While true, the plus side is EVs are still a net CO2 win even when charged by dirty fuel.
As part of the same topic, I think we’re going to see PV-covered EV cars in the not too distant future; not because they don’t need charging (they’re about 10% of your instant needs on the move), but because adding PV reduces the pressure on the grid, and will significantly reduce the need to install power lines to sunlit car parks.
Home, multi-storey, and hotel/motel parking will still almost certainly still need power.
Right now? Houses sure, but I tried to get them in my last apartment, and the answer was “no” even though I owned my own flat. (Owning a flat != owning the building the flat is in).
But this is about what I expect soon-ish, and “soon-ish” both continues the cost decline of PV (including thin flexible panels that would make them suitable for more than just the Cybertruck), and also makes it likely that every roof suitable for PV will already have it (because exponential growth).
There’s some overhead to the charger-battery system to actually generate any sort of a charge. I’ve had the exact same assumption multiple times, and as soon as I do some research with the math breakdown.. its clear why we don’t see these in production yet.
With current efficiency PV systems, the math requires a fairly large surface like the entire flatbed of a Cybertruck to actually generate a meaningful charge.
This also requires the vehicle be parked somewhere its going to get a good amount of sun, excluding parking garages, a lot of urban areas, etc.
Ok, as you’ve done the maths: what charge do you get? Not the ~10% I got from a back-of-the-envelope calculation?
Model 3 is about 5m by 2m, and is apparently rated for 241 Wh/mile
4m * 2m * 1kw/m^2 * 50% * 20% = average power 800 W
(50% because the panels are flat, 20% because cell efficiency)
241 Wh/mile * 60 miles/day = average usage 602 W
I’m not sure what fraction of the day people drive for given that I’m not a driver, but I’m eyeballing 5-10%. I acknowledge professional drivers — taxis etc. — can’t possibly rely on PV alone, that PV can only supply a fraction of what they need (my 10% guesstimate), but I still think this should help with the general public. Or are my assumptions way off?
First you missed a decimal point or two here and scrambled the units of measure-
“ 241 Wh/mile * 60 miles/day = average usage 602 W”
The correct math is
241 Wh/mile * 60 miles/day = average usage 14,460Wh, or 14.46kWh
Further answer - The consensus from people who know this better than you & I, have these cars, and in some cases have tried.. is basically - it won’t charge much, and it’s way more expensive than the electricity it is going to generate.
Note there are AC-DC inverter losses of 10-20%. plus input->battery charge losses which are non-linear and very bad at the low end. For example a Tesla won’t even take a charge if the input is below the ~300-500W range in good weather. In cold weather say Northeast US winter, the floor is closer to a 1kW input as there is a heating system to get the battery put to temperature for charging that is going to eat almost all of that.
So just taking the parent example, even if we assume plastering the car in PV will generate 800W peak
1) This will probably not translate into any charge in winter weather, but possibly allow you to keep the car battery from being fully cold soaked, best case
2) In good weather you are probably looking at post-inverter input to charger at 700W, with charger losses meaning about 400-500W making it to the battery. So that is, in an efficient Tesla about 2 miles of range for every hour of peak sun. Depending on your location, orientation and time of year you might expect peak sun hours of 3-6 hours/day. So grand total 6-18mi/day of range added making a lot of happy assumptions and not moving your car during lunch. This amount of charge per day could be acquired in 1-2 minutes at a supercharger and worth about 30-75cents. Or charge at a L2 charger in your own garage in 12-36 minutes.
I don’t understand your math, I’m sorry. Why would you divide by 24 hours? There is not 24 hours of sun for your PV to capture and put into the car. Peak solar generation is 3-6hrs/day depending on region and time of year.
4m length * 2m width * 1kw/m^2 insolation * 50% loss due to the panel area being calculated by ground area and it not tracking the sun and therefore not getting peak output * 20% cell efficiency = average power 800 W
My BOTE calculation above should use 25% instead of 50% for day-night average of PV panels horizontal to the ground. Can’t edit now, though. 25% is the planet-wide average for day-night and seasonal variation because that’s the ratio of the surface area of the Earth to the area of a disk intersecting the same flux of sunlight at 1AU (4πr^2 : πr^2).
With that correction, that’s 400 watts average over 24 hours (as in: no not merely the peak at noon); which means 24 h * 400 W = 9.6 kWh per day.
If you drive 60 miles per day, and each mile consumes 241 Wh of energy, then you consume 14.41 kWh of energy per day.
> I think we’re going to see PV-covered EV cars in the not too distant future
They're already popping up in the development stage [0]. 12 km/hr peak solar charging sounds really quite good, and I like the overall design. The company was founded by students who won the World Solar Challenge [1] in 2015, it's pretty neat to see them taking that experience and running with it.
Right now yes, but solar is still getting cheaper. I’m expecting it at some vague point in the not too distant future, but not in a surprise tweet from Musk tomorrow.
But solar panels you can slap on a building roof or make into covered parking will always be cheaper than form factor panels installed in the roof of a vehicle. Other than super specific fringe use cases, there is just no reason to justify installing these expensive specialty panels in cars instead of installing them at homes/workplaces/retail centers and just charging EVs at those locations.
There is also the cost savings in scaling. The cost for a solar system comes down even more when a business can install a whole row of solar panel covered parking, either offer EV charging as a perk or charge for charging, and use the rest of the electricity to power normal electric operations. For the driver, buying the electricity as needed from home or work panels would always be cheaper per total watt usage than buying in-car roof panels, since again, the specialty nature of them means they will always be more expensive and less efficient than their stationary mounted counterparts. You can never scale up car roof solar because you can never install move than one car roof’s worth at a time. You also can’t ignore the inverter power loss that is much worse at the lower power a car roof system would have.
Then of course you have the downsides of long term sun damage to your vehicle to get that minimal charge, instead of protected under a solar panel covered parking or in a garage with solar mounted on top. You have the higher rate of damage by being installed on a moving vehicle instead of a stationary object on a building or parking structure roof. You have the lower rate of return, since panels are rated for 20-25 plus years and most vehicles don’t stay on the road that long. You have the downside of sub optimal charging angle and all the time the vehicle spends in a parking garage during the day, as opposed to a stationary panel that is pointed at the sun 365 days a year. At the end of the day, it will always be cheaper and more efficient to have stationary solar panels.
Given that most people charge their EVs overnight/during work anyway, car roof panels would only really provide value during “road trip” situations, where you are driving close to or beyond at full charge per day. Since you are only getting at best 10-15 additional miles over an entire day in the sun (and more realistically less than 10), it would do very little to reduce range anxiety. And that is not even calculating how much the additional weight of the panels would reduce range.
It sounds good on paper but it is highly unlikely to translate to a real world benefit.
I don't have access to that article but does it take into account the notion of marginal power? That any increased load is going to be disproportionally dirty?
It's not a given that increased load is going to be disproportionally dirty. Wind farm output typically peaks at night. Wind is a great match for night time battery energy vehicle charging and it has the lowest life cycle CO2 footprint of any electricity source:
I do like your emphasis on marginal effects. As renewables and BEVs grow it will be a balancing act to pick the most marginally effective resources for emissions abatement. California may soon reach a point where an additional dollar invested in solar doesn't abate as much CO2 as the same dollar invested in transmission, storage, or wind -- even if solar has the lowest instantaneous generation cost.
Is this generally true, or does it depend on geography (e.g. being near the coast)? Where I am in the midwest, it seems that the air normally gets very calm after sunset.
I think that it is generally true for land based turbines. Turbines are so tall that wind conditions at hub height can't be easily estimated by what we experience on the ground. Here's a somewhat dated study that shows hourly patterns for wind generation on the ERCOT grid in Texas, which has the largest wind fleet of any state:
"The Relationship between Wind Generation and Balancing-Energy Market Prices in ERCOT: 2007–2009"
See Figure 5. Hourly generation reaches a minimum from about 1:00 to 5:00 PM and reaches its maximum around 1:00 AM.
Offshore wind output changes less from short term day-night cycles, and generally achieves a higher capacity factor. It is also more expensive to build than onshore wind and no large projects have yet been built for the US, though several are on the drawing board.
They are a net CO2 win, but not enough of a win if you charge them on dirty fuel. So making sure we don't make people too relient on charging on dirty fuel is important.
As part of the same topic, I think we’re going to see PV-covered EV cars in the not too distant future; not because they don’t need charging (they’re about 10% of your instant needs on the move), but because adding PV reduces the pressure on the grid, and will significantly reduce the need to install power lines to sunlit car parks.
Home, multi-storey, and hotel/motel parking will still almost certainly still need power.