Some of you may have read about me mentioning my uncle before, as the now retired head of the EE department at BGSU (and now working on a few different EV race teams) he's very involved in EV stuff and last night I was talking to him about the tesla pack tear down thread and he mentioned that he's currently testing some new cells they're thinking about switching too (he's also testing some lead acid / Li-ion hybrids for a different mfg).
Yea it'd obviously take new hosts but 4000mAh out of a sub 26mmØ cell sure would be nice for us.
“22700”
Ears perk-up. That sounds like I could fit those in my Trustfire X6 if I bored out the tube. [mic’s the inside of the tailcap threads] Nope, just slightly too big to still be structurally sound. :_(
Interesting. I figured that since Tesla was now past the bootstrapping phase and actually investing in cell manufacture, it was only a matter of time before they started considering other form factors. Particularly since the 18650s reign in the laptop market also seems to be on unstable ground.
B42, I wondered the same thing but figure that it may well have to do with Panasonic’s proprietary equipment for manufacturing their batteries and the sizes it can handle without major equipment redesign. All the larger size lithium ion batteries seem to come out of China where labor costs allow more hand work. Japanese labor is as expensive as in the USA and western Europe so things are as automated as possible. Fully automated assembly sometimes means considerably less flexibility in regard to things like product size.
Who knows what sizes of batteries Tesla may be producing and using once their joint venture with Panasonic battery plant in Reno goes into production. Presumably all new state of the art machinery will be installed in the new plant.
This 22700 size sounds like a perfect replacement cell for lights & lantern i have that used the three-AAA round cell carriers. ( which measures just over 22mm in diameter) some of them are long enough inside to handle a cell 7 cm long.
It’s a simple empty space between cells issue, as stated in the link of the original post:
“We know that the smaller teh cylindrical cell are the more you can stack in the same volume… so increasing their diameter appear to me to be a strange desicion… I need more details about that…”
The larger the cell, the more empty/wasted space between.
For our flashlights, this isn’t usually an issue, because we often don’t have cells side by side. In fact, we’d probably prefer them bigger. But this is likely the reason the larger format cells haven’t gained popularity. Flashlights aren’t what these cells were ever designed for, we get the hand me downs from laptops/cars/whatever…
If you think through the math, and the construction of the packs (ie lots of extra space for cooling/safety), you’ll realize that its not a consideration.
It’s all about tradeoffs. Pouch packs have few intrinsic advantages over 18650s other than the obvious flexibility in form factors. The same electrode materials and chemistries are in play, which means the same tradeoffs in capacity vs power delivery. There is a tiny weight advantage for pouch packs, but it isn’t that much, because most of the mass in a cell is the electrochemical bits, and that weight advantage comes at the cost of some physical strength and durability. This can be offset somewhat by the fact that the cells are then wrapped in a substantial metal enclosure, but I’m guessing that Tesla probably takes some advantage of the mechanical strength of the cells and sub-packs to maximize strength and rigidity of the battery pack, which is itself a major structural component of the chassis.
As I understand it, the only real inherent advantage of pouch packs is that they have a bit more give for the expansion and contraction of the electrochemical components during charge and discharge, which helps preserve the physical integrity of the electrodes. That has its own downside though, you need to make other containment allowances for gas accumulation.
The main thing RC packs have going for them for high charge/discharge rates is their chemistry/construction (same as the 18650s targeted at use in power tools), and the fact that people have lower expectations of their lifespan than they do of the most expensive component of a rather expensive car targeted at the luxury consumer market, so, you can push them hard. Tesla, on the other hand, babies their cells in their packs.
Also keep in mind that when it comes to charging, just getting enough power to the vehicle is a challenge. Charging a 60kWh pack at a modest .33C requires delivering ~20kW. If I’ve done my math right, that will take >80 amps of current at 240V and >160A at 120V. That’s a shitload. A typical home might have, what 200A service. An electric stove is, what, 40-50A? So point being, the charging current the pack supports isn’t the limiting factor, power delivery (and probably cooling) is.
Under driving conditions, Tesla doesn’t seem to be wanting for power output either since acceleration sounds pretty zippy (sure, there is always going to be someone who wants MORE). I guess higher charging rates would allow more energy recovery through regenerative breaking, but thats probably not a decisive win either.
As the cylinder size increases, the surface to volume ratio decreases. The surface area of the cell is what contributes to the cells ability to shed excess heat.
This surface to volume ratio thing can be a very big deal when taken to extremes For example our Sun only generates about 275W per square meter from Nuclear Fusion at it’s core. That is NOT a lot energy, in fact it is about as much energy generated by a compost pile. Yet the core of the Sun is at millions of degrees C. The reason for that is that the generated heat, as small as it is per cubic meter, can’t get out. The Sun is so huge that it’s surface area to volume ratio is so tiny so as to not allow the energy to be released until it reaches a high enough temperature to allow for a state of equilibrium. That is, the amount generated equals the amount released.
Just did some simple number crunching for energy density. The 4000 22700 is right on par with a KP 5200 26650 in terms of energy density but has the advantage of being much slimmer. Still not as dense as a 3400 Pana but good for the higher capacity still.
For a range of cell sizes, it calculates the efficiency of packing them into a 100x100unit container (ie the battery pack). The model aligns the cells on a square grid. A hex grid would be more efficient, but I’m not sure how to model it, and I’m pretty sure the result would be similar.
The efficiency doesn’t vary much. The inefficiency comes at the edges, when an additional cell won’t fit in the remaining space.
Furthermore, when packing 18, or even 22 or 25mm cells into a tesla pack, the difference wasted space at the edges for the various options at the edge is going to be quite small, since the cell diameters are ~1/100th of the overall pack width.
In case you have a sense that the packing efficiency would be better if the cells were even smaller. No, not on this side of infinitesimal, as shown in the table for a 0.01unit cell diameter (1/10,000th of the size of a pack size).
I could be missing something, though. Anyone else want to chime in?
Guys what kind of coolant are they pumping threw these things? Normal car antifreeze? Some sort of special antifreeze, or are they somehow heating it (even when the vehicle is off) so they can use distilled water or what? Does this coolant flow directly over each cell or threw passageways or what?
Anyone got a link to explain how they do the cooling?
The cells aren’t immersed. You can see the cooling here. It’s ribbon of aluminum channel wrapped in that orangish kapton tape. The channel is woven between the rows of cells. I don’t know what they are using inside, but I doubt it is significantly different from your average coolant. I don’t know what sort of heating/cooling “plant” they hook it up to, but it sounds like it may share the same heating/cooling passenger compartment.
