Solar freaking roadways? Do the math

In the last few months, I have found myself asked many times about a concept for solar roadways. Folks from Idaho proposing them have gotten a lot of attention with FHWA funding, a successful crowdfunding and even an appearance at Solve for X. Their plan is hexagonal modules with strong glass, with panels and electronics underneath, LED lights, heating elements for snow country and a buried conduit for power cables, data and water runoff. In addition, they hope for inductive charging plates for electric vehicles.

This idea has come up before, but since these folks built a small prototype, they generated tremendous attention. But they haven’t spoken at all about the cost, and that concerns me, because with all energy projects, the financial math is 99% of the issue. That’s true of infrastructure projects as well.

There are two concepts here. The first is, can you make a cost effective manufactured road panel? Roads are quite expensive today, but they are just asphalt gravel and other industrial materials whose cost is measured the range of $50 to $100 per ton. A chart from Florida suggests that basic rural asphalt roads cost about $9 per square foot, all-in, including labour and grading (it’s flat there) and about $4/square foot for milling and resurfacing. Roadway modules could be factory made (by robots) but still would require more labour to install, but I still think it is a very tall order for a manufactured surface to not cost a great deal more, even an order of magnitude more than plain road. Paved roads need maintenance, and that’s expensive. It is proposed that these panels would be cheaper to maintain as you just swap them out, but I am again skeptical of this math. Indeed, one of the major barriers to proposals for electric roads (which can charge cars) is that putting anything in the road makes it prohibitively more expensive to maintain.

I won’t say this is impossible — but it’s all about the math. We need to see math that would show that the modular manufactured pavement approach can compete. I’m happy for that math to include future technologies, like robot assembly and placement (though realize that we’ll probably see road construction with simpler materials also done by robots even sooner.) Let’s see the numbers, how cheap can it get?

All of this is without the solar panels inside (or the electronics.) Because the solar panels have their own math. The only synergy is this: If the modular roadway can be made so that it costs only a bit more than other approaches, it offers us “free land” to put the panels, and it’s connected land in long strips to run power wires.

How valuable is free land? Well, cropland in the USA costs an average of about 10 cents per square foot. 23 cents in California. 3 cents/square foot in the rural west. Much more, of course, in urban places. The land is not that important, so the other value comes from having a nice, manufactured place in which to put solar panels.

Today solar panels are still costly. They are just getting down (primarily thanks to cheap Chinese money) to our grid price. Trends suggest they will get lower and become cost effective as a variable source of power. But until they get really, really cheap, you want to use them most efficiently.

To use solar panels at their best, you don’t want to lie them flat (except in the tropics) but rather you want to tilt them just just a bit below the angle of your latitude. Conventional wisdom also points them south, though it’s actually better for the grid and most people’s power demands if you point them south-west, losing a few percent of their output but getting more of it to match peak demand. Putting them flat costs you 20 to 30% of their output. (You can also have them motorized and gain even more, but it’s usually not cost-effective, and will become less cost-effective as panels get cheaper and motors don’t.)

To use solar panels at their best, you also want to put them where it’s very sunny. Finally you want to first put them where the local power comes from coal. When you have gotten rid of most of the coal, you can start putting them elsewhere. You can put panels in less sunny places which have power from hydro, nuclear or natural gas, but you’re really wasting your money. The ideal places are Arizona and New Mexico, with tons of sun and lots of coal. And lots of cheap, fairly low-value land.

To be fair, the biggest cost of the panels will soon be the hardware they are mounted in, along with the wires and electronics to connect them, and so perhaps these road modules could compete by being cheap hardware for that. But it seems not too likely.

In cities, rooftops provide another source of free land, much of it slanted about right and pointed in roughly the right direction. With lower cost than tearing up roads. But to be fair, right now one of the bigger cost elements is getting permits to do the construction and electrical work. Roads are far from bureaucracy-free, but at least it scales — you get permits for a big project all at once, not one house at a time. But we can solve that problem for houses if we really want to as well.

So my challenge to the solar roadway team is to show us the math. No, we don’t need to see what it cost to make your prototypes. I am sure they are very expensive, but that’s beside the point. I want to see a plan for how low the cost can go in theory, even assuming future technologies. And compare that to how low the cost for the alternatives can go in theory. And then factor in how things don’t get to that theoretical point due to bureaucracy, unions and other practicalities. Compare panels in the road to panels by the side of the road, tilted and not being driven over. Look at what paved roads cost in practice to what they could cost in theory to get an idea of how close you can actually get, or come up with a really convincing reason why one approach is immune from the problems of another.

And if that math says yes, go at it. But if it doesn’t, focus on where the math tells you to go.

solar roads

I agree. What's the gross value of the electricity generated by these tiles? What's the cost of the cables and grid interties taking the power off the tiles, and do these wiring/control items suffer excessive wear from traffic passing overhead? How do you clean the tiles to keep their conversion efficiency as high as possible -- there's another labor content cost question.

Efficiency (both cost efficiency, and as a driver, power conversion efficiency) says we should locate solar farms as close as possible to major power distribution lines and hubs. From that list, pick the regions where the sun is strong and the land is cheap and set them up there.

Setting up solar roads as a distributed network running through places where nobody lives is probably the wrong approach.

Very poor economics

Even a cursory analysis of this sort of radical implementation shows it can never be economic for most if not all roads.
Arterial roads are designed to take maximum loads in the range up to about 16-18000lbs per axle (commercial road trucks) and any modular unit must carry this same load. This means the road underlay has to have very similar load bearing characteristics to existing road designs. So implementing this sort of modular add on is only really replacing the Asphalt top. I would suggest that it would be close to impossible to get the cost structure to match the cost of Asphalt.
Your comments on loss of efficiency are right on too….but the surface is thicker and less translucent than the typical glass, so I’d suggest overall that the loss of efficiency runs closer to 50%+.
Perhaps there is some application as shown in the image of a path replacement, but even here I doubt it. Overall not a great idea IMO.

How do you value life?

If we say that it's a winter evening and it have just snow heavenly and a family is driving out on the landscape and it´s very slippy on the road so they drive of the road and crash what make them all die. If the solar roadways have been there the snow have been melt away and the road lines have been clearly visible so that the crash never happened. So now is the question to you, how much do you count the life that this road maybe can safe is worth except all the societal cost that comes with a car crash that ended with death?

Value of life

As much as it’s not a popular idea with the public, society constantly makes cost/safety tradeoffs which value human lives. You will find many articles on the web about it. Typical US values range around $5M these days, or sometimes $50,000 per year at the lower end.

There are many ways to keep roads clear, and there are other ways to prevent accidents from ice on roads, ranging from going slower to robocars to better tires to better road designs.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Post new comment

His name is Brad Templeton. You figure it out.
Please make up a name if you do not wish to give your real one.
The content of this field is kept private and will not be shown publicly.
Personal home pages only. Posts with biz home pages get deleted and search engines ignore all links
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

More information about formatting options