Tesla Q3 Earnings // Battery Updates // Full Breakdown and Implications
发布时间 2023-11-08 14:59:57 来源
摘要
In the Tesla Q3 2023 earnings call, Drew Baglino made a statement about the 4680 production ramp that contained a lot of new ...
GPT-4正在为你翻译摘要中......
中英文字稿 
Welcome back everyone, I'm Jordan Geesege and this is The Limiting Factor. In Tesla's Q3-2023 earnings call, Drew Baglino made a statement about the 4680 production ramp that contained a lot of new information. Today I'm going to break down what he said line by line, what the implications are for the 4680 ramp, and what the impact could be on Tesla's cost of goods sold.
大家欢迎回来，我是乔丹·吉斯奇，这里是《限制因素》。在特斯拉2023年第三季度收益电话会议中，德鲁·巴格里诺发表了关于4680生产扩产的声明，其中包含了许多新信息。今天我将一行一行地解析他所说的内容，探讨对4680生产扩产的影响，并分析对特斯拉的销售成本可能带来的影响。
In short, after 3 years of waiting, it appears 4680 cell production is finally poised to deliver on the exponential growth and cell cost reductions that were teased at Battery Day. Before we begin, a special thanks to my Patreon supporters, YouTube members and Twitter subscribers, as well as RebellionAir.com. They specialize in helping investors manage concentrated positions. RebellionAir can help with covered calls, risk management, and creating a money master plan from your financial first principles.
简而言之，经过3年的等待，4680电池的生产似乎终于有望实现在电池日所承诺的指数级增长和电池成本降低。在我们开始之前，特别感谢我的Patreon支持者、YouTube会员和Twitter订阅者，以及RebellionAir.com。他们专门帮助投资者管理集中头寸。RebellionAir可以帮助进行认购期权、风险管理，并从您的财务基本原理中创建一项资金主计划。
On screen is Drew's full, unedited statement from the earnings call. All I've done is chunked it down into seven smaller statements so that we can address each one individually.
屏幕上显示了德鲁在收益电话会议中的完整、未经编辑的声明。我所做的只是将它分为七个较小的陈述，以便我们可以逐个予以处理。
The first statement was, quote, 4680 cell production in Texas increased 40% quarter over quarter. Congrats to the Texas team for producing their 20 millionth cell off of line one, end quote. This roughly aligns with my projections in the last quarterly earnings call video. In that video, I said that based on the progress Tesla was reporting, it appeared that each month Tesla was seeing about a 21% improvement in production rate. And if that continued, they should reach 20 million total cells produced in October. However, although that ended up being correct, it doesn't exactly align with Drew's new comments from the Q3 earnings call. Drew set a 40% quarter over quarter increase, whereas I showed a 67% increase. That means although I got the average growth rate correct, the shape of the growth curve is likely different than what I had guessed.
第一份声明是这样说的，引用：“德州地区的4680电池产量季度环比增长了40%。祝贺德州团队成功生产出了第2000万个电池，该电池是从一号生产线上生产出来的。”这与我在上一个季度财报电话会议中的预测大致吻合。在那个视频中，我表示，根据特斯拉的报告情况来看，每个月特斯拉的生产率提高约21%。如果继续这样增长，他们应该可以在10月达到总共2000万个电池的产量。然而，尽管最后的结果是正确的，但它与德鲁在第三季度财报电话会议上的新评论并不完全吻合。德鲁提出了一个季度环比增长了40%，而我展示的是增长了67%。这意味着尽管我对平均增长率有所准确，但增长曲线的形状可能与我猜测的不同。
Drew's next comment was, quote, Texas is now our primary 4680 production facility end quote. This is something that's been coming for a while because as Tesla advised in the Q4 2022 earnings call, in Texas they had four lines and various stages of commissioning and install, which is about 100 gigawatt hours of capacity. That means Texas has more than 90% of Tesla's 4680 production capacity, making it their primary production facility. As for the other 10% in California, we'll talk more about that later in the video.
德鲁的下一个评论是：“引用，德州现在是我们主要的4680生产设施。”这是一个早就预料到的事情，因为正如特斯拉在2022年第四季度收益电话中提到的那样，在德州，他们有四条生产线，并且处于各种各样的调试和安装阶段，这相当于100千兆瓦时的产能。这意味着德州拥有特斯拉90%以上的4680电池产能，使其成为他们的主要生产设施。至于剩下的10%在加利福尼亚，我们将在视频中稍后详细讨论。
Drew's third comment was, quote, where heavily focused on quality scrap is down 40% quarter over quarter. With the increased volume and yield improvements, sell costs consistently improved month over month within the quarter. Although we have a lot more work to do to achieve our steady state goals, and that is our priority. End quote. A 40% quarter over quarter improvement to the scrap rate is exactly what was reported last quarter as well. So Tesla's making huge advances in efficiency on a quarterly basis. However, they still likely have a long way to go. As part of the course for new battery lines, that a lot of material ends up in the dumpster for the first year or two of the ramp. That's because, as I've said in the past, the components within battery cells are the thickness of a human hair, which makes them difficult to manipulate at high precision and speed. That means a lot of ways for defects to occur and a lot of scrap until things get dialed in.
德鲁的第三条评论是，引用他所说的：“在质量方面，废品在季度内下降了40%。随着产量的提高和产能的改进，销售成本在季度内持续改善。尽管我们还有很多工作要做才能达到稳定态的目标，但这是我们的首要任务。”上一季度也报告了40%季度改善的废品率。因此，特斯拉在季度基础上在效率方面取得了巨大的进展。然而，他们可能仍有很长的路要走。作为新电池生产线的一部分，很多材料在第一年或两年的推广期间最终被丢弃。这是因为，正如我之前所说的那样，电池单元内的组件厚度只有人的头发那么细，这使得它们难以在高精度和速度下操控。这意味着会出现很多缺陷的可能性，以及大量的废品，直到事情得到完善。
So why is Tesla's current focus on quality or scrap rates and cost rather than sell volume? My guess is that there's a few factors involved. First, their current production rate is higher than what's needed to support the ramp of the Cybertruck. As this video from Joe Techmyer shows, there's very few Cybertrucks currently coming off the production line. Meanwhile, the 4680 lines are producing enough cells for about 500 to 1000 Cybertrucks per week. Although they are continuously increasing the production rate of the 4680 line at Texas, they currently have the breathing room to focus on quality because they're ahead of the Cybertruck ramp. The focus on quality will mean further reductions in scrap rate and therefore reductions in cell cost to help improve the profitability of the Cybertruck ramp.
那么为什么特斯拉当前的重点是质量或废品率以及成本，而不是销售量呢？我猜想这其中涉及到几个因素。首先，他们目前的生产速度高于支持Cybertruck生产提速所需速度。正如Joe Techmyer在视频中展示的那样，目前生产线上的Cybertruck非常少。与此同时，4680生产线每周能生产足够的电池供应500至1000辆Cybertruck。虽然他们正在不断提高德克萨斯州4680生产线的产能，但目前他们有空间集中精力提升质量，因为他们领先于Cybertruck生产提速。专注于质量将进一步降低废品率，从而降低电池成本，帮助提高Cybertruck生产的盈利能力。
Lastly, what are the steady state cost goals that Joe referred to? I can't speak to what their short term goals are, but later in the video we'll talk about their long term goals, when they might hit them and the implications.
最后，乔所提到的稳定状态成本目标是什么？我不清楚他们的短期目标是什么，但在视频的后面我们将谈论他们的长期目标，以及何时可能达到这些目标和其潜在影响。
Drew's next comment was, quote, the Cybertruck sell with 10% higher energy density than our Model Y sell started production on line two in Texas, end quote. The 10% energy density increase is old news from last quarter. And if you'd like to know more about that, check out my last quarterly update video. The comment that production started on line two in Texas is new information and is very useful for assessing the ramp of the 4680. It confirms that the 10 and 20 million cell milestones they hit in June and October were for the most part from just one line. That matters because it means they're getting all the new technologies they unveiled at Battery Day dialed in on the first line. And now that they're copying that line, the ramps of the subsequent lines will happen rapidly.
德鲁的下一个评论是：“Cybertruck的能量密度比我们的Model Y销售高出10％，在德克萨斯的第二条生产线上开始生产。”这个10％能量密度增加是上个季度的旧闻。如果你想了解更多，可以查看我上一次季度更新的视频。生产在德克萨斯的第二条生产线已经开始，这是新的信息，对于评估4680的推进非常有用。这证实了他们在六月和十月达到的1千万和2千万电池量产里程碑大部分来自于仅仅一个生产线。这很重要，因为它意味着他们正在第一条生产线上逐步调整他们在电池日揭示的所有新技术。现在他们正在复制那条生产线，后续产线的扩建将会迅速进行。
On that note, many people are interpreting that Drew's statement that the Cybertruck sell starting production on line two means that the first line wasn't running the Cybertruck sell. I took Drew's statement at face value and I don't believe he was implying that line one is running a separate version of the 4680. Let's take a look at Drew's next statement to explore that further.
在这一点上，许多人解释Drew的说法是指Cybertruck销售将从第二条生产线开始，意味着第一条生产线没有生产Cybertruck销售。我按照Drew的原意理解，我不认为他是在暗示第一条生产线正在生产一种独立版本的4680。让我们看一下Drew的下一句话，以进一步探讨这个问题。
Drew said, quote, this quarter we convert to building 100% Cybertruck sells to simplify and focus the factory as we ramp all four lines in phase one over the next three quarters. End quote. The interpretation many people are making of this statement is that line one is producing an older version of the cell and will be converted this quarter. However, I have a different view.
德鲁说，引用他的话，本季度我们将转换至百分之百的赛博卡车销售来简化和专注于工厂，同时在接下来的三个季度内逐渐推出第一阶段的全部四条生产线。很多人对这个声明的解读是，第一条生产线正在生产旧版本的电池，并将在本季度进行转换。然而，我持有不同的观点。
Let's start by taking a look at the history. First, when Drew says the Cybertruck sell, my view is that he's referring to the generation two cell that was shown at the Cybert rodeo 18 months ago at the 4680 production line in Texas as both a photo display shown here and a deconstructed cell shown here. As you can see, it appears the top is welded shut and there's a spiral pattern at the center of the lid. That's as opposed to the generation one cell that's shown in this freeze frame from battery day that has no spiral pattern at the top and is crimped. I was able to get a hold of one of those cells early last year from the Kato road facility in California, which I sent to UC San Diego for a tear down. It's also the cell that Sandy Monro found in the Model Y from Texas at a time when the 4680 factory at Texas hadn't started production yet. That is, we know the generation one cell was definitely produced at Kato Road and there's no evidence that generation one cells were produced in Texas at a volume and quality sufficient for vehicles.
让我们首先来看看历史。首先，当Drew提到Cybertruck销售时，我认为他指的是在18个月前在德克萨斯州的4680生产线上展示的第二代电池单体，如此处展示的照片和拆解的电池单体。从图片上可以看出，顶部是焊接封闭的，并且在盖子的中心有一个螺旋图案。这与第一代电池单体形成对比，第一代电池单体在电池日的冻结画面中显示，顶部没有螺旋图案，并且是卷边的。去年初我能够从加利福尼亚州的Kato Road工厂得到一个这样的电池单体，并将其发送给加州大学圣地亚哥分校进行拆解。这也是Sandy Monro在得克萨斯州的Model Y中发现的电池单体，当时德克萨斯州的4680工厂尚未开始生产。也就是说，我们知道第一代电池单体绝对是在Kato Road生产的，并且没有证据表明第一代电池单体在德克萨斯州以足够的数量和质量生产用于车辆。
Beyond the cells and packs that were torn down, there's also circumstantial evidence that Kato Road was the only location where the generation one cell was produced. In mid-September, Cleaner Watt reported that Tesla was shutting down 4680 production at the Kato Road facility for upgrades and within the same week we saw Tesla pull the Model Y from their order page that we know contained generation one 4680s. It reinforces the idea that Kato was the sole source for generation one cells that went into vehicles.
除了被拆除的电池组装线和电池组件之外，还有间接证据表明Kato Road是唯一生产第一代电池的地点。在9月中旬，Cleaner Watt报道称特斯拉正在关闭位于Kato Road的4680电池生产线进行升级，并且在同一周，我们看到特斯拉从订单页面中撤下了搭载第一代4680电池的Model Y。这加强了Kato是第一代电池投入车辆的唯一来源的观点。
On the other hand, as I said earlier, there is evidence that Tesla intended to produce the generation two cells at Texas with a new cell can based on displays at the cell line at Giga, Texas. In light of that information, my view is that the 20 million cells produced online one in Texas all used the generation two architecture and that line one doesn't need to be converted because it's already producing that cell.
另一方面，正如我之前所说，有证据表明特斯拉打算在得克萨斯州生产第二代电池，该电池基于在得克萨斯州吉加工厂的电池生产线上展示的新电池罐。根据这些信息，我认为在得克萨斯州生产的2000万个电池全部采用了第二代架构，因此第一条生产线不需要进行转换，因为它已经生产该种类型的电池。
So when Drew says they're converting to building 100% Cybertruck cells this quarter to simplify and focus the Texas factory on ramping all four lines there, what did he mean? In my view, in the first part of the sentence where Drew says, convert to building 100% Cybertruck cells, he means across Tesla's 4680 manufacturing facilities in Texas and Kato Road. As he said in his last statement, which I'll cover in a moment, they're retooling the line at Kato Road for next generation cells. Then in the next part of the sentence when he says, simplifying and focusing the factory, my view is that he's talking about the entire factory at Texas, not just the small pocket of the building that builds the 4680 cells. If Giga, Texas had to use both generation one cells and packs and a generation two cells and packs in the same facility, it would add unnecessary complexity with no additional benefit. It would make more sense to use the generation two cells produced in Texas for all vehicles produced in Texas. That's not to mention the cost savings from moving the packs around the factory, rather than shipping the generation one cells 1700 miles from Kato Road in Fremont, California.
当Drew说他们正在转型，本季度将100%将生产线改为生产Cybertruck电池时，他的意思是什么？在我的理解中，Drew在句子的第一部分中说的“转型为100%生产Cybertruck电池”，是指特斯拉在德克萨斯和卡托路4680生产设施中进行的转型。正如他在最后一次讲话中所说的，他们正在重新装备卡托路的生产线，用于生产下一代电池。然后，在句子的下一部分中，当他说“简化和专注于工厂”时，我认为他指的是整个德克萨斯工厂，而不仅仅是负责生产4680电池的小部分建筑。如果德克萨斯超级工厂必须在同一设施中使用第一代电池和电池组以及第二代电池和电池组，将会增加不必要的复杂性，而且没有额外的好处。将在德克萨斯生产的第二代电池用于所有在德克萨斯生产的车辆将更有意义。更不用说节省成本，而不是将第一代电池从加利福尼亚州弗里蒙特的卡托路运送1700英里。
Not if I'm wrong, and line one in Texas does need to be converted from producing generation one to generation two cells. If that's the case, there's a chance the 5gWh run rate that was likely reached in October could plummet this quarter. That's because they would have to upgrade line one in Texas from a battery cell can that's crimped to a battery cell can that's laser welded. That sounds simple conceptually, but in reality, it would be a big undertaking. I can't see a scenario where it wouldn't mean shutting the line down, replacing the necessary equipment, and recalibrating all the machines for a different cell design with a different number of parts and slowly re-ramping production. If Tesla's lines were capable of rapidly shifting between cell designs, I would expect that their Kato Road line would be that adaptable as well.
如果我没记错的话，德克萨斯州的第一生产线确实需要从一代电池转换为二代电池。如果情况是这样的，今年第四季度的电池产量可能会出现暴跌，之前很可能已经达到每年5个GWh。这是因为他们需要将德克萨斯州的第一生产线从使用被卡住的电池罐转换为使用激光焊接的电池罐。概念上听起来很简单，但实际上，这将是一项庞大的工程。我无法想象在这种情况下不关闭生产线、更换必要的设备、并重新校准所有机器以适应不同的电池设计和零部件数量，并逐步重新启动生产。如果特斯拉的生产线能够快速转换电池设计，我预计他们的卡托路线生产线也应该是如此适应的。
However, that doesn't look like it's the case if it's going to require re-tooling for the next generation cells. With all that in mind, I'm left with the conclusion that it wouldn't make sense for Tesla to have spent the last year ramping the first 4680 line at Texas only to re-tool it now just as the Cybertruck is ramping. It makes more sense that they've now got line one running well, and they're now copying it to line two, rather than ramping both lines on a new cell design.
然而，如果为下一代电池需要重新调整工艺，那看起来情况并不是这样。考虑到这一切，我得出的结论是，对特斯拉来说，不应该在德克萨斯州过去一年里投入大量时间和资源来推广第一条4680生产线，然后再重新调整它，尤其是在Cybertruck推至顶峰之际。更合理的解释是，他们现在让第一条生产线运转良好，并且正在复制到第二条产线，而不是在新的电池设计上同时推进两条生产线。
Drew's next comment was, quote, phase two of the Texas 4680 facility is currently under construction. The additional four lines incorporate further capital efficiencies over phase one, and our target is for them to start producing in late 2024. End quote. In other words, Tesla is expecting the four current lines at Texas, which have a total production capacity of 100 gigawatt hours to ramp quickly enough that they're already lining up the next set of production lines. I'll cover that further when we get to my updated projections.
德鲁接下来的评论是：“引用一句话，德克萨斯4680工厂的第二阶段目前正在建设中。额外的四条生产线在第一阶段的基础上进一步提高了资金效率，我们的目标是让它们在2024年末开始生产。”换句话说，特斯拉希望德克萨斯目前的这四条生产线能够快速启动，它们的总产能达到100千兆瓦小时，以至于公司已经在筹备下一批生产线。我将在谈到我的最新预测时进一步解释。
With regards to the further capital efficiencies, my assumption is that what Drew is referring to here is the simplified manufacturing process and reduction in factory footprint that was teased at Investor Day. Although the slide says G.F.N.V. or Gigafactory Nevada, they're likely installing the same or similar equipment at Gigataxas for phase two.
关于进一步资本效益的问题，我的假设是Drew在这里所指的是投资者日活动中所提到的简化制造流程和减少工厂占地面积的事情。虽然幻灯片上写着G.F.N.V.或“内华达超级工厂”，但他们很可能在德州超级工厂的二期也安装了相同或类似的设备。
On that note, how much 4680 production capacity is Tesla building outside of Texas? First, Tesla has 100 gigawatt hours planned for Nevada. I don't expect that production to come online until 2025 because the factory hasn't been built yet. As for Gigabrelin, we do know that Tesla originally planned to ramp 100 gigawatt hours there, but there's no indication from the last nine months that 4680 lines have been or are being commissioned there. That may be because it makes more sense to focus on U.S. production due to the massive tax incentives there. So for now, I'm going to count Berlin out for cell production ramps for the next year or two. That assumption could be wrong, but if Tesla provides updated guidance in an upcoming earnings call, I'll update the table and cover it in my quarterly video.
关于这一点，特斯拉在德克萨斯州之外正在建设多少4680生产能力？首先，特斯拉计划在内华达州建设100千兆瓦时的生产能力。我预计这个生产能力要到2025年才能启动，因为工厂还没有建成。至于吉加布伦，我们知道特斯拉最初计划在那里扩大100千兆瓦时的产能，但在过去九个月没有任何迹象表明在那里建立了或正在建立4680生产线。这可能是因为在美国专注于生产更具有意义，因为那里有巨额税收激励措施。所以目前，我打算在接下来的一两年内排除柏林进行电池产能扩张。这个假设可能有误，但如果特斯拉在即将到来的盈利电话会议中提供了更新的指导意见，我会更新表格并在我的季度视频中涵盖它。
Shanghai is a different story. There's plenty of battery capacity being built in China, and the cell and pack prices are low there. So Tesla doesn't necessarily have to build 4680 production in China at any time in the near future and can focus their energy elsewhere. So overall, between Texas and Nevada, there's about 300 gigawatt hours of 4680 capacity planned in the next two to three years. And that doesn't include any additional capacity that Tesla is sure to use from third parties. To put that production capacity in perspective, Tesla will probably only use about 150 gigawatt hours of battery cells this year. So 300 gigawatt hours will give them several years of runway.
上海则完全不同。中国正在建设大量的电池容量，而且电芯和电池组的价格也很低。因此，特斯拉不一定需要在最近的将来在中国建设4680生产线，他们可以将精力集中在其他地方。因此，总体来说，在德克萨斯州和内华达州，未来两到三年内计划建设大约300吉瓦时的4680容量。这还没有包括特斯拉肯定会从第三方获取的任何额外容量。从生产能力的角度来看，特斯拉今年可能只会使用约150吉瓦时的电池电芯。因此，300吉瓦时的产能将为他们提供几年的支持。
With all that in mind, I don't expect Tesla to have any cell constraints in the next two to three years when they choose to ramp products like the compact vehicle, Cybertruck and Semi. Moving along, Drew's final comment was, quote, lastly, in Kato, we're retooling to enable large-scale pilot runs of our next-generation cell designs. Kato's long-term goal is to be the launch pad for new cells, one generation ahead of our mass production facilities, enabling faster iteration and smoother rampups of new cell designs. End quote.
考虑到这一切，当特斯拉选择加速推出紧凑型车、Cybertruck和半挂卡车等产品时，我并不认为在未来两到三年内特斯拉会遇到任何电池供应不足的问题。接下来，德鲁的最后一句话是：“最后，在加东，我们正在重新调整以实现下一代电池设计的大规模试制。”加东的长期目标是成为新电池的试点基地，领先于我们的大规模生产设施一个世代，以便更快地迭代和顺利推出新的电池设计。
There's three major points here. First, the Generation 1 cell line in Kato Road will need downtime for retooling to produce next-generation cell designs. Second, what do they mean by large-scale? My guess is that the upgraded line will still have the roughly 10 gigawatt hours of capacity that Elon advised of in the Q4 2020 earnings call. It would be impressive if they could expand that beyond 10 gigawatt hours, but it wouldn't necessarily provide much additional benefit if they're just using the line for test runs.
这里有三个重要的点。首先，加托路(Gato Road)的第一代细胞产线将需要停机重新装修以生产下一代细胞设计。其次，他们所说的大规模是什么意思？我猜升级后的产线仍然会有大约10千兆瓦时的产能，这是埃隆在2020年第四季度收益电话会议上提到的。如果他们能将产能扩大到10千兆瓦时以上，那将非常令人印象深刻，但如果他们只是将该产线用于试跑，这并不一定会提供太多额外的好处。
Third, what are the implications of Kato Road becoming a launch pad for new cells? As a bit of background, in the past, Kato Road was a research and development facility rather than a production facility. The 10 gigawatt hour pilot line was built at Kato Road because that's where some of the earlier development took place for the 4680. But based on my interpretation of reporting from Cleaner Watt and from Tesla's earnings calls, it appears that the pilot line at Kato Road ran into unexpected issues that may have forced the final development of the 4680 production process to occur in Texas. So in the interim, Tesla leaned on the pilot line at Kato Road to mass-produce 4680 cells for use in the Model Y.
第三点，加藤路成为新电池的发射平台有哪些影响？作为背景说明，过去加藤路是一个研发设施，而不是生产设施。10吉瓦时试验生产线建在加藤路，是因为之前在那里进行了4680电池的早期开发工作。但根据我对Cleaner Watt和特斯拉财报的报道的理解，似乎加藤路的试验生产线遇到了意外问题，可能导致最终的4680生产工艺开发转移到了德克萨斯州。因此，在此期间，特斯拉依赖加藤路的试验生产线大规模生产用于Model Y的4680电池。
What Drew is saying here is that it will no longer be used for mass production and can go back to focusing on R&D. What kind of R&D? In my view, the main focus will be on chemistry improvements. So far, Tesla hasn't focused on improving the cell chemistry of mass-produced 4680s because the focus was on just getting 4680 production working. That's because even slight changes to cell chemistry can cause dramatic changes to things like the cohesion and adhesion of the active materials in the battery cell, which in turn can mean weeks to months of adjustments to the production process to consistently produce high-quality battery cells. Now that 4680 production is going smoothly in Texas, the 10-gWh pilot line at Kato Road can be used to work those kinks out of new chemistries on production-scale equipment before passing them along to Texas and other locations for actual mass production. That should allow relatively seamless production ramps for new chemistries after they've been tested on the equipment at Kato Road.
德鲁在这里所说的是，这种技术将不再用于大规模生产，可以重新专注于研发。那么，是什么样的研发呢？在我看来，主要的焦点将放在化学改进上。到目前为止，特斯拉并没有专注于改进大规模生产的4680电池的化学成分，因为当时的重点仅仅是让4680电池的生产正常运作。这是因为即使对电池的化学成分进行微小的改变，也可能对电池中的活性材料的结合和粘合性产生巨大的影响，进而可能需要几周甚至几个月的调整来稳定地生产高质量的电池。现在，在得克萨斯州，4680电池的生产正在顺利进行，位于卡托路的10千兆瓦小时试验生产线可以用来在生产规模设备上解决新化学成分的问题，然后再将其传递给得克萨斯州和其他地点进行实际的大规模生产。这将使得新化学成分在卡托路的设备上经过测试后能够相对无缝地进行生产扩展。
As for the impact of the chemistry changes, as I pointed out in my last video, they could easily push the energy density of the 4680 above 280Wh per kilogram and potentially above 290Wh per kilogram in the next few years.
关于化学变化的影响，正如我在上个视频中指出的那样，它们可以轻松将4680电池的能量密度提高到每公斤280瓦时以上，未来几年甚至有可能达到每公斤290瓦时以上。
Before we move on to projecting where the 4680 ramp will go from here, several people asked me how the ramp of the 4680 so far compares to Tesla's original production ramp plans. Let's take a look. First, at Battery Day, Tesla showed this slide, which showed 100gWh in 2022. But due to poor wording during the presentation and a lack of notes on the slide, at that time we didn't know whether they meant cells consumed from all sources or if the figures were just meant to be for in-house production. In retrospect, it's clear that Tesla was talking cells consumed from all sources. So although many people assumed this was projecting the 4680 ramp, that doesn't look like it was the case.
在我们讨论4680电池的进展以及未来可能去向之前，有几位读者问我关于4680电池目前的生产增速与特斯拉最初的生产计划有何差距。我们来看一下。首先，在电池日活动上，特斯拉展示了这张幻灯片，显示2022年将达到100GWh。但是由于演示过程中措辞不当以及幻灯片上缺乏注释，当时我们不知道他们是否指的是所有来源的电池消耗量，还是仅限于内部生产。回顾起来，很明显特斯拉指的是所有来源的电池消耗量。因此，尽管很多人认为这是在展望4680电池的生产增速，但事实似乎并非如此。
However, Tesla did give guidance in the Q2 2021 earnings call that there was around a 50% chance that they would hit 100gWh run rate by the end of 2022. That means their guesstimate from mid-2021 is looking like it'll be about 2.5 years off if they managed to hit 100gWh run rate in late 2024 or early 2025. Although Tesla was over-optimistic about the 4680 ramp and that was nail-biting and frustrating at times, it looks like they're on track now and it's quite an achievement.
然而，在2021年第二季度的盈利电话会议中，特斯拉确实发表了关于到2022年底他们有约50%的机会达到100GWh运营水平的指导意见。这意味着，如果他们成功在2024年底或2025年初达到100GWh的运营水平，那么他们从2021年中期开始的预估将会延误大约2.5年的时间。虽然特斯拉在4680电池的扩产方面过于乐观，有时令人焦虑和沮丧，但看起来他们现在已经在正确的轨道上，并且取得了相当大的成就。
Overall, even at this early stage in the 4680 ramp, they already have what's probably the most productive battery line in the world and it's using a dry battery electrode coating technology that wasn't expected to hit the market for at least another couple of years. If they can continue to ramp the first line at Texas and rapidly clone it, then we'll finally see the explosion of battery cell production that was expected in the last 2-3 years after battery day.
总体而言，尽管在4680电池的扩产过程中仍处于早期阶段，但特斯拉已经拥有世界上可能是最高产的电池生产线，并且使用了一种干法电池电极涂覆技术，该技术预计至少还需再过几年才能投入市场。如果特斯拉能够继续扩大在德克萨斯州的第一条生产线，并迅速克隆它，那么我们将终于看到在电池日后的过去2-3年中所期待的电池电池生产的爆发。
On that note, let's do some projections and guesswork on where things are going next. As I showed earlier, on screen is the original projection that I showed back in July that used a 21% improvement rate, which turned out to be relatively accurate. If the 21% improvement rate holds, Tesla will hit 30 million total cells produced in early December, 40 million in January, and then 50 million in February. Hopefully, Tesla lets us know if they hit those specific milestones so we can follow along, but I wouldn't be surprised if the next update is a doubling or 40 million cells just before the next earnings call. We'll talk about the projections for deeper into 2024 in a moment.
在这个基础上，让我们做一些预测和猜测，看看未来的发展方向。正如我之前展示的，屏幕上显示的是我在七月份展示的原始预测，使用了一个21%的改进率，结果相对准确。如果这个21%的改进率保持下去，特斯拉将在12月初达到3000万个电池单元的总产量，1月份达到4000万个，然后在2月份达到5000万个。希望特斯拉能告诉我们是否实现了这些具体的里程碑，这样我们就可以跟进，但如果下一次更新是在下次财报电话会议前翻一番，生产达到4000万个电池单元，我也不会感到惊讶。稍后我们将讨论更深入到2024年的预测。
Before we get into that, is continuing to project the 21% improvement rate into next year reasonable. The bull argument would be that Tesla's now reaching the steep part of the S curve where output skyrockets, and that'll become pounded by the fact that Tesla is now starting to spool up subsequent production lines.
在我们深入讨论之前，我们需要考虑一下将21%的改进率延续到明年是否合理。看多一方的论点是特斯拉现在正在进入S曲线的陡峭阶段，产量将大幅增加，而且特斯拉现在开始启动其他的生产线，这将进一步促进产量增长。
As for the bear arguments, I see three. First as Drew said in the earnings call, the team is currently focused on quality. It could slow the 21% improvement rate.
就看看那些持看跌意见，我看到三个。首先，正如德鲁在盈利电话会议上所说，团队目前专注于质量。这可能会减缓21%的改进速度。
Second, there may come a time in the next year or two that the 4680 ramp progresses so quickly that material supply becomes a challenge. For example, what if the 4680 ramp is dependent on the cathode plant at Austin, or their lithium refinery at Corpus Christi, and those projects run into delays? I imagine Tesla has fallback plans, but it illustrates the point. The factories and mines that support the 4680 ramp are as important as the 4680 production lines themselves.
其次，在接下来的一两年内，可能会出现这样的时刻，即4680电池生产进展如此之快，以至于物料供应成为一个挑战。例如，如果4680电池生产受到奥斯汀的正极工厂或者克里斯蒂港的锂精炼厂项目延迟的影响，会怎么样呢？我想特斯拉会有备用计划，但这也说明了一个问题。支持4680电池生产的工厂和矿山与4680生产线本身一样重要。
The third bear argument is that because the current 5gWh run rate from line 1 at Texas is only at 20% of its potential run rate, they may run into more issues that affect the shape of the production S curve. However, the counter argument to that would be that we know the first 4680 line is capable of at least roughly 5gWh per year at steady state. But the key word there is steady state. Tesla probably did burst production runs on that line to test its maximum output before deciding to spool up the second line. That would have given them some indications of how the line handled faster production rates and therefore given them insights into what the improvement runway looks like for the 4680 ramp.
第三个反对意见是，由于得克萨斯州1号生产线目前每年仅达到其潜在产能的20%的5GWh运行速率，他们可能会遇到更多影响生产S曲线形状的问题。然而，对此的反驳是，我们知道首条4680生产线在稳定状态下至少能达到大约每年5GWh的产能。但是关键词是稳定状态。特斯拉可能在该生产线上进行了爆发性的生产来测试其最大产量，然后再决定启动第二条生产线。这将为他们提供一些线路如何处理更快生产速率的线索，从而为4680扩产的改善道路提供见解。
Overall, in light of the bull and bear arguments, and given that there are so many other potential variables, my view is that the most pragmatic way to project the 4680 ramp is to look at past performance. With that in mind, I'll continue to use the 21% month over month improvement rate until we get new information or advice from Tesla.
总的来说，在牛市和熊市的观点下，并考虑到还有许多其他潜在变量，我认为最实际的预测4680放大的方法是观察过去的表现。在此基础上，我将继续使用每月21％的改善率，直到我们收到特斯拉的新信息或建议为止。
Now that we have a projection in hand for the production ramp, let's layer in some of the other information from the earnings call. First, Drew said that the first four production lines would ramp over the next 3 quarters. Some people might argue that he meant that the first four lines would be fully ramped in 9 months. If that happened, it would be excellent news. But my view is that the intent of his wording was to set up the next sentence where he says that the next four production lines from phase 2 will start production at the end of 2024, which is 3 quarters from now.
现在我们手上有了生产斜坡的预测，让我们将收益电话中的其他信息加入进来。首先，德鲁说前四条生产线将在接下来的三个季度内加速。有些人可能会认为他的意思是前四条线将在九个月内完全加速。如果发生这种情况，那将是个好消息。但我的观点是，他用这种措辞的目的是为了下一个句子的铺垫，他说第二阶段的下四条生产线将在2024年底开始生产，也就是说从现在开始的三个季度后。
Let's look at why that is using the projection that we looked at earlier. At a 21% improvement rate, phase 1 should hit a 74 gigawatt hour run rate by the end of 2024. At 74 gigawatt hours, phase 1 will be 74% ramped, and at that point, the production S curve will be starting to asymptote, meaning that the improvement rate will start to slow as phase 1 reaches its nameplate capacity. So if Tesla wants a continuous ramp over the next couple of years, then I expect phase 2 to start ramping before phase 1 starts losing steam. This is why I don't think Drew meant that phase 1 would be fully ramped in 9 months ahead of phase 2, which would effectively create a pause in production increases between the two phases. If that's correct, it leads me to believe that Tesla is also projecting roughly a 21% improvement rate, because they're planning on starting the ramp of phase 2 when phase 1 would be in the steepest part of the projected S curve. That's when phase 2 would need to start in order to create a continuous ramp.
让我们来看一下为什么要使用我们先前提到的预测方式。以21%的改进率计算，第一阶段应该在2024年底达到每年74千兆瓦时的生产水平。在74千兆瓦时时，第一阶段将会达到74%的增长，并且在那个点上，生产的S曲线将开始逐渐趋于平缓，这意味着改进率会随着第一阶段达到满负荷容量而开始减缓。因此，如果特斯拉想在接下来的几年里持续推进增长，我预计第二阶段将在第一阶段开始减速之前开始推进。这就是为什么我认为Drew所指的并不是第一阶段会在第二阶段之前的9个月内达到完全增长，那样会在两个阶段之间造成生产增长停滞。如果我的推测正确，那么我认为特斯拉也在预测大致上21%的改进率，因为他们计划在第一阶段达到预测S曲线的最陡峭部分时开始推进第二阶段。这就是为什么第二阶段需要在那个时间点开始，以实现持续推进。
The second piece of information to layer in from the earnings call is that Tesla expects to hit volume production of the Cybertruck in about 18 months. Even if Tesla was able to beat that timeline and fully ramp the Cybertruck by the end of next year, at an average pack size of 130kWh and 250,000 vehicles per year, the Cybertruck line would at most be using about 33 gigawatt hours of cells per year in December of 2024. That's as compared to the projected 74 gigawatt hour run rate of the 4680 lines in Texas.
从财报电话会议中获得的第二个信息是，特斯拉预计在大约18个月内实现Cybertruck的大规模生产。即使特斯拉能够提前完成这一计划，并在明年底全面推进Cybertruck生产，以平均130千瓦时的电池组规格和每年25万辆的产量计算，到2024年12月，Cybertruck生产线每年最多只会使用大约33千兆瓦时的电池。与德克萨斯州的4680生产线预计的74千兆瓦时产能相比，差距很大。
In other words, even in a bullish scenario for the Cybertruck ramp, at the end of next year there could be around 40 gigawatt hours of spare run rate for 4680 cells in Texas. If that happens, what could Tesla do with those cells? First, the margins of Tesla's energy storage products are currently 24% and increasing, while the margins for the automotive business are around 18% and falling. From that perspective, it makes more sense to use the 4680 cells to drive further cost reductions in vehicles, because that's where margins are the weakest.
换句话说，在 Cybertruck 生产扩张的看涨情况下，到明年年底德克萨斯州可能会有大约40千兆瓦的4680电池备用产能。如果发生这种情况，特斯拉可以做什么？首先，特斯拉的储能产品利润率目前为24%，并且在增加，而汽车业务的利润率约为18%且在下降。从这个角度来看，更合理的做法是使用4680电池进一步降低车辆成本，因为这是利润率最低的领域。
Second, the high energy density nickel-based battery cells that are being produced in Texas are better suited for vehicles than grid storage products. Although Tesla will likely manufacture LFP 4680 battery cells in the future with lower energy density and higher cycle life that could be used for storage products, I don't see that happening in the next 18 months.
其次，在德克萨斯州生产的高能量密度镍基电池更适合用于汽车而不是电网储能产品。虽然特斯拉可能在未来生产能量密度较低、循环寿命更高的LFP 4680电池用于储能产品，但在接下来的18个月内我认为不会出现这种情况。
That means the best use case for the 4680 cells outside of the Cybertruck and grid storage are for the Model 3 and or the Model Y. Why not the semi-upcoming compact vehicle, S and X, or powerwalls? The semi-compact vehicle haven't started their production ramps yet, and the SX and powerwalls don't have enough installed production capacity to absorb a 40 gigawatt hour run rate of cells by the end of next year.
这意味着4680电池在Cybertruck和电网储存之外，最适合用于Model 3和/或Model Y。为什么不是即将发布的紧凑型车辆S和X，或者Powerwall呢？紧凑型车辆尚未开始生产扩展，并且SX和Powerwall的安装产能不足以吸收到明年年底40千兆瓦时的电池生产能力。
That's as opposed to the Model Y and or the Model 3, where 40 gigawatt hours of cells would equate to roughly 500,000 vehicles per year. If Tesla delivers 1.8 million vehicles this year, they only have to grow production by 27% in the next year to absorb those additional gigawatt hours.
这与Model Y和Model 3不同，因为40千兆瓦时的电池足以支持大约50万辆汽车的年产量。如果特斯拉今年交付了180万辆汽车，那么他们只需要在明年生产增长27%就能吸纳这些额外的千兆瓦时。
With all that said, although it's interesting to talk about product level choices, as an investor my main concern isn't so much about which product each cell will go into, which I'm sure Tesla will make the right decision about, but rather how quickly the 4680 lines scale and how much that will reduce production costs.
尽管说了这么多，虽然谈论产品层面的选择很有趣，但作为投资者，我的主要关注并不在于每个电池将用于哪种产品，我相信特斯拉会做出正确的决策，而更关注的是4680生产线能够快速扩大规模以及这将如何降低生产成本。
On that note, let's take a look at how much the 4680 cells could reduce the production cost of Tesla vehicles. We'll look at costs without Inflation Reduction Act tax credits and then with those tax credits. Currently the average price for battery cells is around $100 per kilowatt hour. That means at the pack level they're about $130 per kilowatt hour. Yes, Tesla might be getting a better deal, but we don't need to be exact here.
在这个问题上，我们来看一下4680电池能如何降低特斯拉汽车的生产成本。我们将分别以未计入《通货膨胀降低法案》税收优惠的情况以及计入该税收优惠的情况来考虑成本。目前，电池单价大约是每千瓦时100美元。这意味着在整个电池组的层面上，价格大约是每千瓦时130美元。是的，特斯拉可能达成了更好的交易，但我们这里不需要十分精确。
The main purpose is to sketch out a rough guide for the impact of the 4680 on vehicle cost. If the average battery pack of the Model 3 and Y are 75 kilowatt hours, each pack currently costs Tesla about $9,750. Tesla's current target for the 4680 is $70 at the cell level, which I estimate would make the pack cost about $90 per kilowatt hour. That would make the cost of a 75 kilowatt hour pack using 4680 cells $6,750.
主要目的是为了勾勒出4680对车辆成本的大致影响。如果Model 3和Y的平均电池组容量为75千瓦时，每个电池组目前对特斯拉的成本约为9,750美元。特斯拉当前对4680电池在单体级别的目标价格为70美元，根据我的估算，这将使电池组的成本约为每千瓦时90美元。这将使使用4680电池的75千瓦时电池组的成本为6,750美元。
$9,750 for the current pack cost minus $6,750 for the 4680 pack would mean a savings of $3,000 per vehicle. Bear in mind that Tesla's probably about 18 to 24 months away from hitting that pack cost because their $70 cost target is likely calculated based on a fully ramped production phase. Even when phase 1 is early in the ramp, perhaps early next year, I expect the in-house cells will already be cheaper than third party options.
当前电池组的成本为9,750美元，减去4680电池组的6,750美元，意味着每辆车节省3,000美元。请记住，特斯拉距离实现这个电池组成本还有大约18到24个月的时间，因为他们的70美元成本目标很可能是基于完全扩产阶段计算的。即使在第一阶段扩产早期，也许是明年初，我预计自产电池会比第三方选项更便宜。
Regardless, if there's one question I have for the next earnings call about the 4680 ramp, it would be when do they expect the 4680 packs to hit cost parity with the average third party options.
无论如何，如果我对下一次收益电话中关于4680电池扩产的问题只有一个问题，那就是他们预计4680电池组何时能与平均第三方选项达到成本平价。
Moving along, we still need to factor in the Inflation Reduction Act tax credits. For each kilowatt hour of battery cells that Tesla produces in the US, they're eligible for $35 in tax credits, as well as $10 for the packs, for a total of $45 per kilowatt hour. $45 per kilowatt hour times 75 kilowatt hours is a savings of roughly $3,400.
继续进行，我们还需要考虑通货膨胀降低法案的税收抵免。对于特斯拉在美国生产的每一千瓦时电池单元，他们有资格获得35美元的税收抵免，以及10美元用于电池组，总共每千瓦时45美元。每千瓦时45美元乘以75千瓦时，大约可以节省3400美元。
If we add in the potential $3,000 of production cost savings, that's a total of $6,400. Currently, Tesla's average cost of goods sold for their vehicles is about $37,500. $37,500 minus $6,400 for production and tax credit savings would make the new cost of goods sold $31,100 per vehicle.
如果我们考虑到潜在的3000美元生产成本节省，总共将达到6400美元。目前，特斯拉汽车的平均销售成本约为37500美元。以每辆车生产成本和减税优惠后的6400美元扣除37500美元，新的销售成本将变为每辆车的31100美元。
Or looking at it another way, Tesla's current average selling price across all makes and models globally is around $44,000. To keep things simple, let's assume that applies to the average price for vehicles in the US specifically as well. A $6,400 cost reduction might allow the average cost of Tesla vehicles sold in the US to drop to $37,600 if those vehicles used the $46,80 sell, or about $41,000 if half the vehicles used the $46,80 sell.
或者换个角度看，特斯拉在全球所有型号上的当前平均销售价格约为44,000美元。为了简化问题，让我们假设这同样适用于美国市场的车辆平均价格。如果这些车辆使用46,800美元的售价，那么6,400美元的成本降低可能使特斯拉在美国销售的车辆平均成本降至37,600美元，或者如果一半的车辆使用46,800美元的售价，那么大约为41,000美元。
That would considerably improve the margins of those vehicles and or allow Tesla to drop the price by several thousand dollars. In summary, nearly three years after battery day, it looks like $46,80 sell production is ramping in earnest.
这将大幅提高这些车辆的利润空间，或让特斯拉能够将价格降低数千美元。总结起来，距离电池日已经接近三年，现在看来46,800美元的销售生产正在认真加速推进。
Tesla isn't out of the woods yet, but the trajectory looks excellent and the actions they're taking, like converting Kato back to R&D and starting on Phase 2 of construction at Texas, inspire a lot of confidence. All continue providing updates on a quarterly basis until they're no longer useful.
特斯拉还没有彻底摆脱困境，但发展趋势看起来非常良好，他们所采取的行动，例如将加拿大Kato工厂转回研发，以及在德克萨斯州开始第二阶段的建设，给人很多信心。我将会持续每季度提供更新，直到这些更新不再有用为止。
But I have a feeling things will be getting more interesting in the battery space over the next two years rather than less interesting as Tesla rolls out more products, launches more production lines, and makes improvements to their battery sell technology.
但是我有一种感觉，在未来两年内，电池领域的事情会变得更加有趣，而不是变得更加无聊，特斯拉推出更多产品、启动更多生产线并改进他们的电池销售技术。
If you enjoyed this video, please consider supporting the channel by using the links in the description. Also consider following me on X. I often use X as a test bed for sharing ideas, and X subscribers like my Patreon supporters generally get access to my videos a week early.
如果您喜欢这个视频，请考虑通过描述中的链接支持本频道。还请考虑在X上关注我。我经常在X上使用它作为分享创意的试验平台，并且像我的Patreon支持者一样，X的订阅者通常会提前一周获得我视频的访问权限。
From that note, a special thanks to my YouTube members, X subscribers, and all the patrons listed in the credits. I appreciate all your support, and thanks for tuning in.
在那张便条上，特别感谢我的YouTube会员，X位订阅者以及所有在片尾名单中列出的赞助者。我非常感激大家的支持，谢谢你们的收看。