I don't believe a word of it. These kinds of announcements tend to show up every few months, and none ever make it to market.

"There are a handful of small hurdles yet to cross in the science, but nobody seems terribly concerned."

If they were so "small" they would be solved. No, they are massive hurdles, and they have no clue what to do, so they put out a press release hoping for more money.

Plus solid sodium in your house? Talk about a fire hazard. You have an external fire which melts the casing, you dump water to put it out - and fireball.

This isn't complete pie-in-the-sky stuff. NaS batteries have been around for decades (http://en.wikipedia.org/wiki/NaS_battery) and in fact Ford was building EVs with them in the late 90's.

It looks like they found an interesting tweak to the technology that might make it suitable for home use. It's possible (likely, even) that it will end up either too expensive or too unreliable to be practical, but it's certainly newsworthy.

Whenever I read an article like this, I add a google news alert for the company and forget about it.

I stopped reading after the first three paragraphs when the article had already managed to drag in paradigm shifts and Socrates.

If you combine net metering with time-of-use charges, it might very well be profitable to engage in power arbitrage.

I.e. you buy from the utility when it's cheap and sell it back to them when it's expensive. In this case you'd actually be providing a valuable service.

So it costs 3 cents a watt/hour (over the lifetime of the battery--much more than that short-term) to store (not produce) the energy, and that is supposed to be better than the seven cents a watt/hour that the electric company will charge for delivery right to my doorstep? I don't get it.

Those are really two different things.

Compare their projected cost with, say, a deep discharge marine battery at $200 for 10 ah, which is 120 watt/hour.

The claimed figures are a huge improvement over what is available today. If they can make this work, it will be a seriously big deal.

There is a long ways to go before that, though.

Houses with power generation already have an indefinitely-large capacity battery on hand. It's called the grid.

If you produce too much power, you push your extra into the grid. If you don't produce enough, you pull what you need from the grid.

Yes, having local battery storage may allow you to time-shift for pricing reasons.

But it won't help power your home. Either you have enough local power production for that or not. If you do, you'll be a net exporter to the grid. If you don't you'll be a net consumer.

How efficient is charging one of these from the grid? Charging the battery at night and using it during the day could help with peak use overloading.

Another interesting idea: Even in the absence of home-based electricity generation, the cost of energy varies based on time of day and based on season. In a lot of areas, power used at 6pm is more expensive than power used at 3am. So in cases where this discrepancy is greater than 3 cents/kW-H, it would make sense to have this battery charge up and 3am to handle your evening power usage.

It's interesting that many alternative energy related ideas tend to emphasize more localized facilities whilst computing moves to huge centralized data centers.

It gets even more ironic considering that one of the popular examples for the long term trend towards outsourcing/centralization is that auto makers stopped generating their own electricity long ago :-)

In both computing and electrical power there's a tension between localization and centralization. As the technology evolves, the relative trade-offs shift. Right now, the rapid development of ever-higher-bandwidth connections tends to favor centralization of computing power (though there's still good reason for a lot of it to be localized in the laptop and cellphone).

With electrical power, the economics have long been mostly in favor of centralization. This started shifting a little with the development of ever-cheaper solar panels, and would shift a lot more were any sort of effective localized power storage possible. And unlike bits, there is a transmission loss for watts.

Though I'm dubious that this technology will ever see the market. When it's for sale to the consumer, then I'll know it's real.

If solar panels become cheaper they become cheaper for centralized facilities as well. So how does that change the equation?

Transmission and distribution costs of electricity are ~30% of the total cost, so if solar panels become cheap enough, the transmission and distribution costs for centralized facilities will become a large part of the cost. Of course, if we get room temperature semiconductors by that time, those costs will come down.

The article (very long...) mentioned the price for a battery at $2,000 and did the cost analysis but I didn't see anything on the product's availability. Did I miss it?

$2,000 is the projected price, I don' think it is all that close to coming to market.

>California-style rolling blackouts would be history.

"Enron-style" might be more accurate.