Tesla's Battery Revolution Just Reached Critical Mass (bloomberg.com)

Tesla is all set to cut the ribbon on a massive battery storage facility in the California desert -- the biggest of its kind on earth. It joins similarly huge facilities built by AES and Altagas, which are both set to launch around the same time. Combined, the plants constitute 15% of the battery storage installed globally last year. From a report: Tesla Motors is making a huge bet that millions of small batteries can be strung together to help kick fossil fuels off the grid. The idea is a powerful one -- one that's been used to help justify the company's $5 billion factory near Reno, Nev. -- but batteries have so far only appeared in a handful of true, grid-scale pilot projects. That changes this week. Ribbons will be cut and executives will take their bows. But this is a revolution that's just getting started, Tesla Chief Technology Officer J.B. Straubel said in an interview on Friday. "It's sort of hard to comprehend sometimes the speed all this is going at," he said. "Our storage is growing as fast as we can humanly scale it."

Comments

Economics

[sjbe]

Without a new breakthrough technology in our pocket, batteries technology should be determined by the real use case. Lithium ion is a good technology when weight is very important, but a lousy technology when does not matter. Why use a bad technology when a pretty good on is on hand?

Several reasons, all economic.

1) Economies of scale. Producing two types of batteries is more expensive than producing the same number of a single type of battery.

2) Standardization. Picking the exact optimal battery type for every application instead of using a standard battery actually results in product fragmentation and added cost. It's actually cheaper in many cases to use a standardized product instead of an optimized one.

3) Excess capacity. If you already are producing a product it's often cheaper to make extras and use those than to build a whole new production system for another product for marginal efficiency gains.

Re:Economics

[AmiMoJo]

I imagine they are planning to recycle a lot of cells that have been used in cars and have maybe 80% capacity remaining after a million miles. As they come onto the market in quantity the price of energy storage will fall even more rapidly.

Re:Why

[bluefoxlucid]

Well, there's Wikipedia...

The cell's energy is equal to the voltage times the charge. Each gram of lithium represents Faraday's constant/6.941 or 13,901 coulombs. At 3 V, this gives 41.7 kJ per gram of lithium, or 11.6 kWh per kg. This is a bit more than the heat of combustion of gasoline, but does not consider the other materials that go into a lithium battery and that make lithium batteries many times heavier per unit of energy.

There's a paper from a DOE lab that suggests:

On a per-unit-mass basis, the Evm values for battery production are quite large, especially when compared to the overall VMA burden. Indeed, the incremental manufacturing energy rate is 13.3 MJ/kg of vehicle whereas the values are 91 MJ/kg of Li-ion battery and 105 MJ/kg of NiMH battery (Burnham et al., 2006).

91MJ/kg for Li-ion battery manufacture to store 0.0417MJ/kg as of 2006. With 6,000 full discharge cycles, that's 250MJ of energy storage in its lifetime, or 2.75 times the energy required to make the battery itself.

It's ten years later; energy cost of Li-ion manufacturer has fallen with newer manufacture technology. Recent reports suggest anywhere from 6 to 10 times energy stored than used to create the damned things. Pumped storage (raising water behind a turbine) is 210:1 and adiabatic compressed air is 240:1.

It gets a bit worse than that: once a battery is expended, you need to remove and dispose of it. That means disassembly and recovery of the lithium, the housing, etc., along with transportation fees for the extreme weight of the thing. Adiabatic CAES requires recertification or replacement of storage tanks, hoses, fittings, pumps, and the like. The latter is going to be easier to improve than the former, so future CAES will likely be more-efficient and require less maintenance, and plants will benefit from these improvements as they upgrade tanks and turbines; future batteries will be more-efficient, but not likely to as great a degree--definitely not without inventing a whole new type of battery.

The actual cost is higher, too. Imagine the cost per kWh to stabilize a grid when you have to have people constantly remanufacturing and recovering batteries, as well as monitoring the station to make sure the battery bank isn't showing signs of failure which could lead to explosion. Compare that to the cost of people remanufacturing what is essentially a large structure (those tanks aren't going to be trucked in and bolted down; they'll be built on-site from plates and seals) 1/24 as often, and monitoring temperature and pressure for lower-criticality events (a damaged battery may run away and explode immediately; an overpressurized tank should have enough safety overhead and valves to fail more-slowly or, preferably, non-critically). It's not all about energy.