Tesla Semi

  • SUPPORT THE SITE AND ENJOY A PREMIUM EXPERIENCE!
    Welcome to Tesla Owners Online, four years young! For a low subscription fee, you will receive access to an ad-free version of TOO. We now offer yearly memberships! You can subscribe via this direct link:
    https://teslaownersonline.com/account/upgrades

    SUBSCRIBE TO OUR YOUTUBE CHANNEL!
    Did you know we have a YouTube channel that's all about Tesla? Lots of Tesla information, fun, vlogs, product reviews, and a weekly Tesla Owners Online Podcast as well!

Scuffers

Active member
Joined
Jun 8, 2017
Messages
187
Location
Mkt Harborough
Country
Country
#81
making many assumptions, specifically:

1) that the 500 miles battery is 1MWh
2) the 300 mile battery is 600KWh
3) that the difference in versions of truck is just battery

that implies that the battery costs $75/KWh - that's REALLY cheap!

can that possibly be right?
 

KarenRei

Top-Contributor
Joined
Jul 27, 2017
Messages
1,649
Location
Reykjavík
Country
Country
#82
making many assumptions, specifically:

1) that the 500 miles battery is 1MWh
2) the 300 mile battery is 600KWh
3) that the difference in versions of truck is just battery

that implies that the battery costs $75/KWh - that's REALLY cheap!

can that possibly be right?
I expect the truck to be "under 1MWh" (thinking ~900kWh-ish), which increases the cost of the battery to $83/kWh (assuming that the margins are the same on both the base battery and the upgrade). But if you assume Tesla has a 25% margin, then that means the production cost is only $67/kg.

It's certainly possible from a raw materials standpoint. The question is about production costs. But, when it comes to EVs... go big or go home! :) Large quantities = low unit costs.
 

Scuffers

Active member
Joined
Jun 8, 2017
Messages
187
Location
Mkt Harborough
Country
Country
#83
I expect the truck to be "under 1MWh" (thinking ~900kWh-ish),
I see your thought process, but consider this, Elon states <2KWh/mile, well, ok, let's be generous, that means the actual figure is between 1.9 and 2KWh/mile - if it was less than 1.9, he would have said that.

then we get to the life of the battery, to run it from 0-100-0% daily will soon kill it, so we have to assume that for a truck designed to be on the road all the time, they will have given it a big enough battery to do the 500 miles without using 100% of the batteries theoretical capacity, they would have added at least 10% buffer (and if it was me, more like 20%).

So, we are left with:

1.9KWh/mile = 950KWh for 500 miles
add 10% = 1,045KWh

Now, I also consider the <2KWh is absolute best case (much like NEDC range numbers), so maybe in the real world, it will only make 450 miles? although if they did this, that would not do their chances with the truck operators any favours, if you say it will do 500 miles, it dam well better be able to..

All that said, I still think it's 1MWh
 

KarenRei

Top-Contributor
Joined
Jul 27, 2017
Messages
1,649
Location
Reykjavík
Country
Country
#84
I see your thought process, but consider this, Elon states <2KWh/mile, well, ok, let's be generous, that means the actual figure is between 1.9 and 2KWh/mile - if it was less than 1.9, he would have said that.

then we get to the life of the battery, to run it from 0-100-0% daily will soon kill it, so we have to assume that for a truck designed to be on the road all the time, they will have given it a big enough battery to do the 500 miles without using 100% of the batteries theoretical capacity, they would have added at least 10% buffer (and if it was me, more like 20%).

So, we are left with:

1.9KWh/mile = 950KWh for 500 miles
add 10% = 1,045KWh

Now, I also consider the <2KWh is absolute best case (much like NEDC range numbers), so maybe in the real world, it will only make 450 miles? although if they did this, that would not do their chances with the truck operators any favours, if you say it will do 500 miles, it dam well better be able to..

All that said, I still think it's 1MWh
1) 10% is too much of a buffer. Tesla usually uses around 5%.

2) You can't assume "<2kWh/mi" means "between 1,9 and 2,0 kWh/mi". If they had said "<1,9kWh/mi", sure, it'd be reasonable to assume "between 1,8 and 1,9 kWh/mi". But with 2kWh you're adding in extra "precision" to the estimate that wasn't given. That's a very rounded number, with no clues given to how much it was rounded.

3) Tesla has, if anything, shown a habit of reporting lower mileage (and thus correspondingly increasing consumption per mile) than they're required to, in order to make the numbers more realistic. See for example the downgrading of Model 3 LR from 334 to 310. Who else but Tesla does that? :)

Don't get me wrong, I'm not ruling out 1MWh or more. But I'm expecting closer to 900kWh. I guess we'll see :)
 

KarenRei

Top-Contributor
Joined
Jul 27, 2017
Messages
1,649
Location
Reykjavík
Country
Country
#85
As for raw materials in the packs as a whole (not just cells)... if we assume:

Li: 0,07kg/kWh
Co: 0,22kg/kWh
Ni: 1,2kg/kWh
Mass: 5,5kg/kWh

Thus:

Non Li-Co-Ni mass: 4kg/kWh

Then with prices:
Li: $75/kg (lithium fraction of carbonate at $15/kg)
Co: $60/kg
Ni: $15/kg
Other: Let's say an average of $4/kg

Then our raw materials costs are: $52/kWh

So the estimated pack numbers are plausible, so long as they keep the manufacturing overhead down. Also note that most of these raw mineral prices have spiked up; eventually, mining will catch up with the demand growth curve (eventually pack recycling will contribute meaningfully to raw material streams as well) and they should begin falling back toward their historic prices again. In that case you might have something like:

Li: $30/kg (lithium fraction of carbonate at $6/kg)
Co: $35/kg
Ni: $10/kg
Other: Let's say an average of $3/kg

Then our raw materials costs are: $34/kWh.

Note that I assumed NCA, but Tesla may be moving to low-cobalt NMCs. If that's so, that'd only be good for prices - trading out more expensive nickel for cheap manganese ($2/kg). So long as that cobalt fraction stays low...

(I really had no clue how to budget the rest... copper ($7/kg), alumium ($2/kg), graphite ($1/kg or more, depends on grade), silicon (depends highly on the form), electrolyte, membrane, various manufactured components.... that's just a WAG ;) Hopefully an overestimate.)
 
Last edited:

Kbm3

Active member
Joined
May 24, 2017
Messages
140
Location
Baton Rouge, La.
Country
Country
#86
1) 10% is too much of a buffer. Tesla usually uses around 5%.

2) You can't assume "<2kWh/mi" means "between 1,9 and 2,0 kWh/mi". If they had said "<1,9kWh/mi", sure, it'd be reasonable to assume "between 1,8 and 1,9 kWh/mi". But with 2kWh you're adding in extra "precision" to the estimate that wasn't given. That's a very rounded number, with no clues given to how much it was rounded.

3) Tesla has, if anything, shown a habit of reporting lower mileage (and thus correspondingly increasing consumption per mile) than they're required to, in order to make the numbers more realistic. See for example the downgrading of Model 3 LR from 334 to 310. Who else but Tesla does that? :)

Don't get me wrong, I'm not ruling out 1MWh or more. But I'm expecting closer to 900kWh. I guess we'll see :)
Especially as the larger the battery size, the more unbelievable the price they hit.
 

TrevP

TOO Founder / Administrator
Founder
Joined
Oct 20, 2015
Messages
6,993
Location
Toronto
Tesla Owner
Model 3
Country
Country
#87
Let’s revisit my prediction in a few months when I said we now know Tesla’s direction with a Model S and X interior refresh.

Fast forward to 14:20 to see what I mean

 

KarenRei

Top-Contributor
Joined
Jul 27, 2017
Messages
1,649
Location
Reykjavík
Country
Country
#89

Scuffers

Active member
Joined
Jun 8, 2017
Messages
187
Location
Mkt Harborough
Country
Country
#90
It looks like someone really smart and expert in physics has found an explanation to the "Tesla promises breaks the law of physics" notion...

http://selenianboondocks.com/2017/11/tesla-semi-part-1/

That analysis also explains the pricing and performance surprise.
he's done the numbers, but I regard them as opn the most optimistic side, I still think we are talking at 1MWh battery here for 500 miles (and last 1M miles)

he has also not factored in the two different final drive ratio's and what looks like a 'clutch' to disengage one of them
 

SuperMario

Active member
Joined
Mar 15, 2017
Messages
28
Location
Ottawa
Country
Country
#91
Does anyone know what is highlighted here, in all six wheels?
It flashed up quickly on the screen during the presentation but Elon seemed to change direction and it wasn't really discussed.
The Tesla Semi is equipped with air brakes, typical of what can be found on a modern class 8 tractor. The components you see here are brake rotors and calipers. The calipers are actuated by small brake chambers. The drive axles have a Maxi type brake chamber, incorporating a spring which applies the brakes when no air is present in the system, a parking brake. So one can assume the vehicle is equipped with an electric accessory drive motor which drives an air compressor(the biggest load), the a/c compressor and power steering pump(unless it has it's own dedicated electric motor)
 

Curt Renz

Active member
Joined
May 22, 2017
Messages
154
Location
Chicagoland
Country
Country
#92
CNBC - yesterday: Growing number of companies ordering Tesla's electric semi

Frederic Lambert @Fred Lambert is the publisher/editor of Electrek and a Model 3 Owners Club member.

Fred performed quite admirably. The hosts asked tough questions about Tesla and its semi truck. Fred responded with excellent and honest answers. It may have been his first time on financial television. Whatever nervousness he may have initially felt turned into confidence as the interview progressed and the questions became tougher. As a retired host of a TV financial news program, I can tell when an interviewee knows his stuff and is being honest. Fred gets an A+ from me. Let's hope we see him more frequently on TV. :)


I knew Fred is Canadian but had not stopped to think that he might be a native French speaker. He writes in English so well that I had not considered that he might speak in English with a French accent. Through emails I have a Belgian friend named Jean Meeus who writes astronomy books. He is also a native French speaker. I once complimented him on his ability to write beautifully in English. He replied that he has no problem reading scientific papers written in English, but has trouble with newspapers written in English. That's because the latter are full of idiom and the former are not. Kudos to Fred, Jean and all who have so well learned English as as second language. ;)
 

garsh

Dis Member
Moderator
TOO Supporting Member
Joined
Apr 4, 2016
Messages
12,476
Location
Pittsburgh PA
Tesla Owner
Model 3
Country
Country
#94
Could the semi and maybe the roadster be using this rumoured new 4416 cell format?

https://www.teslarati.com/tesla-4416-lithium-battery-cell-model-3-next-gen-s-x/
That was information relayed poorly by Ben Sullins. Somebody told him that the Model 3 "would have 4416 cells". He took that number to be some sort of cell model number. It was actually the total number of cells in the long range Model 3's battery pack.

https://teslaownersonline.com/threa...and-next-gen-s-x-fleet.4367/page-3#post-33666
 

Ctesibios

Active member
Joined
Mar 4, 2017
Messages
26
Location
Gatineau, QC
Country
Country
#95
That was information relayed poorly by Ben Sullins. Somebody told him that the Model 3 "would have 4416 cells". He took that number to be some sort of cell model number. It was actually the total number of cells in the long range Model 3's battery pack.

https://teslaownersonline.com/threads/tesla-could-be-using-a-“4416”-battery-cell-in-model-3-and-next-gen-s-x-fleet.4367/page-3#post-33666
Well, it would make sense to use the 2170 since the gigafactory is rigged up to produce them. The semi would require around 50 000 of them!
 

Scuffers

Active member
Joined
Jun 8, 2017
Messages
187
Location
Mkt Harborough
Country
Country
#96
The 4416 cell story is just another example of what's wrong with Journalism, too little time thinking about the info before blasting it out as fact.

As said, the 4416 number is actually the number of cells in the 3LR pack, not the type of cell - 2170