Loaded cars produced first...

[Troy]

Everything has completely changed. Until recently, Tesla was planning to start production with P and D versions in September 2017. Based on Elon's latest statement, D production won't start until January 2018 and P until April 2018. Therefore many D and P buyers including owners and people who waited in line won't get their car before single-motor buyers who reserved much later.

In fact, I have just completed working on the new version of the Model 3 Delivery Estimator and I was surprised to see that around 70% of all single motor buyers in the USA including Eastern states, will get their car before anybody gets a D or P version even in California. I don't expect even employees to get a D version before 2018. To me, it is obvious that they have changed the plan because there is a problem related to the production of D and P versions.

 

[zenmaster]

It's obvious to me too. Weird how many likes are given to the strange alternative narrations.

 

[TexasEV]

To me, it is obvious that they have changed the plan because there is a problem related to the production of D and P versions.

Just because it's obvious to you doesn't mean your assumption is correct. Maybe you don't know as much as you think you do.

 

[Troy]

The current situation is the opposite of what they were planning to do. The plan was to deliver all employee reservations first and then continue with owner P and D reservations in California. However, my calculations show that a non-owner from New York who reserved on 7th April 2016 will get his single motor car first before any Tesla owner from California who waited in line gets his dual motor.

Dual motor:
I'm calculating 64,000 single motor deliveries before any dual motor. The following list shows the completed delivery percentages in different regions before they start making the first dual motor cars for employees or owners.

USA, California Single motor 75.67%
USA, Pacific excluding CA Single motor 77.42%
USA, Mountain Time Single motor 86.43%
USA, Central Time Single motor 77.16%
USA, Eastern Time Single motor 73.38%
Canada Single motor 65.60%
Europe excluding the UK Single motor 12.49%


Performance
I'm calculating 130,000 single and dual motor deliveries before any performance version is delivered. The following list shows the completed delivery percentages in different regions before they start making the first performance versions for employees or owners.

USA, California Single motor 75.67%
USA, California Dual motor 24.00%
USA, Pacific excluding CA Single motor 77.42%
USA, Pacific excluding CA Dual motor 29.84%
USA, Mountain Time Single motor 86.43%
USA, Mountain Time Dual motor 30.75%
USA, Central Time Single motor 77.16%
USA, Central Time Dual motor 30.88%
USA, Eastern Time Single motor 73.38%
USA, Eastern Time Dual motor 25.53%
Canada Single motor 65.60%
Canada Dual motor 30.07%
Europe excluding the UK Single motor 12.49%
Europe excluding the UK Dual motor 22.41%
China or Hong Kong Dual motor 37.62%

Check out this thread to see whether or not my estimates, predictions, and expectations are accurate:
Prediction Thread - "You Called It"

 

[Bokonon]

As I mentioned in another thread, I don't have a background in manufacturing. That said... isn't it also possible that Tesla flipped its production order because of "the machine that builds the machine" (TMTBTM) rather than "the machine" itself? In other words, couldn't it be as simple as taking Elon's words at face value, that the complexity of the Model 3s being produced will increase in tandem with the complexity of the manufacturing lines themselves?

If achieving volume production is the paramount goal for 2017, then the start-simple approach makes sense if they are significantly more confident about quickly ramping up RWD production than they are about doing the same for AWD or performance production. Yes, this confidence is a function of the confidence that they have in their supply chain and the components it provides, but it's also a function of their confidence in their manufacturing process and equipment (i.e. TMTBTM).

Given the amount of emphasis that Elon has placed on the manufacturing process, volume production, and "physics first" principles over the past year, my impression is that Tesla doesn't want to "build" TMTBTM more than once -- at least not at a fundamental level. As a result of prioritizing TMTBTM over "the machine" itself, they have two choices: (1) start Model 3 production only when TMTBTM is ready to produce all possible configurations (which some might be quick to call a "delay"), or (2) start Model 3 production when TMTBTM is ready to produce the simplest possible configuration, while simultaneously continuing to build and enhance TMTBTM for more complex configurations, and bring those configurations online as they become available.

I'm sure if Elon had his way, option (1) would be available in July, and high-margin (and super-s3xy!) P75Ds would start rolling off the line, but clearly that is not the case. Instead, he seems to be prioritizing the near-term Model 3 production timeline and long-term production process over the content (and margins) of the initial product. I trust that whatever calculations led to that decision demonstrated that it was in the company's best interest to take this path.

 

[UnitaryExecutive]

I don't think it's that surprising. In the past they produced the loaded cars first to make as much money as possible after production tooling and ramp, so Elon just assumed they'd do the same. Then they hire Peter Hochholdinger to help with the M3 ramp and he sets them straight and turns a company that has defined their own manufacturing processes and done it like a startup to a streamlined manufacturing machine. Part of this is focusing on getting one variant right to start instead of trying to produce tons of different combinations.

 

[Troy]

Let me summarize. Tesla has abandoned the following 4 plans:

1. Highly optioned cars first
They are doing the opposite. The single motor will be delivered first, then the dual motor and finally the performance versions.
2. Employees first
Most employees won't get their cars until 2018 because no D or P will be produced in 2017.
3. Owners before non-owners
Again most owners won't get their cars until 2018.
4. Near the factory first
Not happening because there are not enough single motor orders on the West Coast. By the time they are ready to make the first dual motor versions, they will already have delivered 70-80% of all single motor orders in the USA.

Why have they changed their plans? There are three theories:

1. Problems with parts related to D and P versions. (My first theory)
2. They will wait until they have more money to purchase the robots for body line #2 and #3. (My second theory)
3. Their new VP of manufacturing says they need to focus on one variant first. (@UnitaryExecutive's theory)

I don't know which theory is more likely. I'm just surprised by apologists trying to argue that highly optioned means premium sound and leather seats instead of dual motors, performance and ludicrous mode.

 

[JoaoD]

Not happening because there are not enough single motor orders on the West Coast.

One could say that there aren't enough dual motor orders too, and in the entire world, since there are in fact no orders at the moment.

 

[WarpedOne]

Most employees won't get their cars until 2018 because no D or P will be produced in 2017.
...
Not happening because there are not enough single motor orders on the West Coast.

Has the sky hit your head already?
That was fast!

 

[acentre]

I am happy either way to see the Model 3 approach production. I would prefer an AWD, but will order a 2wd to get it as soon as possible. I can always order another AWD in the future. Those in the Northeast should now receive a Model 3 by the end of this year. Many of us are also glad to capture the Federal and in some cases, the new NY tax credits.
One screen, no hud, no awd ,no.... let us not lose site of the electric future.
Many happy ev miles to all....

 

[dsvick]

Let me summarize. Tesla has abandoned the following 4 plans:

1. Highly optioned cars first
They are doing the opposite. The single motor will be delivered first, then the dual motor and finally the performance versions.
2. Employees first
Most employees won't get their cars until 2018 because no D or P will be produced in 2017.
3. Owners before non-owners
Again most owners won't get their cars until 2018.
4. Near the factory first
Not happening because there are not enough single motor orders on the West Coast. By the time they are ready to make the first dual motor versions, they will already have delivered 70-80% of all single motor orders in the USA.

The only thing that has changed is that dual motors will not be produced first and that P will not be available until next year. Within those changes all of the other things may still hold true.

• Cars with more of the options that are available day 1 will get produced first.
• Employees will still get to configure first, if they want D or P they'll have to wait for delivery.
• Owners will still get some sort of priority but, again, if they want D or P they'll need to wait for actual delivery.
• Near the factory first will still be done, but it will be done with RWD not dual. I'm sure there are still a few thousand people near the factory that will be perfectly happy with RWD - so that should cover the first month or so of production.

I think the biggest impact on this will be that more and more people will be allowed to configure as they ramp up and need to find RWD orders to produce. After just a few months they should have a nice database of where people are and what they want. That should let them streamline production going forward.

 

[Troy]

THIS is what Tesla said. Listen from 48:13 to 49:08. Now, they have done a complete U-turn on that. Even people in China will get their car first before an employee or owner in California gets a P version.

Transcript:
Jonathan McNeill - VP of Global Sales & Service
In terms of S and X as a bridge to Model 3, we are talking through and thinking through that. Just as Elon mentioned earlier, the quickest path to receiving a Model 3 is being a Tesla owner. We've agreed that Tesla owners are receiving priority in terms of production, and so you can run the math I just mentioned. If 93% are new to Tesla, 7% of the reservation holders are Tesla owners. And the fastest way to get production vehicle even in 2017 is through Tesla ownership and so we're finding that there's a good conversion rate of folks that are coming in to test drive an S or an X who are Model 3 reservation-holders and are motivated to be Tesla owners now so that they can receive their Model 3 earlier.

 

[DFibRL8R]

This is very similar to the initial release of Model S with the most highly optioned cars released first. Earliest (non-signature) cars were 85's with air suspension. I wanted a 60 so I waited about 4 weeks beyond what I would have if I ordered an 85. Seemed like an eternity

The "D" is an option that significantly increases the complexity of the vehicle and the chance of mechanical issue as well as complicating the production line. If the majority of buyers will want a RWD especially in Coastal CA, that should be the version that gets produced first.

The "P" is an option for those who want the most 0-60 performance from their Model 3. This is counter to the mission of sustainable transport that drives Tesla. The P motor is less efficient. Tesla has already proven they can make incredibly fast machines with the P100D. Now to make a great middle market vehicle for the masses with further refinement (D, P) to come for those who want it.

This is also why I ordered two Model 3's. An early version of whatever they make first and a later version configured with whatever awesomeness comes next.

 

[SureValla]

The "D" is an option that significantly increases the complexity of the vehicle and the chance of mechanical issue as well as complicating the production line.

Are you suggesting that the D will be less reliable? Is there any evidence to suggest this is true on the S? I'm not saying you are I'm just curious as it may be a factor on whether I get AWD or not.

 

[WarpedOne]

Are you suggesting that the D will be less reliable?

Yes.

Say you have one part that fails with X % chance. A system containing one such part will not fail in (1-X) cases.
A system containing two such parts will not fail in (1-X)(1-X) cases.
(1-X)*(1-X) is less than (1-X) for all non-zero X i.e. less reliable.
Actual difference between the two may be big or very small, next to non existant, depending on the size of X.

If drive units fails in 1% of cars, 99% RWD cars will be OK, and 98% D cars will be OK.
If drive units fails in 0.1% of cars, 99.9% RWD cars will be OK, and 99.8% D cars will be OK.
If drive units fails in 10% of cars, 90% RWD cars will be OK, and 81% D cars will be OK.
If drive units fails in 50% of cars, 50% RWD cars will be OK, and 25% D cars will be OK.

Simple math.

 

[DFibRL8R]

Are you suggesting that the D will be less reliable? Is there any evidence to suggest this is true on the S? I'm not saying you are I'm just curious as it may be a factor on whether I get AWD or not.

Actually, in terms of reliability where failure = stranded on the side of the road, I would say the AWD is likely more reliable overall. I don't own a "D" but I believe you can limp along with a single drive unit if one fails. But in terms of likelihood of a drive unit failure, I agree with WarpedOne, the more drive units you have, the more likely one will fail. This is especially an issue in the early days of production where Tesla's reputation for quality is under scrutiny. Also simplifying the production line is good in terms of volume production.

Although I have read many accounts of drive unit problems, I've got about 90k miles on my current drive unit. The original was replaced very soon after delivery and has performed fine since. Car has never left me stranded.

 

[WarpedOne]

I don't own a "D" but I believe you can limp along with a single drive unit if one fails.

Depending on the type of the failure.
Drive units are not independant, RU is master, FU is slave, controlled by the logic in RU.
RU gets the power command from main computer and decides who will serve the request. If this control fails or if any connections go bad, you are stranded.

If FU fails without locking, the RU might take over all the work.
If RU fails, you are probably stranded.

 

[N5329K]

Does this announcement mean that 2WD orders will be filled first, starting from west to east?

Yes.
Robin

 

[SureValla]

Yes.

Say you have one part that fails with X % chance. A system containing one such part will not fail in (1-X) cases.
A system containing two such parts will not fail in (1-X)(1-X) cases.
(1-X)*(1-X) is less than (1-X) for all non-zero X i.e. less reliable.
Actual difference between the two may be big or very small, next to non existant, depending on the size of X.

If drive units fails in 1% of cars, 99% RWD cars will be OK, and 98% D cars will be OK.
If drive units fails in 0.1% of cars, 99.9% RWD cars will be OK, and 99.8% D cars will be OK.
If drive units fails in 10% of cars, 90% RWD cars will be OK, and 81% D cars will be OK.
If drive units fails in 50% of cars, 50% RWD cars will be OK, and 25% D cars will be OK.

Simple math.

What is the failure rate of drive units on the S?

 

[DFibRL8R]

What is the failure rate of drive units on the S?

I don't know that anyone but Tesla knows the answer to this. It also depends on how you define failure. Personally, I define failure as stranded on the side of the road or an incident that occurs while driving that places you at risk.

To my knowledge, there were relatively few of this type of failures reported. There were many owners who experienced excessive amounts of noise coming from the drive unit and had it replaced. My failure was intermittent, had periods of reduced power or brief power cut that was remotely diagnosed by service as symptoms of drive unit issues. There also seemed to be more of these issues early in production years than more recently but without knowing the exact number and types of problems, it is hard to say.