Updated dry cathode battery - your thoughts?

[SCTesla]

You don't know this. No one outside of Tesla does.

I doubt it will, though, since they've been testing it for awhile.

-Crissa

They always do. Again, we have history to show us.

Sponsored

 

[Crissa]

They always do. Again, we have history to show us.

No, no they don't.

Your 'history' only shows two instances yet we know Tesla has completely changed battery cells four times, overall chemistry four times, and incremented battery chemistry more times than both of those combined.

This would be an incremental change, so less likely to need a longer safe margin as switching cell size had.

-Crissa

 

[Jager]

The current in-house (wet cathode process) Giga Austin 4680-ramp is only a few weeks in front of the Cybertruck ramp.

It's reasonable to assume that changing 4680 production from wet (cathode) process to dry process will be a significant, disruptive factory change. They can't do that on one of the existing, active battery lines without significantly impacting 4680 output, and thence impacting Cybertruck production.

Another confounding factor is that when the first 4680 dry cathode production process goes live, it will almost certainly see lower yield. That would be typical with a new manufacturing process, while discovery happens to understand what line tweaks are necessary.

The Cybertruck is not yet profitable, even at the higher Foundation Series pricing. Achieving profitability - targeted for the end of 2024 - depends on several things, including Cybertruck volume, 4680 yield, and 4680 volume. All those directly influence per-vehicle unit cost.

Even if the 4680 dry cathode process brought significant energy density improvements - it doesn't - Tesla could not snap their fingers and magically change their factory lines.

Profitability is the lodestone for Tesla. That drives everything else.

There's plenty of economic motivation for Tesla to get its Giga Austin 4680 lines switched to the dry cathode process. But there are multiple moving parts involved. They'll make that changeover slowly - probably with only one line at the beginning, while they work on yield - and for some time I expect to see a mix of both wet-cathode and dry-cathode 4680 cells being put in Cybertrucks.

There are lots of user-can-notice improvements that Tesla will be bringing to its Cybertruck over time. The dry cathode 4680 process is not among them.

 

[ChristoN]

What a bs response. No discussion when you're not even going to be reasonable.

They haven't changed a significant part manufacturing as vehicle like this without a version change. Like a LR. So don't go stating this is totally normal for them either.

If they change the battery and it gets 20% more and the 20k foundation price goes away, what do you expect people to feel? Bboozled? That they rushed the CT just to get it out due to pressure? Heck yeah. If it gets more range and there isn't a. Upgrade/recycle option it's in the wrong if you ask me. They promised more, have given terrible customer service, Charged a premium and now are swapping the battery mid production line year one.

If you think it won't bug us to get a 20% range version same year having paid a premium, you're wrong. If you don't see a problem with it that's fine. It doesn't mean we aren't justified. If no range difference no big deal.
As

You’re acting like I’m a Tesla exec ?. They will definitely do something like this, they lowered the price of the refreshed model s/x by more than 30% after a year. Did they compensate anyone? Nope! Just a big hit to those first customers. They did the same thing with HW3 to HW4. No announcement, they just started putting it in new vehicles one day, hence the obsolete hardware clause in your agreement.

I’m not saying I agree with it, but Rivian and GM are offering products with more range and once Tesla opens up a second assembly line, they need to attract demand and if they have increased range they won’t hesitate to offer it, regardless of what it will do to your resale value.

You knew the specs of the car before you bought it. Doesn’t matter what they said 5 years ago, it’s irrelevant.

But just to be clear, nobody knows what this will do to the range, or charging curve, or anything other than cost. It’s all speculation. Until it comes from Tesla themselves, it’s anyone’s guess. My guess is that this completes the 4680 and they can now optimize it for all the goodies that were promised, e.g. 400+Wh/Kg. But who knows how long this will take. Could be 4th quarter or 2027 lol.

 

[SCTesla]

No, no they don't.

Your 'history' only shows two instances yet we know Tesla has completely changed battery cells four times, overall chemistry four times, and incremented battery chemistry more times than both of those combined.

This would be an incremental change, so less likely to need a longer safe margin as switching cell size had.

-Crissa

3 times. It's always with either brand new batteries (Panasonic 2170s) or batteries that Tesla produced... 4680 gen 1 and the Cybercell.

edit: I actually forgot about the new chem 18650. They were severely software nerfed and Tesla improved the range and charging curve after more data(thanks TMC).

Cybercell was an incremental change from Gen 1 4680 and we saw exactly what I'm talking about.

Find me an instance when they haven't done this with a new battery chem/type. Most other "new" batteries like the LG 2170, CATL battery were relatively old batteries used in other EVs when Tesla began deploying them.

 

[mongo]

The current in-house (wet cathode process) Giga Austin 4680-ramp is only a few weeks in front of the Cybertruck ramp.

There is no in house wet cathode production. Tesla buys cathode rolls from suppliers in China.

It's reasonable to assume that changing 4680 production from wet (cathode) process to dry process will be a significant, disruptive factory change. They can't do that on one of the existing, active battery lines without significantly impacting 4680 output, and thence impacting Cybertruck production.

The cell manufacturing jelly roll stage will not be impacted assuming the dry coated rolls maintain the same dimensional tolerance as the wet processed version.

At this point, they should have 4 lines running with 4 more being installed. With each line projected to produce 20 GWh/yr of cells, one line runming at 50% capacity is sufficient to support 1,500 trucks per week.

 

[ChristoN]

The current in-house (wet cathode process) Giga Austin 4680-ramp is only a few weeks in front of the Cybertruck ramp.

It's reasonable to assume that changing 4680 production from wet (cathode) process to dry process will be a significant, disruptive factory change. They can't do that on one of the existing, active battery lines without significantly impacting 4680 output, and thence impacting Cybertruck production.

Another confounding factor is that when the first 4680 dry cathode production process goes live, it will almost certainly see lower yield. That would be typical with a new manufacturing process, while discovery happens to understand what line tweaks are necessary.

The Cybertruck is not yet profitable, even at the higher Foundation Series pricing. Achieving profitability - targeted for the end of 2024 - depends on several things, including Cybertruck volume, 4680 yield, and 4680 volume. All those directly influence per-vehicle unit cost.

Even if the 4680 dry cathode process brought significant energy density improvements - it doesn't - Tesla could not snap their fingers and magically change their factory lines.

Profitability is the lodestone for Tesla. That drives everything else.

There's plenty of economic motivation for Tesla to get its Giga Austin 4680 lines switched to the dry cathode process. But there are multiple moving parts involved. They'll make that changeover slowly - probably with only one line at the beginning, while they work on yield - and for some time I expect to see a mix of both wet-cathode and dry-cathode 4680 cells being put in Cybertrucks.

There are lots of user-can-notice improvements that Tesla will be bringing to its Cybertruck over time. The dry cathode 4680 process is not among them.

Your floccinaucinihilipilification of this achievement is without merit. The Chinese report that first leaked this information prior to it being disclosed by Tesla already stated that they would be moving the wet cathode 4680 cells to the model y and prioritizing the fully dry cells to the Cybertruck. Given that the report was accurate with regard to the achievement, I would give it more weight now as to its other assertions. They aren’t going to mix batteries that can now be entirely made in Texas.

For reference.

*Text by Li Zinan**
**Edited by Cheng Manqi**

Tesla won't be launching new vehicle models for sale this year, but they might have a new battery.

We have exclusively learned that Tesla plans to mass-produce and install 4680 batteries using entirely dry electrode technology by the end of the year. This will be the "complete" version of the 4680 battery.

In 2023, Tesla already used 4680 batteries in the Cybertruck, but that was a compromised version that didn't achieve the goal of reinventing the battery manufacturing process by using dry electrodes to replace traditional wet electrodes, thereby eliminating the process of wetting and then drying the electrodes, speeding up production and reducing costs. Currently, Tesla's 4680 batteries only use dry electrode technology for the lower-cost anode, while the more expensive cathode is still sourced from LG Energy Solution and two Chinese battery companies using traditional wet cathode technology.

Now, Tesla is just one step away from the "complete" 4680 battery. We understand that the design of the 4680 battery with dry cathodes has recently been finalized, which is the step before large-scale production. Tesla's battery department will now focus on improving production yield and efficiency, and expanding capacity.

According to our information, Tesla had already produced dry cathode electrode rolls in the laboratory by the end of 2022, but due to technical misjudgments, they were unable to mass-produce them. After the project leader, Drew Baglino, left in April this year, Tesla adjusted its technical approach, and breakthrough in dry cathode mass production was achieved.

"Making dry electrodes can change Tesla," said a Tesla-related person.

The 4680 battery is Tesla's self-developed battery announced in 2020. Tesla claimed at the time that the new technology could reduce battery costs by 50%, leading to cheaper new cars.

The key to significant cost reduction is the dry electrode process. The current power battery's positive and negative electrode materials are in a slurry state, known as "wet electrodes," which requires drying and solvent recovery processes in battery production, taking a long time and requiring large factory floor space. Dry electrodes eliminate these processes, greatly reducing plant footprint and capital investment per unit of battery capacity, while improving battery performance and production efficiency.

The dry cathode is the most important and challenging part of the entire 4680 process, and cathode materials are also the most expensive component in the battery, accounting for over 35% of the cost.

We have also learned that while overcoming the dry electrode challenge, Tesla is continuing to improve the packaging process and material ratio of the current compromised version of the 4680 battery. These batteries will be prioritized for the Cybertruck. Currently, Tesla's self-produced 4680 batteries have a weekly capacity sufficient for only 1,000 Cybertrucks.

Informed sources say that among the material suppliers for these 4680 batteries Tesla is producing in the U.S., there are almost no Chinese companies left. Some of their production equipment has also been switched to Japanese, European, and American companies, and they are seeking alternative solutions for processes still using Chinese equipment.

On the other hand, Tesla also plans to have LG Energy Solution directly supply complete wet cathode 4680 batteries in the second half of this year. These batteries will be installed in Model Y vehicles produced in the U.S.

The 4680 battery is core to Tesla's efforts to enhance the competitiveness of its next-generation products, and the dry cathode is the core of the 4680. Since the project's inception, Tesla has spent 5 years on it.

**Past the R&D hurdle, now mainly engineering optimization issues**

A simple rule for success in manufacturing is to make better things than competitors with less money.

Over the past 5 years, Tesla had a clear cost reduction and efficiency improvement path: using larger integrated casting machines to shorten body manufacturing time, designing next-generation cars with fewer parts that are easier to manufacture, using dry electrode technology to compress battery manufacturing costs, using "unboxed" technology to reduce vehicle assembly time and costs, and building more efficient Gigafactories.

Tesla has invested billions of dollars and significant engineering resources in these hardware and manufacturing innovations. If the plan is realized, Tesla would maintain industry-leading gross margins and operational efficiency as it did 3 years ago.

However, since April this year, Tesla has begun to pause these efforts: suspending the manufacture of the $25,000 budget car model, halting research on larger integrated cast bodies, and the next-generation Gigafactory in Mexico, which was supposed to start production this year, has yet to begin construction.

The 4680 battery also once seemed to be in imminent danger. In April this year, the project leader, Drew Baglino, Tesla's Senior Vice President of Powertrain and Energy Engineering who had worked at Tesla for 17 years, resigned. The battery team was also cut by about 25% that month, now with fewer than 800 people remaining.

We followed up on the layoffs and mentioned that at the beginning of this year, Tesla management had set clear assessment targets for the battery department: if by the end of this year, the cost of Tesla's self-produced 4680 batteries is not lower than that of suppliers like LG Energy Solution and Panasonic, Tesla would consider abandoning self-production of 4680 batteries. This is not a low standard goal; if Tesla can truly achieve self-production costs lower than LG, this would be the cheapest battery produced on U.S. soil.

The recent progress in dry cathodes has given the battery department a chance to achieve the year-end goal.

After Baglino's departure, Tesla's battery team decided to take a slightly higher cost route that could be mass-produced faster.

Before this, Tesla's engineering team had already made dry cathode electrode rolls in the laboratory by the end of 2022, with the next step being to replicate this on the production line.

**The current general process for power battery production is to coat paste-like positive and negative electrode materials onto foils to make electrode sheets, then roll them into electrode rolls, place them in battery cases, inject electrolyte, and complete welding.**

However, when mass-producing electrode rolls, Tesla was stumped by the calendering process. "The dry cathode process encountered major challenges. The cathode powder is very hard and sometimes damages the electrode equipment, which we didn't anticipate," Musk said during the 2022 earnings call.

An engineer who previously worked in Tesla's battery department said that Tesla only had a small number of customized and well-tuned dry cathode calendering equipment. Each time this equipment broke down, it took at least 45 days to repair, causing the battery team to waste a lot of time.

He recalled that when Baglino was still in office, he had to face weekly interrogations from Tesla's former CFO Zach Kirkhorn, "On the cost bar chart, only the cell department was pulling to the top, while all other departments were at the bottom."

There were discussions within the battery department about leaving some of the engineering challenges encountered with dry electrodes to be solved after the calendering and winding processes, as there was less room for modification and optimization of the calendering equipment.

This person said that if they did this, things would be much simpler. Japanese equipment companies have mature solutions, but Tesla would need to replace some production equipment, involving more equipment expenditure and other R&D uncertainties. This proposal had not previously been approved by Tesla's senior management.

Recently, Tesla's battery department began to try this previously unapproved plan, switching to another set of equipment to solve electrode roll yield problems after calendering and winding, and has made progress.

A person close to Tesla said that the reason for the recent increase in confidence in the battery department is that they believe they have now passed the most difficult stage of R&D, and the main issues going forward are engineering optimization.

Tesla can improve efficiency through some detailed improvements. As one Tesla insider mentioned, when defective electrode sheets appeared on the production line before, ordinary battery companies would only cut off tens of centimeters of waste, but Tesla would cut off meters of electrode sheets. This resulted in Tesla having to bear higher defective product losses when producing batteries, and it was also more difficult to accurately locate problems.

"This is mainly because we don't have enough engineering staff to do testing," he said. But fortunately, these tasks can now be improved by having more basic technical personnel do more testing, "The route is clearer now, and it can be solved with time."

**Tesla is expected to produce some batteries in-house, reducing dependence on external battery suppliers**

Before the 4680 battery, Tesla had never fully produced batteries in-house. Tesla's initial batch of mass-produced cylindrical batteries was produced in cooperation with Panasonic. By 2020, Tesla, which was already selling 500,000 cars a year, had formed a relatively stable battery supply pattern:

All prismatic batteries were supplied by CATL, mainly used for standard range versions of Model 3 and Model Y. CATL supplied not only the Shanghai factory but also exported to Tesla's U.S. factories. Cylindrical batteries came from Panasonic and LG, used for long-range versions and high-end models.

Since Tesla began in-house production of 4680 batteries in 2023, it has changed its battery supply chain, reducing dependence on external suppliers, including Chinese suppliers.

The "compromised version" of 4680 currently in mass production at Tesla's Texas factory uses wet cathodes from LG and two Chinese companies, as well as Tesla's own dry anodes.

If the dry electrode 4680 battery is successfully mass-produced and achieves a high yield rate, Tesla will be able to fully produce 4680 batteries in-house, no longer relying on suppliers to produce core components.

External suppliers will lose some of Tesla's incremental orders. In Musk's optimistic vision in 2021, Tesla would produce at least 30% of its car batteries in-house in the future.

As Tesla begins to produce more batteries on its own, another opportunity for Chinese industry chain companies is to provide equipment. However, according to our understanding, Tesla has not yet contacted Chinese equipment manufacturers for core processes such as coating, calendering, and winding of dry cathodes. The core suppliers come from Japan and Germany.

However, whether Tesla's dry cathode 4680 can maintain cost leadership remains uncertain.

This depends on the speed of yield improvement and expansion of dry cathode 4680 production. If the yield is too low, it will offset the benefits of reducing production steps such as drying.

At the same time, since Tesla proposed the 4680 solution in 2020, many other companies in the industry have followed suit, such as LG Energy Solution, EVE Energy, and Panasonic. Although CATL is not optimistic about the 4680 route, they also have technical reserves and are researching dry electrodes. Even if they are a bit late, when other companies can also mass-produce dry electrode 4680 batteries, Tesla will face a new round of manufacturing efficiency competition. When companies are at the same generation of technology, extreme cost compression is not Tesla's forte.

CATL and other manufacturers are also exploring the next generation of batteries, namely all-solid-state batteries, which also require dry electrodes. The industry consensus is that solid-state batteries are the best form for automotive batteries.

While conquering the dry cathode 4680, Tesla is also deepening cooperation with Chinese companies such as CATL. CATL began co-building a lithium iron phosphate battery factory for energy storage with Tesla in Nevada earlier this year. Tesla purchases ready-made battery production lines from CATL, while CATL provides some technical and procurement support, earning technology licensing fees.

CATL and BYD currently have battery costs of about 0.33 yuan/Wh, still the cheapest and most cost-effective batteries globally.


- END -

 

[HaulingAss]

You aren't counter anything I said. Just that it's worse for consumers to get new batteries. I never said it's a dud or failure. Tesla is really conservative with new batteries and always has been.

That's it.

And it doesn't matter, there is no significant downside to most consumers who love the capabilities of their new cars. There is a core group of people constantly harping on fast-charging speeds as if battery electric cars are not quite ready for prime-time. It's a false narrative. Most motorists wouldn't even notice if their EVs charged a couple minutes faster.

 

[SCTesla]

And it doesn't matter, there is no significant downside to most consumers who love the capabilities of their new cars. There is a core group of people constantly harping on fast-charging speeds as if battery electric cars are not quite ready for prime-time. It's a false narrative. Most motorists wouldn't even notice if their EVs charged a couple minutes faster.

My only point is that Tesla batteries are always crippled for validation. New batteries are always a step back compared to season ones for consumers. Usually less dense, typically software locked for slower charging speed and in the case of the 18650, for range as well.

I never said they were getting a shit product, but the CT has been launched for 8 months and we are still awaiting additional charging speed increases and we've already had 2....because it was a new battery.

 

[HaulingAss]

My only point is that Tesla batteries are always crippled for validation. New batteries are always a step back compared to season ones for consumers. Usually less dense, typically software locked for slower charging speed and in the case of the 18650, for range as well.

I never said they were getting a shit product, but the CT has been launched for 8 months and we are still awaiting additional charging speed increases and we've already had 2....because it was a new battery.

I don't think you know what "crippled" means. Tesla doesn't deliver new batteries until they are ready to rock and roll. Sure, they might make small incremental improvements after that, but it doesn't completely change the nature of a road trip, it's just a small incremental improvement.

You are making a big deal over nothing.

 

[SCTesla]

I don't think you know what "crippled" means. Tesla doesn't deliver new batteries until they are ready to rock and roll. Sure, they might make small incremental improvements after that, but it doesn't completely change the nature of a road trip, it's just a small incremental improvement.

You are making a big deal over nothing.

Now you're arguing semantics. You agree with me that new batteries are crippled, protected, restricted...what ever term you want to use while Tesla validates in the real world. As I stated to someone else, Tesla has done it with every single NEW battery. My only point is that 1. Most new batteries are to save cost for Tesla (See 4680 vs 2170) and 2. Getting a brand new battery is always a step back for the consumer...At first. Just ad the CT, 4680 MY, 2020 18650, and the launch of the Panasonic 2170 showd us, yes they eventually get software updates to improve, but also get new revisions to improve.

 

[lonsomecrow]

The current in-house (wet cathode process) Giga Austin 4680-ramp is only a few weeks in front of the Cybertruck ramp.

It's reasonable to assume that changing 4680 production from wet (cathode) process to dry process will be a significant, disruptive factory change. They can't do that on one of the existing, active battery lines without significantly impacting 4680 output, and thence impacting Cybertruck production.

Another confounding factor is that when the first 4680 dry cathode production process goes live, it will almost certainly see lower yield. That would be typical with a new manufacturing process, while discovery happens to understand what line tweaks are necessary.

The Cybertruck is not yet profitable, even at the higher Foundation Series pricing. Achieving profitability - targeted for the end of 2024 - depends on several things, including Cybertruck volume, 4680 yield, and 4680 volume. All those directly influence per-vehicle unit cost.

Even if the 4680 dry cathode process brought significant energy density improvements - it doesn't - Tesla could not snap their fingers and magically change their factory lines.

Profitability is the lodestone for Tesla. That drives everything else.

There's plenty of economic motivation for Tesla to get its Giga Austin 4680 lines switched to the dry cathode process. But there are multiple moving parts involved. They'll make that changeover slowly - probably with only one line at the beginning, while they work on yield - and for some time I expect to see a mix of both wet-cathode and dry-cathode 4680 cells being put in Cybertrucks.

There are lots of user-can-notice improvements that Tesla will be bringing to its Cybertruck over time. The dry cathode 4680 process is not among them.

Great description. Thanks! So no substantial range enhancement?

 

[Mini2nut]

Those of us who are holding off for non-FS truck‘s will be rewarded with a lower price, less bugs and hopefully a dry cathode 4680 battery pack.

 

[lonsomecrow]

Those of us who are holding off for non-FS truck will be rewarded with a lower price, less bugs and hopefully a dry cathode 4680 battery pack.

But only problem is you have to wait at-least till 2026 unless you have day 1 reservation. after FS, its going back to the start line.

 

[Crissa]

Cybercell was an incremental change from Gen 1 4680 and we saw exactly what I'm talking about.

No, no we didn't. The post-delivery Cybertruck never was set to a slower charging rate than the 4680 Model Y.

Basically what I'm saying here is... don't worry about it.

-Crissa

Sponsored