http://opalkelly.com/

Author Topic: The uBeam FAQ  (Read 168023 times)

0 Members and 2 Guests are viewing this topic.

Offline EEVblog

  • Administrator
  • *****
  • Posts: 23752
  • Country: au
    • EEVblog
Re: The uBeam FAQ
« Reply #775 on: June 01, 2017, 03:51:10 PM »
My take on this
http://liesandstartuppr.blogspot.com/2017/05/whats-in-picture.html

Off the shelf transducers?
Didn't they have the world's best ultrasonic transducer experts working for them, and a world class in-house transducer manufacturing facility?
 

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #776 on: June 01, 2017, 04:02:34 PM »
My take on this
http://liesandstartuppr.blogspot.com/2017/05/whats-in-picture.html

Off the shelf transducers?
Didn't they have the world's best ultrasonic transducer experts working for them, and a world class in-house transducer manufacturing facility?

I couldn't possibly comment on any of that...
 

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #777 on: June 02, 2017, 01:22:46 AM »
 
The following users thanked this post: EEVblog, Kean

Offline StillTrying

  • Frequent Contributor
  • **
  • Posts: 732
  • Country: gb
  • 100% Brand New and High Quality, in theory.
Re: The uBeam FAQ
« Reply #778 on: June 02, 2017, 10:38:06 AM »
The 2 videos were interesting sort of, at least they've got somewhere after $30m and 6 years, - about $0.1m and 6 months worth to be exact. :horse:





I wonder if you could estimate the US frequency from the interference pattens seen on the LED viewer.

Perry flipped the switch on a large white box, about the size of a ceiling tile. A quiet hum filled the conference

That'll be the cooling fans inside the 2kW transmitter box.

The technology is at least a year away from commercialization

You could say that again.

One challenge could be the perception that always-on ultrasound could be unhealthy to humans.

No worries, they've now added 'optical tracking lasers' to it.  :palm:
 :horse:

« Last Edit: June 02, 2017, 11:45:35 AM by StillTrying »
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 23752
  • Country: au
    • EEVblog
Re: The uBeam FAQ
« Reply #779 on: June 02, 2017, 11:25:49 AM »
Life comes at you fast!
http://liesandstartuppr.blogspot.com/2017/06/someone-was-paying-attention.html

Quote
Technically, it's still hard to say exactly what's being done, and there's not much to add beyond my earlier articles. The video does make it look like off-the-shelf Murata devices are being used and focused into a tight beam straightforward. Efficiency and safety questions are dodged.

No beamforming?
In the video she specifically says there is beamforming and device tracking.

Not that it makes a difference of course, the whole idea is 100% guaranteed doomed to failure just based on the efficiency and basic air saturation physics.
How any investor cannot understand this basic fact is amazing.
 

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #780 on: June 02, 2017, 04:10:31 PM »
Life comes at you fast!
http://liesandstartuppr.blogspot.com/2017/06/someone-was-paying-attention.html

Quote
Technically, it's still hard to say exactly what's being done, and there's not much to add beyond my earlier articles. The video does make it look like off-the-shelf Murata devices are being used and focused into a tight beam straightforward. Efficiency and safety questions are dodged.

No beamforming?
In the video she specifically says there is beamforming and device tracking.

Not that it makes a difference of course, the whole idea is 100% guaranteed doomed to failure just based on the efficiency and basic air saturation physics.
How any investor cannot understand this basic fact is amazing.

That's not what I meant to imply, poor wording on my part if that's what you've taken from it. I mean there is beamforming, but simplified to center line only. (I've updated the post wording to make it clearer)

Imagine a regular phased array with N elements on each side, so you need to address N^2 elements, which if you have a device of N = 30 to 45 is a lot of separate drivers to allow you to steer the beam anywhere. Now imagine instead that you have a collection of concentric rings each 1 element in width, all of the elements in that ring tied together so they are driven by the same signal, but each ring can be driven differently. In that case, instead of N^2 drive elements you have (roughly) N/2, so you end up with a drive electronics reduction at a factor of 2N which is pretty substantial in a large array. The downside to that is that your control over the focus is only down the center line of the system and lateral steering is not possible - however you also ensure that you have really strong focus along that line. Basically, if you can live with the limitations, it's a lot easier and cheaper (electronics wise) to build. It's common in therapeutic ultrasound where you can mechanically scan the transducer to move the focus laterally. e.g.

https://www.researchgate.net/publication/253749971_Design_and_Fabrication_of_a_Wide-Aperture_HIFU_Annular_Array_Transducer_for_the_Treatment_of_Deep-Seated_Tumors


 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 3639
  • Country: gb
Re: The uBeam FAQ
« Reply #781 on: June 02, 2017, 05:18:44 PM »
Some points on the videos...

o The LED panel demos were quite interesting.

o We have no idea what functionalities the black bricks on the back of the phones offer: obviously energy collection is one (you'd hope), but is there some energy storage too, and something to manage device tracking?

o There looks to be a camera on top of the square array with an NVIDIA logo below it. What does that do? Is it just for R&D, or is it part of the device tracking functionality for multi beam forming?

o That's a very big box of electronics in the beam forming video.

o How efficient is it?

o How much power is needed to get the green power indicator to show "charging"? And was it really charging, or just dribbling in a bit of power?

o The key question: how long does it actually take to charge the phone?
 

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #782 on: June 02, 2017, 06:41:30 PM »
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 23752
  • Country: au
    • EEVblog
Re: The uBeam FAQ
« Reply #783 on: June 02, 2017, 07:22:28 PM »
Some points on the videos...
o How efficient is it?
o How much power is needed to get the green power indicator to show "charging"? And was it really charging, or just dribbling in a bit of power?
o The key question: how long does it actually take to charge the phone?

Nope, the key question is always the efficiency.
Basic air saturation physics limits this to sub 1%, or maybe a few % if you are talking ideal conditions, and let's get real, it will never be ideal conditions in practice.
No charging technology will ever be the least bit practical in the market at this sort of efficiency. Nor should it be, on a mass scale it would be awful for the environment and would likely end up being banned.
 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 3639
  • Country: gb
Re: The uBeam FAQ
« Reply #784 on: June 03, 2017, 06:50:41 PM »
A few further observations

http://liesandstartuppr.blogspot.com/2017/06/what-does-it-take-to-switch-phone.html

Interesting. I've just tried a selection of Android phones. The on-screen charging indicator seems to come on (and stay on, i.e., not just an initial attempt, sanity check, go away again) at wildly different USB charging currents. One "charges" at 2mA!
 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 3639
  • Country: gb
Re: The uBeam FAQ
« Reply #785 on: June 04, 2017, 02:26:21 AM »
I am also wondering if those LED panels showing the beam patterns are active or passive considering the reasonably high ambient light.
 

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #786 on: June 04, 2017, 03:48:47 AM »
A few further observations

http://liesandstartuppr.blogspot.com/2017/06/what-does-it-take-to-switch-phone.html

Interesting. I've just tried a selection of Android phones. The on-screen charging indicator seems to come on (and stay on, i.e., not just an initial attempt, sanity check, go away again) at wildly different USB charging currents. One "charges" at 2mA!

It's definitely device dependent, though 2mA is insanely low - 10mW?!

iPhones do seem to need more, so it's worth keeping an eye in the demos of when iPhones and used and when Android is used. It was an Android phone that was bought at the store and they do seem to use them more.
 

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #787 on: June 04, 2017, 03:52:44 AM »
I am also wondering if those LED panels showing the beam patterns are active or passive considering the reasonably high ambient light.

That's a very, very, very good observation and something I've been wondering myself.

The LED panel is cool, but also shows the size of the beam (remarkably large given the ~8mm wavelength) and that grating lobes do exist.

Interesting fact: Did you know that if you measure a grating lobe angle from the main beam, and know the transmitter element pitch, you can calculate the frequency of the wave?
« Last Edit: June 04, 2017, 04:11:56 AM by PaulReynolds »
 

Offline mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 9683
  • Country: gb
    • Mike's Electric Stuff
Re: The uBeam FAQ
« Reply #788 on: June 04, 2017, 04:26:33 AM »
I am also wondering if those LED panels showing the beam patterns are active or passive considering the reasonably high ambient light.

That's a very, very, very good observation and something I've been wondering myself.

The LED panel is cool, but also shows the size of the beam (remarkably large given the ~8mm wavelength) and that grating lobes do exist.

Interesting fact: Did you know that if you measure a grating lobe angle from the main beam, and know the transmitter element pitch, you can calculate the frequency of the wave?
I think it's quite plausible  that the panel is passive - could  be as simple as a bunch of transducers with a LED across each. If they couldn't do this without active help then they stand zero chance of charging a phone.
They have screwed up by using red rather than white LEDs though, as the latter would have given much more light for the power
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 3639
  • Country: gb
Re: The uBeam FAQ
« Reply #789 on: June 04, 2017, 08:16:11 AM »
I am also wondering if those LED panels showing the beam patterns are active or passive considering the reasonably high ambient light.

That's a very, very, very good observation and something I've been wondering myself.

The LED panel is cool, but also shows the size of the beam (remarkably large given the ~8mm wavelength) and that grating lobes do exist.

Interesting fact: Did you know that if you measure a grating lobe angle from the main beam, and know the transmitter element pitch, you can calculate the frequency of the wave?
I think it's quite plausible  that the panel is passive - could  be as simple as a bunch of transducers with a LED across each. If they couldn't do this without active help then they stand zero chance of charging a phone.
They have screwed up by using red rather than white LEDs though, as the latter would have given much more light for the power

I tried it a few hours ago, admittedly I didn't spend long, but I could not get an LED to illuminate visually within mm of a transducer transmitter/receiver setup, and with amplitude maximised on a scope. I was driving the tx with 40kHz at 20V p-p square wave from an AWG.

Transmitter is a Murata MA40S4S that I had in stock, I didn't have the receiver so used another of the same device again as an rx. Not sure how much that matters (I am a total noob on acoustics).

I'm and RF guy, but the wavelength at 40kHz is very short acoutically, even though it's VLF to an RF boy, and so phase changes and nulls/maxima are easily shown on a scope, quite an eye opener.
« Last Edit: June 04, 2017, 08:28:39 AM by Howardlong »
 

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #790 on: June 04, 2017, 09:49:23 AM »
I am also wondering if those LED panels showing the beam patterns are active or passive considering the reasonably high ambient light.

That's a very, very, very good observation and something I've been wondering myself.

The LED panel is cool, but also shows the size of the beam (remarkably large given the ~8mm wavelength) and that grating lobes do exist.

Interesting fact: Did you know that if you measure a grating lobe angle from the main beam, and know the transmitter element pitch, you can calculate the frequency of the wave?
I think it's quite plausible  that the panel is passive - could  be as simple as a bunch of transducers with a LED across each. If they couldn't do this without active help then they stand zero chance of charging a phone.
They have screwed up by using red rather than white LEDs though, as the latter would have given much more light for the power

I tried it a few hours ago, admittedly I didn't spend long, but I could not get an LED to illuminate visually within mm of a transducer transmitter/receiver setup, and with amplitude maximised on a scope. I was driving the tx with 40kHz at 20V p-p square wave from an AWG.

Transmitter is a Murata MA40S4S that I had in stock, I didn't have the receiver so used another of the same device again as an rx. Not sure how much that matters (I am a total noob on acoustics).

I'm and RF guy, but the wavelength at 40kHz is very short acoutically, even though it's VLF to an RF boy, and so phase changes and nulls/maxima are easily shown on a scope, quite an eye opener.

From the Murata data sheet (link below) at 20Vp-p you are getting 120dB so around 20 Pa RMS, that's less than 1 Watt/m^2, and it's 1cm in diameter so maybe 50 to 80uW acoustic out if you are lucky. Not enough to power an LED at all. The transmitter and receivers are slightly different IIRC but not so much you shouldn't still see a response. I'd just look at the result on the oscilloscope. Drive them about 5x harder (100 Vpp, 25x power), then put 10 of them together, and you might (accounting for less than perfect reception efficiency) light an LED.

Acoustic wavelength at 40kHz is around 8.5mm. Yes, you'll get minima/maxima in the near field.


http://www.murata.com/en-sg/api/pdfdownloadapi?cate=&partno=MA40S4S
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 6167
Re: The uBeam FAQ
« Reply #791 on: June 04, 2017, 08:03:30 PM »
I think it's quite plausible  that the panel is passive - could  be as simple as a bunch of transducers with a LED across each. If they couldn't do this without active help then they stand zero chance of charging a phone.

We know they have zero chance of charging a phone.

The question is how much it takes to light up the "charging" indicator on an Android phone. I think it's not much because they sell solar panel phone phone chargers that only put out fractions of a Watt.



Is iPhone different? In the video she seemed very insistent that they were going to buy an Android phone.  :-//

The LED panels are very pretty but are almost certainly actively powered.

Ultrasonic transducers are quite mature technology, they aren't going to reduce the size of that brick and nobody wants that even if it works and you don't have to hold it in the air by the edges for it to charge.

Conclusions: She's turned from queen dreamer to queen con artist. The lies in that video are very deliberate.

« Last Edit: June 04, 2017, 08:07:58 PM by Fungus »
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 23752
  • Country: au
    • EEVblog
Re: The uBeam FAQ
« Reply #792 on: June 04, 2017, 08:50:11 PM »
I am also wondering if those LED panels showing the beam patterns are active or passive considering the reasonably high ambient light.

That's a very, very, very good observation and something I've been wondering myself.
The LED panel is cool, but also shows the size of the beam (remarkably large given the ~8mm wavelength) and that grating lobes do exist.
Interesting fact: Did you know that if you measure a grating lobe angle from the main beam, and know the transmitter element pitch, you can calculate the frequency of the wave?

Simplistically, assuming say a 1kW input signal into the TX array, and assuming it can beamform that to even 0.1% efficiency into an area the size of the LED panel, that's 1W into the entire handheld device. More than enough to power LED's passively.
Even assuming one large beam the same width as the TX panel, the RX LED panel is maybe 1/4 the size?, so even at 0.25W that would be still be enough to do the LED's.
So I'd say it's possible.
It would also be possible to combine all the RX transducers to generate an internal power rail, and then also measure the power level on each RX transducer to map to a LED array actively, rather than just passively power each LED.

But like you said, if they were getting 5W or even 2.5W into the "brick" phone adapter, wouldn't they be shouting that from the rooftops?

After all:
« Last Edit: June 04, 2017, 09:06:13 PM by EEVblog »
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 23752
  • Country: au
    • EEVblog
Re: The uBeam FAQ
« Reply #793 on: June 04, 2017, 09:10:55 PM »
Remember this?
Whatever happened to tripling  the staff?
Isn't the new HQ practically empty?

 

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 3639
  • Country: gb
Re: The uBeam FAQ
« Reply #794 on: June 05, 2017, 01:57:06 AM »
Well I had a bit more than five minutes today. I did manage to get an LED to light but I needed to rectify the rx with Schottky diodes: using an LED to self-rectify didn't work, not sure why.

There was little if any benefit using a full wave rectifier compare to using a clamp diode plus single rectifier diode.

I could get it to light visibly with about 1mA at a distance of 2cm, so as you say, it's still near field. As you need to quadruple the radiated power to double the distance, I'd need 16 to manage 8cm(!) however I could also add gain to the receive element too in a similar fashion.

It looks like there are 900 transducers in the beam forming unit. Say it's 1024 to make the maths easier. That gives a possible gain of 30dB over a single transducer by my calculations and a range of 64cm.

20Vp-p is the maximum the transducers I have are spec'd for. They don't look exactly like the ones in the video because there is a plastic protective mesh in front of mine but the size looks similar. At 20Vp-p I am measuring (real) power into the device of around 100mW. At 2cm distance, I had about 2mW, giving it a 2% efficiency. However, at 2cm this subtends only a small solid angle from an unmodified beam tx, so perhaps only 15-20% of the transmitted power appears at the rx anyway. So beam forming and reasoanably sized apertures on the receiver are essential facets for this to work.

It'd be an interesting experiment to put a larger array together for both tx and rx on a parabolic surface and see what can really be achieved, and at what distance. I only have 30 transducers in stock though!

I was thinking about their brick phone attachment, and assuming it's real, in itself that would have to have a reasonably large aperture, so perhaps 40 sensors. Each would need MPPT and power aggregation which although not rocket science is going to be expensive to fabricate.

That camera thing is an Nvidia Jetson which looks like it's for visual device tracking. They do like making hard work of things! If it needs visual indication of where the target device is, and the sensors are on the rear of the phone, the phone will have to be used face down for a ceiling arrangement, and you won't be able to hold it in a normal fashion to make a call or use the screen. Even wall mounted, assuming nothing's in the way, you'll have to figure out new ways to hold your phone.

In its current form and key use, as a phone charger, this remains practically speaking a non-starter.
 
The following users thanked this post: PaulReynolds

Offline Howardlong

  • Super Contributor
  • ***
  • Posts: 3639
  • Country: gb
Re: The uBeam FAQ
« Reply #795 on: June 05, 2017, 02:11:52 AM »
I managed to "charge" my phone with ultrasonic power transfer...

 
The following users thanked this post: nugglix, PaulReynolds

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #796 on: June 05, 2017, 05:56:25 AM »
I managed to "charge" my phone with ultrasonic power transfer...



Superb stuff Howard, nicely done. I'll likely do a blog post and link to your work here.

Something to note - you can drive the Murata transducers at more than 20Vp-p, I think you can do 60V, maybe even as high as 100V. They in part limit to 20V as that's in the 115 to 120dB range which is the limit in almost every country in the world for ultrasound (and likely the US as well, OSHA have changed the public numbers since 2015 from 145 dB to 115). No idea about longevity at those amplitudes, piezo materials break down at some point, and if >80% of the power is lost as heat, that's not going to be good. Also note that if an ASIC is used, the node chosen will impact maximum supported voltage so can only go so high.
 

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #797 on: June 05, 2017, 05:58:43 AM »
Remember this?
Whatever happened to tripling  the staff?
Isn't the new HQ practically empty?



According to LinkedIn, all the staff who were in that office (3 of them) left at the end of last year. None had even made it a full year at the company. It appeared to be an 8500 sqft office, and when I last drove past it was not occupied.

They may even have left in January this year. Why would they leave when the company is on the verge of this awesome success?
« Last Edit: June 05, 2017, 06:37:23 AM by PaulReynolds »
 

Offline StillTrying

  • Frequent Contributor
  • **
  • Posts: 732
  • Country: gb
  • 100% Brand New and High Quality, in theory.
Re: The uBeam FAQ
« Reply #798 on: June 05, 2017, 10:52:30 AM »
It seems a bit odd that they're still playing with early prototypes, when they've been gearing up for mass production for nearly 2 years.  :horse:

From the brightness of the LEDs on the panel I'd guestimate the recovered power varies between about 0.3W and 0.8W, - nearly enough to charge a phone if you turned it off and held it up in a funny way for 6 hours.  :)

3178 views:
vimeo.com/218093800

Offline PaulReynolds

  • Contributor
  • Posts: 31
  • Country: us
Re: The uBeam FAQ
« Reply #799 on: June 05, 2017, 12:19:32 PM »
It seems a bit odd that they're still playing with early prototypes, when they've been gearing up for mass production for nearly 2 years.  :horse:

From the brightness of the LEDs on the panel I'd guestimate the recovered power varies between about 0.3W and 0.8W, - nearly enough to charge a phone if you turned it off and held it up in a funny way for 6 hours.  :)

3178 views:
vimeo.com/218093800

You don't understand! It's a new paradigm! Stop thinking like an engineer and free yourself from that box you are in! It's not about charging faster than a wire, it's about trickle charge!

It must be nice for everyone in the press to forget what you claimed you had 2 years ago, get to keep working, then deliver a fraction of what was claimed and say "See I was right!"

In the USA Today article the head of their tech advisory board literally admits they had far less than this 18 months ago, same time the company was claiming "faster than a wire" and "ramping to production". And it looks like they are using off the shelf transducers (or simple variation thereof) and not anything groundbreaking.

The video is interesting - in the first few seconds it's "charging" at an angle of incidence far more steep than they claim is possible. There's a noticeable lag between the phone moving and the beam catching up - who gets insonified during that time it's not on the phone?



 
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf