Tesla Model S Battery Pack Teardown

We’ve heard a lot about the Tesla Model S over the last few years, it’s a vehicle with a habit of being newsworthy. And as a fast luxury electric saloon car with a range of over 300 miles per charge depending on the model, its publicity is deserved, and that’s before we’ve even mentioned autonomous driving  driver-assist. Even the best of the competing mass-produced electric cars of the moment look inferior beside it.

Tesla famously build their battery packs from standard 18650 lithium-ion cells, but it’s safe to say that the pack in the Model S has little in common with your laptop battery. Fortunately for those of a curious nature, [Jehu Garcia] has posted a video showing the folks at EV West tearing down a Model S pack from a scrap car, so we can follow them through its construction.

The most obvious thing about this pack is its sheer size, this is a large item that takes up most of the space under the car. We’re shown a previous generation Tesla pack for comparison, that is much smaller. Eye-watering performance and range come at a price, and we’re seeing it here in front of us.

The standard of construction appears to be very high indeed, which makes sense as this is not merely a performance part but a safety critical one. Owners of mobile phones beset by fires will testify to this, and the Tesla’s capacity for conflagration or electrical hazard is proportionately larger. The chassis and outer cover are held together by a huge array of bolts and Torx screws, and as they comment, each one is marked as having been tightened to a particular torque setting.

Under the cover is a second cover that is glued down, this needs to be carefully pried off to reveal the modules and their cells. The coolant is drained, and the modules disconnected. This last task is particularly hazardous, as the pack delivers hundreds of volts DC at a very low impedance. Then each of the sixteen packs can be carefully removed. The packs each contain 444 cells, the pack voltage is 24 V, and the energy stored is 5.3 kWh.

The video is below the break. We can’t help noticing some of the rather tasty automotive objects of desire in their lot.

We’ve shown you a Model S teardown before, but without a video. For comparison, take a look at teardowns of a Nissan Leaf pack, and the NiMH pack from a Ford Fusion hybrid.

Via Hacked Gadgets.

78 thoughts on “Tesla Model S Battery Pack Teardown

          1. “Shelf life” really only refers to it growing outside its useful purpose. Fuses and a quality BMS go a long way to make things useful even when reusing for things like storage solutions. Capacity tends to go down over time, and internal resistance of the cells tends to rise which will limit its safe current discharge capability, but there isn’t a defined “shelf life” in terms of it no longer working at all.

          1. Who gives a shit who’s doing it, if it’s getting done, that’s all that matters. And I’m sure whoever’s doing it has gotten pretty good at getting them apart quickly. Sounds like a non-issue to me.

        1. You have to remember, it’s one thing what company execs say, and another thing what they do.

          Of course Mr. Musk says recycling is important, because the press would hang him if he said otherwise. However, for the business case he has, it’s cheaper to make batteries out of virgin materials bought in bulk – that way you don’t have to invest in the recycling and refining equipment. If somebody else wants to do it, they can – it’s simply not Tesla’s problem and not Tesla’s money.

          1. Depends on the material and ease of recycling though. Some goods are extremely easy to recycle, especially in bulk. Unclear on how easy these batteries are to recycle though.

          2. >” Unclear on how easy these batteries are to recycle though.”

            Quite hard. Pulling them apart is the easy bit. The cells themselves need to be disassembled in inert atmosphere because the contents burst to flames on exposure to air. The cell layers are separated, ground up, and then refined back to pure lithium, aluminium, cobalt, manganese etc.

            It’s all technology that Tesla doesn’t have, and doesn’t want to invest in because the payback would be decades.

          3. Do they really have to be disassembled in a more or less controlled way or could it be possible to shredder them under inert gas? Then somehow ‘wash’ (not with water) the flammable electrolytes and process the rest? There is no metallic lithium in LiIon batteries.

        1. Even better reason : warranty period.
          A significant number of newer cars are now released without a drain plug on the transmission. Why? It saves about a cent, and statistics indicate that the warranty period will expire before a problem arises, and then it is someone else’s problem. One cent times one hundred thousand cars… that could go into my bank account please.

          1. I’m sure it is worth more than a cent.

            Besides, the dealerships replace the fluid through the lines, you flush the torque converter that way. If you’re dropping the pan, siphoning is just as easy. Why install a plug that won’t be used and might leak?

          2. Even better reason: heat…(close to lithium cells – not recommended)

            And honestly lots of bolts and screws seems to be good enough for rockets, huge steam locomotives, u-boots, etc. it can’t be bad.

    1. They look pretty easy to recycle. with a decent jig to hold the battery pack vertically 2 guys with the right tools would strip take all the screws out in a few minutes, pull the cells out in a few more minutes then send them all off to their respective shredders to recycle.

          1. http://batteryuniversity.com/learn/article/recycling_batteries

            “The recycling process begins by removing the combustible material, such as plastics and insulation, with a gas-fired thermal oxidizer. Polluting particles created by the burning process are eliminated by the plant’s scrubber before release into the atmosphere. This leaves the clean and naked cells with metal content.”

            “To reduce the possibility of a reactive event during crushing, some recyclers use a liquid solution or freeze lithium-based batteries with liquid nitrogen; however, mixing Li-ion starter batteries with the common lead acid type still remains a problem as a charged Li-ion is far more explosive than lead acid.”

            “Battery recycling is energy intensive. Reports reveal that it takes 6 to 10 times more energy to reclaim metals from some recycled batteries than from mining.”

          2. Of course it is no good idea to mix up different chemistries. Why mix up at first, what you want separate in the end. Lithium lead alloys are of very limited use.

        1. https://www.technologyreview.com/s/414707/lithium-battery-recycling-gets-a-boost/

          “Larger batteries that might still hold a charge are cryogenically frozen with liquid nitrogen before being hammered and shredded; at -325 degrees Fahrenheit, the reactivity of the cells is reduced to zero. Lithium is then extracted by flooding the battery chambers in a caustic bath that dissolves lithium salts, which are filtered out and used to produce lithium carbonate. The remaining sludge is processed to recover cobalt, which is used to make battery electrodes.”

      1. I’m sure even many of the “bad” (worn out) cells would work like a charm in less demanding applications like portable electronics. Put the used cells in a test jig to reject the total duds. Then use the cells to make aftermarket laptop batteries, USB battery packs, and whatnot.

        Or just advertise to the DIY off gridders and they’ll buy entire battery modules if the price is right.

        1. The demand on the battery is not that high. Ok, sometimes high peak power for a very short time, but in average it is a 5h to 10h discharge. The range is ~500km and an average speed of >100km/h is rarely achievable, in city driving the average is mostly in the range of 30km/h.
          I would not want that worn out car battery cells in any portable application (Laptop,USB or torch) where capacity is an issue. It can be acceptable for a stationary application (solar storage)

    1. Where’d you get the 2208 amps?

      5.3 kWh is the capacity, not the power rating. Tesla pulls 568 kW power out of a 100 kWh pack for the “ludicurous mode” which pulls as much out of the batteries as they can give, which implies the cells are rated for 6C discharge, and the maximum safe discharge would be 1235 Amps.

      1. I doubt ludicrous mode is pulling every last mA out of the batteries, I’ve watched several videos of the P100D launching and it maintains a remarkable amount of traction which is a clear indicator that its limiting power output to keep the tyres from spinning. 568kW (760+HP) is more than enough to break traction on all 4 wheels with a big instant torque hit like the Telsa’s deliver.

        1. Of course it must limit torque to the wheels to avoid wheelspin. Dyno results for the P100D posted to youtube show it doesn’t approach maximum power output (almost 600 hp to the wheels) until 50 mph or so. For the P100D to hit 60 mph in the measured 1.1g. At 60 mph though, it’s power limited to around 0.7g, implying the time up to 50 mph must be close to a constant 1.3g. Holy frack that’s pretty amazing traction performance. My Subaru, with its winter shoes on normal pavement, can’t even manage around 0.5g.

          1. sigh. goofed with the less-than and greater-than again… Fine. eff HTML. Remember FORTRAN?
            should be:
            …to hit 60 mph in the measured .LT. 2.4s it must pull an average .GT. 1.1g. …

          2. Your subaru doesn’t have a thousand pounds of batteries in the undercarriage.
            Cool… so if I put cement a thousand pounds of concrete on the floor I can corner better?

            Unfortunately, at least in my corner of the universe, coefficient of friction doesn’t work that way. I’d be better off putting the summer rubber back on to get back up to its usual 0.7-0.8g (yes, I’ve measured it).

          3. >”Cool… so if I put cement a thousand pounds of concrete on the floor I can corner better?”

            No, but you get better traction.

            >”coefficient of friction doesn’t work that way.”

            Not the -coefficient- but the absolute tractive force. Add more mass on the axle, and you can apply more torque to the wheel before it starts to slip.


            “Traction between two surfaces depends on several factors:”
            “-Normal force pressing contact surfaces together.”

            Plus, the weight distribution of the vehicle makes a difference. If a FWD car the acceleration causes the front to lift up as the mass of the car tries to pivot around the back wheels, which reduces the normal force and therefore traction at the driving wheels. In a RWD car the opposite happens: the rotational inertia pushes the rear wheels to the ground and gives them more traction.

            Plus, winter tires on dry pavement have shit traction.

          4. Whoah! You mean in your universe the force weighing down on an object doesn’t get multiplied with said coefficient of friction to get how much force has to be applied between it and a surface before it starts slipping? Do tell me more! Have you got flying pigs over there too…?

          5. Of course adding weight will increase traction and potential tractive force (and even your dyno results). But it is naive (or disingenuous) to say it will increase your acceleration.

        2. The reason why I think the Ludicurous mode is pulling all the juice the batteries can give is because of two reasons.

          1) it works only when the battery is full. 95% full. With a lower SoC the acceleration drops and the ludicurous mode becomes unavailable.
          2) it gets faster acceleration every time they add more batteries. From the 90D to 100D the acceleration improved from 2.8 to 2.5 seconds.

          Those two are indications that they’re running it on the limit of what the batteries can supply. The motors may be rated for 568 kW but the batteries don’t seem to manage that much.

          1. And here’s further evidence:
            The earlier P85D battery pack was protected by a 1300 amp passive fuse, while the P90D pack has an active 1500 amp fuse with a more precise cutoff, which enables the whole ludicurous mode to work – they’re pulling right up to the limit which wasn’t possible previously because it might have blown the fuse.

            The reason why the ludicurous mode only works with a full battery is exactly because the battery voltage drops with the state of charge, so with a hard limit at 1500 amps they can only get so much power out. That’s why every time they add more batteries, they get more voltage, they get faster acceleration.

            The 85D battery has 16×6 cells in series for 96 times the cell voltage, and if it’s charged up to 4.3 volts then that’s 412.8 V x 1500 A = 619 kW theoretical power output (7C discharge rate). In reality the actual maximum power output will be less because of the ESR of the battery cells which causes a voltage drop, plus the motor drive electronics which also introduce more loss.

            So why don’t they lift the 1500 Amp limiter? Because the batteries would blow up – it’s the batteries that are limiting the acceleration because they’re running them as hard as they dare.

          2. There’s probably at least some headroom left, if only because they want some kind of margin of safety. But certainly not a hugely useful headroom, just enough to be sure they don’t cause any fireworks at “max” power.

    2. I have a 1320ah bank of lead acid-cells for off-grid – 12 x 2VDC cells cabled in series to a nominal 24VDC.

      One of the mandated warning signs says “DANGER. Prospectve fault currents of up to 10000 amps can be expected”

      Yes, ten thousand amps. Hmmmmmm.

        1. The amp-hour rating of lead acid batteries is measured over a 20 hour discharge. It gets much worse in a short circuit. A regular 60 Ah car battery can supply around 400 amps for 30 seconds before the voltage crashes down. That’s just 3.3 Amp-hours in total.

          Likewise, if you were to measure the CCA of a 1320 Amp-hour lead battery, it’d probably be around 9000 amps for 30 seconds.

          1. Not that I would destroy my family’s energy bank for trivial (yet spectacular) reasons, but 9000 amps for 30 seconds… makes me think of those jacob’s ladder videos on youtube, and whether I could make something quite so entertaining from +ve to the -ve terminals and post it on facebook……….

    3. You don’t even have to be zapped, the 0.5MW this thing can easily deliver can cause an arc flash powerfull enough to seriously burn you with just IR, light and UV radiation…to top it off, it would probably set your clothes on fire…

    1. Being cut by a piece of metal should be the least of your worries when disassembling a pack like this…
      If you drop something conductive onto those busbars and it shorts them, it’s time to run like fuck, the pack will literally self-destruct :P

    1. They are not, otherwise you would see proper 1000V tools like these:


      However, I also don’t know of many average auto shops that have regularly have sets of special high voltage tools like that, unless they moonlight as electricians or otherwise regularly deal in high voltage. That doesn’t obviate the fact that they should have said high voltage tools when working on these cells but it’s just not something you would generally commonly see in your average auto shop.

      However, this seems to be a specialty EV auto shop, so, yes, they really should have these tools on hand.

      Lastly, they are selling those, what, 15 modules for $1,375.00 each. Implying a value of $20,625.

      They are about $1,200 to $1,250 each on eBay and other used markets.

    2. Is the battery floating, or is one end connected to the casing?

      If it’s floating (or connected via high impedance for earth-fault detection), the risk of electric shock is much lower.

      Insulated tools are still a great idea though, if you drop a screwdriver near the output terminals you’re likely to be sprayed with droplets of not-quite-vaporised screwdriver.

    1. Many of these cars would likely be sold at dealer auctions. Insurance companies tend to sell or auction the cars that are totalled, regardless of the reason the car is declared a total loss. Some owners choose to ‘buy back’ the car to fix themselves, often times they do not. Repair shops and dealerships sometimes buy these ‘wrecks’ to obtain rare or ‘cheaper’ OEM parts, especially parts which are not readily available or would be otherwise cost prohibitive.

      there are other safety concerns when dealing with wrecked electric cars, but I imagine the actual disposition process here in the US is the same as any other car in the end.

  1. There’s a correction that needs to be made here for technical accuracy. While the outer aluminum can looks like a standard Panasonic battery the inards for the Tesla batteries are exclusive to only Tesla. Panasonic manufactures the Tesla batteries on dedicated production lines, the chemical composition of the electrolyte is different than in the standard Panasonic battery, there are also additional steps and quality control checks used during production that are exclusive only for the Tesla batteries. The OSHA required Safety Data Sheets for the electrolyte reveal it’s a far more corrosive make up, where with the standard Panasonic batteries the electrolyte used is far more flammable. Testing for quality control is extensive during every step of production, Panasonic is absolutely anal in only allowing 100% perfect batteries to leave their facilities, right down to final microscopic inspections just for cosmetic defects in the final product, zero defects of any kind are allowed to pass.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.