+86 400-003-5559 CN


Tesla’s New Tabless Batteries Unlock New Levels Of Performance | Hackaday

tagsresistor smd 220

Through cooperation with manufacturers such as Panasonic, LG and CATL, Telsa is one of the world's largest battery buyers. As the demand for electric vehicles and power storage continues to grow, their infinite desire for more batteries cannot be satisfied soon.

, Tesla has been working on various projects to upgrade battery technology to a new level in order to achieve the goal of reaching 3 TWh in annual output by 2030. The most interesting aspect is the announcement of Tesla’s new watch 4680 battery, which will be produced by the company itself. Let's take a look at why the 4680 is so exciting, and why setting the table is so important.

Tesla is different among electric car manufacturers because they firmly insist on using cylindrical cells in their battery packs, while other manufacturers have largely used prismatic designs. Starting from the ancient 18650 popular among notebook computer manufacturers and flashlight manufacturers, Tesla later switched to using larger 21700 batteries. The larger size means that each battery has a larger capacity. To construct these batteries, long, thin sheets of anode and cathode materials are stacked on top of each other with a separator material in between, and then rolled into a "jelly roll" to fit inside a cylinder. The anode and cathode usually have a small lug in the center of the rolled sheet to transfer power to the terminals on the battery case.

These small tabs block the cylindrical hole in a variety of ways. They act as a bottleneck for the current flowing into and out of the battery, because despite the large area of ​​the anode and cathode, all current flowing into and out of the battery must pass through a pair of tabs that are only a few millimeters wide. The electrons from the outer area of ​​the jelly must travel a long distance to reach the battery terminals. Among 21,700 batteries, the electron path length is up to 250 mm. Larger path length means greater resistance, which has a corresponding impact on thermal performance. In addition, these fins hinder efforts to efficiently and quickly produce anode and cathode sheets, and production machinery must be stopped and repeatedly started to cope with outstanding features.

When switching from the 18650 battery to the larger 21700 design, Tesla had previously improved performance, but efforts to further increase the battery size have encountered obstacles. Although larger batteries can store more energy and save production costs, heat dissipation issues mean that charging time and discharge rate will be adversely affected. Larger batteries mean longer path lengths, while higher resistance means less power output per battery and slower charging speeds. Even with Tesla's fast charging technology, many people still think that the charging speed of electric vehicles is too slow, so this is not a trade-off worth making.

Enter the "table" battery. The entire anode foil and cathode foil are not laser patterned, but are processed, and basically have many tiny tabs throughout the length, instead of connecting a small battery tab to the anode and the cathode. . It replaces the step of manually attaching a separate tab later in the manufacturing process.

When the anode, cathode, and separator are all rolled up, these many smaller tabs flatten out to form a "ribbon spiral," thereby creating a larger contact area between the active battery material and the casing. This means that the path length of electron propagation is greatly shortened. Tesla’s offer is 5 times lower than the previous design. This is due to the fact that electrons can now move directly to the battery terminals instead of having to go through a more round path to the center of the sheet to reach the single-terminal connection.

The final result is a 4680 battery with a diameter of 46 mm and a length of 80 mm. This is different from the five-digit nomenclature, but no one at Tesla can figure out why the tail of the 18650 battery is zero, so the company deleted it from the name of the new battery. It is said that due to the larger size of the new battery, its energy is 5 times that of the earlier design. Tesla claims that it is better that they can provide up to 6 times the power due to the reduced electrical path length of the table structure, which can provide better thermal performance. It is estimated that the conversion of 4,680 battery cells in a Tesla car battery pack could result in a maximum range gain of 16%-an impressive number considering the automaker’s already impressive numbers in this field. For example, the upcoming

Use 4680 batteries.

The new design also brings production benefits. By laser patterning the anode and cathode for direct connection instead of connecting separate tabs, the material can be processed through a continuous rolling process, which is no different from papermaking technology. This guarantees a huge increase in production speed, enabling the machine to move at a continuous high speed without having to constantly accelerate and decelerate in order to fix the tabs to each anode and cathode plate. Tesla's goal of producing more batteries to meet demand cannot actually be achieved with current production technology. Therefore, improving processing speed and production speed is the key to solving this problem. This will also result in lower costs, which is an important part of the company's efforts to produce more accessible electric vehicles at a price of $25,000.

These new batteries with higher energy density and high power output will attract a huge market for hackers and manufacturers. However, in Tesla's keynote speech, the continuing theme is that they simply cannot provide enough batteries to meet their needs. We suspect that because Tesla retains all its power for internal use, it will take several years for the desktop battery to enter the open market. Due to the benefits provided, other manufacturers may scramble to develop similar technologies, but this will take time, and at the same time, those who want the best cylindrical batteries will have to wait for new Teslas to appear in their local destruction field. .

The table technical announcement is only a small part of the Tesla Battery Day announcement. In order to achieve the company's lofty goal of increasing battery production to meet global demand, it is working hard to make progress in other areas such as anode and cathode chemistry and production technology. If electric transportation and Powerwall grid storage can truly change the world, then similar projects will have to be rewarded-otherwise we simply don't have batteries to fit in our cars!

Thank you for drawing without y axis. :(

This is a relative comparison, so it doesn't matter.

Correct. The label doesn't matter at all :)

Of course there is, it just has no labels!

B ^)

There is a y axis. It is labeled "Boost Time Increase" and has a scale ("0") on it. There are also some very faint horizontal lines suggesting (but not sure) that we are looking at a linear rather than a logarithmic scale...

The wider "tab" should also greatly help cool the battery and speed up the charging?

Yes, wider tabs (forms;) will help heating in two ways: First, less heat is generated by the resistance in the jelly roll, because the electrons only need to traverse ~40mm to the tab end instead of the larger ones Set the jelly roll on the roll ~600mm format as the center click. Second, the copper bottom conductor and possibly similar aluminum top conductor will act as a fast heat pipe for the battery. Aluminum and copper also have excellent thermal conductivity. The heat will be so great that the efficiency of the battery pack will be improved, and the cooling requirements will be greatly reduced.

As far as I know, due to the low resistance, the battery does not strain heat during charging and discharging (because electrons do not have to travel far). Heat is the limiting factor for charging and discharging (bad ignition), so it will reduce the time required for both while maintaining a similar heat output.

The change in current direction seems too obvious. Is there a separate innovation that makes this construction method possible, or is it just a situation that no one has tried before?

Dave also explained everything in detail :-)

Thank you! ask

Has Tesla considered a kilovolt laboratory series supercapacitor that can be charged in a few seconds?

The interesting thing about having such a "wide label" is that, to be honest, it is nothing new at all and has been on the market for a long time.

Both capacitors and batteries are suitable.

However, frankly speaking, there is no need to use such continuous labels. From a traditional single label to a single label at both ends, the battery performance can be more than doubled without worrying about folding the label into the package. But having more than two is of course still advantageous.

However, I want to know whether the near-continuous copper tube actually has an advantage, or it will only add more copper volume at both ends than its performance. Diminishing returns is one thing after all. Moreover, if you only increase the weight and reduce the density, then "wasting" copper on... is not a worthwhile thing.

That is, in the end, how often should a label appear, and how wide should it be for best performance?

Although, to be fair, electric vehicles should use more supercapacitors because their charge/discharge losses are much lower and their power density is much higher. However, the capacitor obviously should not completely replace the battery, but should be supplemented for acceleration and regenerative braking.

I want to know whether the super capacitor can be used as a "buffer" between the battery and the engine for regenerative braking/acceleration of spikes. However, given the performance of this new cell design, you can get all the benefits of caching through a highly simplified cell/packet structure.

The copper plate produced by the lugs at the bottom and the aluminum plate at the top also act as heat pipes, which is a very low resistance battery. Cooling will be greatly reduced/simplified.

The "accelerated" label design in their patent makes this work possible where others have failed. Instead of cutting the tabs at the backing paper spacing, the tab spacing bends upward (when drawing the figure) to create a "flat" tab end, as opposed to the bumps produced in other cases.


Compared with capacitors, the problem with capacitors is that their discharge curve is very steep.

Therefore, in fact it must be a separate group with a bidirectional DC-DC converter between them.

However, this does not prevent us from adding some regular 450 Vdc aluminum electrolytic covers to make things smoother.

Yes, increasing the number of lugs on the battery itself will help keep the battery cool.

But the same is true for super capacitor banks. (Because of such a battery pack, the battery does not really need to consume a lot of power. (Although charging is still a "problem", fast charging is bad for the battery, not only because of temperature-related issues.)

Just wait until someone manages to make a capacitor in the GF range.

I don't think I want to sit on it.

However, because our battery is too large, the battery can already meet everything needed for inrush current. We don't have a motor that can handle 2000A. The electric motor and the battery are fully matched, which can reach 0-60 times in 3 seconds per second (the latest and most expensive car is less than 2 seconds). I think the super capacitor used in the car is just a waste. Maybe it's a Formula One car or something, but it's not aimed at ordinary consumers.

In other words, although what you said is correct, it is unnecessary for consumer cars, and the small (small) benefits are not worth the complexity or cost.

Until you understand the manufacturing cost of the battery.

The impact of the product on the environment.

The weight of the vehicle increases, leading to increased road wear.

Now, some companies don't care about these details.

For example, Koenigsegg's battery discharge temperature is as high as 60C, which is quite difficult in terms of discharge efficiency. They, if anyone should use capacitors. But they did not...

Now, capacitors can provide many kA of output current and easily provide hundreds of kW of power.

But this is what you call "no need", and I agree.

But to get any meaningful amount of power from the battery, it needs to become very large.

And this kind of emissions will also be quite inefficient.

The battery has good efficiency at low discharge rates. For most lithium-based batteries, it is necessary to reach above 2-2.5 C, just like climbing Mount Everest. A battery can indeed provide enough power, but it burns more power due to the heat in the battery. Literally, this is what our scope becomes called heat.

A small capacitor bank can reduce the peak power consumption to a more reasonable level, thus ensuring that we do not waste the acceleration range.

It can also reduce the overall weight of the vehicle, not to mention the price.

The bidirectional DC-DC converter is not difficult to develop or manufacture. If it becomes a bottleneck, the capacitor bank is not a problem to a large extent. Because the energy required by the capacitor bank is about the same as the energy required by the vehicle when driving at "typical" highway speeds at its maximum rated weight.

Similarly, in terms of output current, our capacitors are logically most likely to be connected in series, so they may only provide hundreds of A or low kA at most, although they may be between 200-500 volts. (This is not very unreasonable. For example, the Tesla Model S has a total of 615 kW motors running at only 400 volts. Therefore, this is 1537.5 A, regardless of the loss of the drive system, so the peak current is higher)

That car uses a typical 18650 battery.

Configure it as 74 batteries in parallel, of which 6 batteries in series form a group.

16 groups of them are connected in series to form a complete battery pack.

This means that each battery cell requires a total of 400 volts, and each battery requires approximately 4.17 volts. (However, as the battery discharges, the 400 volt voltage will drop. More current is required to obtain a power output of 615 kW.)

But this also means that our 1.5 kA is only shared among 74 batteries. Or after fully charged, each battery is 20 amps.

If we generously say that the capacity of these 18650 batteries is 3 Ah, then they will be discharged at a temperature of 7C, which is much higher than 2-2.5C, and the discharge efficiency has climbed the wall...

Now, most people do not step on the pedal, but say that the 7C is a fully charged battery.

However, it can be said that there is 3.5 volts left in each cell.

Only use 2C (6 amps) among them. Then, we get 150 kW of power.

If our power transmission system is magical and the efficiency is 100%, then the speed from 0-100 km/h will take 11.6 seconds. (0-62.5 mph)

When driving a Tesla, few people step on the pedal.

in the end.

Batteries are very suitable for energy storage, but they lack the ability to operate efficiently.

> "With the fully matched motor and battery, you can get 0-60 times in 3 seconds"

Well, "perfect". Tesla's "ridiculous mode" only works when the battery is fully charged and exceeds a certain charge threshold, because they are moving towards the battery's discharge current limit. They filed a lawsuit in Norway for false advertising promoting accelerators because the Model S will reduce power when the battery is not fully charged.

If the chemical nature's inherent rapid charge/discharge efficiency is low, it is inevitable to heat up and limit the charge/discharge. If conductor loss is the main cause of heat generation during fast charging/discharging, then Tesla’s improvement makes heat generation 1/6 of the current problem, and the use of supercapacitors does not increase cost and complexity.

Supercapacitors are a useful supplement to battery packs, and their ESR and ESL are much lower than batteries. However, their energy density is much lower than batteries. Supercapacitors have high power density because of their low equivalent series resistance, so they are very useful for covering short-term current bursts. For example, the largest supercapacitor battery from Maxwell on the DigiKey website is 2.85V and 3400F. Quite big. However, in terms of battery, it is a 1.4Ah battery. It is larger than two D batteries (60mm D x 140mm L). This is not a good battery. But it can provide a 2000A burst without damage.

Yes. Supercapacitors are somewhat moderate in energy density.

In most applications, this is not a technology that can replace batteries.

However, due to their high power density, they do complement the battery well.

This reduces the peak current drawn from the battery pack, because the main power required for acceleration will come from the capacitor bank.

It is the same as regenerating electricity, where we can plug the energy back into the capacitor bank. And because of the extremely low charge/discharge loss, we can get a larger range.

Simply removing the acceleration-related load from the battery pack may increase the cruising range by itself. After all, compared with capacitors, battery chemistry itself is not that impressive in terms of ESR. (Adding more labels will only reduce the conductivity loss, and it will not have much effect on our chemical-related losses.)

But I have also seen that supercapacitors that target energy density are often less noticeable in terms of power density. Because they have fewer tags, and more tags will take up more space. (More tags also increase costs, and energy/dollars are the same thing.)

I looked at a bunch of capacitors and found that the energy density is about 35 kJ/kg, and the power density is also good (about 4-8 kW/kg), so a car driving 2 tons at a speed of 100 km/h has the same energy Density kinetic energy, because we can store it in a capacitor of about 44 kg. (Or about 175-350 kW output power (240-480 hp).) Although it is best to maintain a good discharge efficiency in the area, our battery can still provide a small amount of power.

But it is clear that we will have a larger capacitor bank so that we will never discharge it to a low enough voltage unless it becomes difficult to handle. (For example, the first volt of a 2.5 V 500F capacitor stores only 250 J, and the next 1.5 V stores 1312 joules (1562 joules in total), so when discharged below 1 volt, most of the energy is negligible. Increase The cost of the DC-DC complexity (to be precise, this line is a different issue, and 2.7 V or 2.8 V capacitors are also available on the market.)

Super capacitors are also quite bad in terms of self-discharge, collision safety, and fire safety.

Any capacitor that stores large amounts of energy under electrostatic tension is basically a bomb. You break the insulator and it immediately explodes.

Yes, compared to batteries, capacitors do have quite a bit of self-discharge.

But it was still so slow that it did not attract special attention for many hours.

In terms of collision and fire safety, they do pose a threat, but batteries also pose a threat.

When the battery is inserted, the main reason why the battery does not explode is due to the contact resistance between the shrapnel and the conductive sheet. Usually leads to self-fusion around the invasion. (After all, unlike the case where the contact area of ​​the lugs on the battery is much larger, intrusion will not keep them in good contact with the sheet.) The same is true for capacitors.

Regarding fire, batteries are not so safe...

Not to mention that both battery packs and capacitor packs can have protective casings around them to prevent physical damage and to prevent fires.

However, even the conventional fuel tank on an ordinary vehicle has its own risk in the event of a collision.

Of course it can, but it will take days or even months-if you retain any energy in the supercapacitor, you will lose a lot. When parking, the capacity of the capacitor will be exhausted, so if there is a kilowatt-hour there, you will lose a kilowatt-hour...

As for the puncture safety, the capacitor is different in nature, that is, the discharge is almost instantaneous and is not limited by the chemical reaction rate because it is electrostatic. A very large very large capacity capacitor is a trigger situation. Smaller faults may fuse, and larger faults will evaporate the material in an instant and completely explode the capacitor. The battery has not yet done so.

What you need to understand is that if the separator is damaged in a chemical battery, as long as there is a separator, the object that pierces the separator will provide electrical contact. In an electrostatic capacitor, different charge carriers are physically attracted to each other through the gap, and the whole thing is lost like a party balloon.

This is why storing large amounts of energy in electrostatic capacitors is a stupid idea. As an energy storage medium, it is like building a huge watch spring with ultra-high tensile strength steel, hopefully it will not get stuck.

When turning off the car, it is best to transfer the energy stored in the car to our battery pack. If we suddenly decide to start the car in the next short period of time, it can wait to do so.

But under normal circumstances, the discharge of the capacitor is not "fast". I left a 500F "ebay special" supercapacitor under a voltage of 2.5V for more than a month without losing more than 0.1V. Voltage. Therefore, I will not worry too much about self-discharge. (This is a supercapacitor, which emits much more electrical energy than batteries. Under constant current charging conditions, it also has linear charging characteristics for batteries.)

Similarly, the capacitor bank does not have to be too large and fair.

Because at full load and driving at a "reasonable" speed, we don't need to store kinetic energy equivalent to a car. Like 100-130 km/h (the typical highway speed in the world is this number).

For a 2-ton car, this is only equivalent to 2.6 MJ of energy, which is approximately equal to 0.72% of kWh. (2 tons are moving at a speed of 130 km/h.)

Therefore, a 1 kWh capacitor bank is only needed in very extreme cases, especially considering that some of the energy used during acceleration can still be provided by our battery bank. (The same goes for the story about regeneration disconnection. The capacitor bank only handles most of the power.)

In terms of security.

Considering the very thin conductive layer that constitutes the capacitor, it will be difficult to contact any intrusive objects. This is the main reason why lithium batteries often do not care about nails or hammers driving into them. The flakes surrounding the intrusion are fused at the intrusion by simply evaporating away.

Although, that is a nail. I haven't seen any tests for forcing through the equipment with a big blade. Since this may reduce the current density on the thin plate due to a larger contact area, it may make the phenomenon of equipment failure more obvious.

take a look

Especially youtube videos about security. They also cited an energy density of 70 Wh.kg. I have seen some of them run in residential applications, and apart from ergonomics, they compare well to the Tesla Wall. But they will be more durable than Tesla and maintain complete performance...

> Very poor contact with any invasion

Likewise, the separator in the capacitor will not do the same thing. In lithium ion batteries, the separator conducts electricity. In a capacitor, it is particularly non-conductive and separates the two halves of the battery.

After the insulator ruptures, the hole itself will leak charge, which will cause the capacitor to discharge itself quickly. Electrostatic discharge is not limited by the rate of chemical reactions such as batteries-all charges rush to the holes in the insulator. If the energy density of the capacitor is high, a plasma arc will be formed, which will emptied the capacitor and evaporate the material in an instant, causing an explosion.

Someone once said: "The best part is that there is no part." I bet someone will say that they will consider these issues in great detail. While considering cost, part weight and power density, it can even be compared to the rocket equation. Although it is impossible to imagine that someone like Elon would have heard of... ;-)

Reducing the number of parts can indeed reduce system complexity and lower prices.

However, since a single component rarely performs/execute a person's application well, people need to find complementary components that together can do better than two components alone.

For example, batteries do not like to output several tons of electricity at a given capacity, but they do have a considerable energy density.

Although the capacitor lacks glossy energy density. But on the other hand, they don't mind putting in literal kilowatts per kilogram with minimal losses.

Therefore, when we want short and very high power pulses, the two tend to complement each other well. It's like accelerating a car.

The disadvantage is that their discharge characteristics are completely different, so one or both of them need to have some DC-DC converters. (For automobiles, the DC-DC also needs to deal with regeneration interruption, but this can be the second DC-DC, but the bidirectional DC-DC is actually not difficult to design and manufacture. (especially if there is synchronous rectification, the actual It has two-way rectification, but only a suitable controller is missing.))

The biggest problem is the power generated by regenerative braking. It doesn't matter that the vehicle can travel at 60 miles per hour, because most cars slow down at roughly the same speed at the legal road speed. The super cap buffer can solve this problem. This problem is even more serious when electric drives are used on trucks. The electric engine in the train solves this problem by feeding power back to the duct.


Capacitors don’t mind simply drawing power.

Unlike batteries, if a battery is charged much faster than 3C, it will sweat.

Even if they "happily" discharge at 7C.

Moreover, compared with batteries, the charge/discharge loss of capacitors is much lower.

In this way, regenerative braking can improve efficiency and is more feasible in practice because it can increase the driving range of electric vehicles in urban traffic. In urban traffic, start and stop are often more common, not just resistive combustion through resistance. Or friction.

I don't understand why capacitors are important here?

The low regeneration efficiency is not about storing electric energy, but about generating electricity. Converting mechanical motion into alternating current through a motor and then rectifying it into direct current is an inefficient place.

Currently, vehicles tend to regenerate at 1/3 to 1/6 of the maximum towing rate. The regenerative power of the bolt is 70kW, and the maximum power consumption is 160kW. In the case of Tesla, they regenerate (77kW) at 1/4 of the supercharging rate (250kW).

In other words, it is easy to store electrical energy in the battery-this is not a limitation or inefficiency-switching to a capacitor will not change the situation.

If the regenerative power is 10 times the power, then yes, you will have a point. Since both the extraction and storage capacity depends on the size of the battery, it will not change for large vehicles.

Musk talked about how to stop the production line to handle the label. A cover design like this avoids this, so it makes the production line move faster because it doesn't have to deal with acceleration restrictions.

In addition, continuous tabs will provide better thermal connectivity, which will also help the battery not to overheat.

Supercapacitors can further increase energy density and can actually completely replace batteries in electric vehicles. This is expected to be achieved within a few years or even three years. It has been used in forklifts and buses; however, in buses, they are too small to maintain a good driving distance and must be charged for a few seconds each time they stop. I actually used a supercapacitor AA battery in my computer mouse, it works normally and can be charged quickly. They provide more than one million cycles without degradation, and do not have the safety issues of lithium batteries, while being able to work at high temperatures without cooling. At the same time, the RTE of the supercapacitor battery is 99% DC-DC.

I suspect that excess copper at both ends can make money. It significantly reduces the overall resistance. Before you consider the high currents involved, the reduction in resistance seems small. In addition, the excess copper will greatly reduce the thermal resistance. Therefore, performance can be improved in three ways. First of all, since lower resistance = lower dissipation, the heat generated during charging and discharging is reduced. Secondly, due to the lower thermal resistance, the heat generated by the IS can be more effectively dissipated from the battery. Third, lower heat actually leads to lower resistance, because the resistance of (most) conductors increases with temperature. All of this means less power wasted, faster charging time, miles of distance when power drops, and significantly higher power density.

So when and where can we order them for our project?

If these products enter the open market, that would be great! I really want to prepare a set for my roamer.

Is Telsa one of the world's largest purchasers of telecommunications equipment?

Yes, it's funny, the first word is misspelled and is a good way to start...

I heard that the trailing zero means cylindrical.

In retrospect, the thickness of CR2032 is 3.2mm, maybe the convention is to measure the height as 1/10mm.

Yes, I think so. They said it was strange that they couldn't figure it out when it was obviously based on cells under 10mm, or even a quick Google search result.


After October 1990, round cells were systematically identified based on their diameter and height.


If Tesla (employees) cannot read Wikipedia, then he may not be so smart.

It makes sense for all button batteries that use a tenth of a millimeter resolution, but just because the battery is slightly larger and deviates from the standard does not, at least not to me.

The discussion on this is only Tesla’s free advertising. This is probably the biggest reason for this.

I am also surprised that this reaches hackaday. For some strange reason, it seems that this completely non-problem with batteries has been hyped up, which has been used in foil capacitors and high-discharge batteries for many years. This may even use paper-wound capacitors more than 80 years old.

These are huge advancements in battery technology, I like to read on hackaday

Well, my opinion is that this reduces manufacturing cost/time more than anything else. For example, when you realize that the volume of the battery has increased 5.5 times, the claimed 5-fold increase in energy storage and 6-fold increase in power are not impressive. It seems that the energy storage density has dropped, which may be due to the volume loss at the end of the battery.

I also don't believe in the so-called thermal improvement. Of course, it seems to dissipate heat faster, but if it doesn't significantly change the self-heating in the battery, then it won't change the amount of heat you have to deal with. It still has to go somewhere, and transferring heat from inside the battery to the terminal is only a small part of it. Basically, what it does is solve the problems caused by increasing the cell size.

In other words, I am not drinking Tesla Kurt aid. They have a strong marketing team, but there is no reason to believe that they have accomplished incredible feats that no one else can.

The bigger step is to allow the use of the central hole of the battery for cooling.

I know that liquids usually don't hold together well, but why not use a non-conductive liquid (better than air) to cool the battery "inside".

When batteries are stacked in series, ducts are naturally formed. This way you can cool the cells from outside and inside; it would be better, wouldn't it?

Gray-yes, it does change the self-heating inside the battery-this is a factor of resistance. As mentioned in the article and Tesla's introduction, since the path that the current must pass is significantly reduced, the resistance is greatly reduced, and the amount of heat generated is also greatly reduced.


Understand that with a single-pole design, the current flows from the farthest layer to the place where the tabs are connected. With a continuous tab design, current must only flow from one side of the layer to the other. The shorter the length of the resistive material means the lower the resistance. Such fairness will inevitably reduce internal resistance.

But my point is that we haven't seen any data showing how much the reduction is, especially when compared to the resistance of the entire battery. For a battery as large as 4680, I guess there may be a conductor resistance of a few mOhm, but if it exceeds 10% of the battery's DC resistance, in other words, it is only enough to offset the heat, I would be surprised. Problems caused by increased diameter. I'm guessing, but I don't mind seeing some real data.

And I mean, this is an achievement in reducing manufacturing costs, not an achievement in battery technology. How to extend the battery life by 16%? I'm pretty sure they are just using cost savings to expand packaging. They must increase it by about 28% to offset the reduced energy density of the new battery.

Note that the graph provided by Tesla marks the y-axis as "boost time increase". No matter what this means, it is not real data, but market hype. Beware of all hypothetical "data" without units and measurements, nor full descriptions. For 4650, the difference may be 5 seconds and 5 hours, or 5 seconds and 5 minutes. who knows?

Generally agreed, yes. But we did mention some numbers.

Tesla said that the current path is reduced from 250mm to 50mm, so the average distance is one-fifth. Since all the heat generated in the battery is proportional to this, only one-fifth of the heat. In addition, since the table design has better thermal conductivity in the entire battery, the overall heat dissipation capacity is also better. Therefore, your calories are only one-fifth and it is easier to remove. Since this is the basic limitation of fast charging, I think we can expect a big improvement. But yes, all of this is now market hype. Let us see what it actually is.

Please note that, in fact, we don't actually need to pressurize so fast. We are already very close or close to the best location (IMHO, this is the best location 180 miles in 15 minutes). Therefore, all we need is a slight improvement and a longer service life, which will undoubtedly provide these advantages.

Telek, sorry for your delayed response, but I want to be more thorough.

I am not a battery expert, but I think that all the heat generated in the battery is directly proportional to the Joule heat in the copper foil, which is incorrect. On the one hand, the current also flows through the electrolyte, the resistance will have a finite resistance, and there may also be resistance from the electrolyte to the lithium and graphite electrodes to the copper anode and aluminum cathode. The chemical reaction that occurs also affects the heat, but during the charging process, it is endothermic (cooling) and is not important compared to the heat generated, so I will ignore it.

The typical copper foil used for the anode in the 18650 battery is about 6-10 microns, with about 18 layers (layer thickness slightly less than 0.5 mm). We can easily calculate the conductor resistance of copper foil:

–Assuming that the foil is only 6 microns thick, using a copper resistance of 1.68e-8 Ohm-m, and assuming the height is about 60 mm (at least 5 mm will enter the terminal, etc.), the foil resistance per m length is 1.68e-8 Ohm -m / (0.000006 m * 0.060 m) = 46.67 mOhm / m.

– Assuming that the average layer length is half of the battery circumference, we can estimate the length of the foil. The length of the outer layer is (18 mm * 3.14) / 2 = 28.26 mm. For 18 layers, it is 509 mm. Please note that when comparing it to Tesla’s 250mm length, they must refer to the average length of the current path, and must also be compared to a 2-label or multi-label design, because even considering the Inaccuracy, I think it is too short for a battery of that diameter.

– Taken together, the total resistance of the entire foil is 46.67 mOhm / m * 0.509 m = 23.8 mOhm.

– Assuming that the resistance of the aluminum cathode foil is about the same, the total conductor resistance of the two is about 45-50 mOhm. Keep in mind that this is made with 6 micron foil, and for power optimized batteries, its thickness may be 10 microns or even thicker.

– Compare it with the typical DC internal resistance of an 18650 battery in the range of 80-100 mOhm, and consider the fact that internal resistance increases with age, which means that in addition to resistance, there are other factors that will affect Conductor of internal resistance.

It is now used for 18650 batteries, and we can certainly get more information. Consider the size of the 4680 battery: as the diameter of the battery increases, the length of the anode will become longer and the width of the electrolyte layer will become wider. Therefore, with the single-tab design, it is possible to reduce the ratio of the resistance generated by the electrolyte etc. to the same ratio as the increase of the anode/cathode resistance.

Looking back, I estimated that the anode resistance should be less than 10% of the total internal resistance, which is very low. It might be more like 40-50%. Therefore, reducing the anode resistance by 80% can reduce the total resistance by about 40%, which is not trivial, but not 80%. So I met you somewhere in the middle.

Thermal characteristics information during charging/discharging:

Internal construction information:

It should also be noted that in this article, there is a linked IEEE paper that provides some insights on the impact of the number of tabs on performance, and suggests that the current density improvement obtained by adding tabs is better for large units than for small units. Is important. It may also be a factor that prompted Tesla to use continuous labels.

It is indeed very smart, and can effectively charge each layer in parallel and discharge in series. The voltage doubler in the battery world.

I don't know where you got this idea. To be precise, they can be charged and discharged in parallel, but this is not like there are separate pieces here, but just a single large chaff rolled up together.

I think I miss it. That said. It might be interesting to create a battery that can be charged in parallel and discharged in series (and vice versa). There is no doubt that people smarter than me are doing this kind of work.

There is no meaning (or very little meaning). Remember, this is mainly about power, and the average distribution/extraction of power is done from each unit, regardless of its layout. If anything, you want more serial because you want higher voltage and less current.

For example, if you bring batteries and use the 96S3P configuration in 288 cells, 96 cells in series with 3 cells in series in each battery pack will get a voltage of 400V. This is what Chevrolet Bolt has. So far, most vehicles have good reasons to adopt the 400V standard. Some will reach 800V.

Parallel charging will not change anything. If you can somehow magically switch it to 48S6P (more parallelization), then you will get a voltage of 200V, but the current is doubled (so it is now four times lost). This backfired.

Serial charging is not better – of course, you can use 192S2P (if we have 50% more batteries – it’s another problem in itself and the cost is higher), but now your voltage is 800V – higher voltages need to be More insulation and isolation of lines and circuits between transmissions. Use this function only when the required power exceeds 400V, mainly for charging considerations.

Remember-if your battery pack with 288 cells has a power of 72kW, then regardless of the series/parallel arrangement, the power of each battery pack is 250W.

My friend built an electric car with Audi V8. This is indeed the part that hinders the usability of the car. He built a battery module using 18650 batteries connected in series to achieve a nominal voltage of 48V. As the series resistance increases, this will hinder the maximum charging current and cause heat dissipation problems. Now the batteries in the module can be charged in parallel...

Reply to Yossi:

But this is done through DIY settings, mass production vehicles are 400V or 800V, so there is no restriction.

Hmm... As a home hacker, I will not be able to buy these products, so it is not very interesting. For all the advancements in lithium-ion battery manufacturing, I personally have seen that average prices have stagnated for the past 5 years. Consumers buy 10s-100s batteries. BMS chips and charging circuits are still expensive. The entire industry has almost stagnated.

It is tempting to sell local BMS and lithium-ion charging boards suitable for different voltages at first, but there is a factor of responsibility, so maybe not laugh out loud.

I am a little confused about this. They claim to have 5 times the energy, 6 times the power and 16% range. However, the new cell is 5.5 times the old cell. Therefore, I don't think the numbers will line up-unless I missed something? Should the range/energy number at least scale with the size? I know that the energy density will increase (this increases our range by 16%), but why is the energy number *less than* 5.5 times the volume difference? That is, it should be 5.5 * 1.16 = 6.4x energy. Similarly, as the heat decreases significantly (the boost chart shows about 1/10 of the resistance), should the power value increase significantly?

I immediately thought these numbers had been turned away by the marketing department.

If they can make lithium-ion batteries with an energy density five times higher than their competitors, you will hear it through different channels.

I am also not sure if I dare to approach this battery.

The power of 4680 batteries may be 5 times that of old batteries, but their capacity is also much larger. There cannot be more than one in the same area. It is more efficient due to reduced packaging materials and interconnections. Coupled with advances in battery chemistry, your mileage has increased by 16%.

Yes, but I have already introduced it. The size of the unit is 5.5 times that of the old unit. Therefore, there should be at least 5.5 times the energy, otherwise it won't make sense-as I mentioned, this will not add an additional range of 16%.

The energy density of the battery seems to be lower, but this is offset by its lower ESR. This translates into lower heat generation, which leads to higher efficiency, and less restrictive cooling systems (also combined with the lower thermal resistance of the new "tab"), which translates to lighter battery packs Or less cooling energy.

All in all, I think it is feasible to increase the range by 16%; energy is just being used more efficiently.

In order to do this, if the density is small, the loss in the current design between the battery and the motor must far exceed 16%, but the efficiency of this new system will increase by 16%. The loss in current designs may be close to 2%. So that cannot be the reason. The efficiency loss between the wall and the wheels only accounts for 20% of the total loss, half of which is in charging, and most of the rest is in the inverter and electric motor. The discharge efficiency of lithium ions is usually above 99%.

I think the missing part is that they don't need as many parallel batteries as smaller batteries, so the same total volume and fewer batteries in parallel can get a larger current capacity.

Power is not the same as energy capacity, so this is not the problem. Using the same technology, twice the size of a battery can provide twice the power and twice the energy. Therefore, there are other things at work.

Fewer cells.


The table design allows the battery to discharge faster.

For a given volume, less volume will be absorbed by the shell.

When you do actual engineering, few things will scale linearly.

Although the energy of a single battery is 5 times higher, because they are larger, each battery cannot accommodate as many as 18650 battery packs. On the contrary, reducing the use of a single battery and the increased efficiency brought about by lower resistance means that you can get 16% of the extra range.

As for why the battery has to be 5.5 times larger, but only 5 times the energy, well, there may be some packaging considerations, different wall thicknesses, etc... Nothing is perfect.

The last zero represents a cylindrical element

So, what is the end of the flat battery name?

The price of Tesla cars is still three times higher than the price required for large-scale market penetration under the traditional ownership and use rights system, and there is no battery label engineering technology that can help achieve this goal. The entire battery technology (among other things) needs to be changed to truly compete with traditional vehicles, and even vehicles running more expensive synthetic fuels, to make them carbon neutral.

three times? So, should a new electric car start at $12,000?

Their goal is to reach US$25,000 in 5 years. In the US, the average price of a new car is US$37,000... Therefore, I think they are actually far below the required price, especially considering the operating cost is 1/ 4. And you can save $1-2k per year. Did not even consider the reduction in depreciation. Literally, in 5 years, you will lose less money by buying Model 3 on almost all other models, and this is just before the price they are targeting drops by $10,000.

> "So, the starting price of a new electric car should be $12,000?"

Yes. This is the entry-level price for a new cheap gasoline car or a used car around 8-10 years old. For example, in Europe, the Renault Logan car starts at $10,400 (approximately £8,000), or if you want to upgrade to a higher level, you can buy a car in the US starting at $15,925 Modern Accent.

Most people will buy new or used cars for less than $20,000. The second-hand market accounts for more than 70% of all cars sold. The illusion of a "mid-priced" car priced at $35,000 or an illusion caused only by considering a new car, including trucks and SUVs that are more expensive by default.

So yes, if you want mass market penetration, you must target a $12,000. The price of electric vehicles is about three times the real public appeal, especially since the batteries are about to run out and need to be replaced, so their second-hand value is very low, so the second-hand value of electric vehicles is low. They are going to the junkyard.

No, most people will not buy a car for less than $20,000. Not far

Yes, there are about 20-25k models in the light vehicle market, but this is less than a quarter of sales. In December 2019, 75% of new vehicle sales were light trucks/pickups/SUVs. That's where the average price is over $40k.

No, the price of an electric car is three times that. Many reports clearly indicate that at today's prices, the current break-even point of owning an electric car is around 5-10 years, and it will be cheaper than an equivalent gasoline car from then on. By 2025, the upfront cost of electric vehicles and gasoline vehicles will be equivalent to more than 80% of the vehicles sold (this means that the break-even point is 0 days). By then, even cheap vehicles, this will still be due to greatly reduced operations and maintenance. Cost, so it reached a 5-year break-even point. So, spend $20,000 to buy a cheaper gasoline car, or spend $25,000 to buy a better electric car, and get back $5,000 in 5 years, and then you can save money.

Similarly, no, the replacement cost of a depleted battery is not very high. Today's batteries (except for the strange leaves) have effective thermal management functions and are expected to last 15 to 20 years. Even within 10 years, the replacement cost will reach 5-10k *max*, and these batteries can still be used for grid storage. Even if they are not, they can be recycled. They will not be sent to the scrapyard.

I have learned that even if you recently said that it was 5 years ago, this is true, but things have changed so fast and the speed of change is still increasing. We are indeed at an industry turning point.

Yes, they do. If you want to talk about the mass market, you are talking about only 75% of the market for second-hand or cheap economy cars that can afford MRSP well below $20,000.

This is the illusion of car price statistics. New cars account for only a small part of the actual market.

>Today's battery... is expected to last 15-20 years.

Do not. Some batteries have a shelf life of up to 20 years (lithium titanate), but they are different from the low-cost, high-energy-density chemicals used in cars. The NMC battery has a history of about 15 years, but because of the high cost of cobalt, companies such as Tesla do not use NMC in cars.

The battery can reach its full shelf life (the number of cycles is very low) only if you do not actually use the battery. Higher energy is a compromise between shelf life and fire safety. Manufacturers optimize more scope at lower cost instead of life (longer time = higher sales). The actual service life of almost all EV batteries on the market is only 10-12 years.

> There are about 20-25k models in the light vehicle market, but this is less than a quarter of sales

Pick cherries by choosing a small part of the distribution. Taking a car between 0-25k is about half the price of a new car sold.

Since people cannot afford to change cars, people have extended the storage period of cars. The average age of vehicles has been steadily increasing over the years. Do as they say, fix it.

>Recover $5k in 5 years,

Unless you lose at least $5k due to depreciation of resale value, no one will pay a high price for a second-hand EV with a failed battery.

It is a myth that maintenance costs are greatly reduced. You still have to regularly replace (battery coolant) hydraulic oil and brake parts, shock absorbers, tires, and if you do not follow expensive regular inspections, you will lose your warranty. In addition, because there is no mature third-party parts market, spare parts are expensive.

More precisely, the US auto market is "bimodal." Some people buy cheap cars, some people buy expensive cars, and some are expensive, which pushes up the average price. In fact, few people actually buy a car at an "average" price.

Furthermore, due to battery life issues, electric vehicles can hardly reach the second-hand market. The technical life of ordinary cars is about 20 to 25 years, and since then they have cracked and repairs are no longer worthwhile. Electric cars will reach the same level in the half-life, because the cost of the battery is about one-third of the price of a new car-far more than the price of an ordinary car in the 10-12 age group that needs to replace the battery. It is cheaper to scrap an electric car and buy another used gasoline car.

I think your numbers are pretty good. Most vehicles outside the southern region will not have a service life of 20-25. Because the body is falling apart, or the repair costs exceed the value of the vehicle, it is difficult to see vehicles from the 1990s in the north at this time.

The batteries in today's cars (don't look at the leaves) are easy to last for 10 to 20 years, and because the complexity of electric cars is greatly reduced, the life of the rest of the car will be as long. We fully hope that the original battery life of most modern electric vehicles will reach 15-20 years, and the replacement amount will be between 5-10k. The second-hand and refurbished markets have begun to rise.

Yes, the market may be bimodal, but relatively speaking, few people buy brand new cheap cars. Yes, there are about 20-25k models in the light vehicle market, but this is less than a quarter of sales. In December 2019, 75% of new vehicle sales were light trucks/pickups/SUVs. That is the average price is $45k+. Therefore, those who are actually buying a new car don't actually have to worry about diving, and if they need to spend $5-10k to replace the battery in 15 years, it won't be a big deal.

"I think your numbers are pretty good. Most vehicles outside the southern region will not have a service life of 20-25. Because of the collapse of the body, or because the repair costs exceed the value of the vehicle, it is difficult to see 1990 in the north. Vehicles of the age."

You may wish to be unclear here. If "Northern" means "New England" here, I can buy it. But definitely not the Central American states. It is 2020 now. Twenty years ago was 2000. There are now *large* 2000 models and early cars still on the road.

"Or the repair cost exceeds the value of the vehicle."

Why is repair cost more important than vehicle value? Repair costs are less than *buying a new car*.

How many are there?

According to statistics, in the United States, the average age of cars on the road is 12 years old, and more than 25% are over 16 years old. However, less than 1% of the time over 25 years. So no, there are not many vehicles over the age of 20 on the *entire* roads in the United States. I can't find a specific figure for 20 years, but from the outside, it is probably 5%.

Maintenance costs are absolutely more important than vehicle value. Because why spend $5,000 to repair a car worth $3,000, and I can only scrap it and buy another used car for $3,000?

But, more importantly, people are more likely to start buying better or newer cars after their old cars die.

Having said that, I am not sure why this is part of the debate. Those who buy a new car (and that's all here) spend at least $25k almost all of the time, and on average 75% of the time is $45k+. Therefore, Tesla's new price point target is the "bottom end" of new car sales.

The cost saved in the process of owning a vehicle will be more than twice the cost of replacing the battery, so this is not a problem at all.

> But less than 1% over 25 years

Yes, it is rare to keep a car between 20 and 25 o'clock, because you basically have to refurbish it to keep it. This is what the statistics show. That is the technical life of ordinary cars: that is when people stop maintaining them and just put them on the ground.

The starting point for technical life and replacement costs is the same: when the cost of buying another bearing is lower than the cost of keeping the old one, the old metal value is scrapped. The average retirement age of cars is actually much lower, between 14 and 15, but this includes crashes and other accidents.

> In December 2019, 75% of new vehicle sales were light trucks/pickups/SUVs.

Yes, but guess what, electric vehicles are not competitive in the truck, pickup and SUV market because of their poor actual towing tax range. Cybertruck made a joke.

Almost all electric cars on the market are cars or hatchbacks. You will make the wrong comparison because SUVs of $45k have been added at the same time, and those who are considering choosing between EVs and conventional cars will still not buy them.

Second, light and medium truck pickups are purchased as work vehicles, and their turnover is high because companies and businesses only keep them for 4-5 years before throwing them on the second-hand market and buying another one.

Okay, here are a few things to sort out.

First, the initial view was that people did not have enough money to buy electric cars. Since the price of 75%+ vehicles is $45k+, this is obviously not the case, and of course there is no need for EVs to be one-third of the current price.

Secondly, in more than 75% of the trucks/off-road vehicles sold, do you think that all the functions *actually used*, and how many functions can be handled by electric vehicles? 5%? 10%? Most people never or rarely use large vehicles to tow or tow. They only like size and "safety", and the ability to travel with their family and luggage, or worse, they only like appearance.

Third, how often do people drag and need the entire vehicle? Today we already have electric cars that can be towed. As long as you can walk around during the day and still place it in an open area (or reach the workplace), then you have enough range. This focus on scope is bizarre and a relic of the old age.

Finally, you are not paying attention to the market. By the end of next year, there will be several electric trucks on the market. F150, Hummer, Cyber ​​truck, etc. These will have the required range and towing capacity. Those will compete in the current ICE market where prices are excessively inflated and will easily be price competitive. By 2025, there will be more than 100 EV models on the road at a price parity, and only a few manufacturers will only adopt EVs.

Yes, only a very small number of heavy users will be satisfied until 2025, but this is only a small part of the market segment.

If you can get the 2009 Prius for less than 120,000 miles for $4,000, your life is much better than buying a new car, and you can save a lot of fuel.

"Three times? So, the starting price of a new electric car should be $12,000?"

Legal electric golf carts on the street.

Instead of having to

Instead of having to keep accelerating

Than the word. You used the correct word here

Instead of having a small battery label

"These new batteries with higher energy density and high power output will attract a huge market for hackers and manufacturers."

I am pretty sure this is incorrect.

4680 does not have a higher energy density; in fact, the energy contained per unit volume is even less than 21700 batteries (otherwise it should be increased by 5.4).

Due to its lower ESR and better thermal performance, it is more efficient.

But the range has increased by 16%? I don’t think the marketers really use the exact numbers.

SMD bottle caps have been doing this for decades. Until now, has no one thought of manufacturing batteries in this way?

In fact, for many years, there are indeed some lithium battery packs with higher C ratings that can do this.

There is really nothing new.

But traditionally, they are flat packs, not round batteries, but to be fair, they are not even new for round batteries...

Nothing new here...

This method of alternately extending the anode film and the cathode film beyond the end of the dielectric film, rolling or folding the film, and then fixing the connection to the end of the cylinder or folding the laminate has been the production method of film capacitors for many years. After the Swiss engineer Max Schoop (Max Schoop) put forward this idea, the technology was called "Schoopage". [1]

Coincidentally, it turns out that Big Clive recently released a YouTube video on the subject with the title: "Will Tesla's table cells use Schoopage?" [2]

1. Film capacitor-Wikipedia, search for "schoopage" and "Max Schoop".

2. Will Tesla's dining table unit use Schoopage? Big Clive.

The cars made by Tesla (reminded me of "Edsel") are impractical, so it is nice to know that they have contributed to the world through battery technology.

Well, Tesla has been in production for many years than Edsel.

I'm not sure to compare production quantities...

>Meanwhile, those who want the best cylindrical battery will have to wait for a new Tesla to appear in their local destruction field.

I wish them chiseled out of the epoxy honeycomb!

We are hackers...

We will do what is necessary!

"But no one at Tesla can figure out why the trailing zero of 18650 batteries"

Tesla engineers must be stupid.

You can also see the nomenclature of button batteries, such as 1632 x 16mm. The height is in feet, obviously the radius is not needed.

But please consider that by reducing the final zero, they will save all the weight! Just like Jeff, the cyclist, "Pearls Before Pigs":

Please be kind and respectful to help make the comment section great. (


The site uses Akismet to reduce spam.


By using our website and services, you expressly agree to our placement of performance, functionality and advertising cookies.