Alex Johns sits on a committee reviewing publicly available specification standards for autonomous vehicle testing in the UK and oversaw the trial of Tesla electric taxis at Gatwick Airport. Here, the business development manager at EV battery insurance and warranty firm Altelium explores the end-of-life battery recycling scheme Tesla announced at its Battery Day last September.
Tesla, the world’s leading producer of electric vehicles, and its Battery Day 2020, stirred up a lot of excitement, especially with the announcement of its end-of-life battery-recycling scheme.
The Model 3 is the world’s best-selling plug-in EV model and more than one million Teslas have been built since 2012. In 2020 the manufacturer is pushing hard to deliver 500,000 units, in spite of COVID-19 shutdowns in its Freemont, US and Shanghai, China factories. The number of Tesla vehicles on the road is therefore increasing rapidly.
The Tesla battery (as supplied by Samsung, LG Chem, CATL and increasingly manufactured in-house) is second-to-none in the automotive industry and is facilitating exactly the behavioural change the company set out to achieve in its mission, “to accelerate the world’s transition to sustainable energy.”
My own first-hand experience of the longevity and excellence of the Tesla battery is extensive, through a previous role overseeing the trial of Tesla electric taxis at Gatwick Airport, in the UK. The initial five Tesla Model S 90D vehicles completed 1.5 million miles whilst stationed at the airport, driving approximately 300,000 miles each during the three-year trial. Yet still the batteries were at 82% state-of-health (SoH).
Examining battery state of health
Initially, Tesla offered an unlimited mileage, eight-year warranty on the Model S. However, that was reduced to 150,000 miles, eight years and 70% SoH in 2017. This means that if your recent Tesla Model S or X started life with a 360-mile range and it drops below 252-mile range in under 150,000 miles and eight years, Tesla will replace it.
It is known that the battery degrades faster in the first year or two. In the case of our trial, the first year dropped about 8% in 100,000 miles. After that, the degradation was almost a straight line of 5% per annum (100,000 miles per annum). At this rate, the vehicles would have easily passed 500,000 miles before they reached 70% SoH— which is really impressive and more than three times the warranted distance.
What every battery and vehicle manufacturer is trying to avoid is a ‘knee-point’. This is when a battery undergoes a rapid degradation.
On the assumption that your battery is well manufactured (i.e. without faults), then knee-points tend to occur if you over-stress your battery. In practical terms, this means supercharging it too much (which raises the battery temperature too much too often) and charging and discharging to the maximum too often.
If you run your battery between 10% and 90% state of charge (SoC) most of the time and rarely supercharge, your battery will last a very long time.
If you ‘rest’ the battery once a week for a few hours at a lowish SoC that will also help. However, even with these precautions taken, it is hard for the everyday user to know that a knee-point is coming. In fact, it’s hard for most people to know their vehicle’s SoH (which is not indicated on the vehicle information systems) although a rough estimate can be derived by dividing the indicated maximum range by the maximum range when new.
What does Tesla consider end of life?
Tesla has announced that it is establishing a recycling scheme for its ‘end-of-life’ batteries. A pilot recycling facility will kick off in Nevada, US, in Q1 of this year, which could lead to turning old Tesla batteries into new Tesla batteries. This begs the question: What counts as an ‘end-of-life’ battery?
Other manufacturers have been looking at this question and the consensus seems to be that a battery is still commercially useful, if used carefully, down to about 50% SoH. Below 50% SoH the probability of a knee-point occurring gets too high for commercial reliability.
So, what is happening to the batteries (which have been removed from Tesla vehicles) between 70% and 50% SoH? This is the question that came to my mind following the Tesla Battery Day announcements.
Owners of pure electric cars are primarily focussed on SoC because, while the cars are still a fairly new and a relatively rare concept on the road, access to charging points can be limited at roadsides. SoC is therefore naturally front-of-mind for the driver because it tells them if they have enough power to reach their destination— and has given rise to the term, ‘range anxiety’.
Indicated SoCs aren’t an absolute measure. Manufacturers adjust the indicated SoC to ensure that the user does not ruin the battery by running it to 0%.
An indicated 0% probably means an actual 2% or 3% SoC because it will be adjusted through the vehicle’s battery management system (BMS) to hold the battery at optimal conditions. System updates will ensure that the SoC the driver sees on their dashboard is what they need to see, to encourage them to charge their vehicle in the optimal way for that particular battery and driver’s charging pattern.
Dealing with absolute measures
Battery SoH (when properly measured) is an absolute measure, and one that will become the currency of battery trading in the future. A battery at 70% SoH may no longer be suitable for use in an EV but it will be very useful in a ‘second-life’ battery energy storage system (BESS) for several years (at least five) until it reaches 50% SoH.
The MWh capacity of battery storage has increased tenfold in just five years, showing the classic dramatic S-curve we have seen in all the major technological innovations such as data storage, network capacity or, more fundamentally, computer power.
Most BESSs are made using first-life batteries, but increasing amounts are being made from ex-EV second-life batteries.
Tesla has told us how its batteries will last for a million miles or even longer (presumably to 70% SoH), but has made no mention yet of using second-life batteries in BESS or any other second life use.
Tesla made its name in big batteries with the Australian Hornsdale Power Reserve, completed at 150MW/194MWh; but it held the title of the world’s largest BESS for just three years. It was surpassed by the Californian Gateway Energy Storage in 2020, which has a storage capacity of 230MGW/hour. However, this is less than half the capacity of the new plant Tesla is planning at Moss Landing in Monterey County, California— at 730MWh.
More impressive still, Tesla’s Moss Landing plant and its predecessors will all be dwarfed by the size of the 1,500MW/6,000MWh installation, which was given approval in October 2020 (also in Moss Landing California, for Vistra Energy).
This dramatic increase in the size and number of large BESSs mirrors growth in the small systems market for domestic use such as the Tesla Powerwall. Together this shows the demand is there for BESSs, and we know that there will be an exponentially increasing number of second-life Tesla EV batteries available on the supply side of the equation. So why is the link not being made?
Tesla batteries have varied quite a bit over the years. Their performance, longevity, safety and cost has varied too. For instance NCA cells are known to be quite expensive and less safe than the latest LFP cells. LFP batteries are cheaper but less energy dense than NMC cells. In the Model S, the 85kWh batteries have a longer life that the newer 90kWh batteries. Therefore, it is necessary to have a detailed understanding of what the second-hand battery is, what is its SoH and how has it been used, before you can comment on whether it would make a good second-life BESS.
it is necessary to have a detailed understanding of what the second-hand battery is, what is its SoH and how has it been used, before you can comment on whether it would make a good second-life BESS.
I would argue that catastrophic failure of a second life BESS battery would tend to be lower than an EV battery because:
- BESS usage cycles tend to be less intense than hard EV use
- Charging (and therefore heat) tends to be at lower rates
- Excessive charge and discharging are not done
- The battery has not only its own BMS to manage it, but also a protection circuit module managing a number of battery packs
- There is usually more than one cooling system in a BESS unit: so if one fails, there is another to step in seamlessly
Automated shutdowns are programmed into the control systems if operating parameters are exceeded. There is no need to get home as there is in an EV.
Tesla Powerwalls
All of these new giant BESSs are built with new battery packs. The same also applies to the Tesla Powerwalls, which use new battery cells. However, the cost of second-life batteries is so much lower than first-life batteries that the return on investment of second-life batteries is both higher and economically attractive in its own right, despite the shorter lifespan of the second life BESSs.
This will also improve the whole-of-life economics and carbon savings of the batteries compared with scrapping them straight after their first lives.
A battery at 75% SoH will last half the length of the time of a new battery but cost far less than half the amount to buy. This is exactly the sweet spot where the ‘missing’ Tesla batteries will sit, not yet ready for complete recycling, but also no longer in cars on the road.
The second-hand value of EV batteries will be part of the business model of new EV sales and we will see second-life batteries used extensively across the world in BESS. It is both economically attractive and environmentally essential.
Environmentally sound batteries from Tesla?
Automakers are acutely aware of the environmental concerns around the electric vehicle. Materials such as cobalt, lithium and nickel all have famously difficult supply chains. At its Battery Day on 22 September, Tesla announced that it has bought a lithium mine in Nevada. It also subsequently secured further lithium from Piedmont Lithium, North Carolina, and is in discussions with Indonesia regarding nickel supplies.
Being able to demonstrate that your vehicle battery will go on to a long and meaningful second life is hugely important emotionally, and rationally, to consumers. It will also be hardwired into the metrics of the vehicles to support environmental credentials, with the data from BESS deals delivering the proof.
In the HGV market this is especially important. Profitability for HGV operators is generated from the load weight: so HGVs need to be as full as possible, for as much time as possible, to operate profitably. This requires the highest density, most expensive, batteries.
Whilst simultaneously being the most expensive to manufacture, they will also reach the end of their effective life in the HGV most quickly (a result of charging and operational patterns), perhaps around 80% SoH. This would be both uneconomical and environmentally unacceptable if the battery couldn’t go on to second life use in a BESS.
Finding second life in stationary storage
Altelium is already providing the information to enable vehicle manufacturers to pivot their batteries from EV to BESS applications.
Advances in Artificial Intelligence (AI) learning and secure data sharing are making this possible. The more information we know about a battery, the higher its value. If we know what the battery has done all its life— its voltage parameters, temperature parameters and C-rates— then we are dealing with a known quantity.
The Altelium data system uses a combination of real-time data and AI-learning. The real-time aspect comes from the Bat Lab at Lancaster University where laboratory batteries are set up to mimic the exact patterns of EV batteries in a variety of use cycles. This data is then enhanced with AI-generated algorithms, which gives the robust predictive data about battery performance.
With this information it is possible to buy the battery with confidence, ‘warts and all’ or to use more modern parlance, in the full knowledge that some fuel cells have failed or will fail, and when. Something that is especially important to be aware of is, if a battery has hit, or will be about to hit, the knee-point.
Holding back BESS growth
Unequal access to knowledge has traditionally skewed markets or held back growth and in this respect the Altelium data is a technology breakthrough, giving equal access to information between buyers and sellers, and with it the ability to access the value in second-hand EV batteries.
The flow of information back to Altelium also provides the data needed on the ground to improve the operating conditions in the BESS and extend the life of the batteries. Real-time data can flag up the need to adjust temperature or ‘shaving’ times across the grid, for example. These adjustments could extend the battery cell life in the BESS by a year or more— which could make the difference between profit and loss.
Extended warranties can then be offered on the batteries, and by providing this security, the second-life use is opened up. It is this convergence of insured warranties, AI-driven big data and battery science that is needed to support the commercial viability of electric vehicles.
But this all comes back to knowing where the batteries are, to be able to use them. Typically, when the EV is 7-9 years old its battery will become available for second-life uses. Although we are a few years away from this happening at any large scale, we are already providing the service between the car manufacturer and the energy storage suppliers. We’re only a few years away from this happening at a large scale across the world.
We must be clear that few Tesla’s are currently going around with batteries at less than 70%— but one day there will be many of them. Will Tesla be tracking and contacting the owners of every vehicle over its lifetime?
Ultimately, we can conclude that Tesla’s batteries, once they’ve reached their end of life use for EVs, can certainly find a second-life use in stationary storage— in fact, energy companies and BESS manufacturers alike are chomping at the bit to lay their hands on them.
