stephenk
Enlightened
I would think that 21700 flashlights would be more ergonomic (i.e. fits better in the hand) than 18650s? Sounds like a positive step forward!
Tesla's gigafactory dropping 18650's and going with 20700 - per Elon Musk
- Thread starter billcoe
- Start date
Help Support Candle Power Flashlight Forum
Drift Monkey
Newly Enlightened
The funny part is that 18650s still have a better size to energy density ratio. The move to this new size must be purely to save money.
Not necessarily. Higher capacity 18650 batteries tend to have less current potential. It is a trade off of capacity for amperage output. I'm betting that they will get the 21700s at about the same energy density of 18650s with higher current capability in the end... if they are spending billions making a huge factory I'm sure they are pouring a bunch of money into getting every bit of power possible out of these batteries at currents that they need to drive large electric motors. Will the new battery packs have more power than a same sized 18650 based pack? I doubt it will exceed that, but also it wouldn't surprise me if they just matched the same output or maybe 5% less with higher amperage capability which could actually work out to be more power under load than the 18650s do.
What the magic tradeoff between higher capacity and higher amperage is I really don't know and how many series/parallel sets of these batteries are wired together and hooked up 18650 vs 21700 I don't know either.
As for 21700s in flashlights I have mixed emotions as AAA "tubes" are too small for constant use IMO, AA is nearly big enough diameter for my hands while 18650 lights are just about right. I guess I've gotten used to smaller diameter lights growing up with most flashlights using D cells I liked the C cell lights they were lighter and more comfortable. I think 21700s will not find themselves into front battery headlamps much the extra weight and size (45%? more) is just too much IMO.
If companies making other battery packs switch to the 21700s enough we could see the 18650 batteries becoming no longer mainstream as 1 layer of 21700s should be close to 50% more power in tool packs allowing a 1 layer pack to be midway between 1 and 2 layer 18650 packs. I think what will be the main driving force is the cost/power ratio of both cells if the 21700s get the cost/power ratio down below 18650 batteries then it will start to erode the use of them in everything that is based primarily on cost/power and size is less of an issue.
Drift Monkey
Newly Enlightened
So using your theoretical reasoning, why not improve upon existing 26650 cells?
In fact, they did release the specs of the 20700 B cells and they didn't match the current GA cells...it only got to 92% of the energy density (4250 mah) with similar current discharge rates. The 20700 A cells give a mere 300 mah advantage @20A than existing 18650s. 21-70 moves even further away from the ideal.
Unless a battery chemistry breakthrough occurs, this cell format is about cutting costs, plain and simple.
Last edited:
FroggyTaco
Flashlight Enthusiast
Per Elon Musk they decided to design a cell in the present as if there was no other existing format that would suit there needs. Apparently 21700 has the ideal capacity/discharge/charge/packaging specifications an electric car needs.
I would surmise a higher capacity 21700 cell would have less desirable discharge & lifecycle characteristics even if it would work amazing for us flashaholics.
Because the market isn't there for them... yet. If they decided to use them in electric cars I would assume that they would push the envelope of that size battery to the max just like they have done in the past with 18650s.
Drift Monkey
Newly Enlightened
It's for cost reasons...as stated above. Froggy is on point...it will work "better" for electric cars and be cheaper, this is the real reason.
Last edited:
I'm guessing that there could be a "need" for a minimum amount of batteries in sets (series) in order to get the desired voltage and if batteries get too large in diameter it could be more difficult to wire them together.
Drift Monkey
Newly Enlightened
New info - the new 2170 cells will have all new state-of-the-art hardware and new battery chemistry! It may even be able to match/beat 18650s in energy density ratio!
And if there is improvements to the chemistry and design they could also be incorporated into 18650 cells likewise such that the main difference is again related to size/volume.
My concern is that other makers of things will be attracted to the new size batteries due to the lower cost/energy ration in devices. If you can get a 2000mah battery for your drill for $75, a 5000mah for $150, or a 3500mah battery for $85 which would you choose? More and more the market on lithium ion powered stuff is about Ampere hours and these batteries IMO could in the future rule the market till someone with a ton of money decides to support an even larger battery size in volume.
Can this factory that will make these 21700s be retooled cheaply to make 25900s or larger at a later time? We could see more interest in 26650s in the future even. I think cylindrical cells will rule the market in the energy/cost ratio for some time until a completely new chemistry comes available that can favor a different cell design.
Will they ever get lithium batteries cheap enough to replace car batteries in use and improve them to a point that they can outlast them in hard use in all climates? I'm looking back at the days car batteries cost 1/3 of what they do now and they were competing well with nicads and even nimh prior to LSD nimh.
kreisl
Flashlight Enthusiast
- Joined
- Jul 5, 2012
- Messages
- 2,244
This youtube clip belongs to this thread, gigafactory opening speech yada 
Last edited:
Ditto
Friend was telling me about new Tesla X model. When he told me about the motor I didn't believe it and had to check it.
2 motors for combined power about 500 kW
Sure pack is big 90 kWh but these motors .... OMG
So you are talking about 6C peak power!!!!!
I am guessing the major factor was in fact high current.
I don't know what is the voltage used so blatant guessing game:
500V @ 500 kW = 1,000 A
500 V / 3.7 = 136 cells per module
90,000 Wh / (3.7 V * 5 Ah * 136) = 36 modules
1,000 A / 36 modules = 28 A/module
So 5 Ah cell has to be able to supply peak 28A current!!!!
(sorry my math is likely not accurate, kinda fudging it)
Ditto
Friend was telling me about new Tesla X model. When he told me about the motor I didn't believe it and had to check it.
2 motors for combined power about 500 kW
Sure pack is big 90 kWh but these motors .... OMG
So you are talking about 6C peak power!!!!!
I am guessing the major factor was in fact high current.
I don't know what is the voltage used so blatant guessing game:
500V @ 500 kW = 1,000 A
500 V / 3.7 = 136 cells per module
90,000 Wh / (3.7 V * 5 Ah * 136) = 36 modules
1,000 A / 36 modules = 28 A/module
So 5 Ah cell has to be able to supply peak 28A current!!!!
(sorry my math is likely not accurate, kinda fudging it)
I'm not sure about all these specs but I did a quick check and people were talking about 300 or 375v on the motors themselves and they use an inverter so the battery voltage may be lower or higher than that you basically have to go by watt hours, calculate the number of batteries total and go from there.
I looked up the information about how many 18650 batteries in a Tesla's 85Kwh pack is claimed to have 16 modules of 444 cells or 7104 cells and a test claims 11.36Wh per battery or about 3.07Ah. Now 444 cells times 3.7v = 1642v nominally or 1864v at 4.2v.
If we take 500kw divide that by 7104 we get 70W and further divide that by 3.7v gives you 18.9A this is for 18650 batteries in the current packs. I suspect the energy density of 21700s will at least equal 18650 batteries and the number of batteries will drop while the current will rise.
The big question is how many batteries will fit in the new "Pack" size of 21700s. I suspect it will be a little larger in at least one dimension and it is possible that the "module" size will change too.
alternety
Newly Enlightened
An overall point that seems to be rather under appreciated: Elon probably does not care even a little bit about how the batteries work in flashlights or anything else. The following has been mentioned in this thread, but here is some of it in one place. For those who want to read one post.
His factory can be fully engaged if he sells the number of cars he is projecting. The factory is for cars. Rather like the A123 thing worked out. They were useful to a niche.
RE format: Cylindrical cells provide a relative easy cooling system. Prismatics would seem to be harder to cool in terms of the required cooling channels, surface contact, and interaction with adjacent cells (Think Boeing). Hexagonal, as has been mentioned is not a practical format. Battery innards are a cylindrical roll. Getting the cells to fit together with adjacent cells kills cooling.
The cells are designed to optimize use in a Tesla car. Built to fit the battery compartment (or a dimensionaly altered one). Built to use the cooling system. Built to satisfy the V/I needs of their electronics. Get the longest range and shortest recharge time. Nothing else is relevant. They are not making consumer batteries. They may, as a side effect, create a standard for other things; but that is not the point. They are simply not bound to any standard. They will be big enough to cause any manufacturer that wants to be supplier to build whatever Elon wants. And it may become a standard. But he will not care. At his kind of usage level, it just is not his problem until demand exceeds his capacity. And at those levels, other manufacturers will build what he needs.
Debates about the cells are quite useless without knowing the structure and chemistry of the cells. There is no reason to assume the construction is like anything we have seen in production. Nor no reason to assume he will not use some esoteric technology. There are many proposed technologies out there. Elon does not back away from using advanced technology.
The format and/or internals may show up in the commercial market at some point. Earlier if he can't get enough cars sold to keep his battery factory busy. He has also been very generous about making his designs open for free use. If he does that with batteries, that could allow some other manufacturers to make them as well.
Short answer: we have no idea what he is doing, but it is not for flashlights.
His factory can be fully engaged if he sells the number of cars he is projecting. The factory is for cars. Rather like the A123 thing worked out. They were useful to a niche.
RE format: Cylindrical cells provide a relative easy cooling system. Prismatics would seem to be harder to cool in terms of the required cooling channels, surface contact, and interaction with adjacent cells (Think Boeing). Hexagonal, as has been mentioned is not a practical format. Battery innards are a cylindrical roll. Getting the cells to fit together with adjacent cells kills cooling.
The cells are designed to optimize use in a Tesla car. Built to fit the battery compartment (or a dimensionaly altered one). Built to use the cooling system. Built to satisfy the V/I needs of their electronics. Get the longest range and shortest recharge time. Nothing else is relevant. They are not making consumer batteries. They may, as a side effect, create a standard for other things; but that is not the point. They are simply not bound to any standard. They will be big enough to cause any manufacturer that wants to be supplier to build whatever Elon wants. And it may become a standard. But he will not care. At his kind of usage level, it just is not his problem until demand exceeds his capacity. And at those levels, other manufacturers will build what he needs.
Debates about the cells are quite useless without knowing the structure and chemistry of the cells. There is no reason to assume the construction is like anything we have seen in production. Nor no reason to assume he will not use some esoteric technology. There are many proposed technologies out there. Elon does not back away from using advanced technology.
The format and/or internals may show up in the commercial market at some point. Earlier if he can't get enough cars sold to keep his battery factory busy. He has also been very generous about making his designs open for free use. If he does that with batteries, that could allow some other manufacturers to make them as well.
Short answer: we have no idea what he is doing, but it is not for flashlights.
vadimax
Flashlight Enthusiast
Poor people. They still believe that switching to nonstandard form factor has something in common with optimizing. Elon Musk just wants bigger share of after sale revenue. It is the same as selling a car that consumes "proprietary" fuel.
It is entirely possible that both are true: better efficiency (whether cost-effective or otherwise) and Elon gets a bigger cut.
Nonsense. Whether the batteries in the car are 18650 or 2170, they are not user-accessible. It makes zero difference to the consumer if the cells use an old standard or a new one.
Exactly. I have yet to see anyone here truly analyze this from an engineering point of view.
Last edited:
alternety
Newly Enlightened
Wrong site - sorry
Last edited:
There's also a Panasonic NCR20700C cell with a 3.5 Ah capacity under development: http://www.sdle.co.il/Default.asp?sType=0&PageId=56631
So capacity wise it sits in between the 3.3 Ah NCR20700A and the 4.25 Ah NCR20700B.
So capacity wise it sits in between the 3.3 Ah NCR20700A and the 4.25 Ah NCR20700B.
vadimax
Flashlight Enthusiast
Switching to battery format that is being produced by no one but you, you are free to rise those batteries' prices at will.
