What a disappointing article. Most of the points they make are just noise (87% vs 92%? come on). The #1 key fact is that $350/kWh DC at 500 cycles is not at

What a disappointing article. Most of the points they make are just noise (87% vs 92%? come on). The #1 key fact is that $350/kWh DC at 500 cycles is not at all impressive. It's somewhere between 50% and 200% more expensive than existing products. Despite claiming to be familiar with a range of battery chemistries, they failed to note that the 2 kW output limitation would not apply to those existing solutions. There are three good things about the Powerwall, only one of which seems to be getting any airtime at all: 1. It offers much higher energy density than VRLA, especially in kWh/kg (kWh/m^3 is somewhat less impressive). This makes it suitable for applications where lower mass is critical, like laptops and cell phones. Residential grid-tie backup is not one of them, however. 2. It's a sealed unit. Taking maintenance-free to another level, this is a "no user-serviceable parts inside" design. There are no terminals to touch (or corrode), no vents for electrolyte or gases, no series wiring to be done. For people who are intimidated by anything involving electricity or chemistry, this is a tremendous marketing achievement. 3. The operating voltage presented to the system integrator is much higher than 48V. As they mentioned, this makes it easier to integrate with grid-tie inverters, which while less flexible than the do-it-all off-grid/grid-tie inverters usually designed to work with 48V batteries, are also somewhat cheaper and more familiar to installers. That's it. If those advantages outweigh the high price per kWh, poor power output, and low cycle lifetime of the battery itself for your application, the Powerwall is right for you. Otherwise, it isn't. Look for a VRLA or (if you're willing to do occasional maintenance to save even more money) FLA solution instead. You could even look at other chemistries like nickel-iron. Lithium just isn't the right chemistry for this application; it's too expensive and the advantages it offers are unimportant.