How does a startup get into batteries?

The first wave tried to produce their own battery cells. While some, such as A123 Systems and Boston-Power, have managed to build factories, it's taken hundreds of millions of dollars to pull off, and most of the others have blown up like a Dell notebook stuffed with shorted 18650 lithium cobalt cells.

Others like Envia Systems have built battery components. Maybe someday that will work out.

The next wave, and potentially the most successful in the long run, could revolve around the traditional Silicon Valley strengths in chip design and software, says Nat Goldhaber of Claremont Creek Ventures. Battery startups will emerge that will focus on producing battery management systems that wrap around cells from mondo-manufacturers like LG Chem.

"We can take your cells and make them longer, and you still get the glory of being a manufacturer," he said. As an added bonus, a successful battery management company could help newer Chinese manufacturers get into higher-value markets like automobiles and motorcycles. A smooth battery management system could homogenize differences between similar, but not exactly matching cells.

Claremont, in fact, has just invested in such a company. Expect to hear a name soon.

Technically speaking, some companies already sell battery management systems. Still, most of the time, battery management systems are an adjunct to another business. Tesla Motors, for instance, sells its battery management to Daimler but is at its heart a car company.

What else might come to the car market in the near future? Alphabet Energy, another Claremont company, has created a chip that converts waste heat into electricity directly via silicon nanowires. The material is far more efficient than traditional waste heat techniques. (The idea originally came from Arun Majumdar, the former UC Berkeley professor now running ARPA-E.)

While large manufacturing companies with glass kilns and furnaces will likely be some of the initial customers, these technologies can also generate electricity from the heat differentials found under the hood of a car. So imagine a hybrid getting a charge from the heat of the engine, or imagine a regular car that runs its electronics off of waste heat. Mileage would increase, but you wouldn't need a whole new engine.

Panasonic has created a waste-heat air conditioner for cars, but it relies on heat exchangers. A waste heat semiconductor would be more efficient, he said.

On other topics:

--Insolar expect to see more innovation on design, deployment, and installation -- the proverbial plumber's crack problem -- as well as innovation in financing, particularly in the residential market. One idea, which is already being touted by Sanyo as well as startup Sunverge, is to combine batteries with solar arrays. With batteries, consumers could get by with smaller, potentially cheaper solar systems. Utilities, meanwhile, would experience less demand for peak power and fewer problems with intermittent flows of power from individual solar systems.

Who knows: maybe someday this could result in a Better Place like swapping system in each garage.

Other ideas coming to market include less expensive, easier-to-install racking systems, as well as racks that can better withstand windy conditions.