KERS, as we've all explained to our grandmas, is the Kinetic Energy Recovery System that's in use on F1 cars to give them momentary bursts of power, like the TURBO button in a video game. The system uses energy recovered from braking, and there's something very satisfying about storing wasted energy and then spitting it back out as an extra burst of speed.

Usually, this energy is stored in a battery or occasionally a flywheel, and then turned back into face-melting speed via electric motors. I was thinking about KERS systems and how they're designed to deliver energy in short bursts, when I realized this may be the one viable application of the long-desired clockwork-powered car.


Clockwork/spring-powered cars have long been experimented with. If you have an Oujia board handy, ring up DaVinci and ask him about it. But for a number of reasons, they're not a viable option. You're always going to expend more energy winding your clockwork car than you will get out of it, and springs just aren't that efficient.

Still, in a KERS-type application, maybe it could work? You could even leverage centuries of self-winding watch development to help take every bit of kinetic energy and put it into that mainspring. Plus, the idea of an F1 car with a big wind-up key is hard to beat.

So, I asked Jalopnik's captive physicist, Dr. Stephen Granade, about clockwork KERS. Here's what he told me:

It's certainly doable in theory. The engineering details will probably get you in the end, though.

The first, and biggest, problem is that springs don't have good specific energy. Specific energy's a measure of how much energy you can store in something per unit of mass. The higher something's specific energy is, the more energy you can store in it for the same amount of weight, and the better the storage method is in terms of how heavy it is.

You know how heavy batteries are? Flywheel systems have similar specific energy values as batteries.

Springs have between a hundred and a thousand times less specific energy.

To store as much energy as a 10 kg flywheel, you'd need somewhere between 1,000 and 10,000 kg worth of springs.

The second problem is that springs work by bending metal and having it, er, spring back. Springs become fatigued and fail over time. Springs are rated for a certain number of cycles before they fail, depending on what they're made of and how much tension is placed on them. For a high-capacity system like you'd need for a KERS, I'd imagine that lifetime would be measured in years. Flywheels wear out over time as well, but they're not under the same kind of bending stress as springs and can last for decades.

I'd say a clockwork KERS is doable, but not practical overall.

This all changes if carbon nanotube springs become a reality. An MIT team had a big press release about them back in 2009 (…) but I haven't seen much about it since, probably because we haven't yet found a good way of producing individual tubes or aligned tubes. We can make nanotubes in unorganized bulk, like a jumbled pile of microscopic pick-up sticks, but not in an organized structure.

Okay, so it's possible but it'll suck. But what if we figure out how to really manufacture carbon nanotubes? How much specific energy would a carbon nanotube mainspring have?

You should be able to match the specific energy of lithium-ion batteries and flywheels with them. If only we could make ordered carbon nanotubes readily, there's a lot of really cool things we could do with them.

Springs that match the energy density of batteries? That's not bad, especially since springs can be "recharged" with relatively simple mechanical processes, possibly even leveraging waste kinetic energy from other sources. So who knows? Maybe a clockwork car based on carbon nanotube springs could be in our future.

And I'll still want a giant wind-up key.