The most important thing here is the 92% efficiency and how "ideal" things had to be in order to achieve that.
Wireless power transfer typically has an efficiency curve that looks like the Washington Monument - or a middle finger, on a chart. Off by a hair and your efficiency is in free fall down the sides. Alignment, distance, frequency, impedance, parasitics...pick a dimension and it's all FU charts.
So I am wondering if they made a system that can actively dial all these in given a real world scenario, or if they hand placed everything to ensure perfect alignment on all fronts, then took an efficiency measurement.
Edited to add: A neat fact though is that when you dial in all those parameters as near to the ideal as you can, you actually can get over 99% efficiency. These are what transformers are. The electrical things, not the AI models.
This can be managed like an induction stove. Many small coils instead of one big one. Small coils outside the optimal position just turn off. Much like an induction stove can detect the pot size
Induction stove coils just need to induce currents above them in what is essentially a giant shorted out single turn secondary. Which is basically useless, hence why it just makes heat. Granted, it is useful for cooking!
Wireless power transfer on the other hand needs carefully aligned coils in order to avoid doing exactly what an induction cooker does - turn usable electricity into heat.
There is no electrical difference between an iron pot that heats up from an inductive load, and circuit that takes that inductive load and directs it to chemical energy storage. The iron pot just doesn't conduct it anywhere useful
Right, and if you bend and mold the iron pot into the shape necessary to conduct it somewhere useful (like chemical energy storage) rather than just turn to heat, you will find than an array of micro coils will hardly work at all.
There are grid-array tiny coil wireless energy devices, but the grid is used to find a "best fit" coil depending on the alignment of the device to be charged, and then uses just that coil.
Again, a cast iron pot acts as a single huge shorted secondary turn. And a shit one at that (hence why it gets so hot), otherwise the primary (the cook top) would burn up too. You want the opposite of this for electrical energy transfer (low loss, low heat)!
The transfer was over a distance of 5 inches.
I suppose it's marginally more convenient to simply park your car over a coil to charge it than to get out and plug it in, but is it really worth it?
I'm surprised there isn't some MagSafe style connector for EVs. Seems like it would combine the benefits of direct connection with the convenience of hands free operation. Part of the reason might be that I think manufacturers like to keep the charger away from the front of the vehicle to keep it from getting damaged in fender-benders.
I'm a volunteer firefighter. It's not literally magsafe, but most of our apparatus has a special female socket for charging the auxiliary batteries that automatically ejects the male plug when the ignition turns on.
I'm not gonna lie, that's some brilliant design. Y'all don't have time to check all the things when you have to get to an emergency site, but I imagine you also need your lights and pumps and such to work.
Very cool. I'd never heard of that. Definitely removes half the hassle. It seems like EVs could implement a plug "pusher outer" that could do the same.
Not to mention that these kinds of systems tend to be very particular about alignment. The amount of effort saved in not plugging in a cable would be very easy to use up if the vehicle has to be precisely parked for it to work.
Also let's keep in mind the claimed efficiency of 95%. That's very good for a wireless power transfer system but if we're talking about 270kW that's still 13.5kW which is getting converted in to heat in the system. That's more power being wasted than a standard home L2 EVSE on a 50A can even deliver, just to avoid plugging in a cable.
I could certainly see specific applications where it'd make sense, like city buses where the pads could be installed at bus stops allowing buses to run their loops effectively infinitely. I'm sure there are also plenty of industrial use cases where a vehicle is constantly moving around a job site but stopping regularly at specific locations. I don't see any reason to desire this as a consumer though.
Here is one reason I can see this being a good thing at least. It would let something like an office park setup charging spots such that there's nothing that the user touches to do the charging. Meaning that you're not going to have people drive off while still plugged in, or not plugging in properly, or having otherwise damaged plugs, etc. So it might make sense for a larger installation to be more maintenance friendly.
> Meaning that you're not going to have people drive off while still plugged in, or not plugging in properly
Are there any electric cars which let you drive while plugged in? I own an EV and have driven a couple others. None of them allow you to leave park while plugged in. I also can't imagine how you'd manage to plug in incorrectly. Is this something people routinely do?
> Meaning that you're not going to have people drive off while still plugged in
This is not possible.
> or not plugging in properly
This is not really a problem.
> or having otherwise damaged plugs, etc.
This is a real issue, businesses offering L2 charging as a convenience have little incentive to maintain their equipment and users often don't put the cables back in the dock when done, inviting damage. Fortunately there's already a solution widely used in other parts of the world, which is for the EVSE (the box most people call the charger that sits in between the mains and the EV and negotiates operating parameters with the actual charger in the car) to simply have a socket on it in to which a user plugs their own cable. This means there's no captive cable to be damaged, each user comes with their own that's in whatever condition they've gotten it in to, and it also makes the installs both smaller and cheaper so more spaces can be given access to power for the same cost.
SAE J3400, the standard that officially adopted the "NACS" connector, brings this to the US. It introduces the "Universal AC Socket-Outlet" where a compatible EVSE will have a J3068 connector, which is the three-phase AC charging standard and conveniently uses the same "Type 2" connector as most European EVs already have. The driver will then be expected to bring their own cable that goes from the J3068 outlet to whatever input their vehicle has, be it classic J1772, new J3400, or even a medium/heavy vehicle that natively accepts J3068 and can make use of three-phase power if available.
This means it's compatible with basically every mass production EV in North America other than the Nissan Leaf. A few of the older '90s EVs that used J1773 MagneCharge won't work either, but the Ranger EV and others that used the Avcon version of J1772 could technically be supported if someone made an appropriate cable.
- Reduce maintenance cost and increasing uptime of car charging parks by removing cabling and plugs operated by the user.
- Simplify integration of autonomous vehicles of different sizes - normal cars, smaller delivery vehicles etc, that could integrate with the same charging areas.
Once you're running a fleet of autonomous vehicles, whatever century that finally becomes a reality, then it makes sense to optimize out of the operation a few people who make at or near minimum wage.
Typically I only plug in when I'm below 60%, if there was a coil, I'd be leaving the house every morning with a fully charged battery. There's probably also benefits for public spaces, and not having to worry about the plug type.
It'll sadly always be at the expense of energy efficiency but I suspect that if it can be brought to 98-99% then it'd be good enough to be a wash from the environmental impact of maintaining and fixing cables from users damaging them when driving off while still plugged in (like they do at gas stations with ICE cars even today). Though I've not looked at any data/info on how often this happens so I can't say anything for certain.
I can imagine it, but all I'm imagining is car companies doing a poor job on the protocol security and I get a $10k bill for power at the end of the month because someone spoofed my car token.
Only if you can imagine all cars with <5 inch ride height and magical coil positioners carefully aligning coils without moving your car or the road beneath.
We don't need magical coil or car positioners, cars do a great job of moving themselves already. What we need is targeting. Similar to how automatic car washes work. Just a "move forward 6 inches .. 2 inches". Non-trivial but not an unacceptably high bar.
Then you could have a dead simple dropping mechanism to lower a charger down to charge - only within <5 inches of the roadway. When you drive off it can just retract like airplane wheels.
I dove deep into this topic after reading your post and discovered something fascinating: electricity doesn't actually flow through wires in the conventional way I was taught. It's more about the electromagnetic field around the wire. This video explains it really well: https://www.youtube.com/watch?v=bHIhgxav9LY
While true, don't get hung up on it, it's kinda like "I was always taught that computers run python, or javascript, or rust, but today I learned that it's actually all just 1's and 0's"
That is actually a super useful and important insight, because the bridge to get there means unpacking interpretation, compilation, and things like compilation targets and CPU architectures. Not saying you're wrong!
Sure, the analogy I am making is imperfect. In reality, you can be a career electrical engineer and never once come across or need to know the true workings of electricity. The models we are all familiar with cover 99.9999% of use cases.
This spurred on some other videos from another youtuber that goes by AlphaPhoenix, he went way into all the details on it including measuring the complete flow of the wave of the electrical power and how it all works.
It goes into a lot of depth of how the wave propagates, including measuring it at effectively "a billion fps" (he goes over how he does this) along the wire and how impedance matching works, etc. It's about 2.5 hours of video IIRC but it's probably the most detailed and easy to understand demonstration and explanation of all of it that I've ever seen.
I wrote off Veritasium when they ran an advertorial about how incredibly safe self-driving cars are, sponsored by Waymo, mostly just citing Waymos own studies.
I had a similar reaction, though just becoming more skeptical of his videos. It did introduce me into a number of others that presented the counter arguments and more complete demonstrations and explanations of the whole thing that fixed a few of my own mis-understandings about it. I posted a sibling comment to yours about AlphaPhoenix who ended up doing the coolest and most complete demo and measurement of what's going on that I've ever seen.
I think it’s useful to point out that in many areas we use a simplified model of the world, because they are much easier to use and the predictions are accurate enough for the intended purposes. But they are not how things actually work, and people forget that.
Yeah, that whole electricity series really put a massive dent in how I viewed all his videos, knocking down veritasium from a 1st rate science channel to a 2nd or 3rd one.
The entire series was predicated on a deeply technical "gotcha" that just caused a bunch of eye rolls from informed viewers.
"wireless charging" as a technology has been pretty baffling to me -- to make it even remotely efficient almost every practical consumer implementation uses some form of mechanical alignment and contact system. As long as you are resigned to do that, why not just use normal mechanical contacts instead?
You might be willing to sacrifice efficiency for the common interface and component degradation over time from repeated physical connection mechanics.
For phones, and other small devices, it's a reasonable solution due to the current and energy transferred involved. For vehicles though, it does not make sense (imho) considering the currents involved, especially as we're approaching 4C-6C with new battery chemistries and management systems.
And the number of phone cables I've replaced, especially beside my bed is already out of control. Switching to cables that plug in at 90 degrees has helped a lot, but it's something that can still wear out quickly.
> As long as you are resigned to do that, why not just use normal mechanical contacts instead?
Maybe a bit far out there, but wouldn't you be able to make both parts (charger and chargee) water-proof easier if power can be transmitted wirelessly?
Because, at least when there are magnets involved for alignment, it's easier. I've never ripped a charger out of the wall by picking up my phone in the middle of the night and forgetting I had a short cable attached to my phone.
Kind of like saying you don't understand why we ever had early keyfobs when you still had to carry a key to start the car. It's not about it being superior in every aspect, it's about ease of use.
Several decades of published research that hint at correlations between EM exposure and health issues coupled with military grade cynicism about the likelihood that industry players had a hand in late breaking claims that said research was inconclusive.
Talking about phone charging in this context is like discussing airsoft pellets in a conversation about war munitions. It's patently absurd. Please do not perpetuate the idea that a low current air core transformer charging a phone is in any way invisibly harmful. It's not.
If this claim was trivially provable it would have been conclusively proven by now and yet there persists hints that chronic EM exposure, even at relatively low power, may cause problems. I vividly recall a cluster of research that blipped into the media landscape for a couple hours a few decades ago that suggested long-term exposure to cellphone broadcast antennas might be problematic and cell chargers necessarily push a hell of a lot more power. To be absolutely clear I'm not claiming to have any answers here, my areas of expertise don't cover this. I would expect any research that comprehensively dispels concerns about EM exposure to make a shitload of noise in the media landscape and to the best of my knowledge that hasn't happened.
You are exposed to greater than 1000W/m^2 of broadband EM every time you go outside; the sun shines WAY outside the visible spectrum, you know. Have you ever ridden in an electric or hybrid electric vehicle? Have you been on a train or ship? Have you walked near a power transformer or driven past a substation? These all produce field effect many many orders of magnitude stronger than we are discussing here.
You are pulling out every logical fallacy in the book in your approach to this question. EM exposure is rigorously studied and safety standards have existed almost as long as we have known about electromagnetism! About the only time that a normal person even comes close to needing to worry about them is when getting an MRI or climbing a radio tower.
Johnny Mnemonic was a work of fiction, not a documentary.
Careful there. In your haste to make your point you're dragging in proven causes of various health issues. The fields coming off of high voltage transmission equipment have been comprehensively proven to cause negative health effects. As to safety standards, MSDS standards have been around since 1983 and somehow they didn't prevent literally the entire surface of the planet from being contaminated with PFAS. You're not going to make a lot of headway here if your chosen tack is to advance coy notions that industry safety standards are rigorous.
To be absolutely clear, I'm not convinced low power EM causes problems. I'm also not convinced it doesn't. What I am absolutely certain of is that given the global scope of the industries that are implicated here, there isn't a legislative body on the planet that would so much as inconvenience them if there was a problem and it was know.
Somewhat related: How unhealthy is it to wirelessly charge your phone (in flight mode) at night on your nightstand not far from your head? Have there been any studies?
There hasn't been a lot of longterm research. Effects will also vary by frequency, intensity, etc. Some frequencies can cause heating, which is a problem for some areas like the eyes. Some frequencies increase glucose metabolism for the cells affected. No idea what that does longterm, but theoretically, that could create an environment for cancer growth.
The health risks of wireless phone charging are massively more psychological than physical. As they will very quickly abut claims of electromagnetic hypersensitivity, most of the legitimate research in this area focuses on the mental health of those who hold such beliefs against the overwhelming lack of any measurable physical phenomenon.
The concerns about the magnetic field may be bunk, but I had to give up using a wireless charger next to my bed because I could hear the coils making a ticking sound. Qi interleaves data with power so I assume the sound is the coils reacting mechanically whenever they interrupt the power to send a data packet instead.
I found that some USB wall chargers emit a high pitched whine when used to power a Qi charger as well, even if they are dead silent when charging a device directly.
Being able to hear the coil whine and finding it annoying is a fine reason to not like wireless charging. But I am pretty sure this is not what most people are going on about.
Is the ticking a function of a mechanical relay in a cheaper device, or the coil not being sufficiently glued down and moving as the magnetic field is turned off and on?
The sound comes from the windings within the copper coils vibrating due to Hall effect in the alternating magnetic field. Yes, the physical construction of the coil matters as to how much it would whine; tighter and more regular windings and appropriate potting/adhesives counteract the physical effects. The easiest mitigation, however, is to just use a switching frequency much higher than humans can hear. AFAIK the Qi standard supports the charger using anything from 80-300kHz and can vary depending on what the charger decides is optimum. Any audible coil whine from Qi would be attributable to harmonic mixing within the physical system of the vibrating coil. Also note that the coil whine could come from either the Tx or Rx coil, so you might easily eliminate it with a different charger that settles on a different frequency or has a better manufactured coil. But if the problem coil is the one in your phone, you might be hosed!
The latter I think, I found the volume varied depending on how the coils were aligned. I'll give it another shot once Qi2 brings Magsafe-like perfect alignment to non-Apple devices.
Claims of electromagnetic sensitivity have been debunked. They show symptoms to fake EM devices, and don’t show symptoms to real but hidden EM devices, so it’s all in their head. It’s the nocebo effect.
Wireless power transfer typically has an efficiency curve that looks like the Washington Monument - or a middle finger, on a chart. Off by a hair and your efficiency is in free fall down the sides. Alignment, distance, frequency, impedance, parasitics...pick a dimension and it's all FU charts.
So I am wondering if they made a system that can actively dial all these in given a real world scenario, or if they hand placed everything to ensure perfect alignment on all fronts, then took an efficiency measurement.
Edited to add: A neat fact though is that when you dial in all those parameters as near to the ideal as you can, you actually can get over 99% efficiency. These are what transformers are. The electrical things, not the AI models.