5 ms·

Robotic gripper uses acoustic levitation for contact-free manipulation

63 points by sahin-boydas 7y ago

flareback 7y ago
That's pretty amazing.
mreome 7y ago
I think acoustic levitation is going to be a technology worth watching over the coming years. I'm very curious what could be achieved with application specific speakers designed/built as arrays as opposed to discrete stock speakers assembled into arrays. It could end up being like comparing a matrix of discrete 5mm LEDs to an LED monitor in terms of precision/accuracy and open up a lot of possible applications.
- lachlan-sneff 7y ago
  You mean, like arrays of microscale acoustic generators? Maybe if the wavelength was small enough, one could manipulate singular molecules with nanometer precision? 3Ghz ultrasound at sea level has a wavelength of about 150 nm. Perhaps the endgame of this would be a phased array that can generate phonons with extreme accuracy.
  - traverseda 7y ago
    I mean remember that the way these things work is by manipulating air molecules. Manipulating single-molecules is going to be difficult since you're always going to be at the mercy of your carrier fluid/gas.
- hanniabu 7y ago
  I really like what they're doing with using it for volumetric displays. They take a small white ball that's a few mm in diameter and they move it around a 3d space really quickly to create the 3d "pixels" while at the same time shining light on it at each position to color the pixel. This is a great video with a demo and explaining how it works: https://youtu.be/tzWP-NL3Lck https://youtu.be/tzWP-NL3Lck
humble_engineer 7y ago
What is the point of this? Its still the same amount of force applied to keep the object suspended in air, is it a sterilization thing?
pudsbuds 7y ago
Does the actual soundwave itself have an effect on these parts? They mentioned something about a thin layer of lubricant. Wouldn't soundwaves also mess with this layer as well?
- jstarfish 7y ago
  It shouldn't; fluid is an insulator and the lubricant has enough (viscosity? tensile strength? I'm not a materials guy) to adhere to the part in question. The soundwaves must be super high-frequency so they should just get absorbed without displacement. I'd be curious to see if something like this could succesfully handle an Etch-a-Sketch or Magna-Doodle without ruining the picture though.
jbay808 7y ago
If they can manipulate multiple objects at a time, this could really speed up PCB pick-and-place operations.
beamatronic 7y ago
Can you theoretically levitate a car with this?
- computerex 7y ago
  In theory but the energy requirements would probably make that infeasible.
  - papreclip 7y ago
    >In theory I don't think so. My understanding is that there's a limit to sound wave amplitude (decibels) beyond which you enter the domain of the shockwave. a shockwave is an irreversible process which couldn't be used to maintain some state the way a standing wave can. Without enough concentrated power, you can't lift an arbitrarily heavy(dense) object If you raise the pressure of the medium, the decibel limit goes up, but air at atmospheric pressure is probably going to fall well short of what you'd need to hover a car
    - adrianmonk 7y ago
      It might just be possible. First of all, if you look at static pressure, obviously hovercraft exist and work. And tornadoes can lift cars off the ground. Even strong winds can. Doing some numbers, my car appears to have a horizontal area of about 7.5 m^2 and a mass of about 1500 kg. With earth's gravity, that means 14700 N spread over 7.5 m^2, which means average pressure of 1960 Pa. Wikipedia says (https://en.wikipedia.org/wiki/Sound_pressure#Examples_of_sound_pressure https://en.wikipedia.org/wiki/Sound_pressure#Examples_of_sou...) that a shock wave happens at 101,000 Pa. So the pressure needed to lift my car would be about 2% of the strongest pressure that can exist in a sound wave. Presumably it's a little more complex than that because sound waves oscillate and can't exert constant pressure. So some factor has to be added for that. But seems like there's about 50X to play with before hitting the limit of shock waves. It's going to be crazy loud, though. Like a gun going off continuously.
- tomcam 7y ago
  I’d like to suggest purchasing a lift from Harbor Freight...
marvin 7y ago
Does anyone know what limits the real-time calculation of these fields?
bdcravens 7y ago
"Acoustic Levitation" Found the name of my band
- adrianmonk 7y ago
  It's going to be awkward playing only songs with suspended chords.
jv22222 7y ago
Here's a fascinating video that explains Acoustic Levitation and shows it in action with various materials: https://www.youtube.com/watch?v=odJxJRAxdFU https://www.youtube.com/watch?v=odJxJRAxdFU

63 points by sahin-boydas 7y ago

That's pretty amazing.

I think acoustic levitation is going to be a technology worth watching over the coming years. I'm very curious what could be achieved with application specific speakers designed/built as arrays as opposed to discrete stock speakers assembled into arrays. It could end up being like comparing a matrix of discrete 5mm LEDs to an LED monitor in terms of precision/accuracy and open up a lot of possible applications.

lachlan-sneff 7y ago
You mean, like arrays of microscale acoustic generators? Maybe if the wavelength was small enough, one could manipulate singular molecules with nanometer precision? 3Ghz ultrasound at sea level has a wavelength of about 150 nm. Perhaps the endgame of this would be a phased array that can generate phonons with extreme accuracy.
- traverseda 7y ago
  I mean remember that the way these things work is by manipulating air molecules. Manipulating single-molecules is going to be difficult since you're always going to be at the mercy of your carrier fluid/gas.
hanniabu 7y ago
I really like what they're doing with using it for volumetric displays. They take a small white ball that's a few mm in diameter and they move it around a 3d space really quickly to create the 3d "pixels" while at the same time shining light on it at each position to color the pixel. This is a great video with a demo and explaining how it works: https://youtu.be/tzWP-NL3Lck https://youtu.be/tzWP-NL3Lck

humble_engineer 7y ago

What is the point of this? Its still the same amount of force applied to keep the object suspended in air, is it a sterilization thing?

pudsbuds 7y ago

Does the actual soundwave itself have an effect on these parts? They mentioned something about a thin layer of lubricant. Wouldn't soundwaves also mess with this layer as well?

jstarfish 7y ago
It shouldn't; fluid is an insulator and the lubricant has enough (viscosity? tensile strength? I'm not a materials guy) to adhere to the part in question. The soundwaves must be super high-frequency so they should just get absorbed without displacement. I'd be curious to see if something like this could succesfully handle an Etch-a-Sketch or Magna-Doodle without ruining the picture though.

jbay808 7y ago

If they can manipulate multiple objects at a time, this could really speed up PCB pick-and-place operations.

beamatronic 7y ago

Can you theoretically levitate a car with this?

computerex 7y ago
In theory but the energy requirements would probably make that infeasible.
- papreclip 7y ago
  >In theory I don't think so. My understanding is that there's a limit to sound wave amplitude (decibels) beyond which you enter the domain of the shockwave. a shockwave is an irreversible process which couldn't be used to maintain some state the way a standing wave can. Without enough concentrated power, you can't lift an arbitrarily heavy(dense) object If you raise the pressure of the medium, the decibel limit goes up, but air at atmospheric pressure is probably going to fall well short of what you'd need to hover a car
  - adrianmonk 7y ago
    It might just be possible. First of all, if you look at static pressure, obviously hovercraft exist and work. And tornadoes can lift cars off the ground. Even strong winds can. Doing some numbers, my car appears to have a horizontal area of about 7.5 m^2 and a mass of about 1500 kg. With earth's gravity, that means 14700 N spread over 7.5 m^2, which means average pressure of 1960 Pa. Wikipedia says (https://en.wikipedia.org/wiki/Sound_pressure#Examples_of_sound_pressure https://en.wikipedia.org/wiki/Sound_pressure#Examples_of_sou...) that a shock wave happens at 101,000 Pa. So the pressure needed to lift my car would be about 2% of the strongest pressure that can exist in a sound wave. Presumably it's a little more complex than that because sound waves oscillate and can't exert constant pressure. So some factor has to be added for that. But seems like there's about 50X to play with before hitting the limit of shock waves. It's going to be crazy loud, though. Like a gun going off continuously.
tomcam 7y ago
I’d like to suggest purchasing a lift from Harbor Freight...

marvin 7y ago

Does anyone know what limits the real-time calculation of these fields?

bdcravens 7y ago

"Acoustic Levitation" Found the name of my band

adrianmonk 7y ago
It's going to be awkward playing only songs with suspended chords.

jv22222 7y ago

Here's a fascinating video that explains Acoustic Levitation and shows it in action with various materials: https://www.youtube.com/watch?v=odJxJRAxdFU https://www.youtube.com/watch?v=odJxJRAxdFU