Tesla's 18650s versus larger format automotive cells

[qwk]

I get my numbers from people on this board. Miles/kwh takes into account vehicle efficiency, I'm talking strictly charging efficiency. 25% loss is terrible.
How so? Have any LEAF packs been damaged from overheating?

So do I. There are numerous owners who get around 3 miles per kwh wall to wheel consistently with the roadster. The leaf is able to match that energy usage, but no one has reported getting any better(leaf forums).

It's hard to compare because of different drivers/roads, but it's not like the Leaf is more efficient to operate.

I didn't insinuate overheating of the Leaf pack. The very few instances of a Leaf in cold climate has the battery losing massive range due to cold temps and heater use. As of right now the Leaf is much less efficient than a roadster, in cold weather. I'm sure Nissan could work on this and improve, but I'm comparing existing vehicles.

 

[JRP3]

The Roadster is likely more efficient when driving than the LEAF, which is taller and heavier. That still does not change the poor charging efficiency of the Roadster and the high energy consumption of the pack. An EV should be able to sit for weeks unplugged and not end up with a dead pack.

 

[vfx]

... The LEAF pack is in the bottom of the vehicle....

As are Coda's.

 

[Doug_G]

An EV should be able to sit for weeks unplugged and not end up with a dead pack.

If it doesn't need active cooling, my Roadster loses a mere .5% SOC per day when in storage mode. Contrary to some reports it doesn't heat the pack when parked, unless it is plugged in and not in storage mode (and below freezing).

 

[RobStark]

How have things changed regarding this topic in the last 2.5 years?

And how is the debate affected by the potential building of the Giga Factory?

Will Tesla choose something the size of 18650 or large automotive format?

What is the upgrade possibilities of the Giga Factory?

Does anyone know how easy it would be to change format or chemistry once Tesla has made this large investment?

 

[JRP3]

Since Tesla is going to be the largest consumer of cells whatever format they pick will benefit from economy of scale. Format size can be changed to some degree, a shape change would probably mean different winding machines, though some can be adapted from cylindrical to prismatic. Most of the machines should not be affected by a chemistry change. I'm pretty sure Panasonic did not change their production machines when going from LiCo to NMC and then to NCA chemistry.

 

[rsquared99]

I seem to recall that Elon said a cell about twice the diameter of the 18650 would be optimal size. Any larger cell would creat thermal management problems. Of course, I've spent too much time at this site, so I may have made that up out of whole cloth.

 

[JRP3]

I remember he said about half as many cells would be optimal, I don't remember him saying anything about the size, but maybe he did. I've been leaning towards the idea of a cell with twice the diameter of an 18650 myself. Actually I measured a "D" cell and it looks to have about 2.5 times the volume of an 18650 in a shorter package, which would make a thinner pack.

 

[RobStark]

I saw an interview in which Elon said a battery " a little larger than what we currently have would be ideal."

But I have not heard or read him say a specific size.

Anyways, thanks for the info.

I have read TSLA bears say Tesla can't significantly reduce cost to make a truly mass market car until Tesla makes its own batteries but shortly after it does it will be making outdated batteries. And it will be locked in for the foreseeable future.

Allowing a major auto maker to make a superior battery pack.

 

[JRP3]

Yeah that's mostly nonsense they are pushing. The 18650 cell is just a container, so it really cant be "outdated", and it can be filled with any sort of chemistry. So worst case scenario would be a new chemistry is introduced that doesn't have any flammable electrolyte in it or oxides so is extremely safe, and it doesn't need to be actively heated or cooled to have a long cycle life and good performance. That means it's easy to build it into larger format cells, but it also means that Tesla can still build that chemistry into the 18650 format exactly as they have been, and eliminate all the temperature management stuff from the pack. I'd say that would only put Tesla at a slight pack density disadvantage compared to large format packs, and they could switch to that format over time.

 

[jerry33]

I have read TSLA bears say Tesla can't significantly reduce cost to make a truly mass market car until Tesla makes its own batteries but shortly after it does it will be making outdated batteries. And it will be locked in for the foreseeable future.

Allowing a major auto maker to make a superior battery pack.

Of course, you could say that about any battery manufacturer, just in general. The thing is that it takes a non-trivial amount of capital to create a battery factory, not that many automakers are in a position to do so. (And after electric cars take 20% of the market, which is when I think they would get serious about it, they will be in even less of a position to do so.) Also any new battery factory is going to be designed to allow a certain amount of flexibility. Sure, you're not going to be able to produce pBa batteries in a LiIon plant or vice versa, but a chemical change and/or size change within reason isn't likely to be a problem.

 

[ItsNotAboutTheMoney]

Yeah that's mostly nonsense they are pushing. The 18650 cell is just a container, so it really cant be "outdated", and it can be filled with any sort of chemistry. So worst case scenario would be a new chemistry is introduced that doesn't have any flammable electrolyte in it or oxides so is extremely safe, and it doesn't need to be actively heated or cooled to have a long cycle life and good performance. That means it's easy to build it into larger format cells, but it also means that Tesla can still build that chemistry into the 18650 format exactly as they have been, and eliminate all the temperature management stuff from the pack. I'd say that would only put Tesla at a slight pack density disadvantage compared to large format packs, and they could switch to that format over time.

http://ev.sae.org/article/11923

Interview with JB Straubel from March 2013 that is a must-have-read for anyone interested in the Tesla thinking on batteries,

 

[TEG]

How have things changed regarding this topic in the last 2.5 years?...

Well, the big nano-cap breakthrough still isn't here. Lithium Ion still is the main choice for EV traction packs these days.
The concerns about lack of cooling in the Nissan (LEAF) pack was somewhat founded as many of the packs aged quicker than projected in hot climates.
The concerns about cells that can ignite when damaged are somewhat founded by the experiences of some Tesla packs that got punctured.
(So it is good to have relatively stable chemistry and active thermal management to avoid those downsides.)
But other than that, they seem to be doing quite well, and volumes of vehicles delivered and used are starting to look impressive.
It doesn't seem like 18650 or large format are a clear "winner" either way. Both can work.
Things just keep getting better year to year as the battery technology improves.

 

[JRP3]

So you think someone in broad daylight was throwing objects on the road in front of the Model S's? And the guy in Mexico purposely crashed his own car at high speed? No reason to waste time on improbable conspiracy theories when simple reality offers a clear answer.
As for "commodity cells" insulating Tesla from some imagined government conspiracy I don't see that as any sort of barrier. They could easily make a law that "laptop cells" are not allowed in EV's. Not that I think it's even remotely likely, but if you're going with paranoid theories you might as well go all the way.

 

[VolkerP]

I remember he said about half as many cells would be optimal, I don't remember him saying anything about the size, but maybe he did. I've been leaning towards the idea of a cell with twice the diameter of an 18650 myself. Actually I measured a "D" cell and it looks to have about 2.5 times the volume of an 18650 in a shorter package, which would make a thinner pack.

Half as many cells would be 100% more cell volume, or 41% more diameter (18->25mm) if cell height is kept constant, resembling the 28650 form factor cells. The cell diameter is limited by the need to safely dissipate heat from the cell core, which depends on the heat production itself at a given discharge rate, and thermal conductivity.

 

[JRP3]

Half as many cells could also be achieved by doubling the specific energy density instead of increasing the physical size. I'd expect a combination of both energy density increasing and physical size increasing, which would allow half the number of cells to provide even larger kWh packs.

 

[ToddRLockwood]

My understanding is that the 18650 form factor was arrived at by the battery industry for its ideal mass-to-surface area, allowing for the most efficient heat dissipation. Am I wrong about this?

 

[stopcrazypp]

My understanding is that the 18650 form factor was arrived at by the battery industry for its ideal mass-to-surface area, allowing for the most efficient heat dissipation. Am I wrong about this?

Not necessarily. The thinness of laptops is also a consideration. That is why with the new thinner ultrabooks they are moving toward 16650s (2mm thinner).

 

[Zzzz...]

Large format cells would be cheaper in the long run. No doubt about that.

Tesla is using cheapest format out there. I think there is a great chance that when large format cells get cheaper then small format ones, engineers at Tesla will make them work

That future is not now, and quite away from now. But it is coming. Here is current, around one month old comparison of prices for large and small format li-ion and what looks like very educated prediction as to when one cross the other:

 

[Grendal]

How have things changed regarding this topic in the last 2.5 years?

I'm fairly certain that Panasonic has shown Elon the battery that he envisions being used in the Model E. He has spoken about it and is making plans based on it. Elon can't hide his beliefs and he believes confidently that it exists and that he/Tesla will be using it. He also implied during the "state of the company" address that there is an exciting secret in batteries that he has been shown if I remember correctly.