The powerwall use case will only grow when people get solar panels and when rural areas in countries lacking power infrastructure eletrify. But while that use case doesn't care as much about weight, bulk, charge rate and % capacity as car batteries, in return they are very sensitive to cost and the market will (depending on how cheaply recycling can be done) likely be filled with used car-batteries with degraded capacity.

If you invent a battery that is 2x as large and heavy as current vehicle batteries but half the cost, then you might think you would be able to sell them for stationary storage, but you are then competing with the price of second hand vehicle batteries which may well be half price already.

One drawback of using lots of lithium batteries in your home is that they might experience thermal runaway and burn your house down. Not having to worry about that should have some additional value.

Most new solar installations now use LiFePO4 which apparently is much safer on that runaway side...

That's good, but is that what's used in "used car-batteries with degraded capacity"?

Used car batteries are conventional unsealed, lead-acid batteries. Lithium-iron-phosphate (aka LiFePO4, LFP) batteries is a different battery chemistry altogether. It is touted as the answer to a deep-cycle, lithium-based electrical storage that the traditional lead acid batteries occupy.

I think they mean used EV batteries - EV batteries that are at the end of their useful life in cars.

That makes more sense.

Also way better cycle life, but the density is slightly lower.

I'm of two minds on this.

On the one hand, you're right that for different applications having different metrics can be acceptable. Sure, a battery for your home that never moves can be heavy and large and that's okay.

But then you have to consider the economics of production. How many of those batteries will you make? What will the factory for them cost to build? And what would it cost instead to just make a bunch more Li-Ion batteries at the existing factory instead?

Your logic makes sense if there is only one battery factory. But there are and will be many.

Economies of scale still apply with multiple factories because you need someone to design and build the equipment used at each of them.

Where do flow batteries (fuel cells) fit in those calculations?

Some of the more exotic battery designs are things like "rust powder held in a silo at a certain temperature" which presumably can be assembled on-site. It seems to me overly reductive to assume that stationary batteries will be "just mobile batteries but heavier".

True (within limits), but the largest use of batteries is batteries that move. Utility scale power storage is the only significant use of battery that doesn't move. (there are others, but they are not probably large enough to develop a battery for, and so will be stuck with whatever they can get from the other markets)

Yes, but stationary energy storage is itself a large use case.

There are also different performance profiles within EVs, where different tradeoffs might make sense. Though, like you, I'm somewhat skeptical that any non-lithium chemistries will break through in that space.