Tesla's new 4680 - We're gonna need a bigger flashlight

Yeah…it’s close to my thoughts as well.

4*18650 lights often went for compromises to increase safety, largely related to users putting different cells in different orientation.
Convoy invented a protection disc that increased flashlight length and weight for a little bit more safety.
A number of makers made use of battery carriers which increased weight, length and diameter for another bit of safety.
Volume of a carrier is going to be similar to that of a 46800 cell so these are really close in size while 46800 offers so much more energy than high-capacity 4*18650 with IR that bests high-current 4*18650. If it doesn’t deliver 60A continuous and 100A for 20 seconds I’ll be disappointed.

46800 is safer in that it does’t have the risk of some cells being inserted with correct polarity and others reversed.

At the same time there’s so much energy inside…and due to low IR this can generate so much power….that any damage or short related hazards are going to be much larger.

Maybe not larger than with 8*21700 lights that are out there but still…

If it is approx 5x the current 48G 21700 capacity, then compared to a 4*18650 battery carrier;

VTC6s: 4 * 3.7V * 3Ah = 44.4Wh

46800: 5 * 3.7V * 4.8Ah = 88.8Wh

It may not exactly meet Tesla’s “5x capacity” claim, but even so, it would nearly double the total energy capacity of current 4x18650 lights.

However, the voltage of the single cell will be limited at 3.7V, whereas the 18650 carriers can be configured to 1S4P, 2S2P, or 4S1P, giving a wide voltage range for lots of different emitters more capable of higher powers.

So a moderate to high-power light using this cell would either need to take the Emisar ‘loads of parallel XPLs’ approach (or the 3V XHP50), or start using some serious boost driver tech.

Yeah, with 300W that these should be able to deliver even GXB100 is not powerful enough. At the same time this driver would suddenly become so much more useful. :wink:

Real Tesla ones? Not until cars using them are totaled. I suppose we may see a small number of them slip out one way or another. It has happened with other Tesla cells.

21700 cells of various quality appeared not long after Tesla started using 2170.

I predict we will see Chinese 4680s sooner or later. They may not be as good as the Tesla ones.

Cells from totaled cars and mediocre cells from small makers are not enough to drive adoption.
We need good cell makers to start making them like they did with 21700.
Now cordless tools won’t be among the drivers for this. I’m concerned whether this will happen. If it does - we’ll end up with much better pop can lights.

I suspect that it’s is because Tesla had Panasonic manufacture most of it’s cells, and the 21700 were exactly the same manufacturing process and chemistry that the 18650, with the 4680 things change a lot, tabless cells, dry electode and cell to pack (no module) plus Tesla’s will of vertical integration and projections of in house battery production ramping orders of magnitude.

Will Tesla ever buy a 4680 from an external battery manufacturer is the question, and if not, will other manufacturers still want to copy this new form factor ?

People who are building new chargers should stop thinking about 77mm, and start thinking about 82mm or even 85mm with protection.

This cell sounds like a nice upgrade for the BLF LT1 lantern :slight_smile:

One 46800 should give the LT1 70% extra runtime, based on the Tesla plan. Even if Tesla falls a bit short, 50% extra runtime should still be within easy reach.

I’m beginning to wonder whether a mainstream charger manufacturer might make the jump to having a common charging module for the electronics and the controls, with plug-in cell holders in a range of sizes, like people already do with hobby chargers. New cell size? Sell a new cell holder, no charger redesign required.

Wait… 300W from a single cell? 35000 lumen soda can lights incoming.

Who’s going to do the first LT1 mod? (I’m not… not that skilled).

People used to make adapters for SLR camera bodies to allow using lenses from other manufacturers.
simple flat rings with each side configured for one brand’s bayonet or thread connection.

It ought to be possible to make new size battery tubes along with adapters for currently existing flashlight heads.

Circle-packing of 4×18mm = 44mm.

46mm is 3mm extra width, about the height of 3 stacked dimes.

But now you don’t need springs to accommodate 4 possibly variable-length cells (sort unprotected vs long protected), but can use a bunch of shallow domes or a big fat metal “leaf” to allow only a few mm of slop to make good solid contact.

I’d say it could probably drop into a Q8 right now by just boring it out and modding the bottom-inside of the tube.

I want one.

Yeh, and the cells inside the Q8 aren’t snug together, but flop around in their respective holes, so the big donk could just drop right in.

TeslaFire cells, 99000000mAH, of course.

These batteries make no sense in flashlights.
You don’t understand that the 40-50% increase in capacity and power is only noticeable in Tesla cars? Actually the larger the cell, the higher the car range will be. 4680 is not an ideal size. It’s just the maximum cell size they currently are able to produce.
But in a flashlight there wouldn’t be any difference between 5x18650 and 1x4680.

But it should fit (more or less) into the space of 4×18650.

As Lightbringer points out it would fit in a 4x18650 space…
and fill the voids around the cells, providing more energy in the same volume - and eventually more available power/current.

Some prefer 8x18650 lights to 4x18650 because of ergonomy.
But others prefer it because of increased runtime.
Tell them that they’d be better off with 18650.

I cant seem to find any info on weight of a 4680
looks to me like a lot of copper

must weigh a ton!!!

They claim the power/weight ratio is better than a tabbed cell.

Tesla 2170s are 250-260Wh/kg.

Back in August Musk tweeted that useful 400Wh/kg batteries are 3 or 4 years away.

That gives an upper and lower bound for the weight.