What Exactly Happened with the Battery Assembly Automation?

[anticitizen13.7]

Didn't Tesla skip making non-production tooling rounds? (As well as beta cars.)

There were such rumors.

Perhaps this is now biting them?

Based on the majority of Model 3 videos and reviews I’ve read, components tooling does not appear to be a problem. With some rare exceptions, delivered Model 3s have much better tolerances and consistency from car to car than Model S or X did in the first year.

Automated pack module assembly is likely to be a different issue: putting parts together is the problem, not the parts themselves.

 

[S3XY]

My guess at the crossover point of "overly confident we knew what we were doing" and "contractor screwed the pooch" narratives is that Tesla spec'd the system they needed and then assumed the contractor was delivering, and spent so much time focusing on all the other stuff for Model 3, assuming that module assembly was "done", that they never closely checked on the contractor.
** CLIP **
Of course it's also possible that had they solved the battery problem sooner by paying attention to it, that 20 other bottlenecks might have popped up to surprise them one after another while trying to ramp production... so no one can really say whether or not this was not the best outcome possible for the allocation of resources.

Or there could have been a lot more Power Inverters to replace

 

[Ulmo]

I believe he walked that plan back. The Board and employees convinced him to keep things simple and bring the Y to market quicker.

It seems like the board is meddling in some clear improvements.

 

[AnxietyRanger]

Didn't Tesla skip making non-production tooling rounds? (As well as beta cars.)

There were such rumors.

Perhaps this is now biting them?

They made them....and then sold them to employees and a handful of prior Tesla owners. One of whom then sold their's for multiples of initial sale price to troll this guy.

Based on the majority of Model 3 videos and reviews I’ve read, components tooling does not appear to be a problem. With some rare exceptions, delivered Model 3s have much better tolerances and consistency from car to car than Model S or X did in the first year.

Automated pack module assembly is likely to be a different issue: putting parts together is the problem, not the parts themselves.

I referred to these:

Tesla Workers Say Skipped Soft-Tooling Phase Ups Danger Inside ‘Production Hell’

Telsa skipping Beta Phase. are you concerned? | Tesla

My thinking is, could these been symptomatic of the reasons that ended up also surprising them at the Gigafactory battery assembly - skipping phases in pre-production results in surprises at production stage that otherwise would have been caught earlier?

 

[30seconds]

I thought the Q3 call was when Tesla first mentioned the issues with battery module automation that was contracted out. From what I remember the contractor was telling Tesla that everything was on track but when the automation was fired up it basically didn't work.

From there Tesla started to do two things simultaneoubeesly - 1. Figure out some interim solution which seems to have become reasonable and is supporting the current production and 2. Have Tesla Grohman come up with a new automated solution.

Based on this call the Grohman solution seems to be almost complete (almost) with a target date to be up and running at the GF in March.

If that goes well then then production should ramp to the next bottleneck. Elon said that they think this next issue will be materials logistics at the Fremont facility.

Anyways that was my takeaway from the last two calls.

 

[Cosmacelf]

30seconds has the best summary.

Many eons ago, I used to work for a software integration company, and programming a robotic assembly process sounds eerily similar to a failed project I was a member of. If you have an inexperienced project manager that hasn’t actually done this type of programming before himself, then this is the kind of disaster you can get. No doubt Elon’s characterization is correct that the system flat out did not work.

By the way, this is unfortunately a pattern with Tesla. Every once in a while, they miss that an outside supplier is screwing up. And every time they get burned, they manage to do it in house, and apparently do it in house well.

I honestly wouldn’t be surprised to hear that they bought a lithium or cobalt mine the way they are going, and started smelting aluminum

 

[jsrawa]

They are gutting 12V on the Model Y, resulting in much less wiring

As others have said, they have abandoned this plan. Elon said that they would move away from 12V about three conference calls ago but two calls ago he stated that the board and other management talked him out of it. While I agree 12V should be abandoned it would just delay the Y further, it’s not worth it at this point. I think modeling the Y off the 3 like the X off the S makes perfect sense.

 

[kengchang]

As others have said, they have abandoned this plan. Elon said that they would move away from 12V about three conference calls ago but two calls ago he stated that the board and other management talked him out of it. While I agree 12V should be abandoned it would just delay the Y further, it’s not worth it at this point. I think modeling the Y off the 3 like the X off the S makes perfect sense.

Tesla Model Y is coming to market sooner using Model 3 architecture, says Elon Musk

Update: Musk said today that they would still aim to change the wiring harness architecture with a carryover from Model 3.

Quote from Earning Call

"No, that's one of the things that we would include. We would aim to switch out the wiring models for (inaudible) the wiring models for a redundant flex circuit that's more in the order of 100 meters or so. But -- and then we'd -- obviously, we'd aim to do that both for the Y, if it's called the Y, and the 3 as well."

 

[ℬête Noire]

By the way, this is unfortunately a pattern with Tesla. Every once in a while, they miss that an outside supplier is screwing up. And every time they get burned, they manage to do it in house, and apparently do it in house well.

"These suppliers are crap for this, let's do this in-house" and doing that successfully is a pattern at Space-X, too. The difference for Space-X is that it seems more cost orientated motivation there, as specialty niche suppliers for the rocket building industry have been operating on some seriously phat margins over the decades.

 

[u00mem9]

Talk Talk Talk. These results calls get more absurd with each new delay. Who cares what Elon says to deflect blame? This is pretty simple: Tesla is currently learning WHY other automakers develop vehicles using their current process.

Why waste time building a 2nd round of prototypes? Just make the designs perfectly the first time!
Why waste time assembling complex machinery and tooling at the makers location first? Just run the first trials after install at the Tesla factory!
Why not take reservations for a product you haven't started designing yet? Just show some bare chassis CAD model of the Model S, claim it's the Model 3, and have people waiting in line!

Don't forget to throw more shade on the old squares running profitable car companies (sorry folks, sustainability has meaning in the financial world too)

 

[dsvick]

Talk Talk Talk. These results calls get more absurd with each new delay. Who cares what Elon says to deflect blame? This is pretty simple: Tesla is currently learning WHY other automakers develop vehicles using their current process.

Why waste time building a 2nd round of prototypes? Just make the designs perfectly the first time!
Why waste time assembling complex machinery and tooling at the makers location first? Just run the first trials after install at the Tesla factory!
Don't forget to throw more shade on the old squares running profitable car companies (sorry folks, sustainability has meaning in the financial world too)

It's called innovation and it rarely works perfectly the first few times. Given more time, data, and experience and you'll see real progress being made. Or, you could just sit back and do things the way they've always been done and get left behind.

 

[u00mem9]

It's called innovation and it rarely works perfectly the first few times. Given more time, data, and experience and you'll see real progress being made. Or, you could just sit back and do things the way they've always been done and get left behind.

That's right. That's why you don't introduce it to the critical path until it works. See cancellation thread for hubris.

 

[ℬête Noire]

See cancellation thread for hubris.

...of overwrought posters and shameless trolls.

 

[Dr. J]

Talk Talk Talk. These results calls get more absurd with each new delay. Who cares what Elon says to deflect blame? This is pretty simple: Tesla is currently learning WHY other automakers develop vehicles using their current process.

Why waste time building a 2nd round of prototypes? Just make the designs perfectly the first time!
Why waste time assembling complex machinery and tooling at the makers location first? Just run the first trials after install at the Tesla factory!
Why not take reservations for a product you haven't started designing yet? Just show some bare chassis CAD model of the Model S, claim it's the Model 3, and have people waiting in line!

Don't forget to throw more shade on the old squares running profitable car companies (sorry folks, sustainability has meaning in the financial world too)

Well, it seems to be working for them so far, considering the market capitalization of publicly traded automotive companies. They may even make a profit someday!

 

[u00mem9]

Well, it seems to be working for them so far

It depends how you define "them".

In my assessment, it breaks down something like this:

WINNING:
Elon -obviously
His cousins -continuously divesting as quickly as they can
The employees of Tesla -great compensation when stock value is included
Current Model S/X owners -paid much less for the car than it cost to produce
Investment Banks -making a killing selling short options to people who haven't caught on that the float is too small for the share price to be market driven

LOSING:
Shareholders -between dilution and the slow reveal that there is no path to breakeven, any retail investors have my sympathy
Model 3 order holders -no reason to think the estimate that was off 1 year isn't off 2 years.
Short sellers -most of the time...and they deserve it

Potential Future LOSERS
Model 3 owners -when development is skipped, the problems that would have been found in development are found by the consumers
Model S/X owners -with 8 year / unlimited mile warranty coverage and a free supercharger network if those things evaporate
Pension Funds run by morons
People who have filled the human weakness for religion with Tesla & Elon Musk

 

[Cosmacelf]

Why are you so worked up about this u00mem99? Even if you are right, why are so stridently telling people about your insights? Do you care THAT MUCH about dumb Tesla owners who don’t know enough not to buy such crappy cars? Do you care about moronic pension funds?

Just seems to me that you looking for validation that you’re so smart, and the rest of us so dumb for not seeing what is so obvious to you.

Anyways, this thread was supposed to be about what is happening with battery assembly automation. Frankly, Elon and JB provided WAY MORE detail about what is going on there than I have ever heard from any company before. It is very detailed, very informative to people who know about software development, system development and manufacturing. There is still some risk that it’ll take a bit longer to get the new automated line(s) working here once they unpack everything from Grohman and set it up, but if you were betting, you’d have to bet that it’ll work sooner rather than later. And, correct me if I’m wrong, but I think they also said that the Grohman line will work better, faster than the original automated line that the system integrator failed at.

 

[ℬête Noire]

Putting u00mem99 on hard-earned Ignore status even removes vision of his quotes out of other's posts. Including my own past ones.

Neat, I think I like that feature.

 

[Dr. J]

It depends how you define "them".

In my assessment, it breaks down something like this:

WINNING:
Elon -obviously
His cousins -continuously divesting as quickly as they can
The employees of Tesla -great compensation when stock value is included
Current Model S/X owners -paid much less for the car than it cost to produce
Investment Banks -making a killing selling short options to people who haven't caught on that the float is too small for the share price to be market driven

LOSING:
Shareholders -between dilution and the slow reveal that there is no path to breakeven, any retail investors have my sympathy
Model 3 order holders -no reason to think the estimate that was off 1 year isn't off 2 years.
Short sellers -most of the time...and they deserve it

Potential Future LOSERS
Model 3 owners -when development is skipped, the problems that would have been found in development are found by the consumers
Model S/X owners -with 8 year / unlimited mile warranty coverage and a free supercharger network if those things evaporate
Pension Funds run by morons
People who have filled the human weakness for religion with Tesla & Elon Musk

By "them," I meant Tesla, Inc. I also think it's working for the vast majority of customers. Maybe that changes in the future, maybe not.

Do you care about moronic pension funds?

^This. It matters not a bit to me whether TSLA works out for anybody. You pay the fee and you takes your chances. My point about market capitalization is that, to date, the stock market does not agree with your [u00mem9's] opinion. That could change, obviously. You can hope/short! (If you can find shares to borrow.)

 

[dsvick]

That's right. That's why you don't introduce it to the critical path until it works. See cancellation thread for hubris.

If you don't test it in real situations you'll never know if it works in real situations.

 

[Carnook]

It's called innovation and it rarely works perfectly the first few times. Given more time, data, and experience and you'll see real progress being made. Or, you could just sit back and do things the way they've always been done and get left behind.

The end result may be innovation, but the delays and rework are just failures. It’s a disturbing trend that we have seen now with the falcon wing doors, the Model X seats, battery modules, Autopilot hardware, etc. Tesla is the common denominator in all these failures not the suppliers. It’s valid to wonder why Tesla keeps failing at getting these innovations to market.