Hacker News new | past | comments | ask | show | jobs | submit login
Hyundai launches car with a roof-based solar charging system (cnbc.com)
36 points by prostoalex 75 days ago | hide | past | web | favorite | 55 comments



I see a lot of folks in this thread doing the math of how this works out, which is great, hard numbers == good. But as one who went to a considerable amount of effort and research to put solar on the roof of our RV camper (which has a lot more roof than a Hyundai), and as one who owns a Nissan Leaf and knows how much an EV draws, I needed no napkin to know it was a gimmick before I even clicked the link.

Cramming a panel on every bit of usable space, using the best panels my money could buy me, got us 410W (26' RV, kinda small). That'll almost run half a microwave under perfect conditions at noon on the summer solstice. It is a long enough way from running a car that I'll use the napkin to wipe burrito sauce away.


I usually point out that solar panel shade structures on parking lots with chargers underneath will work much better than a car with solar panels on it.

That being said, 780 miles (1300km) of free range by just parking outside is pretty cool!


What a beautiful comment

  1. Complement audience
  2. Establish credibility
  3. Set up punchline ("I needed no napkin")
  4. Explain first hand experience to make point
  5. Deliver punchline ("Wipe burrito sauce")


I get your point. So is roof-based solar pure marketing?


I mean, they're not lying. I just don't think it's worth the opportunity cost for the measly added mileage you get. If they can put solar up there, then I'd rather have a full-glass roof.

EDIT: And now that you make me think about it more, it's just dumb; or maybe an interim solution at best. Because solar panels on the roof of my house is much more scalable, efficient, and for the same power output, less expensive (no one does crash-tests on my house roof). The "interim" part comes from admitting that residential roof solar is quite uncommon, but hopefully not forever.

Let me edit some more with an extreme example: my electric scooter that I ride work every day. I could squeeze some panel on there somewhere, and add a teensy bit of extra mileage. But I won't, because remember that RV solar I talked about? Most of the time it's just sitting there doing nothing, so I ran an extension cord from the inverter to the scooter and electric bicycles. Now our Personal Electric Vehicles are solar-powered, and all I did was run an extension cord. (Tiny though it might be, the RV solar can easily handle a scooter and some bikes.)


I get your arguments and when I think about it, I agree with you. It is similar to Tesla roof tiles: they are expensive, not so efficient, take longer time to install than regular solar panels. They just look good on your roof (also disputable). You pay extra for the aesthetics ;-)


Rooftop solar requires landlord permission and wiring a reserved parking space, which works if you own a house and garage but it's a pretty big impediment to wider apartment coverage.


I completely get where you're coming from. Buuuut, in this case it's as if you don't have availability at your apartment for a Level 2 charger, so Hyundai gives you one of those little hand-cranked generators.


Yeah, it would take a lot of squirrels on exercise wheels to add up to what this unit can do.

But none of these are really going to move the car much more than pushing it.


It would be a lot more useful to just put up solar panels wherever the car is being charged - it's not like the car is going to be able to gain any reasonable amount of range from the roof-based solar between charges anyways, so there isn't much benefit from having the solar panels on the car itself. Why put them on something that moves when you don't have to, for the cost of the solar panels on the car you could probably build much bigger solar panels over your parking spot or garage. Interesting gimmick, but I don't see how it is useful unless you don't have anywhere else to put solar, and even then it isn't saving you much.


I don't think for it to be useful it needs to completely power the car. Consider an average user using it to commute to work 5 times a week, where outside of commuting, the car is driven maybe a little over the weekend.

Reality is that most of the time the car isn't doing anything. So if solar can give your car a few extra miles everyday you commute, it seems potentially worth it then.


Plenty of people put there car outdoors but in places where solar on the ground is not feasible. I do. I leave my car for 3-4 days without using it. That is 18 hours of charge, that's a lot of energy.


Sure, so I suppose it can noticeably extend your time between charges if you rarely use your car, leave it outside in the sun for long periods of time, and don't have access to charging anywhere you regularly park your car. Seems like if that is the case though, an electric car isn't a great fit - regular access to charging is pretty critical unless you want to wait around at a charging station for an hour every time you need to fill up.


I don't see how the math could ever work out. GNI in the mohave is on the order of 1 kW/m2 for the daylight period. A car roof, perhaps 1 m2 if you are generous, won't be oriented well much if any of the day.

So is there any more than PR value to this?


This thing is very much a marketing gimmick, but the idea is not horrible.

A Toyota Camry is 4.9 meters long and 1.8 meters wide. That’s almost 9m2, if they can cover ~2/3 of that with panels you’re talking around 6m2. You don’t get to tilt the panels, but you can put them on the sides which makes up for it.

Space is limited so using 30+% efficient panels is reasonable. That’s 1.8+ kw in full sun which is about 5.4 miles of range per hour. These numbers are appropriate, but when you start talking 40+ miles extra range per day that’s not worthless.

Use fewer lower efficiency panels and 20 miles per day should be cheap. This is well below that.


it's not like people haven't thought about this a lot over the years. I've worked in both industries (solar power and on the electrical engineering side of automotive) and have seen these calculations over and over.

On space: the car may be that size but the roof is smaller. The car tends to get dirty, be in absurd orientations, etc. Plus don't you want to park in the shade if possible? Ever driven in a city?

You're suggesting multi-junction cells which cost an order of magnitude more than mass production polysilicon cells. For a vehicle where they are trying to trim the BOM by pennies (e.g. resorting to 8-bit CPUs in the ECU).

I mean, great, more power (ahem) to them, but I can't make the math work.

It's like the BIPV roof that musk keeps promising for delivery by 2016 or sooner: people have been making them for years and year and yet somehow they don't get deployed. Why? Well physics and economic factors like labor to start, as well as safety (which is the killer issue for me) and, quite significantly, aesthetics, which a lot of people seem to care about (I don't). Having the idea, and even making them, is the easiest part of getting them deployed.


BIPV = building-integrated photovoltaics (for those, like me, who haven't run into the acronym before)

as for my 2¢, regardless of the technical merits, i'm all for more solar. it's earth's ultimate energy source. let's cut out the intermediaries, particularly the unrenewable ones.

regarding economics, it may not make a lot of sense on a cost basis, but as others have noted, it may make sense from a value basis (i.e., higher profit margins).


So far as you say it’s never really worked out, but that does not mean things will stay that way.

I suspect when a ~2,500$ option lets people keep the AC running when forced to park in the sun people will seriously consider it. And at that point we are talking about significant power over time. The use case is different, but Solar conversion kits are quite popular for RV’s even if it’s not enough power to run the engine.


It makes more sense to put solar panels on a roof than in a garage, and even more sense to put them in a field where it is easier to install and maintain.


There are advantages to putting the energy source as close to the storage as possible. DC solar cells charging a DC battery, instead of a solar plant converting to AC for transmission, then converting back to DC to charge a battery. Line losses and conversion losses add up.


Maybe a more useful application would be to use the panels to run air conditioning and keep the car cool in a hot parking lot.



It is hugely helpful. My car is still very hot on hot days, but I'm no longer being burned quite as easily by my steering wheel, belt buckle, seat, etc.

The difference is quite noticeable. Especially on days/mornings/etc where it's relatively cool but the sun is hot. The air circulation really does wonders.


That runs only the fan, I seriously doubt that Toyota micro panel can handle aircon


The amount of power required for that is already so easily provided by the existing battery storage in an EV with very minimal impact to the range. I don't see how it could be worth the cost or effort.


Their claims are somewhat surprising, and I guess they reflect advances in solar cells. My 2014 Prius has a rooftop solar panel, smaller than depicted, and uses it solely to power a fan while the car is off.

Then again, the battery pack in the Prius (not Prime) is so small, perhaps it wasn't worth trying to connect the two.


"With six hours of charge per day, the vehicle could increase travel distances by 1,300 kilometers per year." (or 800 miles per year)

One of the things I do to keep myself accountable for all the subscriptions in modern life is to annualize them, so it's not "just $10/month!" but "$120". But in this case, I think we want to flip it around; at 6 hours of charging you're going to get... approximately two miles per day. Three if you're inclined to round up through the pile of lossy conversions I'm pushing that number through. And that's only going to be in play when the battery is not already full, and it probably won't work while you're charging, either, because the charger will be saturating the battery's incoming charge rate.

I don't find anything particularly unbelievable about that, but it's a marketing stunt, not a good engineering idea, as evidenced by trying to present this in terms of "km per year". That's worthless, swamped in the noise of how your driving style will affect your mileage on an electric car (or any car).


Two miles per day is still not nothing. It's energy that requires zero infrastructure, no new charging stations, no tax incentives, no panels on anyone's house for their HOA to get stompy about.

But it's not just about range. Long-term parking of EVs is problematic because if a lead-acid goes kaput after sitting unattended for a year, you're only out $100. But if that's your traction pack, it totals the car. Solar charge makes long-term parking feasible, and enables telematics modems to stay awake, etc.


"Two miles per day is still not nothing."

I didn't say it was nothing, I said it was below the noise floor, which is a much more interesting way of putting it. It's not worth putting a big solar panel on the car and pretending it's going to do anything if the effect of the solar panel is literally swamped by whether or not you had coffee this morning and its effects on whether or not you have a lead foot this morning. Since "literally" has been devalued, let me highlight that I mean it in its original sense; it doesn't take that much of a commute before that will be the case. Or any number of other things that will just swamp this signal, like whether you had to use the electric heater or AC and how much of it you needed, whether you got stuck behind a semi doing 65 for a good portion of your commute or were cruising by yourself at 80 in the left lane, whether the wind was blowing against or with you today, and so on. It's not zero, but it's lost in the noise.

"Solar charge makes long-term parking feasible, and enables telematics modems to stay awake, etc."

https://forums.tesla.com/forum/forums/how-much-battery-charg... suggests drain can vary in the 1-10 mile/day range just sitting there. This wouldn't even cover that entire range, with the optimistic assumptions on how much solar this is producing. YMMV depending on a lot of exact conditions, but it's certainly not the obvious win you'd hope where the solar can overpower normal drain by a clear order of magnitude. (I'm not loving trying to compare two amounts of energy loss in terms of "miles" here, though they're probably roughly comparable.)

It makes me wonder if they've got some other optimistic assumptions in there we weren't thinking of in terms of the car not draining very fast, or even if the number was generated simply by dividing the amount of power it generates by how much power/mile is necessary. Arguing in favor of the latter is that marketers like the bigger numbers, and especially the positive numbers rather than negatives. That post suggests it is likely to be the case that many real customers will at best notice that this solar panel slows down battery drain from just sitting there, but due to cloudy days, winter sun, etc. is not actually capable of overcoming it!


I definitely read these sentences too quickly: "Hyundai added that 30% to 60% of the car’s battery could be charged using the solar technology. With six hours of charge per day, the vehicle could increase travel distances by 1,300 kilometers per year."

I read those as charging 30-60% of the battery with six hours of charge per day, but I see now that it's six hours a day for 3.5 kilometers per day, and that 30-60% of the battery totally can be charged. Meaning that a car left in the sun for a long period of time can never exceed 30-60% charge.

My bad. Your breakdown makes it more clear that this car is in the same range as my Prius, just marketing the result differently.


The idea of your vehicle sitting there and gathering "clean" energy has a feel-good thing to it, so it's definitely useful as a marketing gimmick to the green people.


Not every car is serving the daily fix of a motorism addict. Some are owned by people who only occasionally need a car but have enough money to skip the hassle of rentals that would be cheaper for their usage pattern. A car that gains a few miles per day instead of trickle-discharging the battery could make a meaningful difference for those drivers.


More productively, if you are worried about this sort of thing, you can recover this amount of energy most likely today if you Google the term "hypermiling", read up, think, and then don't do anything stupid but tweak your driving style with what you read. Bonus, you can do this with any type of car, including whatever it is you are driving now.

This will cut your energy usage by wildly more than anything this solar panel can do.

However, please let me re-iterate, don't do anything stupid. That term will lead you to some stupid things to save trivial amounts of energy.


As long as it doesn't involve going 5km/hr down an offramp hill because you don't have to use gas but there's a massive gap in front of you and a hundred cars behind you forced to go your speed (which is far more challenging at low speeds, especially with a standard vehicle like mine). It's a really selfish and inefficent move IMO and I see it all the time in the city from a small group of people.

If you want to save gas go into neutral and go as fast as everyone else or don't do it unless you're alone.


I learned recently that some older cars also had this feature. A 1992 mazda 929 had a solar sunroof option to power some fans. Quite a few early 2000s Audis had this as an option as well.




I still wonder how better suited to vans the idea is.


Pretty much just as unsuited. Even if a van has 3 times the area available to solar panels it's still a negligible amount of energy...

By the way, the best book and/or course I've read for building intuition for these kinds of questions is "Physics for Future Presidents" by Richard Muller. He used to teach it as a course at CMU and I believe the lecture series is still available to watch or listen to online. Very worthwhile.


I know the amount of energy is small. But IF a company spends time on solar over a normal car, doing so on a van seems 3x better.

Even if it doesn't power the motor, a van can have appliances, tools that can be powered by a .9kW setup (I think that could fit on an average van roof)

Also, they have more flat surface for batteries at the bottom too.


I thought Tesla or Mercedes would have one by now. Diesel generator optional.


They don't because the energy available isn't enough to appreciably add to the range of the vehicle.


To be fair, I am mostly surprised there are not yet EV or hybrid vans, versus PVs on the roof. Seems like stop-and-go, or lots of idling, fleets would benefit signifigantly.

I imagine vans use the battery for more than propulsion, so PVs on the roof might be welcome. Charging from a generator is probably still more cost-effective, except for some niches (e.g. boon-docking). Lots of van-dwellers have PVs on their roof.


It's obviously marketing and gimmicky, but I just wanted to note that I've used a portable 85W solar panel to keep my little 4-banger econobox running when my alternator failed on a road trip.

Saved me the cost of a tow and expensive repair wherever the tow truck took me to. Instead I was able to drive back to civilization proper and have the alternator rebuilt locally for $75.

So I think having solar panels for assisting the alternator on an otherwise conventional econobox could at least be useful in terms of redundancy. It would improve the MPG slightly in the daytime as well, offloading the alternator.

In a hybrid that can operate entirely off batteries, a rooftop solar system could get you out of an out of gas in the middle of nowhere bind if you've got plenty of time, food, and water on your hands.


> With six hours of charge per day, the vehicle could increase travel distances by 1,300 kilometers per year.

So a six hours charge adds 3.56 km. This does not seem very useful to me if one uses the car to commute and charges every night anyway.


Using 4.43 KM per kWh (based on the plug-in hybrid model [1]), that means the six hour charge is 0.8kWh.

At 12 cents per kWh, that means if you get six hours of solar every day (365 days), you'll save about $44 per year in electricity costs.

The average person has a car for 6.6 years [2], so that adds up to a whopping $289 in savings (assuming constant electricity prices, anyway). I wonder how much extra the solar panel costs?

[1] https://www.greencarcongress.com/2018/02/20180209-hyundai.ht...

[2] https://www.cnbc.com/2017/05/28/car-owners-are-holding-their...


Yeah I went straight to that line too. That's a little over two miles. But I think the salesmanship could use some work, because they're talking in averages. On a summer day, full sun, you might hit 5 miles of extended range.

I wonder if that range number includes or excludes self-discharge rate? How about reduced AC usage on hot days? That number may be more significant than the charging capacity (didn't Toyota do a vent-only solar roof for this reason?)

Certain battery chemistries degrade more when they reach a low state of charge, so this trickle charge might end up protecting long-term range retention as the batteries age, maybe?

Friend had to have his Prius towed because they stopped driving it for a while.


We should stop and consider that Hyundai engineers know the energy math behind this design, and they know it's ridiculous, but management chose to pursue this line regardless. This says a lot about their perceived market.


How much does this add to the cost of the vehicle and how long would it take for it to pay for itself? Would it take longer than the average lifespan of the vehicle?


2 miles a day is about how much our Tesla m3 loses by self discharge and parasitic loads.


why not double the area by installing panels on hood and top of the trunk too? are panels very heavy? at least on hood since its not opened often.


they should have tried this on a crossover type of a vehicle that has twice the roof surface compared to a sedan


A gimmick for the innumerate.


Stop with the math guys, that solar panel simply feels right. Feels sell very well.

/Yes, its a tax on stupid.


The real question is, how strong does the ramp that I'm going to drive two wheels up on to tip the roof to the sun need to be? Can I build it out of 2x4s and 3/4" plywood and how far apart do the offsets need to be? Is it OK to leave a car parked for long periods at a 30 degree angle?




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

Search: