AUSTIN, Texas -- For its next trick, Google will begin flying 84-foot-long airborne wind turbines, said Astro Teller, the head of Google X, the company's lab of "moonshot" projects. The enormous kite-like turbines will take flight next month, Teller said at the keynote capping off the Interactive portion of South by Southwest.
Google has been working on Project Makani, as these wind turbines are called, since buying a company of the same name in 2013, but until now, the search company has been testing only 28-foot-long models. The turbines to be introduced next month will be full-scale models.
Why flying turbines? Wind speed is faster and more consistent at higher altitudes, but it's impractical to build taller ground-based wind turbines due to their weight (we're talking hundreds of tons), Teller said. That's why Google is focusing on airborne turbines instead. "There's an enormous benefit to going up higher," said Teller, speaking at the Austin Convention Center.
The Project Makani kites look like the wingspan of a large airplane minus the cabin in the middle. Each kite has eight propellers that it uses to take off and a tether that keeps it attached to the ground. After the kite ascends to the limit of the tether, which is more than 1,400 feet, the propellers stop climbing, Teller said. At that point, they begin serving as flying wind turbines and the kite starts doing large circles in the sky. Combined, this generates 600 kilowatts of energy that is sent continuously down the tether.
"If this works as designed it would meaningfully speed up the global move to renewable energy," said Teller, whose title is "captain of moonshots."
Google has been flying its 28-foot models in Pescadero, California -- one of the "gustiest" places in the world, Teller said. Wind speed and wind direction there can change drastically in a matter of seconds. Despite the harsh conditions, Teller said Google failed to crash a Makani kite after more than 100 hours of flying, and that's not exactly a good thing for Google.
Teller's keynote focused on the benefits of failing fast. Failing in an experiment allows Google to learn something new and improve its projects. CEO Larry Page told Teller he wanted at least five kites to crash, but since none did, the company now goes into testing of the full-scale model without that knowledge.
"We're all kind of conflicted about that," Teller said. "We didn't want to see it crash, but we also feel like we failed somehow. There's magic in everyone believing that we might have failed because we didn't fail."