Designing Swappable Batteries

This article reviews the design of swappable batteries. When plug-in charging is replaced with swapping, it leads to some non-obvious differences. Through the prism of practical situations, the readers can understand what is important in designing swappable batteries including the development of its concept, choosing the optimal form factor, and working out external parts and battery management systems (BMS). Meanwhile, swappable and fixed batteries share essentially one inner design, so we aren’t looking at it.

Disclaimer: This article is written by a technology journalist rather than an engineer, so I plead to forgive my arbitrary use of terminology. Please contact me if you notice a factual error.

Future-proofing the concept

Working with a client on a future battery’s concept, engineers should think ahead to several decades. Unlike fixed batteries that can be redesigned with each new generation of vehicles, swappable batteries inherit outer design, power output and data exchange protocols of their precursors for maximum utilization purposes. It’s typical of swap operators to mix modern batteries into their stocks of older ones and offer them at different prices.

Another aspect is, the choice of compatible types of vehicle can eventually be changed. A frequent situation is when a swap operator launches with just one type of vehicle and eventually adds more. In India, some of the existing operators had launched with e-scooters to later expand to three-wheeled e-rickshaws while other companies went backwards. As to European and Asia-Pacific markets, foreseeable vectors are from seated scooters up to to city cars and down to standing scooters and bicycles.

In each type of vehicle, a different set of forces impacts the battery’s locks, connectors and inner parts. For example, in a seated scooter, a module usually sits tight in its compartment in a vertical position, accessed for swaps from the top. An electric bike or standing scooter might have the battery fixed externally on the frame. Meanwhile, in cars, we’re seeing batteries positioned and moved horizontally.

Yet another aspect to be considered is geography. Many of today’s operators have long-term plans of scaling up their networks to other markets with different sets of natural factors. For example, China (the home market of Nio) and Taiwan (Gogoro) feature mild maritime climates, however, the companies are well trained to deal with troubles such as heavy rainfall, flooding, salt water and earthquakes. In this year, Nio has entered Norway where it has to sustain operation in severe frosts, snow storms and icy rain. Meanwhile, Gogoro will shortly expand to Israel, India and Indonesia where the weather is 10°C to 20°C hotter than in Taiwan and the air is dusty and dirty. In India and Indonesia, batteries are exposed to much vibration and shocks due to poor road quality. This is why I’m following these companies’ progression in the new markets with keen interest.
The most challenging part, to me, is to ensure the battery management system remains contemporary. Almost all swappable batteries around the world belong to what is known as ‘smart’ batteries meaning they’re equipped with programmable IoT-enhanced controls which makes swap operators dependent on the shift from older digital technologies to newer ones. Digital technologies evolve fast and who knows when things thought to be smart today will become dumb and lose compatibility with other parts of the digital environment tomorrow. For example, it’s time to weigh the necessity of 5G and 6G compatibility of the BMS.
(Give me feedback: would you find it useful to have a battery future-proofing checklist?)

Swappable batteries inherit outer design, power output and data exchange protocols of their precursors.

Use case: Gogoro’s four generations

Gogoro claims that at least some of the first-generation modules produced in 2015 are still in service. Currently, three generations of batteries are offered at different price tags, compatible with all its stations and vehicles in service regardless of when these were made. While the company estimates that its third-generation modules will last for some twenty years, it has already presented a solid-state fourth-generation prototype, fully interoperable as well. Expectedly, it will be brought to mass production by 2025 or 2026 to be used side by side with older LNMC batteries. With the higher energy density of solid-state modules, small city cars can be allowed into the scheme, as is envisioned by its headquarters.

Reaching the end of life in mobility, these packs are gradually re-purposed for other applications. The Taiwanese operator is working on a backup system for power parking meters, traffic lights and 5G cells and other applications where portability makes sense.

All this wouldn’t be possible if Gogoro’s initial design developed seven years ago wasn’t so ageless.

Battery packs by Kymco, Power.Global and Honda aren’t interoperable with Gogoro’s module, the oldest of them all. Yet, it’s evidently a prototype for them.

Form factor

Apparently, the industry will need a few more years to work out the optimal form factor for each type of vehicle. It is visible that about ten typical designs are in use by now. However, these will hardly last forever. Some of them can die out when (perhaps “if” is a better word) interchangeability standards similar to sim cards emerge. Quite possibly, new and more successful form factors will be invented.

Shifting from one form factor to another is challenging yet possible. The mobile industry has presented an example of resolving this issue with style when it seamlessly switched from full-size sim cards to minis, then micros and nanos. (c) Wikipedia.

Micro-vehicles: manual swapping by untrained private users

In swap services for private e-scooters, e-bikes and trikes, even micro-cars, batteries are moved by service users. For this reason, they shouldn’t be heavier than 10 kg. Being the most visible part of an operator’s brand, a battery should have an attractive look. Its shape should also ensure against incorrect plugging. The case must have strong protection against vandalism, theft, shock from falling and water ingress.

Micro-vehicles: manual swapping by trained users

Swap services for professional drivers such as rickshaws, delivery couriers or municipal staff allow to put down some of those safeguards because this group of clients can be trained. For practicality, modules can be made bigger and heavier. In the majority of instances, it lies in the range 10 to 15 kg. Batteries can also have pragmatic simpler outer designs.

Batteries used by trained clients such as rickshaw drivers can feature standard tethered plugs instead of special connectors for simplicity reasons. (c) Battery Smart.
Italia-based XEV offers the heaviest manually swapped batteries known by now, supposed to be handled by service assistants not customers.

Robotic swapping for cars

In today’s battery swapping systems (BSS) for passenger cars and light commercial vehicles, batteries are manipulated by robots. It helps to eliminate risks inevitable in manual swapping such as falling, vandalism or theft. In other aspects, the aforementioned principles are applicable.

One interesting question is, what size of modules is optimal for cars. Should a car carry a single or a modular pack? Chinese companies such as Nio, Geely and Aulton New Energy operate large single batteries. On the contrary, San Francisco-based Ample uses mini-modules sized like a shoebox. At the current level of chemistry development, each holds about 3 kWh. A single battery is key to very fast swaps, as short as one minute, achieved by the Chinese companies. On the other hand, Ample’s small modules allow for much flexibility in regards to a vehicle’s size. Three or five modules is enough for a micro-car while 30 or even 60 are used in a full size truck (if GMC Hummer was equipped with Ample’s batteries, a 100% top-up could take less than twenty minutes without detrimental effects of fast charging). Moreover, owing to mini-modules, Ample has developed the most compact swap station in the market, the size of one parking lot.

Interestingly, a few months ago, battery manufacturer CATL entered the swapping market with a form factor called Choco Pack sited in between full-size batteries and Ample’s minis. With a capacity of 26.5 kWh, one module suits a city car. A large SUV or a van can carry up to three of them. CATL’s product was presented only recently, so there is no information yet on how it does.

Batteries of Ample, CATL and Nio.

Robotic swapping for heavy-duty trucks

In trucks, Chinese companies such as XCMG and Geely have the longest history of swap services for trucks. At these companies, a vertically extended LFP battery is positioned behind the cabin. It’s hardly a perfect solution because of a high centre of masses and reduced space for a payload. A different design was suggested by Australian startup Janus Electric, expected to enter the market later in this year. In it, batteries are attached to a vehicle frame’s flank where we used to see a fuel tank in ICE trucks.

XCMG vs Janus Electric.

Use case: Vandal resistance

Swappable batteries of scooter sharing operator Tier Mobility are protected from theft and vandalism by an electronic lock (car trunk’s type) and watertight aluminum cases. The battery’s handle is designed to withstand 380 kg of horizontal pull force and 700 kg of vertical pull force before it breaks off. Also, the module must survive a fall of five feet onto a hard, wet and muddy pavement, tolerate grease and sand on connectors and locks.

On the contrary, Gogoro’s battery handles are less strong than a swap station’s locks. If a thief tries to forcefully pull it out of the slot, all they get is the handle. A serviceman can replace the handle right at the station without taking the battery to the warehouse. In the worst cases, the battery can be remotely disabled.

Smart BMS

While manufacturers of all kinds of batteries increasingly adopt smart BMSs, the adoption levels at swap networks is already very high. In fact, a swappable battery is no longer a vehicle’s part. Instead, it’s more of an external device connected to it. While being moved from a vehicle to station then to another vehicle, it keeps monitoring its state and adjusting the discharge, storage and charging modes. It also exchanges data with the operator’s cloud platform.

Effectively, all major BSS operators use programmable connected BMSs saying their role in the prevention of battery degradation is even greater than that of the quality of the cells. For example, such a system must be able to automatically lower the power output in case of a danger of overheating and degradation or fire. The inventory of batteries is an expensive asset, so the perfection of a BMS strongly impacts an operator’s financial bottom line.

The BMS must also be capable of dealing with risks specific to BSSs. For example, in manual swapping, there’s a non-zero rate of users inadvertently drop the battery. A shock from falling can cause short circuiting or a crack in the battery case. Water can leak in. In all such cases, the BMS should be able to detect the damage on its own, alert the cloud platform and disable the battery.

In turn, data exchange helps to improve efficiency on the level of the network of stations. Data supplied by batteries is used to predict where and when a certain user will come in for the next swap and understand what type of a battery is optimal for their driving style, identify locations for installing additional swap stations. Based on the record of a battery’s performance, it can be reprogrammed in case of any signs of early degradation to extend its life in mobility.

(c) Squared Away.

Use case: artificial intelligence of BSS

Horace Luke, head of Gogoro, described the role of smart batteries as follows:

“Think of the battery network as a neural network. It’s a great learning machine. Each time you put in a battery, we learn from that (in order) to improve consumer experience. We know exactly from people’s behaviour when they gonna put in the (battery) or how much it has been used. And we can predict when people gonna come back. Actually, we’re smart about when to charge a battery and which battery will go to who. If you’re a guy who likes to squeeze the throttle hard, we’re gonna give you the battery that performs because I want to put wind on your face. If you’re a user who rides slower and longer ranges, I’ve got a battery for you too. We condition and program each battery based on how you use it.”

Use case: smart batteries mandated in India

In 2023, this trend towards smart batteries will be fixed in India’s future regulation of the battery swapping market. The draft document mandates an IoT module, thermal sensors and automatic disabling in case of a thermal runaway. Operators are also required to keep a record of parameters of batteries for life. Remarkably, all major Indian swap operators are already using smart batteries – an evidence of its strong impact on the system’s overall efficiency and loss reduction.

This is my first article on swappable battery design (possibly, it’s the first ever publicly available article on this topic). So I need your feedback to know if this work makes sense for you and how it can be improved. I also invite you to follow my b-swapping blog on LinkedIn.

Roma Nazarov is a mobility journalist focused on battery swapping systems. His special interest lies in the areas of research, regulation, conceptual development and consumer uptake.

Leave a Comment