4 ms·

Tesla's greatest innovation is simply leadership. Everyone knew castings were the way to go, but noone did it. Now people like Volvo are doing them, and I expec

by _ea1k 3y ago

Tesla's greatest innovation is simply leadership. Everyone knew castings were the way to go, but noone did it. Now people like Volvo are doing them, and I expect a lot of others to do it as well. Everyone knew that 48V would be a big benefit, but noone did it. TBH, in a few years, we may be saying similar things about drive-by-wire without a mechanical backup. But it is too early to say definitively.

p_l 3y ago
What everyone was waiting for was for Italians to make the huge casting press so that castings could be done in mass production at automaker margins.
- _ea1k 3y ago
  Despite Tesla's boastings, large casting machines have existed before. Automotive use is the new part.
  - p_l 3y ago
    Essentially it was question casting of this size being automated at car maker scale. Bigger castings were integrated before on assembly lines, but at higher costs and complexity before Italians made the large mostly automated caster that Tesla called GigaPress (iirc) - it's essentially COTS factory equipment.
  - panick21_ 3y ago
    What's actually new and interesting here is the combination of size and speed. Sure casting existed, but for it to be the backbone of a super fast production line like that of a Model Y, they needed to do a lot of work. Larger parts were often dipped into liquid, because air cooling resulted the part bending. Tesla internal materials team worked on new alloys to improve this, while working in coordination with the machine manufacture to get up the cycle times. And they are continuing to work with them. Tesla is using newer more advanced cooling systems in Texas then they did in the California factory. Because cooling is one of the limiting factors in Cycling these machines.
- panick21_ 3y ago
  Except it was Tesla that went to the Italian casting press manufacturer and the made it a project together. Tesla asked for the biggest they had and asked for bigger and basically bought 5 years of machines, allowing them to do the necessary investment. And Tesla also used its own internal materials team to develop alloys to optimize the process to make it practical for high volume production lines like Model Y.
CharlesW 3y ago
> Everyone knew castings were the way to go… The jury's still out on whether gigacasting is courageous, or just moves costs from production (which will not be passed to consumers) to repairs (which will). > Everyone knew that 48V would be a big benefit, but noone did it. Audi did in 2016 with the SQ7 TDI, no?
- bagels 3y ago
  Car A costs $5k less to manufacture than car B because it uses castings. Car A would be wiser to build because you can get better margins, or lower prices. It'll take a lot longer for buyers to catch on to the fact that the car costs a lot more to repair.
  - biaachmonkie 3y ago
    The question is really if the castings result in more cars being written off as total loss. Any collision severe enough to damage them would likely be severe enough to also total out a car without them as well.
- _ea1k 3y ago
  TBH, the repairability concerns seem to be overstated so far. There are still conventional crash structures in front of and behind them. It takes a lot to damage the casting beyond repair. You are right, though. The jury is still out in that they haven't really been used in the real world for long. > Audi did in 2016 with the SQ7 TDI, no? I didn't know about that one, but it was still a dual architecture. Going full 48V is a bit different, assuming Tesla really did that.
- hinkley 3y ago
  Clearly no MGB fans at Tesla. The unibody is cool unless you live somewhere with snow, and then rust doesn’t just make your car ugly, it makes it unsafe. And expensive to fix. Which is why I was always more of a Triumph fan.
Turskarama 3y ago
The mechanical backup for a drive-by-wire system is different to all those other things in that it's a critical safety issue. If _anything at all_ happens to the electronics while driving then it's important that you can stop the vehicle safely. That mandates manual brakes and steering, even if they're barely effective at all without power it's a lot better than nothing.
- dreamcompiler 3y ago
  The cybertruck has two separate steering motors for redundancy. Passenger jets have been doing "x-by-wire" for decades. Properly engineered it's as reliable as mechanical linkage.
  - Turskarama 3y ago
    Two motors is great and all, but that only provides motor redundancy and won't help if something happens to the electrical system. Planes are fly by wire but that's because they literally have to be, no amount of gym time will let you move the rudder on a 737 by hand.
    - networkchad 3y ago
      [dead]
    - masklinn 3y ago
      The 737 is not fly by wire.
      - Turskarama 3y ago
        Ok fair enough, it is hydraulic assisted instead but for the point I'm trying to make it's essentially the same thing in so far as it requires power.
- _ea1k 3y ago
  I tend to agree, TBH. People bring up fly-by-wire, but even that is simpler. AFAIK, those systems aren't variable ratio the way CT is. CT literally will have different steering ratios depending on the speed, so one steering wheel angle will have different effects on the actual steering angle depending upon the speed. How will that work in the real world? Could there be bugs? Will people get used to it or will it lead to mistakes? I'm looking forward to getting a chance to try it, TBH.
phkahler 3y ago
>> Tesla's greatest innovation is simply leadership. Totally correct. We were making 48v power steering systems 20 years ago for OEM development projects. They simply don't have the will/commitment/leadership to bring change like that to market. Elon sets a direction and people go that way. I can imagine someone at GM listing all the obstacles as a reason they can't change, and Elon looking at them and saying "no, you're the obstacle" or something like that.

by _ea1k 3y ago

p_l 3y ago

What everyone was waiting for was for Italians to make the huge casting press so that castings could be done in mass production at automaker margins.

_ea1k 3y ago
Despite Tesla's boastings, large casting machines have existed before. Automotive use is the new part.
- p_l 3y ago
  Essentially it was question casting of this size being automated at car maker scale. Bigger castings were integrated before on assembly lines, but at higher costs and complexity before Italians made the large mostly automated caster that Tesla called GigaPress (iirc) - it's essentially COTS factory equipment.
- panick21_ 3y ago
  What's actually new and interesting here is the combination of size and speed. Sure casting existed, but for it to be the backbone of a super fast production line like that of a Model Y, they needed to do a lot of work. Larger parts were often dipped into liquid, because air cooling resulted the part bending. Tesla internal materials team worked on new alloys to improve this, while working in coordination with the machine manufacture to get up the cycle times. And they are continuing to work with them. Tesla is using newer more advanced cooling systems in Texas then they did in the California factory. Because cooling is one of the limiting factors in Cycling these machines.
panick21_ 3y ago
Except it was Tesla that went to the Italian casting press manufacturer and the made it a project together. Tesla asked for the biggest they had and asked for bigger and basically bought 5 years of machines, allowing them to do the necessary investment. And Tesla also used its own internal materials team to develop alloys to optimize the process to make it practical for high volume production lines like Model Y.

CharlesW 3y ago

> Everyone knew castings were the way to go… The jury's still out on whether gigacasting is courageous, or just moves costs from production (which will not be passed to consumers) to repairs (which will). > Everyone knew that 48V would be a big benefit, but noone did it. Audi did in 2016 with the SQ7 TDI, no?

bagels 3y ago
Car A costs $5k less to manufacture than car B because it uses castings. Car A would be wiser to build because you can get better margins, or lower prices. It'll take a lot longer for buyers to catch on to the fact that the car costs a lot more to repair.
- biaachmonkie 3y ago
  The question is really if the castings result in more cars being written off as total loss. Any collision severe enough to damage them would likely be severe enough to also total out a car without them as well.
_ea1k 3y ago
TBH, the repairability concerns seem to be overstated so far. There are still conventional crash structures in front of and behind them. It takes a lot to damage the casting beyond repair. You are right, though. The jury is still out in that they haven't really been used in the real world for long. > Audi did in 2016 with the SQ7 TDI, no? I didn't know about that one, but it was still a dual architecture. Going full 48V is a bit different, assuming Tesla really did that.
hinkley 3y ago
Clearly no MGB fans at Tesla. The unibody is cool unless you live somewhere with snow, and then rust doesn’t just make your car ugly, it makes it unsafe. And expensive to fix. Which is why I was always more of a Triumph fan.

Turskarama 3y ago

The mechanical backup for a drive-by-wire system is different to all those other things in that it's a critical safety issue. If _anything at all_ happens to the electronics while driving then it's important that you can stop the vehicle safely. That mandates manual brakes and steering, even if they're barely effective at all without power it's a lot better than nothing.

dreamcompiler 3y ago
The cybertruck has two separate steering motors for redundancy. Passenger jets have been doing "x-by-wire" for decades. Properly engineered it's as reliable as mechanical linkage.
- Turskarama 3y ago
  Two motors is great and all, but that only provides motor redundancy and won't help if something happens to the electrical system. Planes are fly by wire but that's because they literally have to be, no amount of gym time will let you move the rudder on a 737 by hand.
  - networkchad 3y ago
    [dead]
  - masklinn 3y ago
    The 737 is not fly by wire.
    - Turskarama 3y ago
      Ok fair enough, it is hydraulic assisted instead but for the point I'm trying to make it's essentially the same thing in so far as it requires power.
_ea1k 3y ago
I tend to agree, TBH. People bring up fly-by-wire, but even that is simpler. AFAIK, those systems aren't variable ratio the way CT is. CT literally will have different steering ratios depending on the speed, so one steering wheel angle will have different effects on the actual steering angle depending upon the speed. How will that work in the real world? Could there be bugs? Will people get used to it or will it lead to mistakes? I'm looking forward to getting a chance to try it, TBH.

phkahler 3y ago

>> Tesla's greatest innovation is simply leadership. Totally correct. We were making 48v power steering systems 20 years ago for OEM development projects. They simply don't have the will/commitment/leadership to bring change like that to market. Elon sets a direction and people go that way. I can imagine someone at GM listing all the obstacles as a reason they can't change, and Elon looking at them and saying "no, you're the obstacle" or something like that.