Many readers of this blog will know the name EEVblog, it's a website and online forum for talking about electronics run by Dave Jones, an Australian engineer. It's host to the uBeam FAQ, which puts together a lot of information on uBeam in one place. In addition to the forum, Dave makes videos where he delves into topics in a pretty entertaining manner, and for his 1000th video, he takes on uBeam.
Gotta say, he does a great job of covering the tech in an entertaining manner, not sure I could have done better myself.
Following uBeam's demo, EEV Blog contributor Howard Long made a very interesting video showing how you can turn on a phone charge light with ultrasound. It's about 4 minutes long, with audio commentary, and gives more info in that 4 minutes than in the entirety of uBeam's demo. If this subject interests you at all, I encourage you to watch this.
From his comments (edited for brevity, read the whole thing here):
I could get it to light visibly with about 1mA at a distance of 2cm ... At 2cm distance, I had about 2mW, giving it a 2% efficiency. However, ... 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.
... That camera thing is an Nvidia Jetson which looks like it's for visual device tracking. ... 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.
It seems an engineer reproduced a basic version of the uBeam demo in a day with about $20 in parts.
The phone charge indicator lights at 1mA, which implies 5mW (5 Volts supply) and so would take about 1000 hours (~6 weeks) to charge the phone - if it weren't for the pesky fact that a phone requires around 500mW to operate, on average, so it would make no appreciable charge effect at all.
Now of course there's only a single element here, not a full array which could emit more power, but the key point is that a charging symbol tells you nothing about whether it is practically charging. You need voltage and current to know the actual power, and you need it at both transmitter and receiver to get efficiency (which he's calculating as 2% in this setup, pretty good actually for through air). Those are key numbers you need to have. From Howard's numbers 100 transmitters will get you that 500mW and maintain charge at a constant level, in an ideal world setup - possible but very large and introduce many questions on practicality and cost.
I like the way Howard also brings out a key point in this video - of course you can send power through the air by ultrasound. That's never been doubted or questioned, here or on the EEV Blog. What is questioned is how much power can be received, the efficiency of that, the safety aspect, the cost of transmitter and receiver, and the practicality for the user.
None of those points were addressed by uBeam, other than the emphasis on slow "trickle charging", implying the "faster than a wire" claims of 2015 aren't going to be happening.
Anyway, bravo to Howard Long for showing how to put together a short, clear, technically accurate demo from which you can actually learn something.
An eventful day yesterday on the uBeam front, with Meredith Perry finally giving a demo of uBeam technology and showing it charging a phone at the Upfront Summit - well more precisely showing a big box and then a light on a phone coming on if it was put in front of it. Essentially a slightly more glitzy version of the "All Things D" demo done in 2011, showing what 6 years and $25 million gets you.
From what we see here, in my opinion, is proof that you can take a non-technical audience and baffle them with bullshit - if you want to know that the phone is charging, you need to do more than turn a screen on. Perhaps there is more not seen here, I'm just going on the info that's public, but you need to show voltage, current (at both transmitter and receiver to get efficiency), and the phone sitting in front of that panel for several minutes and see the actual charge level increase over time. But that isn't what they showed - and if it isn't, please enlighten me and tell me what is the difference between what's shown in that video, and what was shown at All Things D 6 years ago.
It seems at least some are not convinced and there are journalists taking a sceptical view, such as Axios (albeit promoted with a tweet that is more sensational than what was shown and the content of the article, and sadly is all that is quoted by most)
This is a science project that is clearly progressing, but not nearly finished yet.
Pretty faint praise after $25 million. There was also this interesting statement:
we're told Perry picked that particular Android for the demo because of its highly-visible charging icon
Why would that be mentioned so specifically by the company, and why does it make me raise an eyebrow?
Now, let's be clear, no-one ever said that transmitting power via ultrasound is impossible, of course it's possible - but is there a way to do so in a safe, efficient, and cost effective manner? That's the challenge, and in any practical sense it had never been shown publicly. In my opinion, it still hasn't. All that has been shown is a screen lighting up.
I'm sure uBeam now have potential funders lining up outside willing to throw money at them, based on this, even though nothing was really shown. And if I'm wrong about that, tell me what was shown that proves it works. What's the charge rate? How long to charge a phone? What is the efficiency? How does this line up with "4 meters, any angle, multiple devices, faster than a wire" touted before? Is it a safe and legal level? (OSHA now seems to have gone back to a 115 dB limit, not the 145 dB from a few years ago, I certainly hope there's no-one in the way of that beam, or there are any grating lobes giving the audience a facefull.)
Now the fact the phone charge indicator comes on proves they are charging at a minimum of 500mW (around 5 volts at 100 mA) needed on the USB port, which is awesome as that's enough to at least trickle charge a phone over about 10 hours. Or does it? Potentially you could access the Qi chipset on the phone to show the charging light when at <500 mW, or other similar bypassing of standard input methods, but in the end there's no way to know without looking at actual charge rate - which isn't shown in any form. If it works so well, I'm surprised those numbers aren't released - "more than 500mW" is a very straightforward statement to make. Or leave the phone in front of the transmitter and see it gaining battery level during the talk. But that would be too easy.
And at what efficiency? At 30% end-to-end it's incredible, at 1% it's very difficult to justify, at <1% it's ridiculous. We don't know those numbers.
How many devices can this charge at a time? What does the system cost? Can it track the phone? What happens at an angle? Was the beam always on, or did it switch on when it saw the phone? What were the safety measures to stop an always on-beam being pointed at someone accidentally? If this is the best case demo today, why were some people saying they had seen a similar working demo years ago? Weren't they moving to production 18 months ago? All questions still unanswered.
I'm really sad, of course, for the senior staff who just left the company over the last couple of months, and what I guess is the closure of the San Jose office (or that's how it appears if you check the LinkedIn profiles). Amazing they would leave just on the verge of a breakthrough like this, but more fool them I guess, what do they know? Passing by on the billions...
Overall, with a skeptical eye, there's nothing new here. IMO, no significant new information, nothing to show commercial success or capability, and no path to a realistic product. But it won't stop investors from piling in without doing significant due diligence (investors, feel free to call me and prove me wrong), and it won't convince anyone with one iota of technical capability that there's more there than they thought a week ago. More of the same, move along.
For those of you with a technical bent, I'm including a more detailed analysis from what I saw in that demo below. Anyone non-technical, you may want to stop now.
Taking a technical look at what's there and bearing in mind this is with a lot of assumptions - the video shows an array that seems to be made up of a (approx) 30 by 30 collection of circular transmitters, and given what I see on stage it's about a 30 by 30 cm panel, so each is a 1 cm diameter cylinder. Very much like the Murata MA40S4S used in car parking sensors and available off the shelf at around $3 each in bulk. Of course they couldn't use them because that would be a $2700 transmitter BOM component right there, but let's use them as a starting point.
Assume 40 kHz, and let's say we can drive much harder because why not, something like 6 times more (120 volts p-p, or approx 16 dB in sound pressure) to be generous so that's 120 + 16 = 136 dB sound pressure level. They are circular, so we lose 2 dB from area, that's 134 dB out, across a 0.09 m2, and at that level that means a peak pressure of 180 Pa and about 37 W/m2 or actual 3.35 W transmitted. Incidentally the capacitance of those devices at 2550 pF means (at P=nCV2f) gives 1.3 kW (900 * 2550e-12 * 120 * 120 * 40e3) so right there is around 0.25% efficient on transmit at best, along with a one bar electric fire (update: this capacitance calculation is not correct, I need to update it). A few million people doing this every day means GW more generation capacity, so I hope I'm wrong or we better start building some power stations. (updated efficiency numbers below - a bit better than here, but still pretty awful).
As a side note, those values of amplitude, if I'm in the right ballpark, may avoid the worst effects of acoustic nonlinearity in the distances shown, but in my opinion (and that of physics), would result in nonlinearity if you tried to increase from there, decreasing efficiency considerably.
Now at 1 to 2 m distance you're probably looking at around 3dB loss in the air (pretty low, yay, but still 50% efficiency), so saying you get all of that power at the phone (about 5 by 10 cm) you'd have an focus gain of around 18 times (25 dB), so now we're at 156 dB (wow, that's loud). Now we convert back to electricity, let's say 30% efficient there (massively higher than the Murata MA40S4S), and around 90% on some awesome conversion electronics, it's about 27% conversion efficiency, and you now get to 450 mW to the battery which is almost enough to charge it. Let's go with that - yay we're charging a phone in about 11 hours. If I'm assuming low numbers, then divide that by about 5 to get a 5% overall rate and 90mW, maybe enough to turn on the charging light (and about 2 days to charge your phone, if you don't move it)
At what efficiency? 0.25% at transmitter (I'm ignoring some losses here, but they're minor in comparison to that capacitive loss), a further 50% in the air, and 27% at the receiver, and you've got 0.034% efficiency. (As noted earlier, not including non-linearity). At 12 c/kWh, that's $2 to charge your phone. Ouch. OK, I'm being mean, let's say it's 10x more efficient, it's 20 cents to charge your phone, only $70 per year done every day, still an ouch. And you can heat your room at the same time with a kW scale transmitter, that costs $7500 because of the high BOM and doesn't make you feel so bad about having spent $1500 on a toaster oven.
As an added note from the original post, I noticed on a Twitter feed that some there indicate that the transmitter seemed to be covered by some form of fabric, which looking again at the video you can see is there. This does not mean that ultrasound can pass through clothing, as was previously claimed, but a thin membrane that is significantly smaller than a wavelength and is of a low enough impedance material will not be 'seen' by the ultrasound, for example a mylar film on the order of 10s microns compared to around 8mm wavelength in air at 40 kHz will likely have a minimal effect. Just as with the membranes or meshes used on car parking sensors like the Murata mentioned above...
I'll add to this as I have time to do so, and check my calcs for any mistakes. Comments welcome on why I'm wrong, and just a disgruntled former employee :)
Edit: Just an update to some of my numbers here. Looking at the Murata data sheet is seems that SPL was measured at 30cm, not at the source, so some modification needed to the calcs. Using Murata's published factors, a further ~10dB needs applied for the diffraction and absorption (BTW that's quite a good document on how those transducers work), so they could be producing as much as 130 dB at source, so I can reduce the applied voltage by a factor of around 3 to around 40 volts, and does reduce the capacitive loss to around 130 W for 3.35 W acoustic transmitted, meaning 2.5% efficiency in that portion of the calculation, so it's overall 0.34% efficient at best, not 0.034%. Yes, that means the sound field could be of greater intensity and higher power, however that would start to push it into the nonlinear regime, and also you'd then be beaming very high sound levels at that cameraman and of course they totally considered safety in this demo...
Interestingly, this means those Murata's can put out over the 115dB level mandated by OSHA, however I'd note that a) the Murata operate at a duty cycle of about 0.4% or less (20 cycle bursts until return signal at up to around 2 m, another good link on car parking sensors), and b) there is a single transmitter, that is as loud as it will get, and decay rapidly after that - unlike a phased array for power which operates at a 100% duty cycle and uses antenna gain to amplify the sound by a factor of several hundred.
Once again there's a ton to write about - Brexit, Theranos, Energous, Erin Griffith's article on Ethics in Silicon Valley, and recent developments with uBeam, but a combination of work plus, hunting for a house, buying a house, getting contractors in, and moving, are eating up all my time. Hopefully next month things will be a little more settled and I'll be back to writing more like a post a week.
In the meantime, I wanted to cover the startup story of the moment, Lily Robotics. Lily is a drone company, promising a simple to use drone (throw it in the air, that's it), that follows you and uses a superb camera to take great videos and stills without a controller - ideal for sports enthusiasts to create videos of themselves doing cool stuff. It looks fantastic, with great demo videos and a strong demand. They raised $1 million in seed funding in mid-2014, and then in mid-2015 started taking pre-orders following some amazing videos and marketing - its pre-order list reached 60,000, at over $500 each, for around $34 million in pre-sales. At the end of 2015 they then raised a further $14 million in VC (no surprises - who wouldn't invest with pre-orders like that!)
Units were supposed to ship to customers in Feb 2016, but that was delayed until summer 2016 - no surprises as hardware is hard, give the newbies a break. Then it was delayed again, this time until December 2016 (time-to-carrot of around 6 months), but once again that date came and went, until suddenly last week they simply closed down with a message to their pre-customers that they were sorry, they couldn't manage to make it, but refunds were on offer. A sad tale, a startup that bit off more than it could chew, and ultimately had to close but sought to return the money to the customers and make them right. Sad until it became public that the same day they shut down, they were sued by the San Francisco DA for misleading business practices and false advertising.
I'll leave the other details to The Register, sUAS News, and the EEV Blog, and hone in on a couple of the most interesting points in this case. Remember that one of the key parts of fundraising is to get VCs to think that there are huge numbers of customers out there for your product, and so once you have 'traction', that they want to invest (de-risked is a term used, others simply wonder why you need a VC once you have customers and profits). If you plan on 'hacking' the system to get the VC money, then you aim to get customers - but what if you have no product to sell? Then go with pre-orders! Show the customer an imaginary future product you plan to make, play up the 'plucky little startup' card, and before you know it you've got $34m in sales and VC's knocking down your door, giving you all the money and time you need to make the product and get it to your customer.
That would be the (mostly) legal way to do it, tell pre-customers it's a planned product, tell them what you are showing them is "hoped for" or "aspirational" and do your best to hit it. Or you could simply show them a faked demo and video and hope you've got time to make it a reality by the date delivery is due - 'fake it til you make it' - and this is what the SFDA is claiming Lily did. In effect, it's a variation on what it appears Theranos and others did, except faking the demos to customers, not to investors (who are still likely defrauded, if this is true).
The customers were led to believe the company had more than it did through their promotional video of the Lily in action, however all was not what it seemed. From the SFDA complaint:
Lily Robotics did not have a single Lily Camera prototype that had all of the features advertised in the Promotional Video. Instead, its co-founders Balaresque and Bradlow, who were present during the filming, brought several prototypes to use during the filming. Some, which looked good on the outside but were not fully functional, were used only for “beauty shots.” Others had some functionality but did not look like the product being advertised. Some were able to film video but even those were merely Lily Camera prototypes with GoPro-branded cameras mounted on them.
This is an important point as it highlights something I've seen happen and I think is more prevalent than most want to believe - showing mockups as working devices, claiming many features and achievements in the product, yet not revealing that not only are they not currently available simultaneously in the same product, but that they may even be mutually exclusive. The analogy would be to claim that your company's new aircraft can fly at 90,000 feet, at Mach 1, with a range of 5000 miles, carrying a 100,000 lb load and leading people to believe it can do all at the same time, when that is impossible. It can be done to investors, though they should have the resources and experience to vet such claims, so let's do it to consumers instead - they're gullible and good natured, let's fleece them!
Of course, I'm being cruel to Lily here, founders never think like that. They're all starry-eyed idealists just looking to follow their dreams and change the world, at worst you can say they are true-believers who wanted to make it all happen, but their reach exceeded their grasp. Let's forgive them, they tried and failed, but at least they tried.
And then you read excerpts of emails from a Lily founder talking about their demo video:
Are you sure that the GoPro lens does not create a unique deformation/pattern on the image? I am worried that a lens geek could study our images up close and detect the unique GoPro lens footprint. But I am just speculating here: I don’t know much about lenses but I think we should be extremely careful if we decide to lie publicly.
The founder was worried that smart people would find out the demo was faked, and explicitly and in writing admits that they know they are lying. It's the equivalent of being found at the murder scene, covered in blood and carrying an axe, with a signed letter about how you have to be careful if you decide to kill someone with an axe. Despite knowing it was lying and fraudulent, they decided to go ahead with it anyway. Why? Because the funding 'game' is structured to incentivize exaggeration, fabrication, and lying, and to punish honesty. Honesty doesn't get you funded, lying does. When there's millions of dollars at stake (amounts that people kill for), why wouldn't someone tell a few lies, especially when if they succeed, no-one will ever know? And that part is critical - they didn't think they'd get caught, and why would they? How many startup founders have you heard of going to jail for this kind of thing?
A further question that springs to mind is why the VC firm that invested after the pre-sales didn't spot this during their due diligence. Surely they learned that the promo video didn't match what was shown? If they didn't, they were either lied to and also defrauded, or it smacks of incompetence if they missed it. If they did find it, then it's even worse, because they're then complicit in the deception. Faced with being labelled incompetent, fraudsters, or themselves defrauded, I wonder how long before the VC in question joins in the complaint against Lily and sues.
Which then brings the next question - who gets paid back first? Normally in liquidation the VCs get pain off first (preferred stock), but if there is debt then that has to be paid first. In suing to get back their $15m the VCs will have to wait behind the customers' $34m of refunds (who themselves are behind a $4m bank loan) - thus surprising the investors that for once, they aren't at the front of the line. This is something I expect we'll see more of later this year, from companies where debt and convertible debt are sitting ahead of the institutional investors. It will be interesting to see how the VC community reacts to these new circumstances, and how they explain it to their LPs.
As for the customers and their refunds? Apparently there is over $25m in the company accounts, with the accounts now frozen other than to pay employees and debts, so once there is at least a chance customers will get some money back. Glad to see consumer protections working, while we still have them that is...
Yesterday afternoon, Elizabeth Holmes, embattled CEO of Theranos, took to the stage at AACC to finally reveal the science and data behind their products and tests. She started with a heartfelt apology for the mistakes the company had made, for the misleading information and marketing both to patients and investors, and proceeded to finally show the actual data and methodology of their tests, warts and all, and then pleaded with the community to help the company move forward and realise the initial promise of Theranos, of multiple, accurate, rapid tests from a single drop of blood. The crowd rose to their feet at the end, and applauded her honesty and humility.
Or at least that's what happened in an alternate reality where shame over fraud and scientific honesty are foremost in the minds of startup CEOs. Instead, what we were treated to was a marketing presentation for a future Theranos product, no apology, no real data, and evasion of any substantive questions. If you have 90 minutes to spare, watch the whole thing, it's a masterclass in abusing the platform given by a professional organisation to do marketing and try to give the company a veneer of credibility to those not versed in the subject.
They didn't just fail to reach the low bar set for them - they didn't even attempt to reach it. To list out what was wrong with this presentation would take a book to cover fully, and there are multiple articles detailed issues with it such as the Wall Street Journal, Wired, and the New York Times. I'll try and cover the main points below - for those of you looking for a history of Theranos, there are some summaries here, here,here, and here.
As someone who has actively organised and managed scientific conferences and talks, I'm stunned by what AACC leadership allowed Theranos to get away with. Theranos committed multiple major faux-pas, and Holmes' celebrity status allowed her to get away with things no other presenter would have been. In a scientific conference the goal is not just to present data, but to withstand the scrutiny of your peers and as such the question component is just as important. The talk was set for 90 minutes, split into 45 minutes presentation and 45 minutes questions. Holmes gave a very polished and well rehearsed talk for an hour - clearly and deliberately done to cut 15 minutes of time from questions. This is an old tactic, commonly used by presenters who fear questions, and one an experienced chair/moderator knows to look for. The session chair was left with the choice of cutting them short and leaving out the component everyone thought they wanted to see, the promised (though never really delivered) final section on actual results comparison, or allowing Holmes to cut into the question time. They went for the latter option, a difficult call to make and one I can understand them doing.
Holmes also committed one of the worst crimes - not presenting what was claimed in the abstract, and instead giving a marketing talk for an upcoming company product. In the Wired article linked above, Stephen Master, professor of clinical and lab pathology at Weill Cornell Medical College. “This sounded like a talk from a manufacturer.” Conferences have commercial presentation sections for that type of talk, and anyone who gives a marketing talk in a scientific slot will never be invited back. AACC leadership should have made comment on this after the talk, stated it was unacceptable, and in being silent they are effectively endorsing this approach.
Her talk was slick and obviously well rehearsed, and covered things in a way that would be ideal for fundraising, but inappropriate for a scientific talk. She used buzzwords that CEOs seemingly have to say, without apparent understanding of their meaning. "Inflection Point" is one of them, which is code to a VC that "we're about to have massive sales, real soon, honest" but everyone else knows is just cover for poor performance to date. For example, Holmes uses it at 42m40s correctly to describe a test function, but tries to pretend at 56m20s that sales of this future miniLab will soon boom. Meredith Perry, Founder and CEO of uBeam that similarly has been questioned about the validity of its technology, said in September 2015 "We’re at a massive inflection point. We are about to head into a completely new phase of growth.” but nearly a year later no sales, no demonstrations, no products. Beware the CEO who promises an inflection point, it's nearly always a false promise.
Instead of answering the many questions scientists had, Theranos presented their future 'miniLab' device, a 'cloud connected' unit that may play well with VCs but fell flat with doctors and scientists who actually use the equipment. Geoffrey Baird, an associate professor in the department of laboratory medicine at the University of Washington, commented in John Carreyrou's Wall Street Journal article:
Every piece of technology they presented has been known for many years, and exists in other platforms largely in the same configuration, or in some cases in much more compact form in competitor’s platforms.
So even the future (non-existent) product they are trying to distract us with does not even move the current state-of-the-art forward in any way. In the initial questions following the talk, all done in comfy chairs and a collegial atmosphere I have never seen at a post-presentation question session, one of the moderators called Holmes on the pedestrian nature of the equipment and the departure from the world-changing claims Theranos had raised $700 million with. Masters noted that the original claim was "Many broad tests, 70 tests from a couple of drops of blood - this falls far short of that... I can buy a point of care instrument today that does a finger stick lipid panel." and received much applause from the audience. Holmes then did what was standard for the rest of the question session, talk a lot, but fail to actually answer the question asked.
Holmes clearly had memorized the line "What we wanted to do here today was..." followed by a few options such as "begin engaging with the community." or "present our future plans to the audience." No-one corrected her that this differed from the abstract of her conference talk, but you could tell people were unhappy with those non-answers.
This continued with all questions designed to elicit information such as "Why present data for potassium with venous not capillary blood?" (a key question about their methods) the reply was "What we wanted to show was..." followed by no answer. When asked about false positives and concordance rate, the reply was that their technique was "Very good" and then no answer, even when asked a second time. It was clear no real answers would be forthcoming.
Whenever pressed to answer questions on previous methodology based on their Edison systems (which were never mentioned once in the talk despite being the main company product to date), Holmes gave the most frustrating answer of all:
In the appropriate forum, we’ll address those. But today we’re hoping to be able to engage on a scientific exchange on this platform.
A canned answer, giving no indication of when that data would be forthcoming or what that appropriate forum would be - ridiculous given that there probably is no better place than the AACC, and that audience, to address them. The real data everyone wanted to see, the topic everyone wanted to cover, was dismissed in those two sentences, and not a peep from AACC leadership or moderators on that.
As Wired magazine points out, Theranos just 'pivoted' - that's when a startup company realises their past approach and technology just can't do what it was claimed, and try to shift to another product or market in an attempt to save the company. It's clear from the reactions that no-one was impressed by the miniLab equipment, and that had this been what was pitched to investors, there would have been no $700 million raise. As one of the moderators, Master, said in Wired:
Does this live up to the hype? The answer to that is no.
Finally, no-one raised the most obvious question which should have been:
"Since you (Holmes) have been banned by the CMS from running a company doing blood tests, and you have made it clear that you will not leave the CEO position, how will the Theranos be able to offer this test and equipment?"
That in itself would have been worth the non-answer answer that would have come.
So in summary, Theranos avoided answering any of the questions scientists wanted answered, have abandoned their old technology, and shifted to a new unproven technology that has no advantages over existing available equipment. They truly live in another reality if they think this will be enough to save the company.
Equally importantly though, AACC allowed its prestigious name to be used by someone being investigated by both the SEC and FDA for fraud, to not present real data but instead a marketing talk and to avoid substantive questions, and to fail to press even on the soft-ball questions it asked. Holmes may not have the capability to show remorse, but the AACC should be ashamed for what it has enabled here.
I could write a post on the deficiencies of TED talks and how they're giving the appearance of detailed scientific education for the masses while simply allowing poorly thought through ideas to be propagated to the public, causing legitimate and well conducted research to be sidelined, but instead I give you John Oliver, and TODD Talks.
The TODD talks starts at 15:48, however just watch the whole thing. As with pretty much every topic he covers, he just nails it.
I've been meaning to post this video for some time. It was sent to me a year or so ago, and I actually couldn't make it all the way through, I think it was some form of "Startup PTSD". Fortunately I got over it pretty quickly, and I just marvel at the insight and genius. Truly awesome work.
