Investor Engineering Discussions

[mongo]

Well, yes, if you are a battery engineer, you care about everything including materials used because of ability to source, cost, etc. But this sub-thread started from the Limiting Factor video where the author was surprised to see 4680 Wh/kg wasn't any better than current Tesla batteries. And my point is that it doesn't need to be as long as overall cost is lower.

But I'm not sure what point you're trying to make since you are making the same point I made in the post you are responding to. LFP batteries are a prime example, crappy Wh/Kg, but overall cheaper to use in a particular car (lower range one).

Like I said, I agree

I would have expected 4680 to be better just based on surface area to usable volume, so it was a little surprising. Definitely agree TCO/ vehicle mass are the critical numbers.

LFP was showing they are using lower gravimetric density cells like you called out (seemingly hypothetically? )

 

[Doggydogworld]

LFP are lower Wh/kg (not quite 100, but down there at 125Wh/kg or so) and lower cost and used for SR packs and energy storage.

Tesla/CATL pack is 125 Wh/kg, but the original BYD Blade pack in the Han was 140 and I've been told later variants are ~150. LFP bus packs were 160 Wh/kg in China a couple years ago, I haven't checked lately but I'm sure they've crept up a bit as well.

LFP makes more sense for most applications, IMHO. Legacy auto is missing the boat.

 

[mongo]

Tesla/CATL pack is 125 Wh/kg, but the original BYD Blade pack in the Han was 140 and I've been told later variants are ~150. LFP bus packs were 160 Wh/kg in China a couple years ago, I haven't checked lately but I'm sure they've crept up a bit as well.

LFP makes more sense for most applications, IMHO. Legacy auto is missing the boat.

Yeah, I was looking for the low end of density.

 

[Oil4AsphaultOnly]

Tasmanian provided this tidbit: Tesla Has Already Delivered 36 Semis to PepsiCo, with Remaining 64 to Deploy by 2023

36 delivered so far, with the rest to be delivered after the new year.

But, the source reuter's interview is more enlightening: https://www.nasdaq.com/articles/exclusive-pepsico-to-roll-out-100-tesla-semis-in-2023-exec

"The company is targeting rolling out the Semis in the central U.S. next, then the East Coast, O'Connell said.

He said Tesla did not help pay for the trucks' megachargers but provided design and engineering services for the facilities, which come with solar and battery storage systems.

O'Connell said that a 425-mile (684-km) trip carrying Frito-Lay products brings the Semi's battery down to roughly 20%, and recharging it takes around 35 to 45 minutes."

They won't share the weight of the truck though.

Edit: They got $15.4m in state subsidies (probably just for the solar panels, megapacks, and mega chargers) and $40k per truck from the federal EV tax credit to help pay for all this, so about $19.4m total.

 

[JRP3]

36 delivered so far, with the rest to be delivered after the new year.

That's not what the original article said:

"PepsiCo said it plans to deploy 15 trucks from Modesto and 21 from Sacramento."

Exclusive: PepsiCo to roll out 100 Tesla Semis in 2023, exec says

PepsiCo plans to roll out 100 heavy-duty Tesla Semis in 2023, when it will start using the electric trucks to make deliveries to customers like Walmart and Kroger, the soda maker's top fleet official told Reuters on Friday.

www.reuters.com

 

[Oil4AsphaultOnly]

That's not what the original article said:

"PepsiCo said it plans to deploy 15 trucks from Modesto and 21 from Sacramento."

Exclusive: PepsiCo to roll out 100 Tesla Semis in 2023, exec says

PepsiCo plans to roll out 100 heavy-duty Tesla Semis in 2023, when it will start using the electric trucks to make deliveries to customers like Walmart and Kroger, the soda maker's top fleet official told Reuters on Friday.

www.reuters.com

Hmmm, it seems we are having issues with journalistic integrity again. My link was posted at 4:33pm EST and wrote: "PepsiCo is deploying 36 electric trucks from Tesla, with 15 in Modesto and 21 in Sacramento, so far."

One of those writers changed the wording.

Edit: I've reached my "articles limit", so I couldn't check the wording on your link.

 

[JRP3]

My link still reads the same: "plans". Try it in a private browser maybe.

 

[Oil4AsphaultOnly]

My link still reads the same: "plans". Try it in a private browser maybe.

Yeah, your article has a 3hr newer timestamp. It's either a correction or an editorial revision. In either case, Pepsi either has 36 now or within two weeks, a difference that eventually won't matter.

 

[JRP3]

Wood based batteries

 

[Buckminster]

Toughest material ever is an alloy of chromium, cobalt and nickel
Might be useful for space.

 

[mongo]

That would have crash test and air bag considerations.

Leave aur bags as is, if no one is sitting there, it doesn't matter. Might have a sensor to supress though.

As long as the seats aren't major lateral supports.
Could just eliminate the seat backs? Nothing suspicious there...

 

[MC3OZ]

This is some tentative speculation about the cost to build Cybertrucks, feel free to improve it.

Glass - I'm sure Tesla is making the glass in house it is the same glass as the Semi windscreen and the probably the same glass as the Model 3/Y roof. overall the Cybertruck glass might be 2X the cost of Model 3 glass.

Stainless Steel - It is very likely that Tesla is doing the "cold rolling" at Austin, still stainless steel is a relatively expensive raw material.

Raw Materials and parts - overall, I think the raw materials and parts for a Cybertruck might cost around $12,000 more than raw materials and parts for a Model 3. This is a combination of size, more expensive higher grade parts and more expensive raw materials. And odiously a guess.

Cost to build - I think capex and production costs are possibly similar to the original Fremont Model 3 line, perhaps a bit higher. A lot fewer body shop robots and stamping moulds, no paint, but cold rolling, casting and stainless steel folding equipment is required.

The production line might be slow to ramp, so I would estimate up to an additional $10,000 production costs for the early units, trending down to the same costs as a Fremont built Model 3 over 2 years.

Battery cells - Tesla has a target of $70 kWh for high nickel 4680s, I'm guess around $100- kWh initially trending down to $80 kWh after 2 years.
The implication here is the base model Cybertruck is not built for the first 2 years.

A a rough ball park figure the M0del 3 costs $30,000 to build with the battery costing around $8,000

So A Cybertruck body would cost $34,000 when fully ramped and $44,000 initially.

200 kWh battery $20,000, cost to build $64,000 margin 30% - retail $83,200
150 kWh battery $15,000, cost to build $59,000 margin 30% - retail $76,700

Base model after say 2 years:-
125 kWh battery $10,000 cost to build $44,000 margin 30% - retail $57,200

Of course this is only a rough ballpark guess, other guesses are welcome.

When the Cybertruck was revealed Tesla may have had a target of $60 kWh for high nickel 4680 cells, that target has been adjusted to $70 kWh.
That is one reason why prices may be higher.

 

[30seconds]

hmmm,

the stainless steel should be cheaper than aluminum in the 3/Y but offset for the larger surface area. Of course there should also be substantial savings since no paint / paint shop is needed.

I've been assuming that we would also see a higher voltage system that should reduce wiring complexity.

Not sure what other major systems savings are possible vs. Y

 

[Ogre]

A a rough ball park figure the M0del 3 costs $30,000 to build with the battery costing around $8,000

A kind of big quibble here. That's roughly the cost to manufacture a Model 3, but that includes a lot of things like labor which are likely to be lower for the Cybertruck. Raw materials are only about half of the cost of the vehicle itself. While the raw materials will be higher, the actual assembly should be easier with far fewer steps. Things like paint are gone which is a huge cost.

So A Cybertruck body would cost $34,000 when fully ramped and $44,000 initially.

200 kWh battery $20,000, cost to build $64,000 margin 30% - retail $83,200
150 kWh battery $15,000, cost to build $59,000 margin 30% - retail $76,700

Base model after say 2 years:-
125 kWh battery $10,000 cost to build $44,000 margin 30% - retail $57,200

Of course this is only a rough ballpark guess, other guesses are welcome.

Now carve $4,500 off the base model and $9,000 off the top model for Advanced Manufacturing Production Credits from the IRA.

I'm also a bit lower on the battery costs, mostly because I think they will be a touch more efficient than you do. I'm thinking the 300 mile truck is going to have a 125kWh battery.

The big question is which models they can sell first so they can get to break even.

 

[Ogre]

hmmm,

the stainless steel should be cheaper than aluminum in the 3/Y but offset for the larger surface area. Of course there should also be substantial savings since no paint / paint shop is needed.

I've been assuming that we would also see a higher voltage system that should reduce wiring complexity.

Not sure what other major systems savings are possible vs. Y

One big one.

They are very likely switching to 48v and not running wiring harnesses, but using rails which can be installed by robots. Wiring harnesses are a huge time killer.

 

[AudubonB]

@MC3OZ: When did Tesla bring glass production in house? Our newest Tesla is a 2017 - still makes use of the AGP Glass company provided in the first Teslas.

I'll go farther out and state that I shall be very surprised if Tesla has found it appropriate to float their own glass. As with metal refining: too much CapEx for too little production.

@Ogre: I devoutly hope you are correct. If Tesla is not leading the way with 48V and rail - who is? It would be wincing to learn another has leapfrogged Tesla there.

 

[MC3OZ]

@MC3OZ: When did Tesla bring glass production in house? Our newest Tesla is a 2017 - still makes use of the AGP Glass company provided in the first Teslas.

I'll go farther out and state that I shall be very surprised if Tesla has found it appropriate to float their own glass. As with metal refining: too much CapEx for too little production.

@Ogre: I devoutly hope you are correct. If Tesla is not leading the way with 48V and rail - who is? It would be wincing to learn another has leapfrogged Tesla there.

There is a YouTube video on Tesla glass from Tesla that shows them transporting a Semi windscreen. I assumed that the video implies made in house.

Maybe they only prototype and design the glass in house and then contract manufacturing when I re-watched the video they mentioned prototyping.

 

[MC3OZ]

A kind of big quibble here. That's roughly the cost to manufacture a Model 3, but that includes a lot of things like labor which are likely to be lower for the Cybertruck. Raw materials are only about half of the cost of the vehicle itself. While the raw materials will be higher, the actual assembly should be easier with far fewer steps. Things like paint are gone which is a huge cost.


Now carve $4,500 off the base model and $9,000 off the top model for Advanced Manufacturing Production Credits from the IRA.

I'm also a bit lower on the battery costs, mostly because I think they will be a touch more efficient than you do. I'm thinking the 300 mile truck is going to have a 125kWh battery.

The big question is which models they can sell first so they can get to break even.

I tried to be conservative which means that I might be a bit high.

The stainless steel is a special alloy but most of the cost would be in the cold rolling. In the long run I agree on lower labor costs.

 

[Ogre]

Another place where Tesla might save some cash and weight.

Converting the braking system to drive by wire (Brembo Sensify). While the cost of the braking system is likely more expensive than a traditional braking system, installation costs (and weight) would be much lower.

I think it’s one of those things which will be more expensive short term but save them a lot of money long term. Particularly if they can use the same channel/ rail based system the rest of the electronics are on.

 

[Oil4AsphaultOnly]

Another place where Tesla might save some cash and weight.

Converting the braking system to drive by wire (Brembo Sensify). While the cost of the braking system is likely more expensive than a traditional braking system, installation costs (and weight) would be much lower.

I think it’s one of those things which will be more expensive short term but save them a lot of money long term. Particularly if they can use the same channel/ rail based system the rest of the electronics are on.

I like this idea of removing the brake fluid completely! but one of the commentors of one of the Brembo youtube videos asked a good question, what's the backup in case of power failure?