[셀] Musk Admits Pursuing Dry Battery Electrode in 4680 Cells Was Mistake

출처 : Batterynews.con (2025.11.7.)

출처 : Battery News

Musk Admits That Pursuing the Dry Battery Electrode Process in 4680 Cells Was a Mistake

Five years ago, during 2020 Battery Day, Tesla unveiled the 4680 cells as the most advanced battery tech in the world. However, CEO Elon Musk now regrets he pursued the dry battery electrode process, as he admitted during the 2025 Investor Day. Nevertheless, Tesla will ramp production to use the larger cells in the Cybercab, Tesla Semi, and the Optimus robots.

The 4680 battery cells Tesla unveiled during 2020 Battery Day left everyone wowed thanks to their incredible performance. The larger cells (46 mm diameter versus 21 mm for 2170 cells) promised to hold 5x the energy of 2170 cells and deliver 6x the power. It should’ve also improved an EV’s range by 16% when used in a pack instead of 2170 cells.

The size was not the only thing that mattered, but also the updated design and chemistry. Tesla promised to use a silicone anode and a high-nickel cathode that should’ve improved energy density beyond what was possible for common Li-ion cells at the time. Besides that, a tabless design would’ve improved current flow and thermals inside the cells.

However, the key change was in using the dry battery electrode (DBE) process, which Tesla started developing after it acquired Maxwell in 2019. The dry battery electrode promised to drastically shorten the manufacturing time and cut costs significantly. However, the caveat was that Maxwell developed the process for supercapacitors and never used it for batteries. Musk thought that was a tiny detail that would be easy to overcome.

Tesla 4680 cell failed to deliver

Alas, it wasn’t, and that has turned Tesla’s entire battery efforts upside down. It’s also the main reason the Cybertruck failed miserably, with the 4680 cells delivering poor performance in almost every direction. They hold less energy than 2170 cells, charge more slowly, and the heat builds up inside, further affecting the duration of charging sessions.

In 2022, a teardown revealed that 4680 cells were just ordinary, but larger, NMC 811 cells. There was no silicone anode, no cobalt reduction, and certainly no dry electrode process. This affected the energy density, with the larger cells being significantly worse than regular 2170 cells. This was obvious when the carmaker started building the Model Y with 4680 cells. However, instead of reconfiguring its plans, Tesla doubled down on 4680 cells with the Cybertruck. The result is well known.

Despite repeated claims of progress and even breakthroughs, the 4680 cells remained a tough nut to crack for Tesla. Talent left and came, entire teams were replaced, and tons of paperwork were filed to patent elusive technologies that worked in the lab but not in a factory. For most of 2024, Musk dodged questions about the 4680 cells, while the work continued in the background.

DBE process finally solved?

Later that year, the Tesla battery team announced that the dry battery electrode process was finally solved once the rolling machines were upgraded with more solid parts. The Cybertrucks should’ve started integrating the DBE 4680 cells in the Cybertruck’s battery packs by now, but this didn’t happen.

In the meantime, other battery makers that tried the dry process have already moved on. Their 46-mm battery cells have already entered production with characteristics far surpassing Tesla’s. The best example is the BMW iX3 “Neue Klasse,” which has amazing charging performance that Tesla Cybertruck owners could only dream of.

Speaking on Thursday during the annual shareholder meeting at Giga Texas, Elon Musk admitted he made a mistake when he pushed for the DBE process. The dry process turned out to be “way harder” than he thought. However, he fell short of saying that Tesla has abandoned this Fata Morgana of battery technology. Maybe it did or maybe it didn’t, but what’s certain is that Tesla still advertises jobs related to the DBE process.

4680 cells are still on the table

At the same time, Musk promised the 4680 cell production will ramp up “very dramatically” in 2026. This will be necessary because the production version of the Semi electric truck will use the larger cells instead of the 2170 cells of the pilot vehicles produced until now. Musk also wants the 4680 cells to power the Cybercab and the Optimus robot, and he thinks both products would scale to millions of units per year.

Such high volumes would require a much larger battery cell supply than what Tesla is able to produce right now. This means that Tesla will have to rely on battery suppliers like LGES, Samsung SDI, Panasonic, and CATL for 4680 cells. That’s because Tesla is now in a bit of a catch-22 moment, with several bad solutions to its 4680 cell battery problem.

First, since the DBE process failed, Tesla’s production costs are high, and volumes are low. That’s because the machines on the production lines keep breaking, and that’s even worse than using a wet process and waiting for the electrodes to dry. At the same time, using a wet process (which Tesla should’ve focused on in the first place) would lower the output because production time would increase significantly