Okay I'll give it a shot...
The normal rollout interval has been 2-3 yrs from 'Reveal' to Mass Production, so perhaps 2021-22 for Tesla Pickup in quantity? It'll certainly depend on bty cell capacity and its likely lower priority than Semi.
I still hold the opinion that the rumoured new “
giant, giant, giant machine that duplicates everything, is modular, is simple on the modular level, and is... gigantic" is to build battery cells, modules, and packs. Not for die cast model Y bodies (which is
foolish btw, Elon already said they will replace a 70-part rear-subframe assembly with 4 parts in the Y, then later with a single cast part when the larger casting machine becomes available. This casting machine
DOES NOT replace the body stamping presses, sheesh. /rant).
So let's
spec up that (Giant)^3 machine (I'll call it the
'G-Cube') for its output, first for N. America: (per Elon's statements on expected long-term steady-state demand)
- 750K/yr Model 3+Y @ 67 KWh => 50 GWh/yr (2021)
- 100K/yr Semi @ 1MWh => 100 GWh/yr (2023)
- 500K/yr Pickup @ 300 KWh => 150 GWh/yr (2024)
So that's a requirement for 300 GWh/yr within 5 years for the first N. America G-Cube (I'm going to lump storage requirements together separately since they have a different rampup curve, and their own steady-state demand).
That's roughly 10x the current cell capacity at GF1. A compounded annual growth rate that yields 10x increase over 5 years is 59% CAGR. That's actually pretty close to Tesla's historical auto production growth rate over the past 6 years, so I call it physically possible to do. Now let's build out a growth table for the next 5 years: (numbers listed are predicted production run-rates achieved by the end of the listed year; assuming* 29.5 GW/yr run rate by end of 2019, up from 28 GWh/yr by 2019.5)
Year GWh
2019 29.5*
2020 47
2021 75
2022 119
2023 189
2024 300
Now let's get on with specing out the G-Cube:
- each G-Cube must be able to produce 25GW/yr in finished product
- Tesla must be able to build 1 G-Cube/yr in 2020-21
- Tesla must be able to build 2 G-Cubes/yr in 2022-23
- Tesla must be able to build 4 G-Cubes in 2024-25
Each G-Cube must produce cash flow which will pay for another G-Cube after 2 yrs operation. Technically that's a CAGR of only 41% but hey this is an estimate (the main point is that this is the exponential part of Elon's vision).
For the first G-cube, let's say consumer facing product is $100/KWh at the pack level (long held as the enabling goal for Tesla's production aspirations). That's $5B in product over 2 years from each G-cube. Traditionally, the cost of raw materials for cells has been ~60% of the finished product, so let's say there's $2B left over (
notice how I'm skipping over the part about robotics and cost of labor?)
So as long as Tesla can spec, build, and equip its first G-Cube for $2B, it pays for itself after 2 years which allows them to
build another G-Cube out of FCF (NO LOANS).
Then, as production efficiency increase (that's the reason you stay 'modular' so you can swap out old designs) the exponential increases. After a further 2 years running those 2 G-Cubes, you can afford 2 more G-Cubes, etc etc... and I haven't even modeled depreciation.
Can they do it? Sure. China has demonstrated they're all in for the tech, and especially for the financing (ie: recourse local debt in Shanghai). Then FCA will pay for the first G-cube in Europe with its $2B/2yr CO2 offsets to Tesla. These may actually
INCREASE as other failed manufacturors join the FCA emissions pool after 2021. I suspect the production ramp for Europe will be approx like this:
Year GWh
2022 25
2024 50
2026 100
2028 200
2030 400
So you see that gets Tesla to a world-wide runrate of 1 TWh/yr for vehicle packs sometime between 2025-2030 (depending on precisely when the first EU GF4 w.G-cube comes online).
In all of this, remember,
storage is EXTRA (I'm calling for 2+ TWh/yr production when including Tesla grid+storage products). As a bonus, here's a rough steady of the state product demand I see for both China and the EU (emphasis on smaller** city cars; no pickups at all; numbers approximate)
- 750K/yr Model 3+Y @ 67 KWh => 50 GWh/yr
- 200K/yr Semi/Bus/Delivery @ 500KWh => 100 GWh/yr
- 1.5M/yr Model 2 @ 27 KWh => 150 GWh/yr (2024)
- Total: 300 GWh/yr
So given a requirement for 300 GWh/yr each in China+EU, and 400 GWh in N. America, that's a 1.2 TWh/yr requirement for the planned Tesla vehicles over the next 10 years.
**Recall Elon spoke about a smaller $25K Tesla during the 2018 AGM (which many commenters now refer to as the
'Model 2'), and naturally it will have much larger sales volumes in the older congested cities of Europe and large cities of Asia).
Add it all up, and that's 6.5M Tesla Vehicles per year production reached sometime between 2025-2030 (again depending on the
SELF-FUNDING rampup). Best part? That's roughly
$280B in annual revenue just from the Tesla Automotive division. Even if TSLA continues to trade at just 2x earnings, that puts TSLA at $3,200/share within 10 yrs.
And remember,
'Storage is extra'. So is FSD it seems, but that 10 year horizon is nearly
@neroden scale time for FSD so...
That's just the numbers from currently planned/rumoured products. I personally plan to be taking Sunday rides on a Tesla bicycle that weighs 20 kg with a hubless SRPM motor and a 2 KWh fast-charging bty w. 10-yr svc-life by 2030.
Cheers!
P.S. So yeah, 15X TSLA upside yada-yada, but I will be selling all my shares at the Opening today, because
Capt Sully tweeted that I should be
'concerned' about Autopilot.
/S