Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Semiconductor Shortage and Car Production

This site may earn commission on affiliate links.
Article from 10 months ago: 29 new semiconductor fabrication plants coming. Will cause $140 billion dollars in spending for manufacturing equipment.

The semiconductor equipment industry is at the end of the whip, getting snapped hard. On the other hand, there's only so much they can do to ramp up their production, so they are also an anchor. In two years, when these fabs come online, the equipment sector will be seeing huge drops in sales, totally expected. Meanwhile, there will be a glut in semiconductor production.
 
“it will take 3 to 6 mths for the auto industry to end up in oversupply, which will put an abrupt end to a 3-year phase of unprecedented” pricing power and profit margins for OEMs. - Patrick Hummel, an Autos analyst at UBS, to investors this morning
This is what I am hearing as well in my Semi industry.
In addition, the current economic situation is leading to a softening in new car demand. I don't know anyone that works in tech right now is looking to spend $70k+ for an EV.
 
  • Like
Reactions: DanCar
1667496835550.png
 
Inventory continues to rise through month of November. Pickup's show highest inventory, cars show lowest inventory, but all are increasing. Article is paywalled.
1670528738374.png
 
  • Informative
Reactions: UkNorthampton

That's the trouble with manufacturing with such high capital costs. Once invested, the capital must continue to crank out product to remain profitable. A multi-billion dollar fab has to produce many billions of chips to pay for itself before making profit. In times of surplus, it is hard to justify cutting production rates, since that will benefit the competition more than yourself.

It's funny, that so many people think that the supply of lithium will limit BEV growth rather than the supply of silicon. LOL
 
That's the trouble with manufacturing with such high capital costs. Once invested, the capital must continue to crank out product to remain profitable. A multi-billion dollar fab has to produce many billions of chips to pay for itself before making profit. In times of surplus, it is hard to justify cutting production rates, since that will benefit the competition more than yourself.

It's funny, that so many people think that the supply of lithium will limit BEV growth rather than the supply of silicon. LOL
Wrong kind of leaves snow chips. Glut of more modern chips, lack of ancient chips used by subcontractors & suppliers of legacy automotive. Bosch might be the one to watch. Do they struggle with chips going forward? Automakers still using it as an excuse even when Bosch seems ok?

1673196775801.png
 
Wrong kind of leaves snow chips. Glut of more modern chips, lack of ancient chips used by subcontractors & suppliers of legacy automotive. ...
Automotive industry had it nice with 50 cent to 1 dollar chips. No one is going to build billion dollar fabs to produce $1 chips. Now the automotive industry has to pay $6 for more advanced chip and they are very slow to transition, except for Tesla.
 
Wrong kind of leaves snow chips. Glut of more modern chips, lack of ancient chips used by subcontractors & suppliers of legacy automotive. Bosch might be the one to watch. Do they struggle with chips going forward? Automakers still using it as an excuse even when Bosch seems ok?

Wow! Not sure what you are talking about. Distinguishing the categories of chips does nothing to understand the problem.
 
Automotive industry had it nice with 50 cent to 1 dollar chips. No one is going to build billion dollar fabs to produce $1 chips. Now the automotive industry has to pay $6 for more advanced chip and they are very slow to transition, except for Tesla.

I'm thinking you don't really understand integrated circuits. A $1 chip, is a $1 chip. There is no scenario where it becomes a $6 chip and designers use it in place of the $1 chip. No, older chips will not be built on a multibillion dollar fab line. But with each new fab line, the old one produces chips that become less expensive, because the line has been amortized. This gets passed down the line like shoes in a family of 10.

With all the hoopla about the chip shortage, I've never seen a realistic explanation of why this happened. Initially, it was a matter of the automakers reducing orders because of the pandemic. But that would have been a temporary perturbation that would have worked its way out, within a year. Yet, in year four of the shortage, it is as bad as ever.

We had a similar problem in 2018, when 0603 size passives became scarce. It was blamed on the mass movement to smaller sized parts, which the manufacturers supported, but failed to realize the auto industry was increasing their consumption, but, unlike the mobile manufacturers, using "old school" sizes. At one point, I had to pay $1 each for a $0.10 capacitor! Six months, to a year later, supply caught up and prices were back to normal.
 
  • Disagree
Reactions: ggr
Subsystems come from suppliers. Certification, integration and redesign required when a new chip replaces old, especially as environmental, longevity and safety aspects are important. If Bosch can't get a certified chip, they can't supply subsystems, no cars built, no spares available. Until all the work is done to get new designs recertified, suppliers can't get back up to production speed of subsystems. It only needs a shortage of one 50 cent chip to stop production. My understanding is that after automotive orders died, some chip factories closed or retooled. It was worth continuing production when factory costs long accounted for. Not after this debacle.
 
  • Like
Reactions: ggr
Subsystems come from suppliers. Certification, integration and redesign required when a new chip replaces old, especially as environmental, longevity and safety aspects are important. If Bosch can't get a certified chip, they can't supply subsystems, no cars built, no spares available. Until all the work is done to get new designs recertified, suppliers can't get back up to production speed of subsystems. It only needs a shortage of one 50 cent chip to stop production. My understanding is that after automotive orders died, some chip factories closed or retooled. It was worth continuing production when factory costs long accounted for. Not after this debacle.

Closed? Not likely... at least, not to an extent that they couldn't reopen four months later. Heck, it takes them that long to actually make a major decision like closing a fab. Then it takes them a couple of years to close it. They have to inform everyone who was using the fab well in advance. I used a Lattice FPGA in a design which went EOL a few years after I made the design. Seems Fujitsu was closing the fab Lattice used for this part. I had six months to order, but could schedule the deliveries over the next three years!

I'm not sure what is meant by “retooled” in this case. The only thing that makes sense is upgrading the equipment to finer processes. Again, that doesn't happen quickly. It takes a year to order the equipment.

The original story was that automotive slacked and cut orders, but various electronics needed for work from home picked up. That was three years ago. It should have been worked out by now. In fact, I am seeing some sign of easing in the shortage. I've seen Xilinx FPGAs in distribution inventory. Not much, but there had been zero on anything more modern than the Spartan 3 chips (nearly 20 years old). I also found inventory on a second source for an RS-422 transceiver chip I use. Unfortunately, there's literally just one source for this part... until now. So now there's two and one is in stock! There's still an opamp that is very hard to get, but I should be able to find a replacement if I try hard enough... or they may start being available by the time I need them for July production.