Roadster battery pack specs?

[gregd]

+1 to active thermal management. It's a no-go for me without it. Heating and cooling. Heat is not only generated by the cells, it is also absorbed by them from the environment. In a region where daytime temps can exceed 40C in the shade, 50C sitting in traffic on black asphalt, etc. is not hard to do. And it does dip below freezing in the winter. So, don't try to spoof the cooling system; please leave it intact. Nissan learned the hard way that "Lizard batteries" aren't the right answer.

Beyond that, whatever capacity can be safely packed in there, the better. 60kwh would be great, more even better.

Curious, what is the source of the Model 3 cells?

 

[petergrub]

The Model 3 cells have very low internal resistance compared to the factory cells (and the LG cells), so I don't feel active cooling will be necessary. Eliminating water cooling allows for even more cells to be packed in which further decreases the pack impedance.

You have too much thermal mass to simply leave out the active cooling system. You will essentially be creating a battery where the cells in the middle of the battery are warmer or cooler than those around the outer edges of the pack. Not to mention the slow rate at which the pack's cells reach temperature equilibrium.

Here are some things to think about if you plan to eliminate the active cooling system.
1) How do you warm all cells to a sufficiently high temperature before you can hit them with higher charging amperage?
2) How do you cool all the cells during use to avoid thermal runaway? Limiting use of available power still generates some heat.
3) Cells degrade at different rates depending on temperature, so you could be setting yourself up for pack imbalance issues as the pack ages

The cells are 5mm taller, but this doesn't appear to be a blocker; we can go to a single sided busbar design ala Model 3, and there's some room between the modules anyway.

I had my doubts, but after measuring tolerances and gaps in the ESS I think you could replace the original cells with 21700s. I'm curious what the single bus bar design is that exists in the Model 3 battery?

I am definitely on board with some of your proposed changes. The Roadster community needs a battery solution, so let me know how I can be of any assistance!

 

[chrisro]

if they upped the cells to the 3.4Ah cels like in the early model S then they are pretty much the same as the NCR B's.
If Okashira is the same person i am thinking he is he has probably pulled apart more tesla battery packs than most. He's also the most likely person to have replacement cells for a reasonable price. If someone threw him one of each cell (or if they own a cell analysier like most DIY battery builders do) we can easily spec out the cells.

The 3.0 cells looks like from Tesla Smart Fourtwo car.

 

[X.l.r.8]

Look visually or you have the build sheets? I’d assume the smart like the rav used model S cells. The smart modules are a lot less powerful but I have no idea about the makeup other than the bms is a lot different but runs the same messaging system and is a 12S module I think. I’d have to unbox one to check. That would support the 3400mah. I believe Gruber has a failed 3.0 pack but do not know of anyone who has physically removed a cell from one to check it. I would love to see your source.

 

[chrisro]

Checked visualy from positive side. But thats it. No test made. Smart is 2400 mAh what I know. Early Tesla S cells from 2012 are same like laters 2016?

 

[X.l.r.8]

I think that’s a lot of assumptions, especially the 3.0 80kw pack using 2400mah cells.
Model s is 3400mah
The logical step is the 3.0 using 3400mah model S cells. Not the 2400 smart cells. Again can you post the source?

 

[chrisro]

R80 cells are not same as Model S - this is 100% sure.

 

[X.l.r.8]

You can say no, I am guessing and have no source. Because an educated guess is 53kw is 2400 mah cells and 80kw = 3400 mah cells. So suggesting that the R80 cells are the same as the 2400 mah cells from a smart car needs some better proof. However if you are suggesting we can rebuild the 53kw pack using available cells from the smart car than that’s a totally different thing and an avenue that could/should be pursued. I had never thought to look at those 2 cells together.

 

[petergrub]

The 3.0 cells looks like from Tesla Smart Fourtwo car.

The Mercedes Smart Fortwo battery uses Lithium ion pouch cells, not 2170 cylindrical cells like in the Model 3.

 

[chrisro]

The Mercedes Smart Fortwo battery uses Lithium ion pouch cells, not 2170 cylindrical cells like in the Model 3.

I mean Roadster R80 3.0 cells.

 

[X.l.r.8]

Smart fortwo uses 2600mah cells
https://evwest.com/support/ncr18650.pdf

At the risk of repeating information again

6831 cells - 69 parallel/99 series
18650/2,2Ah Li-ion cells (safety sheet they are LiCoO2).
375V/151Ah/53kWh

6831 cells - 69 parallel/99 series
18650/3,4Ah Li-ion cells
375V/234Ah/80kWh


So how do you get 80kw out of 6831 2,6Ah cells arranged in the same configuration ?.
generously you could get 65kw

I would go for that if it were a straight swap of cells but
$799 for 330 cells
that's 20 modules for $16k
No guarantee its going to work. Add in time and materials and its not looking so good.

No guess work, but known numbers.

I'm not looking to argue just constructive discussion, all be it done a good many times in the pages of this site.
Guessing or indeed hoping that one cell 'looks therefore it is' the same as another is not a great way to construct a pack.
If there were to be a collaboration on a new pack @okashira should be involved as I followed him for quite a while on other forums and I would trust his identification of cells almost unquestioningly. It feels like its the slowest race to build a pack that tesla could easily win if they could be bothered. and after crunching numbers $20k for a new pack from Tesla with warranty would be a steal

 

[chrisro]

Tesla 2.0 cells - NLGFK55
It looks like roadster has Sanyo/Panasonic UR18650FJ cells. I will confirm as soon as I check the coding.
I found similiar looking and parameters with QIHEA2E.
3.0 cells I will open in few weeks.
Thanks

 

[chrisro]

UR18650FK - sorry for mistake.

 

[miimura]

It was widely reported that LG got the contract to supply the 18650 Roadster 3.0 cells.
Like this story from July 2016 - Tesla Roadster 3.0 340-mile battery upgrade now shipping

 

[X.l.r.8]

It would be nice to know what the cell chemistry is, I would assume that Tesla :
Did lots of R&D on the cells although unlikely real world testing as a complete pack
Used a proprietary chemistry with knowledge gained from having their model S cells built, however I would love to know why they simply did not use the model S cells, there must be a reason. They already had the thermal management sorted as the smaller capacity cells almost certainly run hotter than the larger capacity so would be controlled.
Thought that they had devoted enough time and energy to the roadster battery problem, assumes they had signed off on the project, and have not seen any reason to develop any further. I hope I’m wrong but with so much going on with Tesla Corp I think we have pretty much been abandoned. They will sell us old parts sitting in the shelf but I don’t think they will be manufacturing any new parts and that includes battery packs. I know there are still a few in the inventory but they will not be sold unless it’s in their interest.
If a way could be found to start simply swapping cells for either the 3.0 or the 1-2.5 packs and preserving the thermal management then someone could clean up.

 

[chrisro]

2.0 Sanyo LiCoO2 cells:

Materials/Ingredients Approx. %
-------------------------------------------------------------
LiCoO2
(Lithium Cobalt Oxide) 25-35
Graphite 7-10
Hexafluoropropylene-
vinylidene fluoride Copolymer 3-15
Lithium
Hexofluorophosphate 0-5
Acetylene Black 0-2
Dimethyl Carbonate 0-15
Ethyl Methyl Carbonate 0-15
Ethylene Carbonate 0-15

 

[X.l.r.8]

The 53kw pack was known to be laptop cells, all things considered they held up remarkably well. i wonder if they counted on new packs being a regular sideline but when it didn't happen they dropped any further research. I don't think anyone expected 2006 tech to work for 10 years in the harsh environment of a car.

 

[petergrub]

It was widely reported that LG got the contract to supply the 18650 Roadster 3.0 cells.

I checked a 3.0 pack, and they are definitely LG cells (LG ICR18650E1T). The part number found on the sidewall of the cells is LGDBE1T1865.

 

[chrisro]

So everything is clear. PEC board is different because battery is 4.35v max. 3.75v nominal.
max charge 2.17A, max discharge 4.65A.

Thanks a lot Peter!

 

[hcsharp]

however I would love to know why they simply did not use the model S cells, there must be a reason.

That has been discussed. Model S cells didn't have the same discharge curve that was necessary for the Roadster. Using Model S cells would have resulted in lower capacity because they couldn't be discharged below 3.5v (approx) which is the lowest the Roadster can start charging. There is also a size difference but they might have been able to overcome that if the discharge curve matched.

They will sell us old parts sitting in the shelf but I don’t think they will be manufacturing any new parts and that includes battery packs.

Tesla is contractually obligated to supply replacement battery packs. They can abandon everything else but not the battery.