This article is a bit confusing because its talking about the theoretical use cases of structural battery packs in EVs, the possible problems, and possible mitigations.
The Tesla Model Y has been using a structural battery pack for over 2 years (since 2022 model year, I think). While Hyundai/Kia may make different choices, the pros and cons from Tesla’s choices can be used as at least one path for what Hyndai/Kia want to keep or what they would want to change from Tesla’s choices.
Here’s the articles cited problems:
However, placing cells inside load-bearing components would also make them hard to replace, offsetting this practice.
In practice almost no service of the battery is done at the shop where the car is. If there is a fault anywhere in the pack, from a cell to a BMS component, the entire pack is swapped.
Most EV drivers rarely use all of their battery range; they only need it on long trips. Therefore, batteries built into load-bearing automotive components would not be needed for daily use; they could be maintained at the ideal charge level for the battery formulation and rarely used except for long trips.
This is such a strange description. This isn’t how EV batteries are used at all today, as far as I know, because it would be very inefficient in regular use. You wouldn’t want to put extra strain/stress/charge/discharge on specific cells in the battery. That would lead to unbalanced packs. Further it would be much less efficient and slower to charge the segregated pack that the article is describing. The more full a cell gets of charge, the harder/more power you need to work it to take additional charge. Also, deep discharge of cells significantly shortens their life. The author’s design description of a segregated pack would also experience much worse cold weather range reduction.
if the carmaker were to use modular battery packs, it would reduce the related cost of repairs across all of the car batteries; if they had better batteries, the need to replace them would be substantially reduced
Several automakers have tried making modular packs the way the author is describing them here. They turn out to be a bad choice because of all the extra stuff you have to do to make it flexible to accept fewer or more modules. This same thing happened to mobile phones. You used to have a removable battery, which was nice. However to make the battery removable, you had to add spring contacts, a separate tray to hold the battery, and a battery door. All of these things took space. The only time you’d ever use these, practically, would be years into the phone’s life you’d open it up to replace the battery once, maybe twice in the phone’s entire life. Manufacturers could just put a larger battery in and make it cheaper. Many phones don’t live long enough to ever kill their first battery.
The author of the article is well credentialed and has worked in the tech and automotive industry for years, so I’m very confused as to the content of this article. Its almost like it was written 3 years ago, even though it is showing a Feb 14th 2025 dateline.
Agreed broadly, but I want to nitpick the phone comparison. I had spare batteries for my smartphone for events like all-day conferences/conventions to swap batteries in the afternoon. I could see battery swapping making sense in urban areas for mopeds and ebikes or even delivery vehicles, but there aren’t use cases for personal cars that make sense to me.
I think the “swapping” may be a different use case the author is talking about. I don’t think the author was referring to an end-user executed swap to simply put in a charged battery.
This would be a service center option where a mechanic would have to take tools and removed panels and connectors to make the swap. Something done maybe only a few times, if ever, for a car during its life.
A structural battery pack is constructed to not be serviced in parts. The author calls this out with his comments on “replacing a single bad cell”. He’s right that this is a concern for structural battery packs. Here’s a Tesla structural battery pack when it was attempted to be disassembled:
There was more of that pink foam wrapping around the cells now exposed. All of that pink foam is needed for strength and its thermal properties because the battery pack is part of the structure of the vehicle carrying load forces.
Clearly replacing a bank of cells would be difficult to do if there was a cell failure, and no wear near cost effective for a consumer to have done on their car. The author is suggesting having some of THIS type of battery, but also another part of the battery in the standard hard plastic modular cases where the whole module could be removed and replaced (“swapped”)
The author is suggesting SOME of the battery be the pink foam type that is unserviceable, and SOME of the battery be behind panels in cases that a technician can swap at a service center when the module has reached the end of its life.
I had spare batteries for my smartphone for events like all-day conferences/conventions to swap batteries in the afternoon.
Sure, but how often are you going to all-day conferences. Once a year? Twice? Is it worth having the possibly 20% less battery capacity the other 363 days a year for that swapability?
Taxies and other service vehicles too. They constantly complain about basically an unpaid break where they have to charge their cars.
Most EV drivers rarely use all of their battery range; they only need it on long trips. Therefore, batteries built into load-bearing automotive components would not be needed for daily use; they could be maintained at the ideal charge level for the battery formulation and rarely used except for long trips
I wonder what the user experience for that would be. If I want some charging target amount for daily use, but the car wants hard-to-reach battery cells to float at 66% or something whereas I want overall 80% of range, maybe the charge selection screen would just suggest “76% for best longevity”? Or hide it all behind “80%” but the ramp to 100% isn’t quite linear?
Something seems missing from the description. What would make the most sense with the articles description is if there were two packs each with different chemistries. One cheap and low density, such as NA-Ion (Sodium-Ion) that could take care of 90% of EV driving needs for trips under, say, 100 miles. The second pack being a much higher density chemistry like NMC (Nickel, Manganese, Cobalt) which could add significantly more range for the same cubic cm of volume in the vehicle, but the battery would have far fewer cycles compared to other chemistries.
The idea being, beat the heck out of the Sodium Ion battery and make it easily serviceable/replaceable at the cost of range for the same volume consumed, while the NMC battery would be gently cared for and only called on in the more extreme demands of range.
If that’s what the author was trying to say, it was not clear.
