Impressive battery in the Tesla Model 3

Discussion in 'General' started by bwilson4web, Apr 20, 2018.

  1. bwilson4web

    bwilson4web Well-Known Member Subscriber

    Screen shots from the Munro video, this traction battery is awesome. I had my doubts about the cylindrical cells, 2170, because the shells add weight but no energy. The volume efficiency, the amount of battery chemicals to shell overhead increases with size. Large format, cells are attractive for energy density until factoring in the mechanical structure needed. In contrast, this collection of cells makes a very strong, stiff structure.
    [​IMG]
    Notice the assembly is potted with an silicon compound making is both sound deadening and stronger.

    The bus bars are on the top so all of the cells are oriented the same way which simplifies automated assembly. The external 'whiskers' are spot welded to the bus bars which are layered on top:
    [​IMG]

    [​IMG]

    [​IMG]
    All of the cells are in parallel which helps to keep them balanced. Now the real question is whether the 'whiskers' also work as a fuse. When a cell internally shorts, the excessive current might blow a 'whisker' like a fuse. The silicon potting compound helps keep the by products from risking the other cells.

    Battery cooling remains an open question. The silicon potting compound would inhibit heat transfer out the sides and even the top. So it may be the bottom of the cells are in a liquid coolant or touching a heat transfer plate.

    Bob Wilson
     
  2. David Green

    David Green Well-Known Member

    I think this battery looks great, I think the whiskers are the cell fuse, but what impressed me is the silicone fill, I am sure that transfers and balances the heat, and makes the battery easier to control temp over the entire pack. My impression is this is going to work much better in heavy load then the S or X battery.. Impressive... But the tech will now be on the open market for others to copy-improve on.
     
  3. Pushmi-Pullyu

    Pushmi-Pullyu Well-Known Member

    These photos are from the Model 3 teardown? I suppose they are, since you referenced 2170 batteries.

    The TM3 battery pack looks very different inside than a Model S/X battery pack! Interesting that Tesla has made such a radical redesign of their battery pack.

    Here's what a Model S/X pack looks like inside:

    [​IMG]
     
  4. David Green

    David Green Well-Known Member

    Looking at the Model 3 Battery, the S and X pack is now a dinosaur. Although they have to figure an easier way to manufacturer the Model 3 pack, putting the filler in there is going to help so much with heat transfer and temperature balance. Now that all automakers will have all the technicals on the model 3 pack from Munro, it will be interesting to see them reengineer it and hopefully make it easier, and thereby cheaper to manufacture. I think we will soon see more advancements increasing pack density, and lowering cost. Its a win-win for EV's.
     
  5. Pushmi-Pullyu

    Pushmi-Pullyu Well-Known Member

    From a June 2013 Green Car Reports article:

    Intumescent goo

    Having radically simplified the cells, Tesla then designed simple and inexpensive fireproofing systems into its battery pack. Among many innovations, Tesla appears to have incorporated a form of intumescent goo that it sprays onto the interior of the pack to aid in fireproofing.

    When exposed to heat, a chemical reaction occurs in the goo that helps cool the heat source, while simultaneously forming a fireproof barrier to protect the rest of the pack.

    In testing by Tesla, this material often cooled cells experiencing a runaway reaction--to the point that many failed to ignite at all--and provided a fireproof barrier surrounding those that ignited.
    I later read that Tesla stopped using the goo; I'm not sure why. It's interesting that, apparently, Tesla has gone back to filling the spaces between the cells with some sort of material, possibly a fire retardant/suppressant.
    -
     
    Domenick likes this.
  6. Martin Williams

    Martin Williams Active Member

    Cheap way of stopping them rattling about perhaps?
     
  7. bwilson4web

    bwilson4web Well-Known Member Subscriber

    A nice introduction to the 2170 cell:


    Bob Wilson
     
  8. Martin Williams

    Martin Williams Active Member

    Bob

    The energy density of cylindrical cells is maximised by making them bigger. Why? Because the energy is directly related to the volume of the cell which goes up as the cube of its mean radius. The packaging required - the aluminium(?) casing - increases as the square of the mean radius.

    So the maximum energy-carrying to non-energy-carrying material ratio is provided by a cylinder which is (a) as big as is practical, and (b) has a diameter equal to its height. (This last bit from memory. I've just polished off most of a bottle of Malbec and don't feel up to even the simple calculus to prove this.)

    So why in heaven's name is Tesla using these dinky little slim 18650 dimensions? Surely he is ordering enough to make them any size he likes? He has his own factory, so he can bash them out any size he fancies!

    The only practical answer I can come up with is that they are potentially unstable, and a small fire is easier to handle than a big one.

    What do you think?
     
    Last edited: Apr 22, 2018
  9. Pushmi-Pullyu

    Pushmi-Pullyu Well-Known Member

    I think you're intentionally trying to lower the signal-to-noise ratio on this forum with all your "invincible ignorance" anti-EV FUD posts.

    If the only consideration for cell size was volume to surface area, with maximum volume being best, then BEVs would have just one giant cell. It seems rather unlikely that even someone practicing invincible ignorance on the subject, as you are, can actually be ignorant enough to believe there are not other considerations involved. That goes beyond merely deliberately ignoring the facts, to outright delusion on the subject.

    More likely, you know what you're saying simply isn't true.
    -
     
  10. Martin Williams

    Martin Williams Active Member

    I was addressing the question to Bob, really. But feel free to explain why YOU think minimising surface area seems not to have been considered. There are other advantages too. Reducing the number of connections is surely one. Connections are pretty unreliable.

    My bet is that Tesla has been pushed into running his cars with thousands of flashlight batteries simply because making them fatter would mean that the internal temperature could pass the thermal runaway threshold before the event could be detected at the surface.
     
  11. Pushmi-Pullyu

    Pushmi-Pullyu Well-Known Member

    Well see, you do realize that there is a serious downside with making the battery cells too large, with too small a surface-to-volume ratio. Problems related to getting rid of excess heat.

    So why "play dumb" and pretend you don't know that? The answer is, of course, that you're not at all interested in meaningful dialogue on the subject of EV batteries. Your main purpose here is to spread FUD about BEVs, and to promote fool cell cars and the "hydrogen economy" hoax.
    -
     
  12. bwilson4web

    bwilson4web Well-Known Member Subscriber

    My understanding is not only volume but adding Al to the chemistry gave the 2170 almost double the energy per cell. Unlike legacy technologies, batteries continue to improve.

    Bob Wilson
     
    NeilBlanchard likes this.
  13. bwilson4web

    bwilson4web Well-Known Member Subscriber

    Being able to isolate, failing single cells really helps in battery pack reliability. Large, single cells all in series means one failure and the pack is toast. I have experience with failing Prius packs.

    Bob Wilson
     
  14. Pushmi-Pullyu

    Pushmi-Pullyu Well-Known Member

    Yup. Tesla's approach, designing the pack so the failure of a single small cell won't have much affect on overall capacity or pack performance, has been proven in years of use to be an excellent engineering choice. Not necessarily the only good one, but certainly superior than the engineering of some other EV makers' packs in certain respects.

    With other pack designs, the Leaf for example, failure of a single cell might suddenly lower the pack's capacity by between 0.5 kWh - 1.- kWh, or worse if there is a failure of a whole string of cells!
    -
     
  15. bwilson4web

    bwilson4web Well-Known Member Subscriber

    You bring up an excellent reliability issue. How does a single, monolithic cell assembly deal with a failed separator?

    For example, how would a fool cell assembly deal with a failed cell?

    Bob Wilson
     
    Last edited: Apr 22, 2018
  16. Martin Williams

    Martin Williams Active Member

    I think the facts of the matter are that in an attempt to get the energy density as high as possible, an unstable battery chemistry has been adopted. The probability of a single cell overheating is sufficiently high that this possibility has to be accommodated, hence the use of thousands of small calls.

    Taken too far, this would result in the overall energy density falling again, so a trade-off point that yields the highest energy density has to be chosen. The result is that the overall energy density of the pack stays at about 150 Wh/kg.

    Unfortunately, the use of very large numbers of cells means that single failures WILL happen, so the pack has to be designed to tolerate them meaning more arrangements to stop the heat from one cell destroying the next and a progressive destruction of all of them taking place.

    Asking how a fuel cell handles a separator failure is similar to asking how an ICE car handles a broken piston rod. The answer is that it doesn't. The result is a useless fuel cell just as the result of an ICE with a broken piston rod is usually a totally destroyed engine. The trick is to make such failures very unlikely to happen.

    Unfortunately, you can't do that with the batteries you have now. They cannot be made reliable enough. Whether you call this approach 'good engineering' or 'damage limitation' is up to you. To me, there is an air of desperation about it though.
     
  17. bwilson4web

    bwilson4web Well-Known Member Subscriber

    The reason for bringing up total pack reliability comes from this YouTube video:


    The fuel cell sketch shows multiple layers in a single assembly. Although hand assembled, trying to fix one failed stack becomes impractical.

    Bob Wilson
     
    Pushmi-Pullyu likes this.
  18. Domenick

    Domenick Administrator Staff Member

    To address an earlier remark, the reason Tesla choose the 18650 cell is because at that time, it was, by some measure, the most energy dense product on the market. The challenge was to package them in such a way as to limit pack failures and fires. Tesla's record speaks pretty loudly here. Yeah, there have been a few fires, usually associated with a crash or violation of the physical integrity of the pack, and there have been the rare cell failure. Overall, though, it's why the original Roadster had a 244-mile range, while automotive giant Nissan could only manage 75 miles, and really why Tesla is where it is today.

    Regarding the Model 3 battery, it was pretty cool to see it torn down to this level, and I'm looking forward to seeing someone take disassembly all the way down to the individual cell level. Also curious to learn what the green stuff is they scraped off. Is it intumescent for dealing with fire suppression, or is it an insulator? Maybe both?

    Watching EVs take such a range hit in winter, and not enjoying the high heat of Arizona summers, I've been wondering a lot about pack engineering with an eye to temperature management. For instance, would aerogel help maintain temperatures, or is insulation not really the problem?
     
  19. Martin Williams

    Martin Williams Active Member

    My point is that if you can make fuel cell stacks reliable enough, you can afford to junk the whole stack if it goes wrong. It happens so rarely that I imagine an extended warranty from the manufacturer might make excellent sense.

    I think also this fast-talking woman is unaware how fast fuel cells have fallen in price - or indeed she was talking before this happened.

    Finally, I think her point about the need for large radiators is a bit doubtful. As they are rather more efficient than ICEs I would think the radiators would also be smaller. Probably she has been told the rather ugly triangular vents on the Mirai are needed for this reason, but I doubt that. Why on earth would anyone put radiators in such a vulnerable position? I expect they are just rather ill-advised (to my mind) stylistic features. If you can put an ICE radiator well inside, then you can do it with ones for fuel cells too.

    The reality on the ground now is that fuel cell vehicles ARE practical. They appear to work well, and the public seems to like them.
     
  20. Domenick

    Domenick Administrator Staff Member

    If you want to watch the battery portion of the Munro video from which the screen caps were grabbed by OP, here's the link.
     
    bwilson4web likes this.

Share This Page