Tesla Model 3

[redpoint5]

Originally Posted by JSH

You seem to have skipped some key parts of that article:

"But Munro’s manufacturing costs for the Model 3 don’t include everything. Design time for engineers to develop the parts, sales and dealer distribution, marketing or any of Tesla’s “selling, general and administrative expenses” aren’t included. Munro’s figure essentially just considers the material and labor expenses needed to manufacture the vehicle.

He said it can be fairly straightforward to crunch numbers on the difference in cost based on kilowatt-hours of the lower-capacity battery, but devising cost for the car without the Premium Trim package would be hard. “What kind of functionality are they taking out? What kind of materials are they taking out for the lower trim model?”

I didn't leave that info out so much as summarize it. We have a pretty good idea of the cost of the car minus 25 kWh of battery capacity. Removing premium upgrades would further reduce Tesla's cost. By what amount remains to be seen, but it seems that even at a high trim level and the standard battery, Tesla can turn a marginal profit by selling for $35k. It may not be a huge 30% profit, but that's why volume is key to profitability. The article makes my point, which is why I linked it.

As long as their products are marginally profitable, (revenue from producing 1 more vehicle is greater than their cost to produce it), overall profitability is a relatively minor concern, especially when profits are reinvested back into the company.

Amazon wasn't profitable for 4 years, and they followed a similar cycle of reinvesting back into growing their business.

Elon's goal isn't to end up with a pile of money from Tesla profits; he already had that pile of money when he started the company.

[RedDevil]

We do SCRUM and Agile for a big project, which would work well if it was the only or primary thing I have to do, but it isn't - while it supposes it is, so it has the full size overhead but very little gain in my case. It kind of ruins my job.

But I see how it would work well without distractions.

[JSH]

Originally Posted by RedDevil

We do SCRUM and Agile for a big project, which would work well if it was the only or primary thing I have to do, but it isn't - while it supposes it is, so it has the full size overhead but very little gain in my case. It kind of ruins my job.

My thoughts as well. SCRUM and Agile can work great when applied to the right type of product and teams. It doesn't work for every type of project or company.

My wife is a design engineer at a medical device company. They use SCRUM and Agile and it works well. But she has 2 projects and the team for each is 4 people.

On the other hand I have 9 active projects plus 5 new model launches from 2019 to 2023 that I'm working on. If I had daily scrum meetings for each project I would spend 1/2 my day in those meetings.

There are departments in my company that use Agile and SCRUM. Manufacturing Engineering is not one of them though we use something similar to the daily scrum meeting when we are running pre-production vehicles on the line.

[roflwaffle]

Originally Posted by JSH

I agree planning for automation from the beginning is much more efficient than automating a formerly manual process. The problem is the automation doesn't work. So either Musk tried to automate things that aren't a good fit for automation, his automation people didn't know what they were doing or Musk didn't give them enough time to work out the bugs before starting production. My bet is on the first and last options.

Yup, we use SCRUM, Agile, and Lean manufacturing (and lots of other buzz word methods of lean manufacturing that happen to be in vogue)

If the Model 3 launch was successful you (and Musk) would be right. But it wasn't and instead it was "Production Hell"

For example, Musk skipped the soft-tooling stage. This involves making cheap, lower grade set of tooling to make sample parts. If something doesn't turn out right you modify the tooling and test again. When everything works you cut the purchase order for production hardened tools and discard the soft tools. That way when the line starts up all the parts fit together.

Musk went straight from rapid prototypes parts based on CAD to production parts from hard tool. He did this because he said the soft-tool stage in the Model S production wasn't useful. It wasn't useful because he didn't wait until all the tweaks were made before ordering the production tooling. Musk's take-away from this was not to follow the industry standard process and take the time to do the soft-tool stage correctly for the Model 3. Instead he just skipped the step and we have the fiasco

It seems very odd for a new company to skip basic steps when they don't know what they are doing. However, Musk doesn't know what he doesn't know and isn't willing to listen to people that do know.

For example Audi also skipped soft-tooling when they started then new plant in Mexico. However,

A. They already know how to do a product launch.
B. They already build the car in other locations.

You have to walk before you run and crawl before either.

The odd thing about this whole situation is that it's only a fiasco compared to what Elon said, not compared to EVs from other manufacturers. For instance, the Bolt was launched 20-months ago and sold 500+cars/month in it's first month out. It's now selling 1,000-1,400+ cars/month. Tesla starting point was a paltry 30 cars/month, but they reached 6,000+ cars/month in 12 months.

Tesla's doing poorly based on their own estimates, but they're doing very well compared to the competition. At this point they're selling more EVs in the US than every other manufacturer combined, and they're about 2,000 cars/month away from selling more EVs than EVs+PHEVs from every other manufacturer combined.

[ME_Andy]

Yep. Tesla releases a performance version that does 0-60 in 3.3s.

Meanwhile, Nissan releases a performance version that has a sport-tuned suspension, high-performance tires, and a custom-tuned computer that delivers "instant acceleration at all speeds." In other words, nothing under the hood is different.

[JSH]

Originally Posted by roflwaffle

The odd thing about this whole situation is that it's only a fiasco compared to what Elon said, not compared to EVs from other manufacturers. For instance, the Bolt was launched 20-months ago and sold 500+cars/month in it's first month out. It's now selling 1,000-1,400+ cars/month. Tesla starting point was a paltry 30 cars/month, but they reached 6,000+ cars/month in 12 months.

Tesla is only doing well if you are judging by sales volume. If you judge the product launch and quality they are way behind established automakers.

The Bolt launch on schedule and delivered cars that meet current industry standards.

Tesla launched late and delivered essentially pre-production cars with a host of quality and fit issues. The panel fit has gotten better with time (likely as they tweek the tooling to fix issues that they are discovering building customer's cars.)

Time will tell how the long term quality turns out on cars build by hand, in a tent, by recently hired temp workers. It my experience that doesn't work too well. (Every time we add a shift we have quality issues even when the workers are trained and shadow experience workers for 4-6 weeks before during the job on their own)

[RedDevil]

If Tesla was an experienced main stream car maker and if they had designed their EV's using the template and designs they'd be accustomed to, they would have had less problems gearing up for production and with initial build quality.

But then their product would be as attractive as the Bolt. It would not be bad - but it would not have half a million people lining up to buy it either.

I see their problems as both inevitable and temporary.

[JSH]

Originally Posted by RedDevil

If Tesla was an experienced main stream car maker and if they had designed their EV's using the template and designs they'd be accustomed to, they would have had less problems gearing up for production and with initial build quality.

But then their product would be as attractive as the Bolt. It would not be bad - but it would not have half a million people lining up to buy it either.

I see their problems as both inevitable and temporary.

There is nothing new or technically difficult about making the Model 3. It is a very conventional car using a run of the mill steel unibody.

[RedDevil]

Originally Posted by JSH

There is nothing new or technically difficult about making the Model 3. It is a very conventional car using a run of the mill steel unibody.

No. It is steel all right, but there convention stops.

edit:
Just look at the links Roflwaffle provided two posts down... Amazing, shocking!
A symphony of engineering indeed.

[roflwaffle]

Originally Posted by JSH

Tesla is only doing well if you are judging by sales volume. If you judge the product launch and quality they are way behind established automakers.

The Bolt launch on schedule and delivered cars that meet current industry standards.

Tesla launched late and delivered essentially pre-production cars with a host of quality and fit issues. The panel fit has gotten better with time (likely as they tweek the tooling to fix issues that they are discovering building customer's cars.)

Time will tell how the long term quality turns out on cars build by hand, in a tent, by recently hired temp workers. It my experience that doesn't work too well. (Every time we add a shift we have quality issues even when the workers are trained and shadow experience workers for 4-6 weeks before during the job on their own)

Quality is almost always going to be worse because they're doing public betas with their investors, employees, and select owners, but that's why they limit release AFAIK. They're also more likely to deliver cars with quality issues as long as they are functional because it's do or die for production now that they've hit their 200,000 US sale, and they're also using their repair centers to touch up cars instead of doing that at the factory.

In terms of quality, my feeling is that people are focusing on it because certain media outlets are focusing on it. I never thought about or looked at panel gaps and the like until the stories about them were shotgunned all over the news cycle, and when I did look at them it turns out my wife's 2014 Plug-in Prius, which we bought new in 2014, has worse issues with alignment and panel gap than the 3 I picked up in May has, and similar issues with the interior fit and finish. I imagine if you look hard enough, many cars have these issues. The thing is, after 10k+ miles, rock chips, small scratches, and dents are more obvious on most cars in most than the little quality issues that made it out of the factory, and the majority of people don't care or notice.

Production has certainly been an issue for Tesla, but my car's great overall, and I imagine most cars out of the factory will be similar. That's of course assuming a reasonable 10' view. If I start going over any new car like it's a show car I'm going to find a bunch of issues. From what I've pieced together, Tesla probably has experienced people in the tent (GA4) because GA2, which was the first heavily automated line for the 3, was a flop and is in the middle of being reconfigured/redone, so Tesla has enough people to cover the tent until GA2 is up. GA3, which is less automated than GA2, is what's currently responsible for the majority of production (80% IIRC).

Having the tent allowed Tesla to increase overall capacity, specifically capacity that's relatively consistent, like the Model S/X lines, because there's little to no automation. I'm guessing opening up sales of a Performance 3 with base tires/wheels/brakes provided enough in the way of margin to offset the higher labor costs of production and more time spent on QA for the higher margin costs. I think the plan is to approach 10,000 with GA2+GA3 running at 8,000+ cars/week and GA4 at 1,000+/week. Tesla's been on a hiring spree for production associates ever since it became apparent that GA2 wasn't going to work late last year/early this year, so I'm not sure if the limitation for expanding beyond 4,000-5,000 cars/week is getting GA2 up, hiring/training enough people for GA2, or both.

Their product launch was certainly a failure compared to their own estimates, but at the same time I think they're currently delivering cars that meet and/or exceed the majority of industry standards. I think everyone's going to have to spend/work a lot to catch up to Tesla on tech/design. The outstanding question IMO is how and when Tesla can ramp 3 production significantly from where they are now. After driving the car for a couple months, my feeling is that the 3 is years ahead of anything else offered today, and that seems to be a consistent theme I've seen, provided of course someone's paycheck doesn't depend on their opinion.

Tesla Model 3 - NextGen Battery - EVTV Motor Verks
https://www.reddit.com/r/teslamotors...iewer/e2ryr7u/
https://www.inl.gov/article/advanced...y-development/

https://www.youtube.com/watch?v=pAS-yjWj9DY