Battery Swapping: EV Accelerant or Technological Dead End?
In a previous post, we explored the subtleties behind a variety of automated charging solutions. One of those options was unlike the others — battery swapping. Given the excitement around NIO, the recent unstealthing of Ample, and China’s endorsement of a national standard for battery swapping, it seemed appropriate to take a deeper dive into the technology. Is there a use case where battery swapping works?
As the prior post explained, battery swapping introduces a hornet’s nest of complexity, from extra components that reduce pack-level energy density to new reliability concerns from increased mating cycles and debris ingress. The biggest problem is convincing OEMs to share proprietary technical details in order to standardize the swapping technology, as both NIO and Ample are trying to do. This seems like a tall order when battery technology continues to improve and fast charging times steadily decrease.
On the other hand, battery swapping may be useful if you need a full charge in less than 20 minutes, which is likelier for vehicles that have to operate 24/7. For these fleets, the negative impact on energy density may be offset by higher uptime through swapping. Swapping older battery packs with newer, more advanced batteries may also protect fleet operators from lower residual values, a major barrier to EV adoption.
In summary, it seems that the ideal use case for battery swapping is high duty cycle fleets that can afford to trade performance of an individual pack for improved fleet-wide uptime. In this scenario, battery swapping companies would likely work with OEMs to incorporate less proprietary and cutting-edge battery technologies into fleet vehicles to bypass the business hurdles.
A deeper dive
To simplify things, let’s consider the use case of autonomous long-haul trucking. (Aurora recently announced its SPAC debut, so this is top of mind!) This year, TuSimple shaved 10 hrs off a typical 24-hr trip hauling fresh watermelons along a 951-mile route from Nogales, Arizona, to Oklahoma City.
If they would have used an electric Tesla Semi with 500 miles of range, they would have needed to charge twice during that time. Assuming future electric trucks are capable of ~1MW charging, these two charge sessions would add about an hour to the trip.
In 2019, the average trucking revenue was about $1.58 per mile. Assuming the truck travels at an average speed of 70 miles per hour, the charging deadtime would have equated to $110 in lost revenue.
On the other hand, battery swapping could take about 15 minutes or less, according to practitioners like Nio and Ample. Two battery swaps would then take half an hour, equating to $55 in lost revenue.
Behind the curtain
However, the calculation isn’t that straightforward. Battery swapping stations would cost far more to install than DC fast chargers, even megawatt-level ones. (As an example, let’s estimate $400k for a single 800kW charger installation vs. $1M for a battery swapping station.)
Vehicle integration NRE would still be required, although the costs for a standardized modification of a relatively known technology (e.g. LFP) might be far less than a custom design for a cutting edge battery chemistry. (Let’s estimate $1M for the former and $10M for the latter.)
We’d also need more battery packs than vehicles so that there’s always a charged pack ready to be swapped. (Let’s say this number is roughly 15%.)
These upfront costs are significant — to support a thousand vehicles, battery swapping stations could cost between $30–50M more than megawatt-level DC fast chargers. Fortunately, the time savings from battery swapping are enormous. A back-of-the-envelope calculation shows that it would take less than two years to pay back the initial CapEx difference!
In light of this, battery swapping looks very promising, but it’s surprising that current battery swap operators are focusing on private vehicles (NIO) or ridesharing and delivery fleets (Ample). All of these applications have significant downtime that can be leveraged for normal charging, eliminating the biggest benefit of battery swapping.
Or perhaps it’s not that surprising, given the limited types of electric vehicles on the market. If the right integration model is found — perhaps leveraging older technology that doesn’t require structural integration with the vehicle — swapping could significantly accelerate the electrification of heavy duty cycle fleets.
As always, this was a back-of-the-envelope pass, so let me know if there’s something I missed. Should the economics be calculated differently? Is there another use case that makes sense for battery swapping? If so, let me know in the comments!
Note: At the time of writing this post, I work at Cruise on charging and energy infrastructure. This post relies solely on publicly available information and represents my own opinion only, not that of Cruise or any other organization.