Wind + battery
It is sometimes claimed that wind energy with batteries to store energy could provide a reliable 100% renewable supply. In principle this seems reasonable: how practicable would it be, and how much would it cost? Let us take as an example the United Kingdom, which has some of the best wind resources in the world.
How much storage do we need?
The Energy Research Partnership finds that windless periods can last up to 3 weeks, and finds that storage capacity of 6-8 TWh would be needed.
Here we will calculate figures for 1 TWh of storage; we can multiply them by anything from 1.2 to 8 to get actual amounts needed.
There are two important factors we need to work out:
- how much it would cost
- how long it would take to build
Let us assume that we choose to use Lithium batteries, since nobody seems to be building grid-scale batteries with lead-acid or other battery technologies and battery builders presumably know what they're doing.
How long would it take to build?
Tesla's GigaFactory 1 can in theory produce 35 GWh of batteries per year. It would therefore take almost 30 years to produce each TeraWatt-hour of batteries needed. To get to the capacity calculated by ERPUK would take around two centuries.
To speed up this process we would need to build several GigaFactory-sized factories.
How much would it cost?
Cost of factories
If we wanted produce enough batteries for an ERPUK-sized store in just a couple of decades we would need 10 Berlin-sized factories, so we need to spend about $40bn on top of the cost of actually producing batteries.
Cost of batteries
The Big South Australian Battery which Tesla built and sold to Australia has a capacity of 129 MWh. The price the Australians paid for it has not been published but Tesla's owner, Elon Musk, promised that if they didn't deliver it on time they would give it to their Australian customers for nothing. Musk later tweeted that if he'd had to do so it would have cost him (or his company) $50M. We can assume that this is roughly the cost price of the battery installation, which gives us a price of $400/kWh. However battery prices are falling, so let us assume that they reach the "holy grail" price of $100/kWh. This then gives us a price of $100bn for 1 TWh of storage, or $600bn for the lower end of the ERPUK estimate. The $40bn we would have to spend on our 10 GigaFactories adds little to this bill.
(For comparison this is of the same order as at least 25 Hinkley Point C reactors, even at HPC's huge first-of-a-kind, poorly-financed price, which would produce 75 GW of power, about 50% more than the UK's current peak electricity demand.)
This doesn't include the cost of the actual wind turbines needed to generate the power being stored.
- See page 10 of Managing Flexibility Whilst Decarbonising the GB Electricity System - Executive Summary and Recommendations by Energy Research Partnership; Aug 2015 [pdf]
- Tesla battery partner Panasonic sees higher Gigafactory output, cites Model S/X demand increase by Simon Alvarez in Teslarati on 10 May 2019 [article]
- Tesla was ordered to stop work on its $4 billion Berlin Gigafactory over environmental concerns by Isobel Asher Hamilton in Business Insider on 17 Feb 2020 [[article]
- Note however that the Berlin factory is intended to produce cars as well as batteries, so this figure may be different from what a battery-only factory would cost. However we are interested only in a rough order-of-magnitude estimate of the costs so this inaccuracy should not be particularly significant.
- Elon Musk wins bet, finishing massive battery installation in 100 days by Timothy B Lee ion Ars Technica on 23 Nov 2017 [link]
The Holy Grail of Battery Storage
by Roger Andrews
in Energy Matters
on 18 Aug 2016
(referencing Holy Grail of energy policy in sight as battery technology smashes the old order by Ambrose Evans Pritchard in The Daily Telegraph on 10 Aug 2016 [article] which is paywalled)
- Assuming HPC ends up costing $25Bn