Hacker News new | past | comments | ask | show | jobs | submit login
Light Year One (lightyear.one)
51 points by cvaidya1986 on June 25, 2019 | hide | past | favorite | 64 comments



I want them to succeed, but it will be difficult. Looks like they're following Tesla's model of starting with the top-of-market so they can "fundraise" via early adopter sales. Will they ever be able to hit mass production? Not sure.

In-wheel motors are interesting. That changes the dynamics of the vehicle by adding what we mech-e's call "unsprung mass". Vehicle handling takes a huge hit. There's a reason there are no cars on the market with in-wheel motors at present. Additionally, you're putting expensive components in a place where they can be more easily damaged. There's also the fact that in-wheel motors can be very dangerous if there's a failure of the control system. Since there are no cars on the road like this, I'm wondering if the team truly has the expertise to tackle these issues.

But I'm not trying to be a cynic. I spent 8 years as a mechanical engineer designing hybrid and electric cars; these are simply challenges that need to be engineered around. This is a cool car, and I hope it makes it to market. We need more of this.


To save others time, "unsprung mass" is the mass of the vehicle not supported by suspension.

Lightyear's claim is that putting the motors in the wheels improves control. Could it be that unsprung mass is reduced by negating the axle?


I would bet that there's more unsprung mass in this configuration, but it's hard to say without knowing the details of the motors themselves. If each motor is, say, 80 pounds or heavier, then this configuration would likely increase the unsprung mass.

Note a distinction between "control" and "handling". Four in-wheel motors will have better control but worse handling. Traction control can be applied to each individual motor, improving control in some cases, but on bumpy or rough roads the car will experience worse handling, as the wheels will "hop" off the surface more.

Imagine a car driving over a speed bump a little too fast. In a standard suspension, the wheels are light enough that the suspension pushes them back down to the road surface very quickly at the far end of the bump. If there is more unsprung mass, however, the wheels have more inertia, and so the suspension is not able to push the wheels down to the surface as quickly. Now imagine that same car taking a fast turn on an uneven road surface. If the wheels aren't pressed down to the surface after coming up off a bump, that wheel will lose traction, and traction is required for sticking the turn. In that case, no amount of traction control can make up for the loss of friction.


> If each motor is, say, 80 pounds or heavier, then this configuration would likely increase the unsprung mass.

Apparently, the motor on the Tesla Model S is 70 lbs ( https://chargedevs.com/newswire/elon-musk-cooling-not-power-... ). With individualized motors, each would likely be less than this. Combined with the battery weight attached to the car's chassis, I would expect this to result in a lower ratio of unsprung mass to sprung, and thus better handling than typical on a non-battery car.


> better handling than typical on a non-battery car

Is that their main competition, though?


I mean that if it's better than what is typical, then it shouldn't be inadequate, and that helps justify this trade-off in favour of efficiency / range.


> Traction control can be applied to each individual motor

This is not a feature of in-wheel motors, but individual motors. And not even much of that, because you can apply the brakes to shift power through the differential to the other wheel (a technique modern all-wheel drive cars including Teslas use for traction control). Is having a motor for each wheel really a huge improvement over that? Maybe latency can be improved a little?


I live in the area and I know some people who know some of the people behind this. My impression is that they're for real. It's not just a marketing pitch or a pretty piece of plastic with a fundraiser, they're actually building a proper solar car.

Parts of the team behind it was part of a university challenge team called "Solar Challenge" and won it a few times. They drove that car through Australia the fastest without fuel of charging. That car was ugly and uncomfortable though.

All that said, I strongly doubt they'll ever be able to spin up production. I bet that everybody's best case scenario is they produce a few cars that don't totally suck and kinda/sorta make good on the efficiency / usability claims, and then some automotive bigco acquires them.


You’d be referring to the World Solar Challenge a “biennial road race covering 3,022 km (1,878 mi) through the Australian Outback”. Basically running across Australia North to South.

https://en.m.wikipedia.org/wiki/World_Solar_Challenge


Yep, thanks!


It's really cool to see this event spawing a real company.

Do you know how the company is financed? What price they're aiming to hit? Have they done market research to figure out whether people actually want a probably very expensive car that is efficient, but not sporty? Does anyone actually need that range (I know I don't)?


Seems like they are sacrificing performance for efficiency. They are claiming per mile efficiency 3x better than a Tesla Model 3 but their 0-60 spec is 3x slower than the Model 3.

The idea that a car could charge itself from integrated solar panels enough to cover the distance of the average daily driving is interesting but you would think Tesla thought about this and decided the math didn't work or they would already be doing it. They are in the solar power business as well after all.


Usually the math leads to putting solar on your garage and charging via grid otherwise. You get the excess for other electricity generation, can put in more panels.

Especially if you want to, you know, park in a garage.

This is an idea well ahead of its time. When solar cells become REALLY cheap after the current growth curve in a decade or so, it will be an afterthought to put on solar vs paint.


That doesn’t matter as long as there is a ‘sports’ mode that sacrifices efficiency for performance. Most times you don’t need the range anyway.


They list their 0-60 performance at 10 seconds. That's the typical acceleration of an average mini van. Seems unlikely you would put that in the specs if you could magically flip a switch and get 3x better.


As someone who is not a car guy, but interested in EVs this is a good sign to me.

Tesla has done great work in exploring the space of how you make an EV that impresses Top Gear viewers.

I'm also interested in seeing similar progress for people who just want to get from A to B cheaply and efficiently and sadly few of the existing car makers seem to be taking EVs seriously yet. I've often thought that a different set of assumptions (e.g. not caring about topspeed) might lead to a better overall car, sounds like they are exploring this area.


Yes, the in-wheel motors are at the center of that tradeoff. Several major auto companies have played with using in-wheel motors but they aren't proven in production yet, as there are still several problems to solve there. I assume Tesla simply did not want to tackle those problems.


Predictably, the math does not work.

They say that at 3.7KW it gains range at 35km/h, and from solar it gains range at 12km/h. This means that they expect to get 1.23KW of power from their solar cells. Top theoretical efficiency possible for any solar cell is about 33% (Shockley-Queisser limit), most efficient you can buy today is 22%. So that means they expect 5.59KW of sun energy to fall on the car per square meter. A typical car is 4.5x2 meters. Let's assume they tile the whole damn thing with highest efficiency cells except the windshield (1x2m) They then have 7m^2, so they expect solar energy to provide 799W/m^2. And it is true, sunlight reaching earth is about 1KW/m^2 at the equator in noon, but that is in perfect conditions and just there, and when cell is perfectly aimed at the sun. The car will not tilt, and some days there are clouds or haze, and we do not all live at the equator. Also, in fact, not the entire car can be tiled with solar cells, and dirt exists, especially on the road.

So the numbers they promise are pretty much at the very limit of theoretical possibility in ideal conditions, and not even remotely likely to be hit in real life.


We have solar cells which have exceeded the Shockley-Queisser limit. It's possible to exceed the Shockley-Queisser limit by having a multiple junction solar cell. In fact the world record for conversion of sunlight is 46%[0]. Of course this was with concentrated sunlight. Without concentrated sunlight multijunction solar cells have achieved 30.2% efficiency[1]. Multijunction solar cells are somewhat expensive though. Based on the current rate of advancement of perovskite solar cells it's possible we could see perovskite silicon tandem solar cells with 30% or higher efficiency in the coming months. In december, perovskite silicon tandem solar cells achieved an efficiency of 28%[2].

Still, something's fishy about this vehicle.

[0]https://www.ise.fraunhofer.de/en/press-media/press-releases/... [1]https://www.ise.fraunhofer.de/en/press-media/press-releases/... [2]https://www.oxfordpv.com/news/oxford-pv-perovskite-solar-cel...


my comment remains: if they commercialized such efficient cells, why bother making a car? Why not take the money and go buy a small-to-medium-sized country to retire in?


More efficient conversion of sunlight electricity doesn't mean the cells are cost effective for putting on a house.

On a luxury car like this, area is constrained and money is no object, so it may make sense to use otherwise prohibitively expensive cells to win a few extra percent efficiency. Similarly, such multi-junction cells are used on spacecraft, where the lower mass per watt makes it worthwhile.


There's this claim on the site:

> In fact, our solar cells provide about 20% more energy than traditional ones.

The numbers they provide are best-case. There's also this:

> Someone driving the national average of 20,000 km/year in the cloudy Netherlands would get about 40% of their mileage from solar energy.

They've clearly thought about this. I agree that solar roof on a car, as it stands today, is a gimmick at best, but the fact that your envelope calculation actually _confirms_ that solar charging is possible is enough for me to not dismiss them out of hand. This isn't utter bunk, at very least.

Edit to add: we're probably doing ourselves a disservice thinking about solar charging in terms of range; instead, consider that parking the car in the sun for a day could offset the cost of running climate control for the next 3-4 days. Someone needs to try it, sometime, so I'm glad they are.


20% more than leading top efficiency cells would be 26% efficiency mass-produceable solar cells. IF they had that, they would not waste their time making a car, they'd be to busy diving into huge pools of money, having just made the world's most efficient mass-produceable solar cell by far, and thus become instant billionaires!


Totally agree how the thermodynamics and the economics doesn't add up. Solar panels on cars never really made sense for a reason. The view factor and efficiency of a solar panel on a car would produce very little energy. I would be surprised if it's more than 500W without some sort of breakthrough technology. People forget that 50 kWh (the standard range of a Model 3) is a shitton of energy. A full roof can barely keep up with those number, let alone a car amount of solar panels.


There's no claim that the cells are operating at 26% efficiency; they just claim that, in aggregate, the system is 20% more efficient than traditional systems. There is an important distinction there.


modern DC-DC is already 96+% efficient. LiPo charging is well understood too. Where is the extra efficiency coming from, other than the hot air out of the marketing department room?


On the flip side it is worth pointing out that 83 w/km (133.6 w/mi) is a little over half the power that most of the best selling EVs like the Model 3 or Bolt use. That rate would probably make it the most efficient non-ultra compact EV on the market. So the panels aren't the only place they expect to make large gains on efficiency.


That too should probably be viewed with skepticism. Unless the car has a way lower weight and frontal cross section it's unlikely to be vastly different than existing vehicles. The motor and supporting equipment isn't going to me any more efficient.


Also after doing a little math the battery is larger than the entry level Model 3 and the same size as the Bolt. Plus this car is longer and wider than both the Model 3 and Bolt (although it is shorter than both so I guess it could be somewhat more aerodynamic). I just don't see where they cut any substantial amount of weight. I am rather dubious on all their claims.


I think I miss something in your numbers. They say 83 Wh/km. which means that they expect to get 996Wh in an hour. How did you reach 5.59KW?


At 12km/h it requires 5.59KW of sun energy to fall on the car (in total, not per sq meter as in original comment).

Actual figure given on website is 5m2 of panels on car. Assume 1Kw per m2 of sun power (near equator, no cloud, midday) @ 22% conversion efficiency gives 1.1kw electrical output from panels.

So one needs 5.59KW sun power input for 1.1kw output.


i am analyzing charging rates. they cite charge rate in km/h for an outlet providing 1.3KW and they cite a charge rate for solar, from which we can work out the power they expect to get from solar


I see, but that does not answer my question though. How is my math wrong? the expected range from solar in 1 hour equates to 996 Wh. You suggest they need 5.6kW of solar to achieve this?


you are forgetting that solar cells are not 100% efficient, and charging is not either. so you MUST use charge rates cites, not consumption rates


No, you must not. The energy required for the car to drive 12kms is 996 Wh. let's round to 1kWh.

Solar is DC. for instance Tesla powerwall is 97% efficiency for DC/DC conversion for charging its battery. So you need basically 1kW of solar to produce the needed energy to drive 12km (per hour). I think you are off by at least a factor of 3.

edit: at 20% solar efficiency that would require 5m^2.


so then your claim is they lose 66% on rectification of AC when they charge from a wall? :)


No, my claim is that their solar claim is not ridiculous ;)

Also, if I do the math on their AC charging. 35km/h gained --> 2905Wh energy. Standard 230AC, 16 amps is 3680W. So I guess the number there is also correct? Take about 85% efficiency, then you have 3000Wh in one hour from 230V AC


I looked into the idea of solar panels on cars, to figure out why they don't do that.

Turns out it would add only a trivial amount of energy, and thus is just a gimmick.

>Although the energy from the solar panel might seem like a large value, it's only 2 percent of the total battery energy.

https://www.wired.com/story/could-tesla-power-its-electric-t...


I agree it's a gimmick -- a very expensive one. That said, it can offset the energy required to run climate control, and if you're leaving your car parked in the sun daily, it can add up. So I'm not willing to dismiss the idea out-of-hand. Somebody needs to try it, sometime. Glad they are.


If and when solar paints and solar windows become super cheap, then it would make sense to slap some on.

But at the moment, panels are expensive and would add weight, which offsets much of the benefit.

I do want such ideas to succeed, but this is about as credible as a cryptocurrency whitepaper.


The weight is saved by using in-wheel motors, no axle, no differential, and so on.

It's definitely a gimmick, but I'm not willing to dismiss them out-of-hand. I've designed and built several hybrid and electric cars with teams 1/4 of this company's size, so I don't take issue with the design or their ability to build 1 or 100 units, but rather their ability to actually bring this thing to market and be successful.


If you want to see actual solar cars, you're looking at something like this:

https://www.cbc.ca/news/canada/toronto/toronto-engineering-s...

Car is extremely light (you can pick it up), spindly, aerodynamic, has none of the roadworthy safety features, uses all energy for driving (no AC, headlights, etc.). These compete in the World Solar Challenge and manage a good speed (50 mph) over long distances powered by the sun. This is a good example of what is possible.

There is no way that a "traditional" car, carrying five, spare tire, safety features, headlights, bluetooth, cup holders, with apparently less solar surface area, is going to manage any sort of respectable distance or velocity powered by the sun alone.


Actually the team behind lightyear won that competition last year in the "cruiser" class carrying 4 people and having various features that you mention. (look for "Stella lux")


They actually won it 3 times in a row, with road legal vehicles (all of them got a license plate and thus passed the tests!). They know what they're doing.


This reminds of the Solar Car Challenge. I remember watching videos many years ago about the cars.

> https://en.wikipedia.org/wiki/World_Solar_Challenge

> https://en.wikipedia.org/wiki/The_Solar_Car_Challenge


The team behind Lightyear won that competition a couple of timer in the Cruiser division with the Stella [0] car

[0] https://en.wikipedia.org/wiki/Stella_(solar_vehicles)


Is it just me or this is thing ugly? I think the price tag associated with it will make it hard for most to justify when you can, in my opinion, get a much better-looking car for less money that has a proven track record and AutoPilot. I am happy to see more EV companies popping up, but I think they need a consumer grade car or something really game-changing to get their start.


A button which when clicked launches an unexpected full-screen video is an unwelcome development in web dev.


Indeed, this is something that needs to be blocked asap by ublock origin or a browser setting. Taking over my screen full-screen with whatever they want to stream to me.. ??? I'm already thinking about streaming a full-screen image of a generic Windows desktop with a Chrome browser. Just to trick people into thinking they are not looking at a full screen video.

Lesson for web designers: do not do this. Now we are talking about a silly web feature while we are supposed to look at whatever message your website is trying to convey.


Ugh, this is so upsetting. Stopped my music. Instantly left the page without reading it.


Looks cool, but in Brighton UK it promises me 56km a day in summertime. Seems... not a lot?

edit my bad this is how much extra it would charge a day. Seems like it could hold an 800km charge which is incredible.

Opinion: Tesla should buy them.


I am sure Tesla has already put in an order to first buy one of these cars. See whether there is any significant leg up that their own car making team can get by buying lightyear or not. Reason being, Tesla themselves could already be working on a car design with solar panels, not as primary source of power, but just to help squeeze out a little more mileage


So I guess this car can't be garage kept?


Looking at it, the "oh duh" moment really hits. Why don't more EVs have solar hoods?


My Prius has a solar panel on the roof. It provides enough power to run a fan while the car is turned off, keeping the heat from being overwhelming on sunny days. It doesn't provide enough power to run the fan all the time, just periodically.


Because the sun provides a tiny fraction of the power required to be vaguely useful, plus, solar panels are heavy, reducing the efficiency of the car.


Panels are heavy, cells aren’t. In the extreme (https://en.wikipedia.org/wiki/Thin-film_solar_cell) ”Film thickness varies from a few nanometers (nm) to tens of micrometers (µm)”

That still has to be mounted on something, but that something easily could be the car’s structure (and it is. The site says ”The roof and hood of Lightyear One comprise of five square meters of integrated solar cells within safety glass”)

But yes, the power budget still will be small.


Because you can't really get that much solar power out of it. This car gains only 12 km of range for every hour it spends in direct sunlight.


So... after 8 hours I would get my entire commute. Sounds great, actually.


8 hours of direct sunlight in absolute perfect conditions.


You will never get back the cost of that solar panel, you will actually gain money if you just use the electrical plug


Because it doesn't provide enough energy to warrant the expense.


Now, that is the car company that Apple needs to buy, if they’re really going into that sort of thing.


What, one that's completely unproven with no customers or product? Seems like an odd choice for Apple, but I'd like to hear your argument.




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

Search: