Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
This site may earn commission on affiliate links.
I'm fairly certain that Chile still has plenty of excess capacity, not to mention a number of untapped sources right here in the US.
True. But remember folks lithium is not the supply constraint in LiOn battery production, it's cobalt. I believe Tesla uses the lithium nickel cobalt manganese oxide (NCM) formulation. Lithium is actually pretty common in the earth's crust.
 
Ok, so I'm keeping my promise of calculating practical beam and column sizes, for no reason in particular, but I've hit a snag. My calculated column spacing of 30 by 50 with the latter being the length of the filler beams has the lightest possible beam at a W33x118. That seems too heavy and the construction images seem to support this. I'm calculating for 200psf and 500psf live loads with 5.5 inch reinforced concrete, but it's the length driving up my numbers. Does anyone remember any info of column spacing being less than this, like 30 by 45? That would bring my numbers into the realm of possibility.

Update: I think I have it figured out.
Something very interesting I would like to share is the decrease in effective length of the column extending from grade to 40 feet (center of 41ft floor beams) to an effective length of 24 feet. This means that despite adding nearly a third more weight, the required column is a W14x398 instead of a W14x550. This is pushing the limits of structural engineering. I double checked with images and was able to consistently get a column spacing of 30 feet by 50 feet, so I reduced the live loads. My calculations are/were based on the following numbers.

23.5 ft Mezzanine floor AND 41 ft floor:
5.5 inch thick standard weight reinforced concrete slab with 200 psf live load.
W21x93 filler beams
W40x331 girders (or W33x354)

71 ft roof:
5.0 inch thick standard weight reinforced concrete slab with 100 psf live load.
W21x55 filler beams
W36x210 girders

Column from grade to 44 feet:
W14x550 (W14x398 required with mezzanine floor)

Column from 44 feet to 70 feet:
W14x90

I think I'm getting oversized shapes because I'm not treating it as a composite floor, which it obviously is.
 
Last edited:
True. But remember folks lithium is not the supply constraint in LiOn battery production, it's cobalt. I believe Tesla uses the lithium nickel cobalt manganese oxide (NCM) formulation. Lithium is actually pretty common in the earth's crust.
For their cars, they are currently using Lithium Nickel Cobalt Aluminum Oxide (LiNiCoAlO2, "NCA") for their cathode material.
They are using NMC for their daily cycling Tesla Energy products, however (from Q1 earnings call).

energy.png

Source: Southern California Energy Summit (dead link, too big to attach)


About typical unit fractions, both contain about 10 times as much nickels as other elements.
(multiply with respective atomic mass units to calculate mass fractions. Lithium = 3, Aluminium = 13, Cobalt = 27, Nickel = 28, manganese = 55)

material.png

Source: Patent WO2013142637A1 - Methods of making metal-doped nickel oxide active materials - Google Patents

As for cost, nickel lists for about $12k/ton, cobalt (compounds?) for about $30k/ton, lithium $6k/ton, aluminium and manganese $1.7k/ton. H2SO4 and NaOH are very cheap.
While I suspect Tesla's negotiated prices will differ materially, I still think the overall situation remains comparable, and nickel and cobalt are the main cost drivers.

Sources:
Nickel - Monthly Price - Commodity Prices - Price Charts, Data, and News - IndexMundi
London Metal Exchange: Cobalt
Tesla: Expect Lithium Prices to Rise | InvestorPlace
 
Last edited:
Also, I would add that based on the last talk of the battery chemistry when they announced the 90kWh battery they said that their newest cell actually uses some silicon in it, as opposed to the synthetic graphite. I don't know how much that might change the weights of the cell and therefore the costs, but I assume that helps somewhat because silicon should be cheaper than the graphite (and synthetic graphite at that, which is likely more expensive than mined graphite.)
 
Also, I would add that based on the last talk of the battery chemistry when they announced the 90kWh battery they said that their newest cell actually uses some silicon in it, as opposed to the synthetic graphite. I don't know how much that might change the weights of the cell and therefore the costs, but I assume that helps somewhat because silicon should be cheaper than the graphite (and synthetic graphite at that, which is likely more expensive than mined graphite.)
Cathode material is the biggest cost driver. Graphite ($2500/ton) and Silicon ($2000/ton) are relatively cheap.

the-long-view-electric-cars-14-638.jpg

Example above for a manganese cell. Numbers are different from the paper linked above ( less nickel)
Source: The Long View Electric Cars - Bernstein Research
 
Wow, that must be a pretty recent sattelite pic, looks like all the roof is done. Too bad they don´t give you a date when it was taken more detailed than '2015'. Pretty cool, checked googlemaps and all you need to type into the search to go there is "gigafactory" :).
 
Wow, that must be a pretty recent sattelite pic, looks like all the roof is done. Too bad they don´t give you a date when it was taken more detailed than '2015'. Pretty cool, checked googlemaps and all you need to type into the search to go there is "gigafactory" :).

Looks to be maybe around 10 days ago based on progress and my search results.. My Username is way too true to be fun..
 
Ah, that´s the way to get a date :). So the pic is already more than 2 months old. Pretty good for sattelite, but I wish we had access to the live cams Tesla has!! On the other hand, I guess from now on the most important stuff will be happening on the inside.
 
Cathode material is the biggest cost driver. Graphite ($2500/ton) and Silicon ($2000/ton) are relatively cheap.

View attachment 90054
Example above for a manganese cell. Numbers are different from the paper linked above ( less nickel)
Source: The Long View Electric Cars - Bernstein Research

It was just a statement that everything helps bring the cost down. If they could get more storage in the same package at even a slightly cheaper cost because of adding the silicon that is a win because you have improved the cell while making it cheaper at the same time. Things don't typically go that way with both moves at the same time. It's a triangle between speed, cost, and how good you want it.

Also typically the graphite costs you see are raw materials from the ground. Synthetic is always more expensive although I haven't found out a good measuring stick to see by how much. The stated reason for using synthetic is that it forms the cell in a more perfect way improving the density of the cell (see, there is that good side of the triangle dragging away from the cost side...)
 
Now that we have a good satellite image, I thought I would throw together a photoshoped image of what I believe to be the final layout of the building. We know that the final building will have a 5,800,000 sq ft footprint, and we know that it will be built in 7 blocks, each with an average of an 828,571 sq ft footprint and that the basic structure of the first block is already completed. Given that there are 4 sections per block, there should be a total of 28 sections once completed. Based on that information, here is a very rough representation of the final building (current structure "block 1" outlined in red):

giga render.jpg
 
Last edited: