France to build 1000 km of roads with solar panels

Answer: "Roads."

Question: "What surface or environment -- by virtue of having no control over shading or solar angle, and by maximising dirt, wear-and-tear, and difficulty of maintenance -- is more or less the worst place on earth to mount photovoltaic panels?"

I mean, I know that roads occupy a stupid amount of surface area, and I understand the desire to do something better with that area. (Tip: just build less of them.) But seriously, as far as solar panels are concerned, they have nothing else going with them.

Many of you probably know EEVblog. In this video he does the math.

Spoilers: as common sense suggests, those roads are not such a good idea (at least with currently available technology).

Even cheaper, just put the panels up along side the road, perhaps on shoulders or other already owned rights of way.

Even cheaper, just give the money to everybody so they can put panels and batteries on their houses and businesses and over parking lots.

That said, I do think roads are a prime place to combine with energy distribution infrastructure. On top of energy distribution, we don't need cars with batteries if they can instead function like slot cars getting their energy from the road surface below. This greatly reduces the weight of cars such that they are much cheaper to accelerate and decelerate and the stored kinetic energy is much lower making them much safer. Basically roads need a safe "automotive third rail". PG&E in California in partnership with the California DoT is the ideal company to pioneer this approach.

Moving batteries around is great short term, but long term we really should be electrifying roads.

That said, I do think roads are a prime place to combine with energy distribution infrastructure. On top of energy distribution, we don't need cars with batteries if they can instead function like slot cars getting their energy from the road surface below. This greatly reduces the weight of cars such that they are much cheaper to accelerate and decelerate and the stored kinetic energy is much lower making them much safer. Basically roads need a safe "automotive third rail". PG&E in California in partnership with the California DoT is the ideal company to pioneer this approach.

Energy distribution is actually a power company's biggest strategic asset and they should be focusing on this, especially in the face of a world where energy production is being massively distributed through such approaches as home solar.

Moving batteries around is great short term, but long term we really should be electrifying roads.

1. Road surfaces need to be resilient to grit and large trucks driving over them in rain and other shitty conditions. Glass is much softer than rocks, and it will undoubtedly crack eventually in the constant heating and cooling, not to mention tree roots and all the other issues that fuck up asphalt. Not to mention you'd need to clean (or replace) the panels so often you might as well make the roads out of gold.

2. In order to actually get the power from the solar panels, you need an energy grid that is as big as the road network (this isn't a big deal for small rollouts, but it is an important note to make). In almost all countries, this is simply impractical (the electricity network is several times smaller than the road network) and would be an enormous waste of money for very little gain.

3. Solar panels are already about 30% efficient at best. However, that metric only applies for solar panels being hit with direct sunlight. If you lay your panels flat on the ground (and not on rotating sunflower-like angled panels) you lose yet another third of the amount of energy you could've gained. Why? What was the fucking point of that?

TL;DR: Please just build solar farms and stop trying to reuse the road system for something it wasn't designed for. You're just wasting taxpayer's money.

Although there is solar farm, not the road..

•Shading. Cars don't occupy 100% of a highway, even when it's stopped bumper-to-bumper. The faster cars go, the larger the gaps between them. For PV, the recovery is instantaneous, and for a large road surface, the power output flat and steady in spite of car movement. The most sunlight will come in from the best angle during the noon hour, when roads experience little traffic. Morning and evening rush hour usually suck for sunlight to begin with (the sun's lower in the sky), so even bumper-to-bumper shading wouldn't be taking much away from the total.

•Dirt. The technical term for dirt on a solar panel is "dusting". It's a factor, and even in places where it rains mud on occasion, a pretty minor one. The rain eventually washes most of the gunk away. A much more significant factor would be heat buildup. PV's put out more power the colder they are. On rooftops they can convect that heat away. On road surfaces, not as much.

•Ruggedness. Having broken more solar panels than anyone I know, I can attest that a PV road surface can be done right. Cars don't jump Mario-style on surfaces but rather apply predictable compression and shear stresses that would have about as much effect on the tempered glass surfaces as your windshield wipers do on your windshield. Even a shattered solar panel still works almost as well as a new one, believe it or not. The silicon cells themselves have a diamond crystalline structure and are tough as nails in compression. If you somehow managed to bend a cell 90 degrees then it would end badly, but that's simply not happening on this road.

•Solar angle. These won't win any awards for most power output of a given surface area of solar panels, but per hectare of land, they're probably just as good as any system. All the sunlight incident on that hectare will hit a solar panel. Given how cheap solar panels are now (<$1/W), and how expensive an urban road is ($2000 a linear foot!), the solar road competes not against a solar rooftop installation (15% efficient) but against a conventional road (0% efficient).

•Economic gotchas. The biggest concern I have with this approach is the balance of system- the miles of copper wire, the mounting hardware, the inverter assemblies, and the places to store and access them outside the weather. Inverters wear out after X years, and so the question is at what point is it just more cost-effective not to replace the inverters? French electricity will probably justify it, but maybe not Pacific Northwest electricity (where cheap hydro keeps it ideal for heavy power users).

•Heat vs. Electricity- Let's face it. We run on electricity to a far greater extent than low-grade heat. Saving a few kWh from cooling and warming a building using stored heat from a roadway is great, but it's a drop in the bucket compared to what that space probably uses in electricity. You can swing the temperature in that building by literally thousands of degrees with electricity, whereas a couple dozen is the most you can hope for with heat capture. Electricity can be dispatched thousands of miles away along existing infrastructure, but that waste heat is limited to a few thousand feet at best.

•Wheel Grip/Slippage [edit]. Conventional PV's have a flat surface since that's how glass is most cheaply manufactured. If modules are small enough (apparently these aren't) then a tile/lowered grout approach will still create a rough enough surface to allow for reasonable braking distance. Glass is really hard to create coarse surface roughness in when transparency must be maintained, so I'd be concerned that minimum braking distance would be increased. Icing conditions would be particularly concerning.