Tesla has been hinting at some revolutionary new battery technology for a few months now, and a Reuters story today revealed that the goals are to create a “million mile” battery design that can finally make EVs on par with the cost of building ICE vehicles. There are lots of elements to what will make this possible, but for the moment I’d like to focus on the one that doesn’t seem to require an advanced degree in chemistry: the switch to “cell-to-pack” batteries.
The cell-to-pack battery design is interesting because of what it eliminates, rather than adds. Current EV battery packs are composed of a number of individual modules ganged together to form the full battery pack. As these packs have developed over the years, the numbers of modules has decreased.
The Model S has a battery pack with 16 individual modules of battery cells, while the later Model 3 reduces that number to four long modules. Having separate modules could be a benefit if they were easy to swap out individually, but they’re all sealed inside the battery pack, so that’s not the case.
More modules means more hardware for enclosing each additional module, connecting that module electrically, connecting it to the thermal management system, and so on, and all that extra hardware adds weight and cost to the battery pack.
The thinking with cell-to-pack batteries, which have been pioneered primarily by the Chinese company Contemporary Amperex Technology Company, Limited (CATL), is that it would be better, cheaper, and simpler to just integrate the individual battery cells directly into the full battery pack, without dividing it up into individual modules.
There’s some real advantages here: CATL claims it can increase the mass to energy density of the battery by 10 to 15 percent, the batteries will take up 15 to 20 percent less space, and it can cut the amount of parts needed to build packs by up to 40 percent.
The new batteries Tesla has been talking about have been developed jointly with CATL, and are expected to employ the cell-to-pack designs, which should be a key part in getting the cost of the battery packs down.
This new approach to battery design will likely require a redesign of Tesla’s thermal management system, which we explored in detail before. Essentially, the Model 3's battery module uses lasagna-noodle-like aluminum cooling conduits that lace through the array of cylindrical cells inside the battery module:
The advantage to this more complex setup is primarily that it allows for better and quicker fast charging, since this design allows the system to remove heat from the battery cells quickly enough to allow for really rapid charging.
This approach, of course, is more labor and resource intensive than just packing cells into a battery pack, which I suspect will make it incompatible with the new simplified battery designs developed by Tesla and CATL.
We can get an idea of what the new cell-to-pack thermal management system may be like from this Tesla patent dated October 10, 2019.
The battery pack shown in the patent seems to be a simplified cell-to-pack design, with some interesting approaches taken to streamlining the overall design.
In this diagram from the patent, we see that cells are made right at the factory in already-clustered subgroups, with an array of cylindrical cells sharing a common, integrated rectangular upper plate that also seems to house the positive terminal tabs.
The panels above it (520 and 522 in the diagram) appear to be layers of a combined circuit board to accommodate the wiring and an integrated outer panel that actually seals the housing of the battery pack.
Instead of the complex lasagna noodle-style of coolant channels, it looks like they’re thinking of just pumping some sort of nonconductive, di-electric coolant through the entire pack, relying on flow between the cylindrical cells. Such a system would require substantially less hardware and manufacturing complexity.
The patent also addresses the possibility of coating the cells with some sort of insulating coating, allowing normal, potentially conductive coolants to be used.
Again, this is just one aspect of the upcoming battery packs, as the fundamental chemistry is crucial as well. But the move to a cell-to-pack design seems a very rational and logical next step in the production of large-scale battery packs, and it’s easy to see the kinds of cost, complexity, space, and weight savings that this new approach will afford.
This patent suggests why we haven’t seen this sort of battery produced on any real scale before--to do it properly, you need to design it from the individual cell level, as Tesla describes in their patent, and so far the EV industry hasn’t quite been ready to do so, or at least hadn’t found it made financial sense, yet.
The final design should be shown at a Tesla “Battery Day” event later this month, so hopefully we’ll get to see how these packs actually turned out soon. I hope they have one cut open to look at, at least.