What's new
What's new

Tesla's new casting operation to replace stamping/welding/various other methods

nyc123

Hot Rolled
Joined
Aug 31, 2018
Appears to replace what was once many parts and processes.

First they tried heavily automating more of the normal production process and apparently ran into the same problems as GM in the 80's or many others, but now the goal this time is to use this aluminum casting machine by an Italian company to make one single part out of what would typically be many.

One obvious issue could be what happens if you later (after consumer use) need to replace one or more parts of the assembly?

Thoughts?

Original story where I saw it:Move aside robots, Tesla bets on aluminium casting | Reuters

More detailed article: Tesla Ditching Industrial Robots for Huge Die-Casting "Giga Press" - Core77
 
I suppose it makes it more difficult for a shop that wants to make a stretched Model Y limo or something like that. But when was the last time you could buy a car frame part? Well, maybe a more precise way of asking that question is "What was the latest model year car for which you could buy a frame part?" as Model T frame parts are probably available aftermarket.

Basically, if that part is broken, the car is totaled. So it would be insane to put anything that's subject to fender bender damage in that casting. I have to believe Tesla's engineers have that much experience.

The well-developed art of frame straightening doesn't work so well on castings (of any sort) and doesn't work as well on aluminum as steel. So I don't expect body shops to do a big business in straightening out Model Y Teslas.
 
I suppose it makes it more difficult for a shop that wants to make a stretched Model Y limo or something like that. But when was the last time you could buy a car frame part? Well, maybe a more precise way of asking that question is "What was the latest model year car for which you could buy a frame part?" as Model T frame parts are probably available aftermarket.

Basically, if that part is broken, the car is totaled. So it would be insane to put anything that's subject to fender bender damage in that casting. I have to believe Tesla's engineers have that much experience.

The well-developed art of frame straightening doesn't work so well on castings (of any sort) and doesn't work as well on aluminum as steel. So I don't expect body shops to do a big business in straightening out Model Y Teslas.

This appears to be just a cradle assembly. The location is in an area that normally is not touched during a frame stretch, not that I can imagine someone needing to do that with a Model Y. Most of the cars manufactured today use such an approach on frame sub assemblies and yes they are supported as a repair part.

This is a smart move on Tesla's part as they are moving a major assembly manufacturing from the frame line to a separate manufacturing area. Great way to reduce assembly line issues with time and space constraints and ultimately lower overall manufacturing costs even though the casting itself might cost more as a single component. Integrated into the vehicle the overall cost is reduced.

I have had serious doubts about Tesla's viability upto this point but I would put this in the category of them actually starting to get a handle on actually building a car.

Automation is wonderful but not the panacea for all manufacturing problems. It is only one tool in the tool box. Moving beyond the hammer as a tool and everything is a nail is a real sign of manufacturing process engineering maturation.
 
This is interesting. However it's not the first time its been tried. In the late 1940's early 1950's Kaiser-Frazer had Dohler-Jarvis die casting in Toledo, Ohio build and run some parts for the (I believe) the left front door for a car in production at the time. It never went past sample parts because of the expense of building new dies every time there was a model change.

Tesla by going with a part that's not subject to (yearly) model changes and if it get damage in an accident, the car will no doubt be totaled. Tesla will no doubt be more successful.
 
Wow...just like the company that produced whole travel trailers by using 2 vacuum formed plastic sheets, glued at a seam, and foam filled.

....All around 1968...:willy_nilly:
 
I wonder if they considered using a method like the heavy stamper stuff used for aircraft bulkheads.

I'll try to find the article I saw about these huge coining/forging machines that had to be restored and started to be used again in relation to the Raptor project.

Heavy Press Program – Monster Machines for Forging Light Metal Components

Essentially, they machined an aluminum "blank" and placed it in a forging die. reducing and shaping at the same time. The result was a very strong and very light airframe part that held the two wings together.

Much more efficient than casting.
 
I wonder if they considered using a method like the heavy stamper stuff used for aircraft bulkheads.

I'll try to find the article I saw about these huge coining/forging machines that had to be restored and started to be used again in relation to the Raptor project.

Heavy Press Program – Monster Machines for Forging Light Metal Components

Essentially, they machined an aluminum "blank" and placed it in a forging die. reducing and shaping at the same time. The result was a very strong and very light airframe part that held the two wings together.

Much more efficient than casting.

this is actually forging, and it works pretty well- especially for things like titanium. But, unfortunately, the number of these presses on the entire planet could probably be counted on fingers and toes, and most dont even have "lead times' - they have slots that open up when a company like Boeing goes bankrupt and ceases production, and not before. These presses in the USA are 60 years old or so. To build a new one is really expensive, and takes years. China recently began using an 80,000 ton press, which cost over 200 million dollars and took something like 5 years to build.
So, sure, Tesla COULD use a press like this- but the costs would be incredible.

Every US military plane has forged parts that come off of one of these presses- titanium, stainless, and steel bulkheads and components- but that is for planes that cost tens of millions of dollars each.

its a bit overkill for a part that can be cast just fine.
 
I read a overview of a teardown of a tesla by a auto manufacturing engineer. He claimed the tesla used way to many parts. The one I remember was the rear wheel well. A car company might form it out of 3 parts welded together to form one piece to be installed and welded into place. tesla used 9 pieces because of poor design and probaly lack of big enough presses to form larger parts in one shot.
Bill D.

An expert dismantled a Tesla Model 3. He found poor design and manufacturing are squandering profits - Los Angeles Times
 
I read a overview of a teardown of a tesla by a auto manufacturing engineer. He claimed the tesla used way to many parts. The one I remember was the rear wheel well. A car company might form it out of 3 parts welded together to form one piece to be installed and welded into place. tesla used 9 pieces because of poor design and probaly lack of big enough presses to form larger parts in one shot.
Bill D.

An expert dismantled a Tesla Model 3. He found poor design and manufacturing are squandering profits - Los Angeles Times

It's stuff like this that furthers my skepticism of a company like Tesla. The idea is that this guy, and the company, is supposed to be doing all these new things that the big manufacturers that have been doing this for many decades have failed to do. I find that idea somewhat hard to believe. I think maybe there are some things that they are doing that are new and cool, but I'm generally skeptical that they are nearly as advanced as the media purports them to be.

From your article:

“If that car was made anywhere else, and Elon wasn’t part of the manufacturing process, they would make a lot of money,” Munro said in an interview. “They’re just learning all the old mistakes everyone else made years ago.”


Human behavior tends to forget the past quickly. How many times have there been people held up as being the next new thing and it never happened? Much more than turns out to be the reality. Theranos is one recent example I remember, and now it's all crickets about that. No "let's be more careful" - nope. Just move on to the next one and hold them up and do it over again.

This is just one of many reasons why there are the "Tesla bears" that short its stock. At this point, I don't give it any more credence than the many others during the dot-com bubble and such that were supposed to change this or that. When it happens, I'll believe it. I remember reading a while back that they lose money on every car they sell. Sounds a lot like all those companies that went bust before.

When I think of advanced manufacturing, I think of something like Toyota, not Tesla, although admittedly I'm not drawing on tons of expertise or detailed knowledge on car manufacturing to say that. I just know that Toyota has tried, and is trying, new things all the time. From reducing the plant size, to even removing automation and replacing it with workers (how's that for moving forward?), I would put my money on them instead of Tesla.

Sometimes I even wonder if Elon Musk is just a crazy guy who's managed to get a ton of money from investors to play with...

Here's an article from just a couple years ago that, in retrospect, seems to make this point. You can see where Toyota is moving in one direction with automation and Tesla is proclaiming that the future is moving in the other. Fast forward a couple years later, and Tesla has come out and said this was a poor decision.

Robots won't replace humans in auto factories anytime soon
 
I read that one of Henry Fords inovations was the way he specified shipping boxes and crates back in the day. he required the crates to be made a certain way with certain size boards. None of this ship them in barrels or on pallets. He used the crate wood as dashboards etc. Not something you would know from looking at the finished car. he also used the scrap wood to make charcoal briquets. As in Kingsford brand charcoal.
Dupont used to be famous for not pouring toxic waste away and polluting the environment. They just saved it and used it to make other chemicals they could actually sell. they might chose a less efficient more expensive process to make a product if the waste was useful enough to offset the extra cost.
Gold miners in Nevada used to compalin about the nasty black gunk that hindered mining. They had to spend lots of money to haul this waste off into giant piles to get rid of it so they could get to the gold. This went on until someone checked the black gunk for other then gold content and realized it was very rich in silver.
Bil lD
Bil lD
 
Last edited:
I read a overview of a teardown of a tesla by a auto manufacturing engineer. He claimed the tesla used way to many parts. The one I remember was the rear wheel well. A car company might form it out of 3 parts welded together to form one piece to be installed and welded into place. tesla used 9 pieces because of poor design and probaly lack of big enough presses to form larger parts in one shot.
Bill D.

An expert dismantled a Tesla Model 3. He found poor design and manufacturing are squandering profits - Los Angeles Times

The manufacturing engineer in question was likely Sandy Munro... He has become kind of famous for his teardowns of brand new Teslas, but my understanding was that he was very well known in the auto industry before he became well known on the internet for his Tesla specific teardowns.

He was very critical of Tesla's early body work and overall quality control in terms of body work, particularly the very inconsistent panel gaps and poor design of assemblies with too many parts. But very complimentary in terms of their electronics and the general design of the 'skateboard' which housed the motor, battery pack and control electronics.

He is quoted in the first link that OP posted:

“We were very critical of Tesla when we first started on their vehicles. The gaps were horrific, the weld spatter was everywhere. Nothing fit,” Munro said during a recent presentation with consultants Frost & Sullivan. But the switch from the Model 3 to a larger vehicle, the Model Y, resulted in a step change in manufacturing improvements, he said.

“This is the biggest casting we have seen in a car company. This is just spectacular,”
- Sandy Munro

source: Move aside robots, Tesla bets on aluminium casting | Reuters
 
I remember in the 1980's? mercedes used forged aluminum wheels with lots of small spokes, turbine wheels. Konockoffs were being sold made by casting aluminum. Looked the same for much less money. They would break apart at speed since cast is no where near as strong as forged.
Bil lD
 
I read a overview of a teardown of a tesla by a auto manufacturing engineer. He claimed the tesla used way to many parts. The one I remember was the rear wheel well. A car company might form it out of 3 parts welded together to form one piece to be installed and welded into place. tesla used 9 pieces because of poor design and probaly lack of big enough presses to form larger parts in one shot.
Bill D.

An expert dismantled a Tesla Model 3. He found poor design and manufacturing are squandering profits - Los Angeles Times

This is a direct result of the culture at Tesla, and SpaceX

They like their engineers to be in their 20-30's. What they are adverse to is hiring the people with 20-30-40 years design experience in the auto industry. If they'd hired any one with body/chassis design experience from detroit this wouldn't have happened.

They could have hired a consultant to oversee the design process to make sure their going down the right path. They won't (same at SpaceX)

Same goes with manufacturing.

And this new casting is going to bite them in ****, it probably isn't going to take much of a prang to total a Tesla, so it will be a nightmare for Tesla, their repair facilities, insurance companies and owners.
 
This is a direct result of the culture at Tesla, and SpaceX

They like their engineers to be in their 20-30's. What they are adverse to is hiring the people with 20-30-40 years design experience in the auto industry. If they'd hired any one with body/chassis design experience from detroit this wouldn't have happened.

They could have hired a consultant to oversee the design process to make sure their going down the right path. They won't (same at SpaceX)

Same goes with manufacturing.

And this new casting is going to bite them in ****, it probably isn't going to take much of a prang to total a Tesla, so it will be a nightmare for Tesla, their repair facilities, insurance companies and owners.

I don't disagree with the idea that they've had some problems because they didn't take on people with industry experience... However if they'd started an electric car company filled with people from traditional auto industries they all would have said "electric cars aren't sexy", and likely made boring ugly electric cars like other auto manufacturers have until fairly recently. Or weird looking sci-fi stuff that no-one wants like the BMW i3.

I feel similarly about SpaceX. They brought in a bunch of people with 'new' ideas about how to tackle getting to orbit. These people were likely (in some cases) less experienced, but they also had a fresh viewpoint, rather than wanting to walk the same road as everyone else.

Both companies have definitely faced their share of challenges, but they seem to be tackling them in new and interesting ways. The startup mentality is also really evident, try new stuff and iterate fast rather than taking an ultra-safe route and moving slowly.

Hard to deny that both companies have had *large* impacts in their respective fields.

Assuming that each mission will take four astronauts on board, the average cost per seat will be approximately $90 million on Boeing’s Starliner and $55 million on SpaceX’s Dragon 2.
source - Why NASA Paid Boeing $2 Billion More Than SpaceX for Same ISS Mission | Observer
 
I don't disagree with the idea that they've had some problems because they didn't take on people with industry experience... However if they'd started an electric car company filled with people from traditional auto industries they all would have said "electric cars aren't sexy", and likely made boring ugly electric cars like other auto manufacturers have until fairly recently. Or weird looking sci-fi stuff that no-one wants like the BMW i3.

I feel similarly about SpaceX. They brought in a bunch of people with 'new' ideas about how to tackle getting to orbit. These people were likely (in some cases) less experienced, but they also had a fresh viewpoint, rather than wanting to walk the same road as everyone else.

Both companies have definitely faced their share of challenges, but they seem to be tackling them in new and interesting ways. The startup mentality is also really evident, try new stuff and iterate fast rather than taking an ultra-safe route and moving slowly.

Hard to deny that both companies have had *large* impacts in their respective fields.

Assuming that each mission will take four astronauts on board, the average cost per seat will be approximately $90 million on Boeing’s Starliner and $55 million on SpaceX’s Dragon 2.
source - Why NASA Paid Boeing $2 Billion More Than SpaceX for Same ISS Mission | Observer

They didn't bring in people with new ideas. They bought in people with no ideas straight out of college. The head of experimantal aerodynamics for the Falco Heavy was 3 years out of college. How could his experince ever stack up against somebody who had been building rockets for years. I did a lot Delta 4 work for Boeing, they used to ask me how Boeing did it on Delta 4. They didn't have the experience to know any better. I was embarrased for them

The guy who designed the rear wheel well had never designed an automotive part before. Better engineers and managers would have caught that f'up at the design stage, and better manufacturing engineers would have rejected the design when it was presented to them. Why didn't they? They didn't know any better.
 
They didn't bring in people with new ideas. They bought in people with no ideas straight out of college. The head of experimantal aerodynamics for the Falco Heavy was 3 years out of college. How could his experince ever stack up against somebody who had been building rockets for years. I did a lot Delta 4 work for Boeing, they used to ask me how Boeing did it on Delta 4. They didn't have the experience to know any better. I was embarrased for them

The guy who designed the rear wheel well had never designed an automotive part before. Better engineers and managers would have caught that f'up at the design stage, and better manufacturing engineers would have rejected the design when it was presented to them. Why didn't they? They didn't know any better.

I do get where you're coming from. Little hard to argue with results in the case of SpaceX though... They are pushing further and faster than anyone else. Could they have gone even faster with more experience help on board? Maybe. Maybe they would have been slower as they might have been battling pressure against new ideas from more seasoned engineers who 'already know how to do this'.

I'm a pretty well seasoned software developer, and time has taught me hard lessons about sticking with technologies that are well proven generally being the best approach... I've fought that battle before against newer folks or people who had outlandish ideas about how to achieve a goal. Sometimes I was right, but sometimes they were right and it took me a while to come around despite their idea being clearly better in the end.

My current belief after all this is that there is always at least some value in naivety. People who "don't know any better" will suggest out of the box solutions that sometimes end up being better. Having a team full of grizzled veterans is always going to produce more conservative and mainstream results from what I've seen.
 
I read that one of Henry Fords inovations was the way he specified shipping boxes and crates back in the day. he required the crates to be made a certain way with certain size boards. None of this ship them in barrels or on pallets. He used the crate wood as dashboards etc. Not something you would know from looking at the finished car. he also used the scrap wood to make charcoal briquets. As in Kingsford brand charcoal.
Dupont used to be famous for not pouring toxic waste away and polluting the environment. They just saved it and used it to make other chemicals they could actually sell. they might chose a less efficient more expensive process to make a product if the waste was useful enough to offset the extra cost.
Gold miners in Nevada used to compalin about the nasty black gunk that hindered mining. They had to spend lots of money to haul this waste off into giant piles to get rid of it so they could get to the gold. This went on until someone checked the black gunk for other then gold content and realized it was very rich in silver.
Bil lD
Bil lD

Floorboards on the model T

Grandfather worked for Ford until the switchover from T to A, told me that when I was a kid
 
If you are making 100k of something, you make it very differently than if you are making single thousands.

Think of how fast you can turn aluminum to chip in hundreds and thousands. Hits a certain point and you make a die and cast it.

There are many things that Tesla has done that will change now that the volumes have increased. If they had done it to begin with it would have been more expensive at those volumes

I am sure that industry guy is correct, but has he ever overseen the manufacture of 10000 of something[rather than 1/2 a million]
 








 
Back
Top