The Tesla 4680 cell has intrigued ever since it was announced. A cylindrical cell that is 46mm in diameter and 80mm high.
- Cell format: 4680
- Diameter: 46mm
- Height: 80mm
- Volume: 0.1330 litres (outer volume)
- Cell case
- stamped and drawn nickel plated steel or stainless 
- 0.5 to 0.6mm thick
Capacity tests :
Capacity: 23.35Ah at 2.5A discharge ~C/10
Capacity: 20Ah at 25A discharge ~1C
26.5Ah (estimate based on 21700 5Ah volumetric energy density) and this fits with capacity of the Model Y pack that uses this cell.
The Laboratory for Energy Storage and Conversion carried out the testing and data analysis of the two 4680 cells reported in this article. The goal of the Laboratory for Energy Storage and Conversion (LESC), at the University of California San Diego Nanoengineering department and the University of Chicago Pritzker School of Molecular Engineering, is to design and develop new functional nano-materials and nano-structures for advanced energy storage and conversion applications.
- Nominal Voltage: 3.7V
- Nominal Energy: 86.5Wh
- Mass: 0.355kg (confirmed for both ells)
The internal resistance is a key parameter and was measured using the HPPC pulse test data .
- DCIR (charge/discharge): 7mΩ +/- 2
- Cell gravimetric energy density = 244 Wh/kg
- Cell volumetric energy density = 650 Wh/litre
- Cathode: NMC811
|Element||Batch 1 Atomic %||Batch 2 Atomic %|
- solvent based slurry and application
- cathode thickness 85µm
- porosity 30-40% (based on estimations by Addionics)
- aluminium current collector thickness 9 to 10µm 
- aluminium “flower” end disc thickness 400µm 
- carbon, no silicon 
- dry based electrode 
- anode thickness ~121µm
- copper current collector thickness 6 to 8µm 
- copper “flower” end disc thickness 260µm 
Same anode thickness (areal loading) on both of the cells tested.
The same silk-like morphology was identified, indicating the dry electrode method.
Does anybody know what separator this cell is using? Any modifications to accommodate the dry anode? Ceramic coated? Thickness? Any special process with respect to electrolyte fill and wetting?
How much of it? and what additives?
What you immediately notice with this cell is that the crimped closure is on the negative, bottom, of the cell (the top of the image on the left). This is in contrast to the 18650 and 21700 cells where the cell is crimped closed with an isolating and sealing gasket around the +ve, top end of the cell.
This means that the positive button on the top of the cell is pre-installed into the empty can shell and is isolated from the can.
Online recording of a discussion around the manufacturing process for the 4680 cell.
The 4680 manufacturing process appears to be:
- roll the anode, cathode and separator (appears they cut into the tabs ready to fold at this point)
- fold the tabs over – this must result in wrinkles on the edge of the active materials
- laser weld the cathode and anode discs to the jelly roll
- fit the positive button into the can with isolator (this might arrive pre-assembled)
- push the jelly roll into the can and weld the cathode disc to the underside of the +ve button though the hole in the anode disc and centre of the jelly roll (very similar to the current 18650/21700 process for the anode connection to the bottom of the can)
- weld the anode disc around the edge to the inside of the can
- fit the can closing disc and crimp this into the can (this step has been changed apparently as the crimp was leaking)
- fill with electrolyte through the hole in the centre of the closing disc
- close/weld the hole in the closing disc
Stay up to date with our posts by signing up to our newsletter.
Based on this we have constructed a schematic that sows the cross-section through this cell.
It appears that this is an old design. The following are a couple of issues described online:
- The can closing process is different. The design in the picture shows a weakness, which makes it difficult to get it work, in the closing process since is missing the gasket between lid and can so even with grooving + crimping process you cannot ensure hermetically sealing, and you waste too much space in the can.
- Even the cathode disc design is different from the picture above. No reason to do it like that, you loss energy density (no space optimization).
It appears that this cell is designed to rupture or un-crimp the bottom disc in the event of the cell going over pressure. The next video shows the cells being assembled into a battery pack that appears to show serpentine side cooling. This would make sense if the bottom of the cell is designed to be the venting path.
Tabbed vs Tabless
The tabless jelly roll significantly improves the electrical and thermal connections. Tranter et al  have analysed this design and looked at the electrochemical and thermal behaviour.
The new design is found to mitigate the ohmic losses experienced around the “jelly-roll” current collectors which are significant for the traditional tabbed case, thus leading to higher efficiency and capacity and reduced heat production.T. G. Tranter, R. Timms, P. R. Shearing and D. J. L. Brett, “Communication—Prediction of Thermal Issues for Larger Format 4680 Cylindrical Cells and Their Mitigation with Enhanced Current Collection“, Journal of The Electrochemical Society, Volume 167, Number 16
- Cell venting
- appears to be from the bottom negative end
- Cell fusing
- no internal fuse
There are a lot of other companies jumping on the 46mm diameter cylindrical cell format:
- LG Chem
- Panasonic GREEN IMPACT – New Environmental Concept for a Sustainable Future
- Tesla 4680 Teardown: Specs Revealed! (Part 2) – The Limiting Factor
- Part 1: Tesla 4680 Teardown // Cell Disassembly // 4 hours in 1 hour – The Limiting Factor
- T. G. Tranter, R. Timms, P. R. Shearing and D. J. L. Brett, “Communication—Prediction of Thermal Issues for Larger Format 4680 Cylindrical Cells and Their Mitigation with Enhanced Current Collection“, Journal of The Electrochemical Society, Volume 167, Number 16
- Not only tesla 4680 battery, which battery manufacturers are in the layout of 4680, TycoRun Energy
- UC San Diego analysis, The Limiting Factor
Note: we will update this page as more data becomes available.
Some data, some estimates and gradually we can piece together what this battery pack is and can do.
The cell manufacturing process is laid out in 14 steps covering everything from mixing chemicals, dryers, printing and electrical testing.
The BMW strategy for 2025 onwards is to use a 46mm diameter cell like the Tesla 4680, but use it in two lengths. The 95mm long cell will be for saloon cars and the 120mm cell for SUV’s.
A recent concept study by Speira, PEM Motion and the Chair of Production Engineering of E-Mobility Components RWTH-Aachen (available at speira.com/publications) outlines potential benefits of an aluminium cell housing based on a 4680 concept cell, comprising an aluminium cell housing compared to a reference LIB cell with commonly used nickel-plated steel housing.