Hacker News new | comments | show | ask | jobs | submit login
Demo of Short-Range Wireless Power Transfer (disneyresearch.com)
203 points by saycheese on Feb 16, 2017 | hide | past | web | favorite | 73 comments

If you're having trouble loading the page, here is the direct link to the PDF: https://s3-us-west-1.amazonaws.com/disneyresearch/wp-content...

Here are the YouTube videos:



TLDR: Demo shows that in a 54 m3 specially designed room that it is possible deliver power to small coil receivers in nearly any position with 40% to 95% efficiency — and 1900 watts can be transmitted to a coil receiver enabling safe and ubiquitous wireless power.

AND nobody's eyeballs got cooked.

from what I observe after work in high power radio wave industry, there is a possibility that long time exposure to high power radio wave can have effect on people's chromosomes, not quite sure what will happen if someone expose too long to this type of device

In this device, power is not transmitted via electromagnetic radiation, but via oscillating magnetic fields. It's completely safe for humans and any electronics not connected to an antenna tuned to the resonant frequency.

EDIT: Obvious in retrospect, but alternating current like this is exactly how you make EM/radio waves, so my comment above is misleading.

However, that's the cool part about this research. They are generating quasistatic magnetic fields, and decoupling the magnetic field from the electric field—similar to how near-field charging pads work, but at room scale. So they are producing very little in the way of EM radiation. From the paper:

> For example, radiative transfer methods have tightly coupled electric and magnetic fields that propagate over long distances and are typically used for radio communication. These far-field wireless power techniques have not found wide spread use, since they are limited to delivering only a few milliwatts of power due to health and safety concerns. In contrast, non-radiative transfer systems such as inductive charging cradles and resonant charging pads can safely deliver 10s-100s of watts of power by loosely decoupling the magnetic fields–which are used to transfer power–from the potentially harmful electric fields. However, near-field coupling is a highly localized phenomenon and transfer efficiency drops off rapidly as the source and receiver are separated by more than a coil diameter. Likewise, it is not possible to strongly couple coils of drastically different sizes.

> Drawing upon recent work using far-field standing electromagnetic waves to generate uniform field patterns in a metallic chamber, we introduce quasistatic cavity resonance (QSCR); which can be used to create near-field standing waves that fill the interior of the resonant structure with uniform magnetic fields, allowing for strong coupling to small receivers contained within.

> power is not transmitted via electromagnetic radiation, but via oscillating magnetic fields

I don't think you can have one without the other, can you?

That's probably true. I admittedly am pretty bad with basic EM physics. Would love someone to explain this to me like I'm five.

A varying magnetic field produces an electrical field.

A varying electrical field produces a magnetic field.

Therefore if you produce either a varying electrical field or a varying magnetic field, you will generate the other, which will then propagate out; this is EM radiation.

I only did a quick read of the paper, but this looks like they are generating a standing wave (think like a guitar string, where the amplitude of motion is fixed at any point along the string) with a wavelength much larger than the size of the room; this lets them capture the electrical field in capacitors in the center of the room while still having a moving magnetic field throughout the room, thus effectively separating the two, which allows for a moving magnetic field (which the receivers can convert to an electric field) without having a high-magnitude electrical field in the free space of the room.

Seems to me more like they used the big copper tube in the center of the room and the aluminum walls of the room itself to guide the current flow. It's as if the copper tube and the room are a big conductor through which electricity flows (actually oscillates) - this changing electric field in the conductor creates a magnetic field in the space around it, i.e., the room. So the electric field is contained and guided through the conducting walls of the room, while the magnetic field permeates the space in the room itself.

Yes, the walls and ceiling conducting are important to this (otherwise there would be a large induced electric field outside the room).

There are capacitive elements in the pole which is important for allowing it to be resonant at a wavelength much larger than the room (the wavelength used is over 200 meters).

The very long wavelength means that, in theory, one cuold design a room with less of the perimeter conductive (a wire mesh would certainly work; they suggest that doors and windows, or even conductive panels that are connected could work).

There are electric and magnetic fields. While you can't have one without the other, one always dominates when those fields propagate. Also, the effects vary depending on distance from the generating source.[1]

This article apparently says that they have made a practical way to transfer magnetic energy to inductors (coils) well beyond the near field range.[2] They do this by creating quasistatic magnetic fields within a specific space. The coils within that space react to the magnetic fields the way coils in a circuit would.

[1] https://en.wikipedia.org/wiki/Near_and_far_field [2] https://en.wikipedia.org/wiki/Inductor

I think you can. Using electricity, you can induce a magnetic field (Electromagnets) and vise versa, using magnets you can create electricity.

Like the other commentor though, I'm not very strong with my physics knowledge, so I can't answer for certain, but I'll be reading about it at the gym tonight I think.

Can you read while running on treadmill or something? That's an extraordinary multitask if you can!

No, I use the recumbent bike machine.

So basically this is the prequel to Children of Men?

Avg whole body SAR limit they use is .08W/kg. For reference, IEC 60601 governs the SAR limit for MRIs at 2-10W/kg [1]. Also, FCC limits cell phones to 1.6W/kg [2].

[1] https://en.wikipedia.org/wiki/Specific_absorption_rate#MRI_s...

[2] https://en.wikipedia.org/wiki/Specific_absorption_rate#Mobil...

You could also heat humans, much more energy efficient than heating the whole apartment or house.

Though it would heat any water or fat lying around.

It's an old idea: https://www.newscientist.com/article/mg15220615-500-not-cook...

very far infrared == very short microwave

It was also a joke on the first season finale of Silicon Valley.

I remember seeing that and thinking it was actually a solid idea

Hmm, didn't think of it like that. Although most things are indeed quite reflective in the infrared. You should choose a frequency that humans would absorb, but walls and furniture etc would reflect. Or you would need to buy "certified microwave non-absorbing" wares only for such rooms.

The idea was to use very high frequency microwaves, so that only skin surface was heated.

I thought the problem was that this also heats ocular fluid and furniture staples causing potential problems with things like macular degeneration or house fires, respectively.

If the penetration depth is << 1 mm, I can't imagine that ocular heating would be an issue. And I'd expect that power levels would be too low to heat metal. But maybe so.

But any sparks from the metal coulf cause fires. Not much power needed for a spark.

I wonder what your DNA integrity check has to say about that after a couple of years of exposure.

That's probably why it's never gone past "hey, why not..." since the 70s ;)

On the other hand, the outer ~0.03 mm is dead. So ???

Wasn't that a thing on Silicon Valley (the TV show)?

From the article: "While QSCR enabled spaces do require purpose-built structures, as the walls must be conductive, it offers a substantial improvement in the tradeoff between range and the magnitude of power that can be safely delivered."

Think theme park and sound stage, not home and office.

Until we start offering conductive walls as an option in new homes :)

Neat stuff!


- pacemaker warning sign on the edge of the doorframe

- Frequency smack in the middle of the AM band (1.32 MHz)

- demo'ing very low power consumers

- very directionally sensitive (just like any other radio transmitter)

- room is set up like a Faraday cage

If you watch the video, they have a 3-coil power receiver that is not directionally sensitive!

It's 3 coils, each at a different orientation. For something like a smartphone, constrained in thickness, it might be difficult to do more than 2 orientations (L x W).

You should still be able to do all three. Consider a block around the height of the device and around half the width. You could have one set of coils going around the small edge, one around the middle width-wise, and one around the middle length-wise.

Unless I'm mistaken, that should give you 3 orthogonal coils.

You are both correct and mistaken.

That does give you 3 orthogonal coils, technically, but the other 2 orthogonal planes will get very little power.

Except that won't work due to the reduced cross section of two of the coils.

If this became widespread, I could imagine two coils unfolding from the back of the device; maybe as part of a stand.

Core issue with the solution appears to be that the faraday cage used by the room would require any wireless wifi/cell access points be inside the room, otherwise they would be blocked.

Or it's to protect nearby receivers. A Tesla Coil I built in middle school wiped out TV and radio reception for most of the block...

Just like a whisper in your ear would overcome someone shouting a few hundred yards away. It's not the radiant power that's the problem but the proximity to the origin of the transmission. And Tesla coils are particularly dirty in their emissions.

You could potentially switch to a mesh or some kind of resonant structure with large enough gaps that cell signals could get through while blocking the rooms designed resonant frequency.

That and the lack of windows might not be to everyone's taste.


> Finally, the high Q-factor and sub-wavelength operation of the QSCR room permits the inclusion of windows and doors, without significantly altering system performance.

But there may be limits on how much area can be covered by windows...

Thin enough silver layer on glass is transparent.

Just coat them with ITO!

Conductive windows exist.

What could you do with this? A small drone that hangs in the air indefinitely?

A huge swarm of flying drone manipulators in a manufacturing facility.

There are already demos of this:


...but drones use so much power that it's unsustainable beyond demos. Beaming power to them suddenly makes this useful because you can keep them flying indefinitely.

I'll predict that if Disney actually have this working beyond a prototype, we'll see demos of drone swarms building complex objects out of children's bricks within a year, and prototypes of tool-wielding drones actually doing assembly within another.

It also increases the weight of the payload they can move around since they wouldn't have batteries (unless the weight of the receiving coils overwhelms that).

Nice idea.

+ A small sphere drone with a dusting cloth around it to vacuum things.

+ Never charge a remote control anymore.

+ Knowing if someone is present by sensors in the floor, furniture, etc.

+ Automated valves on the heaters.

+ Charging locks (not requiring wires should be better w.r.t. security).

+ Easier light placements. They can be put in any cabinet (where it is normally dark, but you don't want to put wires).

+ Blinds, curtains control

+ Speakers integrated in couches / chairs. Perfectly located. Vibrations etc. is also cool as experience.

+ Clothing. So called smart "shoes", "belts", "umbrellas", etc. only hinges on wireless charging taking off.

Thanks! I left out a few words - I specifically was thinking about what Disney Imagineers might use this for at a Disney park. Are they developing some new tech for use in a new land, perhaps? [1]

[1] https://en.wikipedia.org/wiki/Star_Wars_Land

A whole room of disney animatronics that are completely wireless and battery powered for crazy interactive experiences would be pretty cool

1.9 kilowatts is a sizable amount of power. In comparison a DJI phantom 4 consumes about 175 watts[0]. One could probably keep a decently large drone in the air indefinitely.


Unteathered completely wireless vr springs to mind as a potential advancement in the tech. Depends on how much the headsets draw can be reduced by with foveated rendering.

The article says they can supply 1900 watts. All modern VR headsets consume a fraction of that.

I mean they can, yes, at a maximum. The article says "80% of the room’s 54 m^3 total volume being able to deliver wireless power to a receiver at over of 40% efficiency" which means it's 760 watts of uncertain power (because it can drop off further than that). That's more in the range of high power computers... which I guess wouldn't be a concern.

So this could totally be used for a HMD already. I wonder if it interferes at all with data transmission protocols that are used for wireless on the currently available HMDs...

Just what the world needs, less power efficiency.

Man that is awesome. Wouldn't mind that. I guess although you'd only use a quarter of it, that pole wouldn't look bad in a corner of a room. But they did say they could shrink it. Maybe you'd just tape some wires on your walls that blend in to the color of your walls for that pole part.

I wish them well.

I would imagine that it'd be feasible to build several rooms within the field, i.e. embed aluminum walls in the exterior of your building. Then the pole could be disguised in a pillar in the middle of the buildling somewhere.

Im sure they could mask the pole by using a dividing wall or something.

Couldn't they setup wires in the corners of walls though or just in the walls too. Is that pole's diameter a contributing factor?

I don't mean literally in the wall like behind sheetrock, I mean if you scored a line in the sheetrock and placed the thick 12 gauge wire in there or whatever.

40% efficient, so let's just generate 2.5x as much electricity! (Oh, and remove all of that extra heat with more air conditioning, too!)

Even at 95%, I wouldn't want a 1900W cool near my body for fear of burns. (Compare 100W light bulb)

Finally, what do other electronics do within this field? It's not like there aren't already lots of coils (inductors) that would now have to worry about significant RF back power!

It depends heavily on what you're doing though, as to what the tradeoffs are. A simple example: remote controls use very little energy and the extra energy required at 40% efficiency would probably easily pay for the energy cost of creating / replacing the batteries. Phones, similarly, use very little actual energy.

Pretty neat demo, sitting inside a loaded cavity resonator. Probably not very practical, but still cool.

Tesla would be proud.

From the uBeam discussions, I thought this was impossible

uBeam's USP was charging via ultrasound, rather than wireless charging per-se.

This don't have any harm to health?

this is all cool, but those fields created all around the room could affect (charge?!:) us as well; we are after all electrical beings before we are biological beings; so how is the presence of such fields all around us for extended number of hours daily for a lifetime (or half of lifetime) is affecting human health?

The electric field is contained by using capacitors in the pole in the middle of the room. The magnetic field is what propagates around the room to power devices.

But I suppose you'll say we are magnetic beings or something?


Applications are open for YC Summer 2018

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact