Sanyo NCR20700B 4250mAh cell is here!

[spinningmagnets]

These are pricey if you buy them at the "one at a time" price, but...contact them and ask about their 100-units price. US-based supplier in southern California (a partner of Luna)

http://www.wakeandvape.com/batteries

 

[Samd]

And the vendors/sellers of Genuine Panasonic GA cells and the other Quality 18650 cells that are over 3100 mAh , here in the U.S.A. is / are .... Who ? ( luna only has 2500mAh cells, the INR 25R )

What prices did you find ?

Shipping costs for how many ?

Does Okashira have a website, does he only sell salvaged tesla cells ?

My research has shown me that the high-volume and high-current cells we ebikers crave are in a market driven by cordless tools (less than an hour ago, I used my 14S string-trimmer/weed-wacker). Laptop computers still use 18650's, but those are low-current, and compete on lowest wholesale price. I am willing to stick my neck out and predict that for the next year (at least) our best bang-for-the-buck will remain the 18650-format.

I think you'd be crazy not to buy GA's or Hg2's out of asia direct.

 

[macribs]

@Hillhater What Tesla did for the P100DL was new cooling lines and new way of packaging the cells.
The cells size is still 18650, and the chemistry remains the same. A new more complex way of putting the pack together is the reason for capacity and performance gain.

 

[Hillhater]

Thats certainly possible, but it would be even more impressive than a cell improvement !
To find space for 750-800 extra cells, yet retain the external pack dimentions (for interchangeability with existing packs) ,..
..that is more difficult...and costly..... than just subbing in 3.4+Ahr cells.!

 

[flippy]

That will probably be the next step, use the more efficient packing method and higher capacity cells will probably kick it up to 120kwh in a year or so.

 

[Samd]

But its still a round cell. Why would there be a more efficient packing percentage guys? I don't follow.


Sent from my iPhone using Tapatalk

 

[Wheazel]

But its still a round cell. Why would there be a more efficient packing percentage guys? I don't follow.


Sent from my iPhone using Tapatalk

I guess you decrease the volume that isnt comprised of cells.

 

[Pajda]

Thats certainly possible, but it would be even more impressive than a cell improvement !
To find space for 750-800 extra cells, yet retain the external pack dimentions (for interchangeability with existing packs) ,..
..that is more difficult...and costly..... than just subbing in 3.4+Ahr cells.!

An easy calculation gives 7936 cells in 100kWh pack, so it is 832 extra cells. It should be also recognized that the original 85kWh pack from 2012 never been a symbol of an excellent packaging efficiency. We can find many designs of 18650 battery packs with thermal management with much better packaging efficiency e.g. from Kreisel bros.

I also see some misunderstanding about this new 100kWh Model S/X pack and the correlation with Model 3 pack with announced 21-70 cells. The main reason for this modification was the increase of capacity and cooling efficiency for higher power draw at all cost. So this new pack most probably uses heatpipe cooling system which cools/heats the bottom of the cells (both cell terminals are now connected from the upper side). Heatpipe system is also far more thermal efficient than the "snake pipeline". But definitely it is not cheaper than actual system. So I believe that there is only minor design correlation between this new Model S/X and Model 3 battery pack.

I have heard many statements that 21-70 (or 20700) is a holy grail of battery technology in all aspects. But there is just no evidence for this statements. The only parameter of actual generation of 21-70(20700), which seem to be comparable to the 18650 is production price and better power and cycle life. But that is it! because there are only two main goals for Model 3 battery pack. To be very cheap and to have a good lifetime even for battery packs with capacity of 50-55kWh only.

The Model 3 pack must be definitely very cheap which means no heatpipe and the whole cooling/heating system must be very simple. The packaging efficiency is secondary objective, because for Model 3 does not make sense higher capacity than 70-75kWh in top versions. So the significantly lower energy density of actual 21-70(20700) than 18650 is not problem here.

Also it is important that actual NCR20700B seems to have better cycle life under heavy load 2C and even 3C than 18650 cells used in Model S/X batteries. This is important because due to lower capacity the Model 3 packs will be operated at much higher average DoD than Model S/X packs.

 

[okashira]

It's possible, if not likely that Tesla redesigned the modules to simply be much larger/heavier.
There is a-lot of wasted space between modules... plenty for another 900 cells.

 

[Hillhater]

.....An easy calculation gives 7936 cells in 100kWh pack, so it is 832 extra cells. It should be also recognized that the original 85kWh pack from 2012 never been a symbol of an excellent packaging efficiency. .

Do we know (for certain) what the changes were between the 85Ahr, and the 90Ahr pack ?
Cell quantity ?
Cell capacity ?
Software changes ?

 

[Samd]

But its still a round cell. Why would there be a more efficient packing percentage guys? I don't follow.


Sent from my iPhone using Tapatalk

I guess you decrease the volume that isnt comprised of cells.

Not sure if you mean for Tesla's layout or for those of us making bike/motorbike packs etc.

My issue is this:


In the top pack using 18mm diameter cells, the interior percentage is precisely the same percent of square area as the lower pack using 20mm diameter cells.
And in fact if you don't just consider the interior percentage, the lower image is less efficient spatially because the ends of rows have less chance of filling the gaps at the side as the battery as you increase the cell size.

The 20700 isn't volumetrically any better for most of us as the press releases asseted. *I think*.

 

[okashira]

You are 100% correct, but it's a really a non - issue with such a large pack such as a the Model S pack. It's almost 6 feet wide and 12 feet long so you can get fill just fine with a larger cell.

I think the 21700 will be awesome for ebike and general purpose as well, just like the 18650 was. It's annoying to fit to existing size cases designed for other stiff, but when you can size the case/frame/whatever for the cell pack size, it will be great. Same with 18650. I come across folks who want a custom pack for their case/frame/whatever, and you can rarely get an ideal fit.

 

[Samd]

Yeah thats spot on Okashira. Thanks for putting up with me - it was driving me crazy to try to figure out if they really were more dense as some folks asserted.
They certainly are less welding

And as a percentage of overall length, assuming the positive ends are the same thickness for both then it will allow for more cell per unit length

 

[Pajda]

.....An easy calculation gives 7936 cells in 100kWh pack, so it is 832 extra cells. It should be also recognized that the original 85kWh pack from 2012 never been a symbol of an excellent packaging efficiency. .

Do we know (for certain) what the changes were between the 85Ahr, and the 90Ahr pack ?
Cell quantity ?
Cell capacity ?
Software changes ?

Of course I am guessing until the first photos of inside of the pack will appear. But we can learn from the last Tesla designs.

1) Tesla always install in highend packs exactly the same full capacity as announced, so "85" pack have ~85kWh full capacity, "90" pack have ~90kWh.
2) Tesla always used already mass produced cells, like NCR18650BE (~12Wh) and NCR18650GA (~12,65Wh).
- Yes, It is known that Samsung, Panasonic and LG already have their 3600mAh 18650 cell samples, but there is no evidence about their mass production.
3) Almost all BEV manufacturers uses for 96s pack 16 modules, so It is most likely that Tesla will stick with this module count too.
-Then the total cell count must be divisible by 16 and it gives for 100kWh pack 7936 or maybe 7952 cells, where the 7936 is better count for dividing into rows in modules.
4) If you need much higher power draw from the same cells, your only help is to use much more effective cooling/heating system and there is not much other options how to do it without heatpipe.
- The power draw for new P100DL is much higher than the 10kWh increase in capacity.

 

[Pajda]

And lets get back to the new cell format 20700 or 21700 (21-70). From the energy density point of view you need ca 5000mAh capacity in 21700 format, or 4750mAh in 20700 format to equal the current NCR18650GA cells unless you are not able to build a 100kWh battery pack in Tesla box.

From what I know it is already in serial prodution 4250mAh in 20700 and maybe few samples of 4750mAh in INR21700-47P model from Samsung. Or anybody knows about something better with real contours? I am not interested in "vaporware".

By the way for 120kWh battery pack you need capacity of ca 5500mAh in 21700, or 4200mAh in 18650 format and the optimistic future outlooks says 2020 but the more conservative outlooks says 2025.

 

[cwah]

still not as good as 18650 for now :/

 

[edrift.eu]

Just ordered 2 sample cells from Akkuparts24.de
They are 9,16€ without tax at the moment.
Will test them and post test on my YT channel
https://www.youtube.com/channel/UCr7ax9-rsonEMP100GZzHzA

Will also have a big test with 26650 cells from 6 brand. One of them actually have 5000mah+, all other are fake!

 

[redilast]

I contacted one of my suppliers and they also have them now in stock now. But I'm waiting on another supplier that usually has them for cheaper to also carry them.

The NCR20700A also look interesting for higher current applications. Looking at the data sheets for the NCR20700B, it looks like these are best for <8A continuous discharge if you want to get 500 cycles out of them.

 

[Hillhater]

Any sign of the Sony/BMZ "3Tron" 21700 cell yet ??

 

[Hillhater]

I have not seen anyone post this test data for the Sanyo/Panasonic 20700..
http://lygte-info.dk/review/batteries2012/Sanyo%20NCR20700B%204000mAh%20(Red)%20UK.html

 

[neptronix]

Shame. I was hoping that these would have an improved chemistry inside them, but the cells are average in every way. The only advantage i can see is that if you put a pack together with these, there'd be less cell welds to make.

 

[macribs]

Well there is also the NCR20700A, lower 3200 mAh but higher C rate.

 

[flippy]

the tesla 21700's have a reported capaciy of 4750mAh. i would wait for a few crashed model 3's to get stripped and see how those sells preform. the extra cm in hight makes a lot of difference.

still, if the 20700's are cheaper it can become the next 18650.

 

[SniperGaulois]

good one, they are 5.95 euros in nkon, can have a 21AH 54v battery for 386 euros that weighs 4kg for 5p13s.

 

[Vanarian]

Well there is also the NCR20700A, lower 3200 mAh but higher C rate.

Didn't know this, that's GOLD ! http://lygte-info.dk/review/batteries2012/Sanyo%20NCR20700A%203100mAh%20(Red)%20UK.html

At maximum rated current they are up to 81°C

30A at only 81°C , what is the magic trick behind this ?

A Sony VTC5A goes 92° C at 30A end of discharge A LG HB2 is 79° C 30A end of discharge, is the temp measured by lygte a full discharge cycle temp?