Tesla 4680 batteries...

[RagingGrandpa]

But will Tesla ever sell their cells?

[InfiniteWheelie]

1 hour ago, RagingGrandpa said:

But will Tesla ever sell their cells?

No, but numerous other companies will.

Edited December 19, 2023 by InfiniteWheelie

[meepmeepmayer]

Do we know any potential specs of those cells? Amps? Capacity?

[InfiniteWheelie]

@meepmeepmayer Well generally speaking they’re 5x the size and have 5x the capacity (25ah). Thanks to the tab-less design, resistance is supposed to be comparable to current cells, whereas normally bigger cells would have much worse resistance.

Nobody is mass manufacturing them yet except Tesla, but other companies are supposedly coming out with their own. How long that takes is anyone’s guess. I also suspect that (unlike with 2170s) the whole supply may go to cars for quite a while since that’s what they’re uniquely suited to.

Edited December 20, 2023 by InfiniteWheelie

[meepmeepmayer]

@InfiniteWheelieI was just wondering, as you drew 1p and 2p batteries, if the amperage would even be enough for our EUCs, or if these type of cells wouldn't be suited because they can't give enough amps. They would have to be very high discharge cells to match the amperage our current batteries can put out.

[InfiniteWheelie]

@meepmeepmayer The short answer is that there’s zero issue using 1p when it comes to power. What matters is amp hours x parallels x c-rate (assuming equal voltage).

Look at the Lynx for example: 5ah cells x 4p x 5c = 100a. Then look at the 20” wheel from my drawing. We don’t know the c-rate of various 4680 cells yet, but let’s assume it’s the same for demonstration purposes: 25ah cells x 1p x 5c = 125a.

As you can see, they don’t need to be “very high discharge cells” just to match current batteries. With the exact same c-rate, you can have 5x less parallel connections for the same power, because each cell has 5x more ah.

Edited December 20, 2023 by InfiniteWheelie

[Planemo]

3 hours ago, InfiniteWheelie said:

Look at the Lynx for example: 5ah cells x 4p x 5c = 100a.

Out of interest, where are you getting your c=5 figure from?

[InfiniteWheelie]

@Planemo Here’s the data sheet. It says the max continuous discharge is 25a (5c).

https://www.dnkpower.com/wp-content/uploads/2022/07/SAMSUNG-INR21700-50S-Cell-Specification.pdf

Edited December 20, 2023 by InfiniteWheelie

[RagingGrandpa]

4 hours ago, InfiniteWheelie said:

Lynx for example: 5ah cells x 4p x 5c = 100a.

While Lynx uses 21700-50S cells that are rated for 5C discharge, Lynx will not sustain such a high current continuously. The cells would overheat; and the ~14kW system power will overheat other components in the EUC as well. 

My point: it's not actually 100A battery current.
More like 50A @ 140V, in rare high-speed hillclimb scenarios. 
 

4 hours ago, InfiniteWheelie said:

you can have 5x less parallel connections for the same power, because each cell has 5x more ah

The cells are physically 5x larger, so yes, in terms of stress to the Li-Ion chemical process, one 4680 is practically no different from five 2170's... 

But! 
... the interconnects also need to be rated for 5x current. 4680 promised tabless bottom connection, but it still needs a top connection lead internal to the cell, which will have to bear this ~60A by itself. And, suitably huge weld strips in the pack, to go with it. 

It's doable, but challenging.

And probably irrelevant, since Tesla won't sell cells 

[Planemo]

1 hour ago, InfiniteWheelie said:

@Planemo Here’s the data sheet. It says the max continuous discharge is 25a (5c).

https://www.dnkpower.com/wp-content/uploads/2022/07/SAMSUNG-INR21700-50S-Cell-Specification.pdf

Holy crap I wasn't aware the 50S had a 25A discharge. I note from another review they hit 85degC at that discharge rate though, pretty scary temps and one would assume a battery cooling system would be needed for an application drawing that much current.

In any event, your formula could be written even easier for a 4P wheel, simply 4x25A=100A

And if @RagingGrandpa is correct, it does mean the 50S wouldn't be stressing at all at an output of 10~15A per cell which is nice. I guess thats what makes them work OK on a 4P setup, which until the 50S scared me a bit.

[alcatraz]

11 hours ago, Planemo said:

I wasn't aware the 50S had a 25A discharge. I note from another review they hit 85degC at that discharge rate though, pretty scary temps and one would assume a battery cooling system would be needed for an application drawing that much current.

Good point. 

Samsung can spec a generous C-rate on the data sheet but as long as they overheat you can't reliably pull that much current. A lab and vehicle environment are very different. 

It's a good cell but it's not going to suit every situation. It's best if we don't treat it as "the last cell we're ever going to need".

For our use case, you can be sure that Samsung isn't going to take any accountability if there were to be any failures.  If they're not accountable there's no incentive to be extremely careful with the specs in the data sheet.

Edited December 21, 2023 by alcatraz