Tesla’s battery is the future. The distant future.
- Ivo Vegter
- 26 May 2015 01:38 (South Africa)
When you’re Elon Musk, you can do no wrong. Granted, this alumnus of Pretoria Boys High has done some fairly cool stuff in his life. Really cool stuff, like found a company to make payments to anyone with an email account almost painless (which is how he made his fortune). He went on to found another company to build high-performance electric sports cars, crash a McLaren F1 supercar without bothering to insure it first, and start yet another firm to build space rockets as a commercial enterprise. That deserves some respect. He appears to be driven by a ruthless, idealistic ambition.
So when his electric car company, Tesla, spun off Tesla Energy, to sell fairly large batteries for residential and commercial use, the fanfare was deafening. This, pundits wrote, would spark a solar energy revolution. It would kill the traditional electricity utility. It spelled the end of fossil fuels!
The insane hype predictably fuelled “crazy, off the hook” demand, and the pre-order quota until mid-2016 sold out in a week. At $3,000 (R36,000) for a 7kWh daily-use battery and $3,500 (R42,000) for a 10kWh emergency backup version, not counting an inverter, installation cost, or any solar panels to drive the system, that’s pretty snappy sales work.
“The issue with existing batteries,” Musk told reporters at a glitzy launch event, "is that they suck. They’re expensive, they’re unreliable… stinky.”
That is certainly true. One of the key reasons that electric cars and renewable power aren’t bigger than they are is that batteries suck. In the case of cars, they offer limited range, require long recharge periods, and are very expensive. A decent electric car (like a Tesla) is a toy for rich, green showoffs. Not including VAT and shipping, a Tesla Model S sedan would cost almost R1.2 million, once imported into South Africa. (It is not available locally.)
In the case of power generation, the biggest problem with wind and solar power (besides the fact that the equipment is nasty to produce and they require large surface area per unit of power produced), is that both are highly variable.
An electricity grid at all times has to produce exactly as much power as is being consumed. Any more or less leads to equipment failures and blackouts. The sun doesn’t shine 24 hours a day, 365 days a year. The wind often blows too lightly or too hard to operate wind turbines. Neither synchronises with peak demand times. That means with renewable energy there is almost always a mismatch between supply and demand. This isn’t much of a problem when you have a base load driven by coal or nuclear, and a peak load driven by gas turbines. (Assuming, of course, that your installed capacity is sufficient and in working order, which in South Africa it isn’t.) But it certainly is a problem when you can’t control supply.
This dilemma could be solved by employing some sort of storage mechanism as a buffer. Pumped storage schemes at hydro-electric dams are well-suited to this task, for example, but they are expensive to build. Batteries would be a great solution, if they weren’t large, toxic, inefficient, and expensive. In short, Musk is right. They suck.
This, ostensibly, is why Tesla introduced what it calls the Powerwall battery pack, sleekly designed to pack 100kg worth of lithium-ion battery into a slick, 20cm-thick package. The smaller, 7kWh unit is designed for daily, residential use. The second, larger battery, at 10kWh, can only be cycled once a week, and is therefore only useful as emergency backup power.
Ironically, Elon Musk’s own solar panel company, Solar City, isn’t as excited about its daily-use battery as customers appear to be. It does not supply the 7kWh battery with its solar sets, because apparently it doesn’t make economic sense compared to just selling excess solar power back to the grid, and running off the grid in off-peak hours when the sun doesn’t shine. Instead, Solar City only sells the 10kWh once-a-week version, for a price of $7,140 (R85,000) including inverter and installation.
They are set to be sold in South Africa, although they won’t be available for some time. Here, two-hour power outages, three times a week are now routine, and feed-in is not yet possible (and probably will not be for years, since Eskom has bigger fish to fry, like preventing a national blackout).
Does that mean the case for a Powerwall battery is more favourable? Yes, but the odds are stacked against it.
Second, the advertised price point of $3,500 (R41,760) for a 10kWh battery does not include an inverter, installation, or maintenance costs, which could easily double or triple the price. Nor does it include shipping costs or import duties.
This makes it cheaper than some (but not all) competing batteries. It is fairly competitive with small-scale diesel generators, but that comes at the expense of flexibility and a much larger initial outlay, which only the rich, with access to loans (or large bank accounts), can afford.
Comparing the battery to a generator is difficult, because of different continuous power delivery capabilities. The Powerwall delivers power at a peak of only 2kW, so at maximum load, a fully-charged 10kWh unit will last at most five hours.
If you want to drive a fridge/freezer, a television and decoder, a computer, essential electronics like security systems, garage door openers, gate motors and cell phone chargers, and a few lightbulbs, 2kW of continuous power isn’t going to get you that far, especially not if you consider that non-linear loads (that is, anything with an electric motor), need two or three times the rated capacity for start-up. And it certainly won’t run your geyser, heaters or stove.
A small (4kW) diesel generator that can deliver that kind of power will only set you back about R10,000, not counting the fuel. If you run it 50 times a year, for 10 years, as the Powerwall is designed to do, the fuel cost will more or less match the price ($7,140, or R85,000) Solar City charges for the 10kWh battery, with inverter and installation included. Still, import duties are not included.
But 50 times a year – once a week – is not good enough. You need a battery that can be cycled at least three times as often. That means you’re limited to the smaller 7kWh battery, which will keep only a bare minimum of electrical gear going during load-shedding, and still cost a small fortune.
You could always double up on the batteries, but that also doubles up on the price. If the load-shedding situation gets worse, and power cuts start lasting twice as long, the Powerwall looks marginally more cost-effective. But if it gets any worse than that, no batteries will keep you going at a reasonable price.
If you have the ability to finance a large purchase, and you’re a well-off South African able to swing a large loan, a couple of 7kWh Powerwall batteries might be attractive. But it’s nowhere near as much of a bargain as the hype would have you believe.
At least a battery is quieter than a diesel genny, and it will probably make you feel all smug and green. That may be worth money to you, but to suggest that it will revolutionise the energy sector is nonsense, even in places where grid power is on the verge of collapse.
Anywhere else, the calculations against ordinary grid power end up making the Tesla Powerwall look outright silly.
According to a recent report by Moody’s, which pegged the capital costs of batteries at $500 to $600 per kWh, there is no risk that anyone will defect from utility power until prices reach $10 to $30 kWh. Another assessment, in Scientific American, agrees that the Powerwall will not kill utilities, nor unleash solar power.
What is needed to spark a revolution is not a minor innovation, slickly packaged, which is all the Tesla Powerwall appears to be. What the world needs to make smart grids, small-scale distributed generation and renewable energy real is a giant leap in battery technology.
Tesla has delivered no such thing. It is at best a linear improvement, which is what batteries (unlike computers) have been doing for decades. Exponential improvement is needed.
There are frequent signs at laboratory-level that such improvements may be around the corner. Every so often, an innovation that either radically cuts charging times, radically improves duty cycles, or radically improves storage capacity, hits the news. The most recent is a new semiliquid battery developed by researchers at The University of Texas at Austin. It combines the power density (W/kg) of super-capacitors, with the energy density (Wh/kg) of lithium-ion batteries. That’s an orders-of-magnitude improvement, but it is far from production-ready.
Batteries that are good enough to make domestic solar power work and challenge the centralised grid may be the future, but they are a distant future.
Tesla’s slick battery might be flavour of the year, but it will not change the world. On the contrary. There has been speculation that because Tesla has never actually made any money, and it will have plenty of battery-makers for competition, high-risk dabbling in this space could destroy the company. DM
Clarification: Several municipalities, including Cape Town, Nelson Mandela Bay (PE) and Ethekwini (Durban), have been running pilot projects with feed-in tariffs for small-scale domestic or commercial power producers. The legalities are still murky, since only Eskom is entitled to purchase power, and if what Cape Town calls "embedded generators" are independent power producers in law, they are required to obtain a generation licence from the National Energy Regulator (NERSA) and a power purchase agreement with Eskom. NERSA has only approved a feed-in tariff for larger installations, such as solar installations producing more than 1MW. If anything comes of these pilots, the case for home renewable installations will become stronger, which may increase demand for batteries like the Powerwall. However, it will also run up against the economics problem that Solar City identified: it may be cheaper to simply sell power back to the grid when the sun or wind is up, and use the grid the rest of the time.
Reader notice: Our comments service provider, Civil Comments, has stopped operating and will terminate services on 20th Dec 2017. As a result, we will be searching for another platform for our readers. We aim to have this done with the launch of our new site in early 2018 and apologise for the inconvenience.