It will be fascinating to see how much this changes after Tesla put their next-gen vehicle into mass production.
They talked the big talk at Investor Day this year, and Elon said it again this past week that "The revolution in manufacturing that will be represented by that car will blow people's minds". [1]
I'm cautiously optimistic. The more we learn about Cybertruck, the more we learn they're seriously moving the yardstick forward.
I work with a big supplier for the conveyors and stuff that moves the vehicle bodies through the plant. The talk you mentioned really impacts them. They are thinking a lot about the new functionalities they need to develop in their products in order to fulfill the Tesla requirements. Not only does Tesla want to change the car body structure itself, they also want to reduce the cycle time per manufacturing step significantly. This causes the conveyor suppliers to develop completely new concepts for moving the car bodies through the factory.
Absolutely. From the Munro interview it sounds like Tesla currently have one Model Y coming off the line every 42 seconds, and they're aiming to halve that.
From what was said, the second best in the industry right now is about one vehicle every 60 seconds.
Elon has said many times that just like Starship and Raptor, the product itself is not nearly as revolutionary (or difficult) as the production system to produce it at scale and speed.
not to mention it wont get approval for driving on many roads around the world (so far) like Australia. due to the danger it poses to whatever it hits.
I have seen some Indian panelbeaters do miracles on a wrecked car - cutting away a not only a rear quarter panel but half of a floorplan (FWD) but welding a replacement in sutu and then using a chassis jig to pull to straight again.
Don't think it will work on your CyberTruck - Tesla repairs are astronomical in normal fender benders.
I was surprised to learn that the production of stamping tools, used to shape panels from sheet stock, involves lots of manual passes to bring them to the quality and finish required by car manufacturers.
After machine milling them roughly to spec there are employees that angle grind, sandpaper and polish every critical surface of these 10 ton chunks of metal.
The one that really surprised me for some reason was PCB manufacturing. The Strange Parts YouTube channel has a few tours of Chinese factories, and I guess I pictured that it's just some sort of giant clean room with a long conveyor belt that goes through all the processes on its own. So I was not expecting it to start with grabbing a blank PCB from a pile stored outside and taking a band saw to it. Such a mix of high-tech automation and human hands picking things up, racking them on the table, and sticking them into the next machine.
The board level electronics assembly is mostly an solved problem that can be automated perfectly fine, but when you get to through-hole almost nobody does it, because (a) the THT capable pick and place machines are uncummon and thus somewhat expensive (b) it often requires more expensive parts that are compatible with such machines (better geometry tolerances, surfaces for the machine to pick them...) and finally (c) the NRE costs of getting that to work.
That said, then there are niches where the automated THT assembly apparently makes sense and all the NRE costs seem to be offset by lower per-unit costs of the PCB material itself. Prime example is mid-range to high-end home AV equipment (Japanese branded if not made in Japan outright), where the large mostly empty single-sided FR-2 boards with truly ridiculous amounts of obviously machine populated wire-links and discrete THT radial components are particularly striking.
Elon was surprised too. The model 3 was supposed to have a fully automated line. He couldn't make it happen. Industrial robots are bad at wiggling stuff into place, and hopeless at handling error cases. https://qz.com/1261214/how-exactly-tesla-shot-itself-in-the-...
Elon is the biggest day dreamer who imagines everything can be solved with technology around the corner in 6 months, but has a hard time acknowledging when the harsh realities hit him in the face, that real world engineering challenges aren't as simple as those in building successful websites.
He's gotten further with subsequent products. A lot of the learnings with model 3 went into the model y and later iterations of the model 3. And of course cybertruck.
Industrial robot usage and other manufacturing innovations are a big reason they are leading the industry in manufacturing cost and speed. Other manufacturers are starting to emulate a lot of the things they are doing with structural battery manufacturing, casting big parts, etc. The likes of VW and others are (by their own admission) behind and struggling with the number of parts, amount of manual work, etc. It might not be as automated as Elon Musk would like but it is pretty impressive.
It will be interesting to see what they do with their upcoming cheaper model. He seems to be hinting at the notion that further manufacturing improvements are going to be a big part of that.
People think automation solves problems - it's the opposite. In order to use automation, you have to solve every problem that the automation could encounter.
It's not clearly shown in this, but the majority of the work is manually adding parts.
Cars are made in smaller volumes than most people imagine.
Compare to toilet paper production... Your family perhaps consumes 100 rolls of toilet paper per year, and perhaps 0.1 cars per year (typical lifetime of a car is ~20 years).
So there is 1000x the benefit to any time saving by automation on the toilet paper production line than on the car production line.
Speaking of inspection, I was actually wondering what they do with rejected units. It would seem hard to fix problems once the car is already assembled, so do they just trash them?
This video does a very poor job of explaining the quality process. Inspection happens at multiple points throughout assembly, and most problems are caught before end of line testing. There are metrics on this and it's common to shoot for 98%+ of defects to be caught in process.
There are only a few problems that would lead to a scrap job - for instance if the body of the car is damaged to the point that they would have to cut and weld new panels, that vehicle would be scrapped.
But even for scrapped vehicles, all of the components are removed to re-build that vehicle.
Basically anything besides serious body damage can be repaired. But you are correct that it is hard to fix problems. Most parts are added in one minute or less - including the engine assembly! But it takes much, much longer than one minute to remove an engine.
I went in with an expectation of it being completely different. And there is much more manual labor in 1939. It's dirtier, and grittier, but mostly for the steps we don't see in the Toyota video, as for example the motor comes in one block, and we don't see how it's built.
But on a high level, both videos look like the same process. You can find many steps, like the stamping, the welding, the final assembly, that look nearly unchanged in 85 years, save for some automation with robots, and some quality-of-life improvements in the assembly, where people don't have to lug around heavy parts.
It's much more relaxed, with little commentary - and really shows every step. Having done some very rudimentary work around manufacturing and testing, it's pretty incredible to see how much work - jigs, robot programming, test software etc - goes into the manufacturing machine.
@5:27 Note the strange welded-aluminum frame the worker sets into the battery box. That's because the i3's battery is cooled by the AC system!
What a tragic car. Discontinued in 2021. The material costs of carbon fiber are dirt cheap, so in theory it ought to beat aluminum for car structure. But once you get away from simple tubes, which can be fully automated, layup ends up being most of your costs. (Not helped by the fact that i3 carbon parts were made on the opposite side the of the planet and shipped to Germany)
BMW's business case for the car had to be "once we make enough of them, costs will come down", but that never happened.
Carbon fibre is cheap-ish tp buy. Carbon fibre parts are the opposite of cheap so, especially load bearing and safety critical parts with complex geometries, those are quite expensive.
The i3 was an interesting car so, and BMW for some reason fumbled the head start the car got them.
Interesting that they attach doors, then remove them, then put them back on again. It raises the question, why did they bother to attach them the first time around?
My guess is that the answer is the paint process. There's some range of allowable difference in paint hue going from one car to the next, even if they're marketed as having the same color. That range is much smaller for a given single vehicle. If the doors are even a slightly different hue, you're likely to notice.
They paint it all together to have exactly the same color (as you well said) and avoid two painting jobs in different parts of the process. There are fewer moving parts this way, following the lean manufacturing principles. Then it is temporarily removed to create more space for the workers (as the video shows).
And the doors move in a parallel line for assembly of the door. Making it easy to maintain a one peace flow and not worry about the various options for each car on the line. And, as already pointed out, a homogenous paint job.
One thing to note is that usually most of the door itself is usually manufactured by some external supplier and the parallel assembly line in the final assembly only mates a subassembly that contains all the complexity of the door (interior paneling, speakers, buttons, DCU, airbags, locking mechanism...) to the outer door body panel.
The space of the options for that "door module" is large enough that it usually gets ordered from the manufacturer some time after the car body gets assigned to particular order (near the start of the painting step) and these modules have to reach the final assembly line in particular time and in particular order. All is well when the whole manufacturing and logistics chain works in the same order that the car orders entered the production. But in practice there is a lot of things that can cause the ordering to become desynchronized, which leads to interesting logistical challenges (especially because if there is a need to do something out of order the order discontinuity is essentially never neatly aligned to size of the containers that the subassemblies are stored and transported in).
The same thing also applies to other subassemblies that have large option spaces, but probably the only other thing with somewhat huge option space that is not manufactured directly by the automaker are completed dashboard modules.
Also, in the traditional approach the rolling chasis and the body tend to be manufactured by separate assembly lines and then "married toggether" pretty late in the whole process. But this is usually done in the same plant, so there is less of an possibility of things getting desynchronized.
Interested in an informed comment on the paint application part. How do they keep those spray rooms clean from paint? I noticed the robots were covered up to protect against the spray but the floor was pristine.
The floor is open grating above a pool of water that is pumped to reclaim the solids. Further the paint is electrostatically charged as it comes off the bell applicator so it is attracted to the body like a balloon to the wall.
A few nitpicks - I know this video is just an overview, so these don't matter all that much, but:
- The paint doesn't "dry" per se, it is baked on. Usually at least twice - once for primer, once for mid, top and clear coat. Sometimes more than twice - 1 prime, 2 mid, 3 top and clear.
- Clear Coat is incredibly important for surface finish durability, it's not just for appearance
- Quality is baked in at the process level, not just inspected at the end of the process.
- Everything is much more complicated than portrayed, and there's way more going on, but it is only a 5 minute video after all.
44 comments
[ 5.9 ms ] story [ 99.5 ms ] threadThey talked the big talk at Investor Day this year, and Elon said it again this past week that "The revolution in manufacturing that will be represented by that car will blow people's minds". [1]
I'm cautiously optimistic. The more we learn about Cybertruck, the more we learn they're seriously moving the yardstick forward.
[1] https://fortune.com/2023/12/06/elon-musk-tesla-sandy-munro-c...
From what was said, the second best in the industry right now is about one vehicle every 60 seconds.
Elon has said many times that just like Starship and Raptor, the product itself is not nearly as revolutionary (or difficult) as the production system to produce it at scale and speed.
Don't think it will work on your CyberTruck - Tesla repairs are astronomical in normal fender benders.
After machine milling them roughly to spec there are employees that angle grind, sandpaper and polish every critical surface of these 10 ton chunks of metal.
That said, then there are niches where the automated THT assembly apparently makes sense and all the NRE costs seem to be offset by lower per-unit costs of the PCB material itself. Prime example is mid-range to high-end home AV equipment (Japanese branded if not made in Japan outright), where the large mostly empty single-sided FR-2 boards with truly ridiculous amounts of obviously machine populated wire-links and discrete THT radial components are particularly striking.
Industrial robot usage and other manufacturing innovations are a big reason they are leading the industry in manufacturing cost and speed. Other manufacturers are starting to emulate a lot of the things they are doing with structural battery manufacturing, casting big parts, etc. The likes of VW and others are (by their own admission) behind and struggling with the number of parts, amount of manual work, etc. It might not be as automated as Elon Musk would like but it is pretty impressive.
It will be interesting to see what they do with their upcoming cheaper model. He seems to be hinting at the notion that further manufacturing improvements are going to be a big part of that.
It's not clearly shown in this, but the majority of the work is manually adding parts.
Interesting that they don't show engine manufacturing at all. That's been automated since the early 1950s.
Compare to toilet paper production... Your family perhaps consumes 100 rolls of toilet paper per year, and perhaps 0.1 cars per year (typical lifetime of a car is ~20 years).
So there is 1000x the benefit to any time saving by automation on the toilet paper production line than on the car production line.
There are only a few problems that would lead to a scrap job - for instance if the body of the car is damaged to the point that they would have to cut and weld new panels, that vehicle would be scrapped.
But even for scrapped vehicles, all of the components are removed to re-build that vehicle.
Basically anything besides serious body damage can be repaired. But you are correct that it is hard to fix problems. Most parts are added in one minute or less - including the engine assembly! But it takes much, much longer than one minute to remove an engine.
Those workers went on strike a few years after that, in 1939, and won. That ended the "Chrysler Speed Up Initiative".
[1] https://archive.org/details/masterhandscomplete4kh264
But on a high level, both videos look like the same process. You can find many steps, like the stamping, the welding, the final assembly, that look nearly unchanged in 85 years, save for some automation with robots, and some quality-of-life improvements in the assembly, where people don't have to lug around heavy parts.
It's much more relaxed, with little commentary - and really shows every step. Having done some very rudimentary work around manufacturing and testing, it's pretty incredible to see how much work - jigs, robot programming, test software etc - goes into the manufacturing machine.
What a tragic car. Discontinued in 2021. The material costs of carbon fiber are dirt cheap, so in theory it ought to beat aluminum for car structure. But once you get away from simple tubes, which can be fully automated, layup ends up being most of your costs. (Not helped by the fact that i3 carbon parts were made on the opposite side the of the planet and shipped to Germany)
BMW's business case for the car had to be "once we make enough of them, costs will come down", but that never happened.
The i3 was an interesting car so, and BMW for some reason fumbled the head start the car got them.
My guess is that the answer is the paint process. There's some range of allowable difference in paint hue going from one car to the next, even if they're marketed as having the same color. That range is much smaller for a given single vehicle. If the doors are even a slightly different hue, you're likely to notice.
The space of the options for that "door module" is large enough that it usually gets ordered from the manufacturer some time after the car body gets assigned to particular order (near the start of the painting step) and these modules have to reach the final assembly line in particular time and in particular order. All is well when the whole manufacturing and logistics chain works in the same order that the car orders entered the production. But in practice there is a lot of things that can cause the ordering to become desynchronized, which leads to interesting logistical challenges (especially because if there is a need to do something out of order the order discontinuity is essentially never neatly aligned to size of the containers that the subassemblies are stored and transported in).
The same thing also applies to other subassemblies that have large option spaces, but probably the only other thing with somewhat huge option space that is not manufactured directly by the automaker are completed dashboard modules.
Also, in the traditional approach the rolling chasis and the body tend to be manufactured by separate assembly lines and then "married toggether" pretty late in the whole process. But this is usually done in the same plant, so there is less of an possibility of things getting desynchronized.