  • For bass, you generally need to displace a lot of air. To do that you are likely going to need a mechanism such as flexture to multiply the moment, as found--even in some piezo tweets.
  • https://www.reddit.com/r/diyaudio/comments/8hq5f4/logitech_speaker_failscam_tweeter_is_just_glued/

  • If you're going to play "We Are The Champions" choose a better medium! Like an F1 Engine! [youtube.com]

  • Congrats MIT, you just reinvented the tweeter.
  • ... WoW - totally _got_ that sound quality coming through in the sample!

    I'm now going to watch some 8k video content samples on my 2k monitor, for the WIN! - take my money!

If all else fails, lower your standards.

Working...