Impressive battery in the Tesla Model 3

When Tesla started building the 2008 Roadster, 18650 cells (made for laptop computers) were the cheapest way to buy kWh of li-ion batteries. Later, the 18650 batteries which Panasonic made for the Model S (and later the Model X), made to Tesla's specifications, were the highest energy density batteries used in any EV. So far as I know, they still are, with the exception of Panasonic's 2170 cells. There has been a lot of argument over whether or not those actually have a higher energy density than the 18650 cells currently made by Panasonic for Tesla, so I won't offer an opinion on the subject.

A higher energy density let Tesla put a high-capacity battery pack into its cars without making them that much larger, which meant Tesla's BEVs had a longer range than anybody else's. To a large extent, this is still true. But to claim that a higher energy density is the only advantage of using small cylindrical cells... well, that's simply not true.

* * * * *

Let me try to create an objective overview, comparing cell types in what I think are the two best types of battery packs used in BEVs:

Tesla's battery cells: Small cylindrical cells (18650 & 2170) from Panasonic

Pro:
Higher production means higher volume, and lower cost from the advantage of scale
Smaller size enables higher energy density, allowing smaller pack size, significantly reducing the cost of the car

body​

Metal case means simple, strong battery pack architecture
Round shape means space between the cells, helping dissipate heat safely
Relatively small size means a high surface area to volume ratio, again helping dissipate heat safely
Large numbers means the failure of a single cell has little impact on the pack's capacity

Con:
Small size means large numbers are required, which also requires a large number of connections
Round shape makes coolant loop architecture more complex, as the loops must be curved around the cells
Higher energy density means greater risk of overheating and fire, which has required Tesla to build various fire

prevention systems into its packs​

Despite fire prevention systems, there have been a small number of battery fires in Tesla cars

GM's battery cells: Moderately large pouch cells from LG Chem

Pro:
Pouch cells are cheaper than other types of cells, all else being equal*
Larger pouch cells means fewer cells are needed per kWh, and thus fewer connections needed
Pouch cells have flat sides, making cooling loop architecture simple
Pouch cells can be packed tightly, side-by-side, with no gaps between the cells, allowing smaller pack size,

significantly reducing the cost of the car body​

Lower energy density chemistry means lower risk of fire

Con:
Pouch cells are soft-sided, requiring supporting frames, meaning complex pack architecture
Packing cells tightly side-by-side yields little area for radiating heat away (most heat must be carried away via

conduction)​

Larger cells limits energy density because more heat builds up in each cell
Larger cells means the failure of one cell has a greater impact on pack capacity
Lower energy density means larger pack size for a given capacity, requiring a larger and more expensive car body

*Of course it's not equal in this case, because Panasonic makes a far greater number of kWh of cells for Tesla than LG Chem makes for GM, so that gives Tesla a significant cost advantage via volume discount.

* * * * *

Overall, I'm not sure which approach has the greater advantage of pack-level energy density. (Perhaps someone else can supply figures for pack-level energy density.) Tesla's use of round cells means there must be some space between the cells, but those cells have higher energy density. GM's cells are flat and are packed tightly side-by-side, but the cells themselves have lower energy density. Perhaps it's best to say that both approaches result in good (or perhaps even superior) pack-level energy density, but use different approaches to get there.
-

 

At ~50% efficient in operation, common sense says that a FCEV's fuel cell stacks must emit less waste heat than an ICEV of comparable power, but still far more waste heat that the much, much more energy-efficient BEV of comparable power.

50% lost, and anyone with even the most basic understanding of science and thermodynamics will realize that nearly all of that lost 50% appears as waste heat, means a lot of heat must be absorbed and dissipated by the car's coolant system and radiator!

You appear to be trying to redefine the word "reality" to mean "delusion".
-

 

However you dress it, using four and a half thousand unreliable batteries to power a car is laughable!

 

Not really crazy. The percentage of the population indulging in this folly remains negligibly small.

 

It will if FCVs keep doubling in considerably less than a year!

Of if Tesla folds - which seems a distinct possibility too.

 

Okay, but 50% more energy dense than what? Certainly not a 50% improvement over the 18650 cells currently being used in the Model S and X, unless the discussion of that subject on the Tesla Motors Club is shockingly misinformed. Perhaps 50% improvement over the Panasonic cells Tesla was using back in 2012, when the Model S was new.

That's the problem with most claims of X percentage improvement in new batteries; claims rarely specify exactly what they are using as the basis for comparison. Rather like TV commercials that compare a product to "Brand X".
-

 

The energy density stays stuck at about 150Wh/kg. But this is 50% more than ...er... battery packs with an energy density of 100Wh/kg.

Satisfied?

 

bwilson4web cited:

Tesla owners that consistently charged to a state of charge between 50% to 75% often reported seeing less overall battery range, presumably because the charge state within each cylindrical lithium ion cell that makes up the overall battery pack is “out of balance”.​

This appears to be written by someone who is very confused on the subject. I don't believe any part of that is true, nor have I seen anyone reporting that they consistently charge to only 50% SoC (State of Charge). Doing so would make no sense, and would offer no benefit. I think the writer is confusing DoD (Depth of Discharge) with SoC; and is also confused regarding the fact that consistently using a lower DoD increases li-ion battery life, rather than decreasing it!

This is contrary to early beliefs that a battery undergoes a “training effect” if not fully depleted and charged to full capacity.
​

The term he's looking for is "memory effect", not "training effect". It is true that you will find many misinformed claims on various EV forums about a "memory effect" if you don't occasionally drain the battery pack to near-zero and fully recharge it. Something similar to the "memory effect" does affect NiMH batteries, which are used in non-plug-in HEVs (such as the non-plug-in Prius models), but there is nothing like a memory effect for li-ion batteries.

More info here: "Five tips for extending lithium-ion battery life"
-

 

Troll bait.

More troll bait.

I'm "satisfied" that you've removed all doubt about your agenda here being to disrupt meaningful discussion regarding EVs, and that you are anything but the EV supporter you claim to be.
-

 

I believe the problem with charging cells in series is that the only way one has of deciding it is fully charged is by measuring the total voltage across them. They are generally charged at constant current and individual cell voltages are not measured.

Unfortunately, the individual cells are not identical, and some have smaller capacities than others. Those with smaller capacities reach a charged state earlier than those with bigger capacities and the total voltage rises. It is possible to get a 'fully charged' voltage from the batteries whilst some of the cells have not been fully charged.

On discharge, some of the cells will discharge before others and under certain circumstances, you can actually get reverse voltages developing across them. This is really bad for them.

So I suspect the best strategy would be to fully charge them always for longer than they 'need', and to not discharge them completely.

It seems to me that 'fast chargers' not only overheat the batteries, they also fail to fully charge them, and repeated use will probably do your battery pack no good at all. I suspect that this sort of worrying about the health of your battery is one reason why battery cars have not taken off as expected.

 

Let me be quite clear.

I am an EV supporter, but NOT a battery supporter. With a huge effort I am convinced you can understand this distinction.

 

I suspect that the more nearly identical you can make the cells, the longer the battery will retain its capacity for the reasons given earlier. So if you happen to get a battery pack in which all the cells were made in teh same batch or on the same machine etc. it will degrade slowly. If youy happen to get a pack incorporating cells made at different times on different machines or from different batches, then you will find them aging much faster. Does Tesla/Panasonic try to make them from the same batch I wonder?

It would explain why different cars show different rates of degredation.

 

They are pretty close to 50% bigger in volume, so all things being equal one might expect them to hold 50% more energy. As they use the same chemistry as the 18650 cells, I can't see there being much difference in energy density. A 50% increase in energy density of Lithium ion cells would surely be headline news.

Could there have been some misunderstanding here, perhaps?

 

Oh come on, "Excoriator". That was your original screen name on TheEEStory forum, and it means, roughly*, "one who is abrasive". You can't really believe I'll swallow such obvious cabbage; you're just practicing your slightly less obvious form of trolling, which given your chosen name "Excoriator", must be intentional on your part.

*pun intended

You've got literally thousands of posts on TheEEStory forum showing how very strongly and stubbornly anti-EV you are. I seriously doubt you're fooling anybody else, either, from some private comments sent my way on that subject.
-