Tesla Model 3 anticipation thread

In My Humble Opinion
421

They do use NCA batteries for this very reason. (Cobalt is very expensive, NCA is a compromise that has the same energy density as pure cobalt-lithium chemistry but needs a smaller proportion of cobalt.)

On the other hand, they are talking about near future production volumes through the gigafactory equaling the entire worldwide production of lithium-ion batteries when they built the gigafactory. Even if they are using only 9% cobalt, that’s enough to easily affect market prices.

Also, the model 3 motor uses rare earth magnets.

422

They seem to be around 200, actually. Pretty decent considering how early things are. Lots of companies ordering tens or so to experiment with. It’ll take a bit longer before they figure out how to integrate the units in their fleet, but once that happens I expect it to really take off.

500 miles, with a 30 min/400 mi charge, is a really fantastic number for a specific reason: drivers must take a 30 min break after 8 hours of driving, and cannot drive more than 11 hours continuously. So a driver can drive, say, 7 hours @ 60 mph for 420 miles, take the legally-mandated 30 min break, and then have more than enough juice to fill out the rest of the day’s 4 hour max. Even going on the fast side, say 70 mph, the numbers work out fine.

423

https://www.reuters.com/article/us-tesla-quality-insight/build-fast-fix-later-speed-hurts-quality-at-tesla-some-workers-say-idUSKBN1DT0N3

424

Tesla’s spin on this is a real knee slapper. “That’s just because we put all of our cars through 500 rigorous inspections and tests!”
Yeah, that “Deming” guy might have a thing or two to say about a production system which relies on that much inspection to catch defects.

425

Agree completely, this is obviously a bad sign. My feeling in general is that I think ‘tony stark’ is going to figure it out. That they are going to get this factory running smoothly, using lots and lots of fast, cool robots, and actually get a decent production rate out of it. Remember all the rockets they blew up at SpaceX? All the heartbreaking failed landings?

Now, it’s going to cost a lot of investor money to do this. Maybe so much they have to ask for more. And the long term prospects - the dollars spent vs dollars returned by people buying their cars - may not work out into the black for a long time, if ever.

So I wouldn’t recommend investing in Tesla stock. And only buy an early Model 3 if you’ve got money to burn : it’s probably not going to be as seamless and reliable as a similar Toyota. But I do think they are going to get it to work.

426

“robots” and “Tony Stark” have nothing to do with statistical process control. All you’re saying is they’re just going to build junk faster.

427

no? The robots can obviously converge their control loops to minimize errors, especially if the systems properly feedback, where the robotic system responsible can review what happened for the steps it messed up and adjust it’s coefficients accordingly. But this kind of fancy control takes iterations to develop right and is state of the art and allegedly Tesla is using brand new approaches to this.

428

you’re only considering final assembly. Tesla only manufactures certain parts of the vehicle, plenty of it is made by suppliers.

here’s an example from the Model S. and that’s hardly exhaustive since I know of several more.

429

Tesla’s using the same factory automation equipment as everyone else is, though. Kuka robots, WTC spot weld controls, Norsdon and Graco and Schucker dispensing systems, etc., etc. The thing is, none of these things make mistakes that are recoverable. Humans make mistakes, processes mean-shift, tolerances coalesce, and then no one catches it. Before you know it, there’s not enough epoxy to hold onto the windshield. Technology alone doesn’t solve the problem.

Whispers in the industry indicate the Tesla has a people-knowledge problem, and it doesn’t matter if you hire one genius, because he’s not an island. It doesn’t matter if he can deliver perfect stampings if the dimensional control strategy in the welding shop was developed by a bunch of orangutans. Or maybe you hire the absolute best purchasing guy, pinched from Walmart based on his reputation for getting the best price, but has no idea how to evaluate the supplier of the M10 bolts that hold the driver’s seat to the floor, because he has no quality systems experience or engineering background.

Tesla has a reality distortion field much larger than the late Steve Jobs.

430

People (mostly) don’t invest in tech companies because they’re profitable. Most don’t even offer dividends. You invest because you expect the company to grow and that the stock price will grow with it.

I hope that Tesla is unprofitable for decades to come. Not on individual products, obviously–I want them to make money on the Model 3 and continue making money on the S/X. And Powerwalls and Roadsters and Semis and the like. But I hope they continue investing every cent and then some into future growth and that they won’t stop until all ground transportation (and maybe some air) is electric and all generation is carbon-free.

Investors will be all too happy to put up with new stock issuances, debt, etc., because they know that Tesla will invest the money into things worth far more than the cash value. When they start having huge cash balances, that’s when the stock will crash because it means Tesla has run out of things to innovate in.

431

Do you think this is an option? If Musk’s estimate of March for ramped up production is accurate then they would have burned through 6 months of unplanned capital expense with no sign of cash flow from the Model 3.

This is not a small amount of money.

432

Again, I’m not talking about unprofitability on individual products. If Tesla can’t make reasonable gross margins on the Model 3, investors will have good reason to be skeptical.

But Tesla’s “master plan” has always been to use previous products to finance the next. This isn’t just raw profits, but also using investor confidence for stock and debt issuance.

Tesla’s current unprofitability isn’t just from capital spending, but if you subtract all the things they’re doing because they don’t plan on being a Model S/X company forever, they would be making money. Examples include more showrooms than they really need and underutilized Superchargers (note here that the Supercharger can be both over and undersubscribed at the same time). Also, the NUMMI plant which is too big for just S/X, and things like software development which costs the same no matter how many units you ship.

The current burn rate is unsustainable in the long term but I think investors will be happy with not making a profit for quite some time as long as they think Tesla is investing in reasonable projects. The Semi and Roadster 2.0 were well received, so if they use Model 3 profits to finance development here I don’t think there’s a problem. On the contrary; if Tesla started accumulating cash I think people would wonder why they weren’t moving faster to get into new markets. The strategy of being first has paid off so far.

433

Are we talking about the same thing? I’m questioning whether or not they have the financial capacity to continue to lose money.

434

except the “previous product” launch they’re in the middle of is an unmitigated disaster. and if they don’t get their shit together, won’t be able to “finance the next.”

It’s also from the immense waste of a 90% re-work rate on cars they’ve been building for 3-4 years now.

how about simply following the core principles of mass production and W. Edwards Deming’s philosophy of statistical process control (Which the Japanese took to heart and enabled them to make very high-quality, cost-efficient cars) instead of pissing all over them and running around in “production hell” like a flock of beheaded chickens. then bragging about their “500 inspections and tests” like it’s a good thing instead of an indicator of a total lack of discipline.

435

As much as I agree with criticism of Tesla on many grounds - quality, truthfulness, etc - I just can’t see that six months (or even a year, IMHO) of delays to the Model 3 is going to impact investors’ tradition of throwing buckets of money at the company.

Last year, their stock price was $180. Today it is $306. If investors were getting cold feet about the company, I think it would manifest itself in ways other than a few guys griping on a message board.

436

Give me a break. A few months delay is not an unmitigated disaster. They’re way ahead of schedule on Elon time. Tesla warned about delays and everyone expected them.

A company looking to break into a market has two basic options:
a) Copy whatever the incumbent guys are doing.
b) Throw out the existing knowledge and do your own thing.

Doing a) means you avoid all the stupid errors that were learned the hard way over time (in this case, the last century of auto development). It also absolutely guarantees that you will fail, because you can’t possibly do the same thing as the others better, and if you’re even slightly worse then they will eat your lunch.

Doing b) means you make lots of stupid blunders, but also occasionally hit on new ideas that were overlooked by the incumbents. You are almost certain to fail, because there are many good reasons for doing things the way they had been done, and the odds of the new ideas benefiting more than the loss from those blunders is low.

However, almost certain to fail is better than absolutely certain to fail. And so b) is the better approach for a company in Tesla’s position, even if it means you do all kinds of stupid things.

SpaceX has managed exactly this already with the rocket business. They are, without question, at the top of their game and others are barely able to even articulate a plan to compete, let alone actually compete. So b) paid off (despite making many serious errors over their history).

With Tesla, it remains to be seen. They’re doing ok for now, and obviously have a lot of positive mindshare. But they are still in the woods as far as development errors go, and it’s possible they won’t make it out in time. Overall I’m optimistic, but even if I weren’t I would still say a) is the worse option.

The Japanese succeeded because Deming’s approach was high technology at the time. It was a risk–they could have failed. It wasn’t known, then, that the approach (which has costs of its own) would pay off. But they took the risk and it did pay off. They would have failed had they simply tried to copy Detroit’s processes.

437

Are you reading anything I’ve said? I said the current burn rate is unsustainable. They can’t lose $600M/q forever. However, no one will care if they lose, say, $50M/q even for many years. It’s just not enough money to care about, and easily raised by any variety of means.

Even if the Model 3 program was wildly profitable and made hundreds of millions in profit per quarter, I would hope they spend every cent of it. And I would be concerned if they did something like pay dividends or do a stock buyback.

438

I’m questioning the depth of the well to get the Model 3 into production. You keep promoting Tesla as if it holds some kind of secret technology that investors will get a piece of once they get it to market. That’s simply not a realistic picture.

439

Well Muskcertainly has an innovative way of restricting the second hand market for his cars.

440

I dunno (kicks tires) it’s got 33.9 million miles on the odometer and the hood latch seems a bit wonky.