Panasonic phasing out of Cylindrical Battery Business
- Thread starter IlluminationDomination
- Start date
Help Support Candle Power Flashlight Forum
SubLGT
Flashlight Enthusiast
Hope Panasonic has a plan B for those extra cells that will soon be overstocked after this
https://read.bi/2IxRXsD
https://read.bi/2IxRXsD
idleprocess
Flashaholic
Tesla initially went with 18650 because it was a commodity:
- Almost every cell manufacturer made 18650
- Every chemistry was available in 18650
- Every new advancement in Li-Ion was immediately available in 18650
- Short of a massive investment in volume production, 18650 lead other formfactors in price
erformance - $/WH, $/peak amp, $/cycle, etc - The 18650 format makes for relatively simple thermal management due to its high surface area : volume ratio; the small diameter results in better heat dissipation throughout the cell
- In a multi-kWH pack, if a handful of 18650 cells degrade or fail the effect on performance will be negligible relative to a pack with fewer, larger cells
Tesla could follow the major automakers into larger-capacity prismatic packs. But for the reasons above, they still use small cylindrical cells in their packs. Their transition to 20700s doesn't really change this: high-density (3.5AH) 18650s are ~12.5WH while their 20700 cousins (4.8AH) are ~17.25WH. I gather that the other automakers aren't as enamored with cylindrical cells and have opted for large prismatics - perhaps to ensure supply via contractual arrangements with suppliers, perhaps out of habit and their traditional spare parts business, and also perhaps out of packaging concerns: most of the electric cars they've produced in the last <10 years since CA lease-only compliancemobiles have given way to mass-market vehicles sharing a platform with an ICE model, thus their packaging requirements might demand space efficiencies that Tesla's clean sheet designs do not. Boeing's first iteration of 787 batteries highlights some of the inherent problems with large prismatic cells - greater surface area:volume efficiency begets unavoidable thermal resistance that's hard to monitor and can't necessarily be managed under high-draw conditions.
18650s might fall out of favor with laptop manufacturers who are steadily transitioning to Li-Po, but I expect cylindrical cells to remain popular with power tool manufacturers. They demand a physical robustness that Li-Po can't offer, and even moreso than Tesla are dependent upon standard sizes. It's possible that the availability of standard prismatics with dimensions similar to cylindrical cells such as the 18650 and 21700 might find foothold in applications where volumetric efficiency is a must, but that brings added cell costs, some sacrifice of mechanical robustness, the efficiency bump is small, and such designs would largely negate the minute amount of cooling that cylindricals offer in the gaps between cells in a hexagonal arrangement.
Tesla initially went with 18650 because it was a commodity:
Tesla could follow the major automakers into larger-capacity prismatic packs. But for the reasons above, they still use small cylindrical cells in their packs. Their transition to 20700s doesn't really change this: high-density (3.5AH) 18650s are ~12.5WH while their 20700 cousins (4.8AH) are ~17.25WH. I gather that the other automakers aren't as enamored with cylindrical cells and have opted for large prismatics - perhaps to ensure supply via contractual arrangements with suppliers, perhaps out of habit and their traditional spare parts business, and also perhaps out of packaging concerns [...]
- Short of a massive investment in volume production, 18650 lead other formfactors in price
- [...]
That's not quite accurate. 18650's do not lead other form factors in price performance (even restricted to cylindrical format). That's one of the primary reasons why other EV manufacturers have chosen larger prismatics instead - which further offer greater design flexibility, e.g. they don't suffer from an ~100μm limit on electrode thickness in cylindrical cells (any thicker and the electrode coating would crack because they are wound in cylindrical cells vs. stacked in prismatics).
The transition from 18650s to larger format 20700s does change it, e.g. it lowers their cost/kWh to be much closer to prismatics (e.g. see the final chart below where "Optimistic"=20720 NMC cells are only slightly costlier than prismatics).
Below are excerpts from a 2017 study that extends the common BatPaC PBCM (process-based cost model) for cell manufacturing costs. The model incorporates all of the steps involved in cell manufacture, e.g.
Below is a chart summarizing their results. There "BaseLine" = 18650 cell, "Optimistic" = 20720 cell, and "Prismatic" = 25Ah prismatic cell. Notice the large improvements that the 20720 offers - esp. in the higher-rate LMO chemistry on the left. There note that the (per kWh) cost of the 18650 materials alone is almost the same as the entire cost of the prismatic - which highlights the reduction in hardware costs afforded by using larger format prismatic cells.
Excerpted from: Comparison between cylindrical and prismatic lithium-ion cell costs using a process based cost model, by Rebecca E. Ciez, J.F. Whitacre, Jnl. Power Sources, 340 (2017) 273-281.
Last edited:
idleprocess
Flashaholic
Perhaps I should have emphasized my point to a greater degree. At the time that Tesla was getting started with the Roadster launch in 2008 and during the run up to the Model S in 2012, 18650 Li-Ion cells were the best choice for the company. Their then highly competitive price : performance ratios along with being an industrial commodity meant that a fledgling startup could make cars using a COTS product using existing supply channels and focus limited financial/engineering resources on more pressing needs.
Their continued reliance on cylindrical cells may be due to tooling / plant synergies between 18650/20700 footprints, easy transition of pack control electronics/cooling/management, or some other business strategy that negates the performance gap between cylindrical and prismatic cells. The fate of the Model 3 will be telling, however I doubt that a ~5-20% cost delta on the cell cost (pretty sure they're not using high-rate chemistry) will be the primary cause of their demise should the Model 3 fail.
Many folks disagree with that, since the same sort of analysis as above shows a great advantage for prismatics.
HKJ
Flashaholic
That is always a interesting analysis, who knows best: The people doing and using the money or the people that sit back and tries to analyze whats happening.
If the world was based on university people, it would probably never move forward, it needs people doing things, even if xx% of university people says it is a bad idea. Sorry, but we got way to many cases with "educated" people says something that do not match with the practical experience optimal solution. I know that they can show experiments that match exactly with what they say, but they do not explain the limitation of the experiments.
With the above, do the analysis show that prismatic cells are best with Wh/weight, Wh/volume, Wh/impact, Wh/safety, Wh/power, Wh/??? or only on some of the parameters?
I believe that cylindrical cells may have an advantage due to mechanical robustness and cooling.
idleprocess
Flashaholic
If you want to dig out figures circa 2003-2008 to demonstrate that Tesla f__ked up, knock yourself out; the 2017 date on your seemingly-paywalled excerpt suggests it's based on analysis a markedly closer to today than when Tesla was planning the Roadster and Model S ~15 years ago. Make sure you look into the commitments that the battery industry likely would have demanded in exchange for a moderately more efficient product and compare that to their early financials.
I don't understand why you are so often skeptical of professional literature. While no doubt there is plenty of academic research on Li-ion tech that never becomes practical, the above is very far from that. Such process-based cost models are widely used in industry for such purposes. If you do a web search on the term you can quickly find links to such applications, e.g. the excerpt below mentions a few.
PBCM is not some pie-in-the-sky research that never made it out of ivory towers. Rather, it is a standard technique widely used in industry. Likely the large cell manufacturers use this (or similar) methods to perform such analyses. The only difference is that they may be able to do slightly more precise analyses since they have access to more precise (cost) data, etc.
The results of the paper I linked are consistent with other analyses and general industry trends. I see nothing at all controversial in their results so I am baffled why anyone would be so skeptical of these results.
Last edited:
Those were your claims, so that is your responsibility. The decisions made by Tesla way back then may have been motivated by other factors unrelated to pure cost analysis. Musk's logic does not always appear to be rational, and is often questioned by experts.
The paper was linked through sci-hub, so is freely accessible.
Last edited:
StandardBattery
Flashlight Enthusiast
Well if you go back and look at what all the experts had said about Musk's 18650 plans early on, you would hopefully come away with a healthy skepticism of 'experts'. Some of those experts might have even commented on the first cellular phone systems.
I have no idea what that vague remark is supposed to mean. Likely it has little if any relationship to the matters I mentioned above.
IlluminationDomination
Enlightened
My thread has certainly turned into an interesting and EGO inflated debate. Who is right, who has to be right and who needs to be right?
Time will tell if it has not already!:shakehead
Time will tell if it has not already!:shakehead
Latest posts
-
Is the 796 bulb a safe application for a reverse light?
- Latest: Hamilton Felix
-
-
-
-
