Yes, exactly my thought when I heard this. And half the no of cells means more or less the 26650 cell as the volume of this is slightly more than twice that of the 18650. Anyone knows if 26650s are sold/used in any volumes? If there are few uses of these, and production capacity, or, the inability to buy 26650 in volume for model S was the reason they were not choosed instead of 18650s, it means that interestingly enough Tesla in Model S/X actually uses a sub-optimal solution with regard to battery pack volume and cooling. I.e., the cooling might be unnecessary effichient in this configuration, and that half the no of cells would yield enough cooling but better capacity/volume. This would come in handy for Gen 3.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Model S Battery Pack - Cost Per kWh Estimate
- Thread starter CapitalistOppressor
- Start date
-
- Tags
- TSLA
Yes, exactly my thought when I heard this. And half the no of cells means more or less the 26650 cell as the volume of this is slightly more than twice that of the 18650. Anyone knows if 26650s are sold/used in any volumes? If there are few uses of these, and production capacity, or, the inability to buy 26650 in volume for model S was the reason they were not choosed instead of 18650s, it means that interestingly enough Tesla in Model S/X actually uses a sub-optimal solution with regard to battery pack volume and cooling. I.e., the cooling might be unnecessary effichient in this configuration, and that half the no of cells would yield enough cooling but better capacity/volume. This would come in handy for Gen 3.
Increasing the diameter of the cell makes the heat exchange and the cell yield exponentially worse. Yes, making them bigger helps if the cooling is unnecessarily good right now and the yield is unnecessarily high on the cells. So by increasing the diameter you increase the packing density a few percent, but decrease the heat exchange and yield pretty dramatically. I certainly don't know, but I wouldn't be surprised if it wasn't optimal now at 18mm.
I heard Elon say in an interview that they use 18650 cells because that is the best that is available in high quantity. If he could have designed the cells himself, he would have made them twice the size.
Can't remember the exact words, but he was referring to the physical size of the cell and not the density.
Can't remember the exact words, but he was referring to the physical size of the cell and not the density.
stopcrazypp
Well-Known Member
Don't forget with larger cells you also reduce the amount of interconnects. There's probably an optimal balance for this and 18650 may not be it.
Unless an 18650 fab can be easily retooled to 22650 og 26650, the switching costs will be huge. A lot of 18650 capacity will be obsolete, and big investments will be needed into the new format.
Does anyone have any insight into what making this transition entails?
By the way, assuming ~30% improvement in energy density, it seems that the right format would be 22560?
EDIT: By the way, is there any reason why Tesla should conform to standard dimensions for this new cell? After all, at the time of a future switch, they would be close to 100% of world demand for that format. If their engineering were to show that the optimal size is 22mm x 73mm (arbitrary example), why wouldn't they just go for that?
Does anyone have any insight into what making this transition entails?
By the way, assuming ~30% improvement in energy density, it seems that the right format would be 22560?
EDIT: By the way, is there any reason why Tesla should conform to standard dimensions for this new cell? After all, at the time of a future switch, they would be close to 100% of world demand for that format. If their engineering were to show that the optimal size is 22mm x 73mm (arbitrary example), why wouldn't they just go for that?
Last edited:
JRP3
Hyperactive Member
I know that at least some winding machines can switch between cylindrical and prismatic cell winding, so I'd imagine changing from an 18650 size to another size should be fairly easy. On the winding side just let the ribbon run for twice as long before cutting. Of course you'd need to stamp and install larger end caps.
If you're talking Gen 3 they'd need new capacity to be built anyway, so it would be a good point at which to shift. It's one of the great bonuses of Tesla's approach: every car sells at least 2.5 times the capacity of other plug-ins and 40 or more times that of hybrids. 20k Teslas is comfortabky more than 800, 000 Priuses, 72, 000 Volts or 50, 000 Leafs. Tesla's a small volume car manufacturer but a large volume battery manufacturer.
this isn't necessarily about tesla's battery technology, but i think it is relevant to the discussion -- has anyone here noted uber's announcement that they plan to spend $375 million to purchase 2,500 driverless cars from google in 2014? as if this isn't interesting enough on its own, read the following blurb from the full article (emphasis mine):
"Due to its low weight and the latest in fuel cell technology, the GX3200 can get up to 750 miles of travel on a single charge, or about 48 hours on standby mode. Like Google’s other autonomous vehicles, the GX3200 is designed to find and dock in the nearest Google PowerUP station whenever it’s not in use."
750 miles on a single charge? anybody here know what kind of battery technology google is using in their driverless cars?
surfside
"Due to its low weight and the latest in fuel cell technology, the GX3200 can get up to 750 miles of travel on a single charge, or about 48 hours on standby mode. Like Google’s other autonomous vehicles, the GX3200 is designed to find and dock in the nearest Google PowerUP station whenever it’s not in use."
750 miles on a single charge? anybody here know what kind of battery technology google is using in their driverless cars?
surfside
The answer to your question is in the part you bolded:
It is a fuel cell powered car and not battery powered.
edit: just read the article and it says electric vehicle so now I am confused.
edit 2: I think the article is fake based on my google search
Last edited:
JRP3
Hyperactive Member
Electric700
Active Member
By July 2023, I hope there are the new generation 5 Superchargers located at convenient points all over the planet, powered by solar and wind energy and able to charge your 2023 Tesla Model A (autonomous flying vehicle) in 30 seconds!
you're totally right. should have done more searching (or better yet -- thinking) before posting. the title of the article (dispatch from the future) would have been a good place to start! the part about uber's second quarter earnings call raised an eybrow given uber is venture capital owned, but i didn't put two and two together. doh!
surfside
Off topic:
What's wrong wit this thread? For the past 2 days there seems to be a Page 31, but when you click it, Page 30 loads. Maybe one more new comment will fix it...
Edit:
I think it did...
What's wrong wit this thread? For the past 2 days there seems to be a Page 31, but when you click it, Page 30 loads. Maybe one more new comment will fix it...
Edit:
I think it did...
Last edited:
There's a fun, um, "feature" of the forum software that when a new page is created it won't "uncreate". Why would you want to "uncreate" a page? When the mods move N or more posts (where N is your posts per page user setting) from the original thread to another thread.
It gets really exciting around earnings time when we have dozens of posts with the same theme and they all have a broken page 2.
I wouldn't expect them to stay wedded to the 18650 per se. They probably don't want really large cells though if they want to retain the production efficiencies that you get with the 18650 cells. But a modest increase in size might save materials costs and simplify the manufacturing process during pack integration.
Where you find the proper balance isn't an answerable question at our level, and it's something Tesla will need to study hard, then put through an iterative learning process as they go into succeeding generations of cars.
- - - Updated - - -
I agree, they'll go with whatever is best.
Another factor is that they will want to have a common cell size across all platforms (though its not strictly required), and switching from 18650 will have costs associated with it. They will have a substantial number of factories devoted to 18650 production at the time that GenIII comes out, so do they abandon that capacity or just eat the costs associated with handling two different form factors at the cell level?
Ultimately though you probably want to devolve into just a single form factor across all platforms, and you want it to be one that you are comfortable designing around across multiple generations. It'll take a lot of thought to optimize all of the various factors going into that decision.
stopcrazypp
Well-Known Member
18650 is already the most popular lithium ion format and with Tesla adding a bunch of demand, it'll only be even more popular. So battery manufacturers are probably going to double up on investing in this format. If Tesla switches to a different size, it would have to be for a very good reason (a really optimal size that has significant improvements in density, module complexity/cost, etc) as they lose a lot of economy of scale advantages.
To me it seems that there is a fairly good chance that Tesla will stick with 18650 on shear momentum alone. It is non-obvious to me that they either need or desire to make a switch. At the same time, the engineering team probably has a good idea of what format they would prefer to design around (assuming it's not 18650), and a cost benefit analysis is definitely in order.
Honestly, if not for the new appreciation for the need for Tesla to vastly expand its battery production capacity beyond what they were envisioning pre-2013, they would simply have locked into the 18650 and this wouldn't be an issue.
But with Tesla now looking to reshape the battery industry in the next few years, it would behoove them to think hard about the form factor they are going to choose going forward. If it's still the 18650, fine and good, but with the production scales they are now looking at a point or two in lowered costs, or additional design flexibility, becomes a big deal.
lolachampcar
Well-Known Member
First, Tesla's needs far outstrip any concern for what existing capacity is or wants to produce.
Second, Elon's comments about having about 1/2 the number of cells as ideal is telling. If you can get G3 with the targeted range (200+ in Telsa speak probably means at least 250 EPA Rated miles) with 4K higher capacity 18650s then 18650 wins out. I think it is the targeted number of cells that will dictate cell packaging.
Second, Elon's comments about having about 1/2 the number of cells as ideal is telling. If you can get G3 with the targeted range (200+ in Telsa speak probably means at least 250 EPA Rated miles) with 4K higher capacity 18650s then 18650 wins out. I think it is the targeted number of cells that will dictate cell packaging.
It will be a hard tradeoff between switching costs and optimization. For two reason I think optimization might win out: (1) The current volumes are quite tiny compared to what is envisaged in the future, so most of the future capacity has not yet been built, and (2) someone in this thread (JPR3?) made the case that switching a factory from one format to another wouldn't be a huge thing.
On the other hand, never underestimate the momentum of path dependency (imperial vs. metric system, QWERTY keyboard etc.).
Switch to right-hand driving, Stockholm, Sweden, 3 September 1967
On the other hand, never underestimate the momentum of path dependency (imperial vs. metric system, QWERTY keyboard etc.).
Switch to right-hand driving, Stockholm, Sweden, 3 September 1967
Similar threads
