Greater capacity, more power, smaller size, lighter in weight, easier to manufacture on a mega scale and with less expensive components are the challenges of designing an EV battery. In other words, it comes down to cost and performance. Think of it as a balancing act where the kilowatt hours (kWh) achieved need to provide the most range, but at a reasonable manufacturing cost. Hence, you often seen a battery pack description listing its manufacture cost coming with numbers such as a range of $240-$280/kWh for example during production.
Oh, and let’s not forget safety. Remember the Samsung Galaxy Note 7 fiasco a few years ago and the EV battery equivalents of vehicle fires and Chernobyl equivalent meltdowns. Spacing between cells in a pack and thermal controls to prevent one cell from igniting another cell, from igniting another cell, etc. in a runaway chain reaction disaster scenario adds to the complexity of battery development for EVs. Of which, even Tesla has had its share of problems.
While EV battery packs consist of three major parts: the battery cells, the battery management system(s), and a box or container of some sort to hold it all together, for now, we will take a look at just the cells and how they evolved with Tesla, but remains an issue for Toyota.
The 18650 Cell
The cylindrical 18650 cell is a lithium-ion type measuring 18mm in diameter and 65mm in length and weighs approximately 47 grams. At a nominal voltage of 3.7volts, each cell can be charged as high as 4.2 volts and discharged as low as 2.5 volts, with each cell storing up to 3500 mAh.
Much like an electrolytic capacitor, Tesla’s EV battery cells consist of long sheets of anodes and cathodes separated by a charge-insulating material and are rolled up and packaged tightly into a cylindrical form to save space and pack as much energy as possible. Those sheets of cathode (negative charge) and anode (positive charge) each have tabs for connecting like charges between cells resulting in a powerful battery---the sum of many as one, if you will.
The 2170 Cell
Like a capacitor where its capacitance (power) is increased by decreasing the spacing between the anode and cathode sheets, changing the dielectric (the aforementioned insulating material between the sheets) to one of higher permittivity, and increasing the area of both the anode and cathode sheets, the next step up in Tesla’s EV cell was the 2170 which came in a cylinder slightly larger than the 18650 measuring at 21 millimeters by 70 millimeters, and weighed approximately 68 grams. At a nominal voltage of 3.7volts, each cell can be charged as high as 4.2 volts and discharged as low as 2.5 volts, with each cell storing up to 4800 mAh.
There’s a tradeoff, however, primarily of which is more about resistance and heat than it was about the need for a slightly larger can. In the case of the 2170, the increase in the anode/cathode plate sizes resulted in a longer path for charge to travel which meant more resistance and thereby more energy escaping as heat from the cells as well as interfering with rapid charging requirements.
The 4680 Cell
To create a next generation battery cell with even greater power (but without increased resistance) Tesla engineers designed a significantly larger cell with what is called a “tabless” design that shortened the electrical path and thereby resulted in less heat generated from resistance. Much of this can be attributed to who is possibly the best battery researcher in the world.
The 4680 cell is shingle-spiral form design that is simpler to manufacture and comes in a package size of 46 mm in diameter and 80 mm in length. The weight is not available but its other voltage characteristics are reportedly similar or the same; however, each cell is rated for around a whopping 9000 mAh, which is why the new Tesla tabless battery Is so good. Furthermore, its charging speed remained conducive to rapid requirements.
So, What Does This All Mean?
While going up in size rather than down in size with each cell seems counter to design desires in a battery cell, the improvements made in the power capacity and control of heat generation of the 4680 over both the 18650 and the 2170 resulted in basically fewer cells with more power per pack in the same size pack as used with earlier Tesla models powered with the 18650 and 2170 cells.
To put this in a numerical perspective, that means only about 960 “4680” cells are needed to fill the same space as the 4,416 “2170” cells in the same amount of space, but with added benefits such as a lower cost per kWh production and a significant increase in power using the 4680 cell pack.
As reported earlier about a comparison to the 2170 cell, the 4680 is expected to provide five times more energy storage with 6 times more power resulting in an increase from 82 kWh to 95 kWh in the newer Tesla’s, translating into an expected increase in road range of up to 16 percent.
Final Thoughts
Bear in mind that this is just the basics on Tesla battery cells and that there is much more behind the tech. But it’s a good start for future articles, as we will take a look on how the battery packs are managed regarding power use, as well as safety issues with respect to controlling heat generation, lost power, and…of course…the risk of EV battery fires.
If you are into All-Things-Tesla, here's your chance to buy a Hot Wheels RC version of the Tesla Cybertruck.
Timothy Boyer is Torque News Tesla and EV reporter based in Cincinnati. Experienced with early car restorations, he regularly restores older vehicles with engine modifications for improved performance. Follow Tim on Twitter at @TimBoyerWrites for daily Tesla and electric vehicle news.
Comments
The "whopping 9000 mAh" in
Permalink
The "whopping 9000 mAh" in the 4680 battery does not sound whopping at all considering the 2170 battery has 4800 mAh, which is more than 1/2 the energy but at less than 1/5 the size. I think there must be a mistake here as otherwise you'd need a battery module way larger with the 4680 to get the same amount of power as with the 2170.
The capacity of Tesla 4680
Permalink
In reply to The "whopping 9000 mAh" in by MrCurious (not verified)
The capacity of Tesla 4680 cell has been told "25Ah", not 9Ah.
A 21mm diameter cylinder with
Permalink
In reply to The "whopping 9000 mAh" in by MrCurious (not verified)
A 21mm diameter cylinder with a height of 70mm equates to ~2,307mm³. A 46mm diameter cylinder with a height of 80mm equates to 5,777mm³. That means the 2170 is ~2.5 times smaller In volume. That is not less than 1/5th the volume, correct? Also the 2170 batteries need coolant jackets along the sides of the vertical batteries requiring for the to be spaces out and take up more room. The 4680 batteries need heat dissipated from the top and bottom allowing for them to be placed closer together and take up less room. Please let me know if my math and/or logic is off.
Volume is calculated by Pi x
Permalink
In reply to A 21mm diameter cylinder with by Trenton (not verified)
Volume is calculated by Pi x r² x height. I think you used r instead of r². The volume of the 46mm cylinder is 5.48times larger than the volume of the 21mm. 24,245mm³ vs 132,952mm³
So it's not a miracle, it's a 5 times larger cell with 5 times more capacity. If the makes everything easier and reduces costs has to be seen.
That was my first thought,
Permalink
In reply to The "whopping 9000 mAh" in by MrCurious (not verified)
That was my first thought, too.
Less heat generated, longer…
Permalink
In reply to The "whopping 9000 mAh" in by MrCurious (not verified)
Less heat generated, longer lifespan or more charge cycles and much safer?
Hi, well...the "whopping" may
Permalink
Hi, well...the "whopping" may be a little on the hyperbole side, but as the resources I found state "...the 4680 is expected to provide five times more energy storage with 6 times more power resulting in an increase from 82 kWh to 95 kWh in the newer Tesla’s, translating into an expected increase in road range of up to 16 percent." That, and the information also reports that the number of 4680 cells to occupy the same space as the 2170 battery system in a Tesla is only 1/4 as many, but results in 6 times more power, I believe this is a significant improvement. An engineer could likely put this in a better and clearer perspective. Any engineers out there who would like to comment? All input is welcome.
Maybe it is best to present a
Permalink
In reply to Hi, well...the "whopping" may by Timothy Boyer
Maybe it is best to present a graph of power per unit weight/volume versus energy storage capacity per unit weight/volume for all three designs.
None of the stated numbers
Permalink
In reply to Hi, well...the "whopping" may by Timothy Boyer
None of the stated numbers agree with the results posited in the article. 9000 is not 6 times 4800. I've tried reconciling individual stats on each iteration with the improvement claims and these assertions simply don't add up. Something's my is amiss.
Given the numbers you
Permalink
In reply to Hi, well...the "whopping" may by Timothy Boyer
Given the numbers you provided, 82kWh with 4x as many cells as the 4680 configuration at 95kWh, simple math puts that at 4.63 times more energy on a per cell basis.
not an engineer but Just do
Permalink
In reply to Hi, well...the "whopping" may by Timothy Boyer
not an engineer but Just do the math
pii x r squared x length
21mm x 70 mm = 96,931 cu mm
46mm x 80 mm = 531,539 cu mm
531,539/96931=5.48
so the 4680 is 5.48 times the volume of a 2170
highest capacity of a 2170 cell currently available is 5000mah @15amp max continuous discharge
5000 x 5.48= 27,400 mah required output of a 4680 to equal 2170 energy density.
so a 4680 would have to have more than 27ah before is is more efficient than a 2170
As far as size goes there is no significant advantage to a larger size over a smaller size battery as the proportion of space between them remains the same. With equivalent chemistries they would occupy the same space per ah at scale. There are less batteries though so they could possibly be more reliable. But take up less space. No
I've been using those 18650
Permalink
In reply to not an engineer but Just do by boo radley (not verified)
I've been using those 18650 batteries for years now, and all the ones I get are 30-35 amps. As an aside, my oldest set is almost 10 years old, and still work perfectly even after thousands of charges. Does that amp range help the math?
Hi
Permalink
In reply to Hi, well...the "whopping" may by Timothy Boyer
Hi
The space occupied comment is a little misleading. They may take up the same volume when assembled in a pack, however if you do the math no of cells x cell volume you will find the actual effective volume of the 4680's is around 17% bigger. This makes the extended range a little less impressive.
Yeah, it's not a matter of
Permalink
Yeah, it's not a matter of simple arithmetic. You just don't have all of the data that Tesla has behind their battery---proprietary info you know. For example, the new battery cell may be designed to avoid losing a significant amount of power due to heat loss compared to the the earlier model of cell. But they are not going to reveal how, of course. The article info comes from last year's battery day in which Elon claimed 6X more power.
They have discussed how heat
Permalink
In reply to Yeah, it's not a matter of by D. Banner (not verified)
They have discussed how heat is managed with the cell, at least in part. The large array of tabs means less resistive heating of the tabs, which means an less power losses.
Hmm.. the numbers don't make
Permalink
Hmm.. the numbers don't make sense.
960 cells @ 9Ah each = 8640 Ah
To get a 95 kWh battery, the V of each cell would need to be ~11V (95k/8640)
NMC cells are ~4.25 V/cell, so that does not make sense.
Are the new Silicon cells of a much higher voltage??
Can anyone tell me if this
Permalink
Can anyone tell me if this battery is the NMC cathode
It seems clear that this new
Permalink
In reply to Can anyone tell me if this by Charlie (not verified)
It seems clear that this new 4680 battery has a rated capacity in the 25AH range and they have eliminated cobalt from the design. Given the inherent space lost in a cylindrical shape (made up for in manufacturing efficiencies), the larger the diameter of a battery, the more efficiently it will fill a specific area. They are literally putting a round peg (lots of them) into a square space, so the fewer gaps you have in that space, while maintaining thermal stability, the better.
That's the real advantage of what Tesla has achieved, here. They have increased the volumetric size of the battery (and hence the capacity) while controlling thermal properties by building a tabless design with very low, and very consistent, internal resistance. This new cell really is a game-changer.
Well done Jason. Now can you
Permalink
In reply to It seems clear that this new by Jason Broom (not verified)
Well done Jason. Now can you tell me the volume loss for 500 4860's versus 2750 2170 cells ?
multiply 500 x 5,48 for the
Permalink
In reply to Well done Jason. Now can you by royal gosser (not verified)
multiply 500 x 5,48 for the equivalent number of 2170s = 2740 they would occupy the same physical space. look up circle stacking in wikipedia. But you already knew that didn't you. the six times more power from a 4680 is due to the lower internal resistance which lowers heat build up and allows higher continuous output. Energy density remains the same.
Hexagonal lattice packing is
Permalink
In reply to Well done Jason. Now can you by royal gosser (not verified)
Hexagonal lattice packing is the densest possible packing of circles in a plane, and it's independent of the diameter of the circle (or for cylinders standing on end); the packing density is ~90.69% of the area. The Tesla batteries are already using hexagonal lattice packing, so you're not going to get any significant amount of change in waste volume between the two.
While they may be hexagonal
Permalink
In reply to Hexagonal lattice packing is by Sean (not verified)
While they may be hexagonal packed, each row of the smaller cells have cooling running between them, increasing the effective diameter by about 15%. Whereas the larger cells are cooled top &/or bottom.
Yes they are NVM the data I…
Permalink
In reply to Can anyone tell me if this by Charlie (not verified)
Yes they are NVM the data I've found is this
Cathode. NVM
Weight. 1355 g
Size.46x80mm
Capacity.9000 mah
Total energy.96-99wh(estimated)
I want to find out this info:
Max capacity.?(per cell)
Amp output.?(per cell)
Voltage.?(per cell)Im very excited about these cells making they're way into the open market and being put to use for p.e.vs
Cause I use the 21700 Panasonic's in my electric skate as 12s8p configuration from 2 12s4p batteries ran in parallel and it is able to max out at 80amps continuous discharge (bms max capacity not batteries)60mph and 80 miles of range per charge. So I can only imagine what the 4680s will accomplish and what new configuration shapes will be designed for best convinience of mounting said battery pack to bikes etc.
Cause my 14sx4p on my ebike did 40mph and 25 miles of range a charge. And that was just a 300$ battery
In other reports, there is
Permalink
In other reports, there is presented capacity of 25000mAh per cell 4680. This is much more in accordance with 6 times increase against 21700.
Whopping? That closes this
Permalink
Whopping? That closes this article for me. One to the next one.
doing the numbers the 4680
Permalink
doing the numbers the 4680 would need to produce over 25 maps to equal the energy density of the 2170. The was an article (can't find it now) that showed that when grouping cylinders efficiently there was no difference in proportion of space between cells regardless of size. e.g. bigger cells have bigger spaces between them. So the main advantage is cost to produce.
(note: the battery number is the dia and length if you want to calc it yourself but it comes out that the 4680 is 5.48 times the volume of a 2170
I looked up available 2170 cells that range from 4000mah to 5000mah and choose one that was 500 and had a continuous discharge rating of 15 amp
Remember its the energy
Permalink
Remember its the energy density that is more important (I believe). Power alone only tells you the current capability (V*I actually with both have same voltage degradation curve) which is more important for charging maybe . So the 4860 has roughly the same energy density as the 2170 as mentioned above and with the total space loss the same, then it seems that the lower thermal resistance of the battery will be the main factor for better energy density......Yes?
Change will to could and I
Permalink
Change will to could and I will agree with your statement. However the only available specs do do not bear this out. I look forward to tests of actual 4680 samples when they become available
Upon further reflection I
Permalink
Upon further reflection I realized that we need to define the metrics used to measure lithium cells. The three most common methods are capacity (energy density), power(continuous output in amps) and internal resistance. They are all related. Capacity and continuous output rating are inversely proportional and governed by the internal resistance which is related to the cathode size. This means that the less internal resistance the greater the continuous output can be. This is usually attained by a larger cathode which in turn reduces the volume of anodic material and therefore the capacity of the battery. All other parameters aside. The 4680 can be made with a physically larger cathode without affecting energy density which lowers internal resistance and therefore allows higher continuous output. This is important to a point as continuous output must match or exceed the output of the motors and accessory electronics.
In summary the 4680 design offers a free boost in continuous output at the same energy density over 2170s. This is supported by the battery data charts which show 5,48 times the capacity and 6 times the continuous energy output. Harnessing this to the best advantage is another matter out of the realm of battery specs. e.g. Do you decrease the weight of the vehicle to increase the range by suing fewer batteries do you tweak the battery spec to produce less output over more capacity. It seems Musk has chosen to to stay with the same energy density and perhaps throttle the output through the current controllers already in the vehicles over tweeting the battery cells. But remember 5000 mah is 5000 mah so any range increases would come from the decrease in weight due of using fewer cells.
In summary make the car go further pew kilowatt due to the decreased weight of the battery pack. I believe they said those saving were in the neighborhood of 6%. We'll just have to see how Musk and his engineers work all of this out.
Please remember that I have come to these conclusions through my calculations and the sketchy data available and statements made by Musk in his battery day video which support my data. I may have missed some finer points of battery physics which could skew this data. If anyone has concrete data or knowledge that support or refute my conclusions I would be more than happy to hear them.
Can I have one for my old TmS
Permalink
In reply to Upon further reflection I by boo radley (not verified)
Can I have one for my old TmS (2015)??
Pagination