2017 Investor Roundtable:General Discussion

[jbcarioca]

To be fair, The F 35 program tried a very similar approach by building the production equipment while flight testing. It was a disaster, even with some of the most advanced computer prototyping on the planet.

To be fair they made zero attempt to actually know what the eventual performance might be. The large disasters in defense procurement have been related to defining production, sourcing and support before defining the product. Precisely zero relationship to our present issue.

 

[erthquake]

To be fair, are you comparing the technical complexity and ease of manufacture of the F 35 to the Model 3?

Because, you know, $100M = $35,000.

 

[kenliles]

To be fair, The F 35 program tried a very similar approach by building the production equipment while flight testing. It was a disaster, even with some of the most advanced computer prototyping on the planet.

Good point.
Maybe Lockheed should've farmed it out...

 

[dakh]

If that Reddit post is true, and the poster didn't misheard/misreport on this, I personally think this skipping beta is risky as well. Essentially, alpha is testing feasibility of the car beta is testing the reliability of the car. If Tesla really skipped beta and go straight with release candidates, there are three scenarios possible, but no matter what the first 10k Model 3 owners/Elon's employees will truly be the beta testers of the car:

1. All the "advanced analytical techniques" worked out perfectly and everything is well.
2. Major flaws were discovered and Tesla needs to call back the cars and modify design and/or production line.
3. Serious flaws happening systematically resulting serious injuries even deaths. All the negatives of 2 plus a big mess.

Essentially I hope the skipping beta part is not true.

If you remember the kind of problems they had previously -- leaking seals, misaligned panels, bearings not getting greased enough -- none of those would be caught in the prototype phase. Maybe the titanium plate deal could, but it didn't happen anyway. They were flaws of the assembly process not the original design. What they need to test is the whole system, not just the car itself. And looks like they lined up the tooling early enough to start testing now. They can put more miles on a much larger fleet of cars in shorter period of time because they can pop out a lot of them quickly and economically while testing production equipment as well. Long haul testing might be more difficult due to shorter time frame but I'm sure they considered that and came up with ways to mitigate that.

 

[MitchJi]

Tesla Model 3 ‘release candidates’ are currently being built – Musk notes ‘almost entirely built with production tooling’

Musk however thinks that the raise will be sufficient to bring the vehicle to “reasonable production levels”, but he also left the door open for raising more money this year. He sees an 80% chance of them not needing to raise any more debt or equity this year and if they end up needing more capital, they plan “a revolver or asset-backed line or warehouse line”.

I think that type of credit is used for inventory in transit?

If so the only chance they see of needing to borrow more money is if the M3 ramps so quickly that they need to borrow money to cover the cost of M3's in transit.

 

[30seconds]

also telsa has shown that they can rapidly change production design/ tweeks. part of skipping beta may be their ability to make changes post-launch

 

[dennis]

Also we know they've changed the trunk opening design, I imagine that would have impact on crash test results, so if they did crash testing, does that mean there have to be some newer builds with the new trunk design already?

They have had a year to change the trunk opening. It was most likely done before "pencils down" at the end of June. Plenty of time for crash testing.

 

[Waiting4M3]

Ok, dumb question. What if it's a HUD, but projected onto the "dash". IIRC the guy they poached from Porsche had design pics of variable dash displays, not on the windshields. Possible? Guess we'll see soon.

That's how Prius does it I believe, or if I'm misunderstanding, if you could point me to a picture of the example, much appreciated.

 

[dennis]

If Tesla is skipping Beta, it means they don't expect high volume for Model 3. Taking such a big risk on a high volume car is suicidal. Warranty costs and service center congestion will drown the company. Bears will love it the most.

Fixing issues during production? Tesla hardly fixed the issues found in lower volume Model S production. Remember the story about some engineer reporting headliner issues, who was pushed down and then out?
Fixing the issues in (suposedly) high volume Model 3 production line will be a bigger nightmare.

This is a great example of a bear with no domain expertise or inside knowledge assuming they are smarter than the people responsible for executing the Model 3 program at Tesla. Like the very experienced head of manufacturing from Audi is going to allow an untested product to go into mass production? Give me a break.

Tesla does things differently. They have extremely smart engineers. They have learned a lot since they are on their 4th major car design and 5th or 6th production line design. Go ahead and second guess. But you don't know more than they do, no matter how smart you are.

 

[EnzoXYZ]

BMW Unafraid of Tesla Model 3, Says One in Five Bimmers Likely to Be Electrified by 2025

Conventional automakers seem to have pretty much worked their way through the five stages of Tesla. First came the denial that the electric upstart was indeed a rival, then we had anger (mostly at subsidies), bargaining (over subsidies), depression at the scale of the Model 3 order bank—and now, in the case of BMW, acceptance.

BMW is fully prepared to admit that Tesla’s pioneering work will potentially help its own quest to sell electric cars. “In many ways, I’m very supportive of what Tesla has done,” Ian Robertson, BMW’s head of sales and marketing, told us at the Geneva auto show. “The world needs that sort of new competitor. That isn’t to say that we’re not going to be very competitive with them as well, of course—but Elon Musk has achieved a lot, and I admire what he has done.”

The Tesla Model 3 is the case in point; it’s likely to be the biggest competitor to BMW’s upcoming mid-size iNext EV. “The Model 3 will come, but I’m not sure of what volume it will come with, I’m not sure of the price point it will come with, and I’m not sure how good the car [will be],” Robertson said, while insisting that the iNext, which we don’t anticipate seeing until 2021, will be “the latest in electromobility” when it comes out.

Seems to me BMW is still in stage 1- denial that the model 3 is a competitor of its 3 & 4 series. Notice how Robertson references the iNext EV as it's main competitor. Sorry to say BMW is far from accepting Tesla will destroy its smaller ICE models too.

Too bad the automotive industry isn't like the phone business in which you can see the demise of an industry in just a few years. I can wait 5+ years.

I've been a BMW fan for so long and it saddens me to see such denial in management. It's no reflection of the brand but its current employed managers.

 

[Waiting4M3]

BMW Unafraid of Tesla Model 3, Says One in Five Bimmers Likely to Be Electrified by 2025

In other news, in 2025, 4 out of 5 BMWs have problem finding buyers in EU. They're already banning cheaper diesels in city centers, writing is so clearly on the wall.

 

[Waiting4M3]

If you remember the kind of problems they had previously -- leaking seals, misaligned panels, bearings not getting greased enough -- none of those would be caught in the prototype phase. Maybe the titanium plate deal could, but it didn't happen anyway. They were flaws of the assembly process not the original design. What they need to test is the whole system, not just the car itself. And looks like they lined up the tooling early enough to start testing now. They can put more miles on a much larger fleet of cars in shorter period of time because they can pop out a lot of them quickly and economically while testing production equipment as well. Long haul testing might be more difficult due to shorter time frame but I'm sure they considered that and came up with ways to mitigate that.

Tesla car production may be automated, but last time I checked FSD is not available and they still need human drivers. So if they have a large(ish) test fleet, I wonder if there are any job openings....

 

[RobStark]

Tesla car production may be automated, but last time I checked FSD is not available and they still need human drivers. So if they have a large(ish) test fleet, I wonder if there are any job openings....

Owners and their Model S,3,X are the test fleet.

 

[Fallenone]

If you remember the kind of problems they had previously -- leaking seals, misaligned panels, bearings not getting greased enough -- none of those would be caught in the prototype phase. Maybe the titanium plate deal could, but it didn't happen anyway. They were flaws of the assembly process not the original design. What they need to test is the whole system, not just the car itself. And looks like they lined up the tooling early enough to start testing now. They can put more miles on a much larger fleet of cars in shorter period of time because they can pop out a lot of them quickly and economically while testing production equipment as well. Long haul testing might be more difficult due to shorter time frame but I'm sure they considered that and came up with ways to mitigate that.

Well imagine they switched the mechanism of the falcon wing door after launch. What a disaster would that be? IIRC, the mechanism (hydraulic to electronic or the other way around, can't remember) was changed during beta testing. Yes Model X was the hardest car ever made. Yes Model 3 was designed to be easily manufactured. But with a complex system like a car with thousands of parts, something wrong is bound to happen. Beta testing's purpose is to screen out major ones so those won't plague production let alone safety.

 

[Wenche]

You would have to ask our president.

Late answer... But I think he has an orange Tesla Roadster

 

[Jonathan Hewitt]

Answering a survey online from Strategic Vision for the Nissan LEAF I got last year. I saw this gem of a question and thought I would share it. The question previous to this one was if I was familiar with the Supercharger Network. I wonder who is paying them to send these surveys out.

 

[Waiting4M3]

Well imagine they switched the mechanism of the falcon wing door after launch. What a disaster would that be? IIRC, the mechanism (hydraulic to electronic or the other way around, can't remember) was changed during beta testing. Yes Model X was the hardest car ever made. Yes Model 3 was designed to be easily manufactured. But with a complex system like a car with thousands of parts, something wrong is bound to happen. Beta testing's purpose is to screen out major ones so those won't plague production let alone safety.

The way I see it, as long as they're building cars now to test, what's the harm if the cars are not built by hand, but rather by an almost complete production line? They will still catch the same problems, if not more, considering people generally think production line is more prone to quality variability than hand build.

Maybe Tesla instead of saying they're skipping beta, they should have said

"yes we're building beta now as planned. In addition, our production line is almost complete, so we thought might as well just build the beta using the production line. If we find anything that needs to be tweaked, our production line can be quickly changed. [add some color on how automated and flexible the new design of the production line is] By July, we'll be ready to build early production version on a production line that is already proved out and ready to go."

 

[dakh]

Well imagine they switched the mechanism of the falcon wing door after launch. What a disaster would that be? IIRC, the mechanism (hydraulic to electronic or the other way around, can't remember) was changed during beta testing. Yes Model X was the hardest car ever made. Yes Model 3 was designed to be easily manufactured. But with a complex system like a car with thousands of parts, something wrong is bound to happen. Beta testing's purpose is to screen out major ones so those won't plague production let alone safety.

I have no details on the door story, but I'll speculate they couldn't get it working in a prototype either. They just hoped they will figure that part out soon enough and proceeded with other steps for the rest of the vehicle. That's the only rational explanation I see -- the issue with the doors wasn't production, it was just making them work well enough at all. Their contractor couldn't demonstrate a working version in time so they had to take it over and ran out of time getting it ready on the same milestones as the rest of the vehicle.

Like it was mentioned here before, this ended up being a blessing in disguise because they now know how to manage contractors better and had the aha moment to optimize both vehicle and production together.

 

[jschwefel]

I wonder how the focus on the Model 3 designed for ease of production impacts this. Could Musk be thinking in terms of a beta version of the production line and that the test car is just media to validate the production line? I could see Musk inverting conventional thinking in such a way.

Suppose you were developing a 3D printer. The actual printable object is of little importance. You can use any number of test files to evaluate the performance of the machine. So any number of test versions of the Model 3 can be used to evaluate the performance of the production line.

Very possibly/likely.

 

[mmd]

This is a great example of a bear with no domain expertise or inside knowledge assuming they are smarter than the people responsible for executing the Model 3 program at Tesla. Like the very experienced head of manufacturing from Audi is going to allow an untested product to go into mass production? Give me a break.

Tesla does things differently. They have extremely smart engineers. They have learned a lot since they are on their 4th major car design and 5th or 6th production line design. Go ahead and second guess. But you don't know more than they do, no matter how smart you are.

Thanks for the "inside" info, that the engineers who worked on botched Model X are no longer with the company.
Yes, yes. Tesla engineers are the smartest while they work at Tesla. But they are the dumbest before they join Tesla, and after they leave and work for others. /s