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

Behold the Terafactory (aka Terrafactory)

This site may earn commission on affiliate links.

jhm

Well-Known Member
May 23, 2014
10,182
39,908
Atlanta, GA
Elon Musk has said that hundreds of gigafactories will be needed to replace ICE vehicles with EVs. I beg to differ and will propose a wild ass theory to prove that only 20 gigafactories or 1 Terrafactory will ever be needed for total world auto domination.

Let's proceed. One gigafactory will have the capacity to produce 50 GWh of battery packs at 2017 energy density level. Let's assume that in the future consumers will only require 100 kWh average per vehicle. At current density this would give you about 300 miles range, but as energy density increases over time this range will grow owing to ever lighter battery packs improving miles per kWh. So put 20 of these gigafactories together and you get nameplate capacity of 1000 GWh in 2017. Behold the Terrafactory!

But how could one Terrafactory be enough to dominate the auto world, you say? Isn't 1 TWh enough only for 10M cars each requiring 100 kWh? Indeed, that is the nameplate capacity in 2017. But let us consider that battery technology will continue to double energy density every 10 years. Moreover, the capacity of the Terrafactory is dictated by the volume and mass of the batteries, not the amount of energy they may hold. Therefore, as energy density doubles every 10 years, so does the TWh output of the Terrafactory. In 2017, 1 TWh. In 2027, 2 TWh. In 2037, 4 TWh. And so by 2050, the great and awesome Terrafactory is pumping out 10 TWh of battery pack, enough for 100M cars each to have 100 kWh and a range over 500 miles. This is total auto world conquest. Behold the Terrafactory!
 
Last edited:
you are assuming that these factories are making batteries solely for automobiles. that's wrong. these factories will be supplying batteries for lots of other non-auto things as well such as grid storage system and residential battery systems. think bigger.
 
Agreed, the fixed location energy storage and management market is potentially bigger than the EV market even if 80% or more of all new car sales are EVs.

Ecarfan, it would be nice to see your reasoning behind this. I am going through a upgrade process now for my own home and it seems that supplying from rooftop PV, the total annual energy supply for our home plus 2 evs would need about 10 kWh of storage for the home, compared with 85 kWh of storage for the MS and maybe 40 kWh for a smaller second vehicle. This assumes we can sell excess to the grid in summer and buy it back as off-peak in the winter. Off-grid, I agree the requirement for home storage would be much greater, but I expect grid companies will see the light and as a viable alternative to the death spiral, will allow peer to peer wind and solar transactions eventually as their new sustainable role, basing their charges on power and distance.

I agree that in the early years of expansion, domestic sales for domestic storage could be relatively larger than this scenario suggests because at first there could be many more existing PV arrays in need than electric cars, but as electric cars grow quickly in number, I imagine they would ultimately outstrip domestic storage capacity. Imagine what you would need for your own home and you can see that in your own case, the vehicle storage will be much more than the car storage, and that is a microcosm of the domestic and most of the transport sector. Indeed, if we ever get V2H (vehicle to home) working, the existence of two cars could make home storage minimal.

If we move out of the home sector, commercial PV storage would be smaller than the domestic storage sector and as many commercial enterprises operate during the day only, less needful of storage. The service industries also have lots of roof area and lots of vehicles they could charge with that resource.

So I see the home storage to be a complex play over time, where PV storage might temporarily provide a immediate secondary market to increase production and reduce costs, but in the long term the vehicle market predominates over domestic and commercial stationary storage.

The elephant in the room is manufacturing. Whether manufacturing all goes to China or to your 3D printer in the basement, who knows? But Elon is ahead of us as usual, and is designing the prototype large gigafactory where the energy is sourced nearby as part of the project, and the storage comes from used car batteries. If Elon is right, then much less new production for batteries would be needed for the stationary sectors if old vehicle batteries are used.
 
.....let us consider that battery technology will continue to double energy density every 10 years. Moreover, the capacity of the Terrafactory is dictated by the volume and mass of the batteries, not the amount of energy they may hold. Therefore, as energy density doubles every 10 years, so does the TWh output of the Terrafactory....

This is an important insight, making the transition to EVs less costly than it would appear, since most people assume 200 Gigafactories will be needed to supply world auto production. Even Elon has said 200 Gigafactories if I remember correctly.

GSP
 
Err, because the "terrafactory" suggests that they are making dirt, while "terafactory" suggests that they are making terawatt-hour scale batteries, i.e. 1000x more than a gigafactory. I'll unfix-it, if you feel strongly about it.

Yes, please. This is work of fiction intended to get us thinking on an Earth size scale. So I am talking about the Earthfactory, the Terrafactory, not merely a terafactory. JB says none of us are thinking big enough yet.
 
Maybe the
Terra sized Terafactory
then
Not quite. The Terrafactory already exists. You and I live and work in it. We breathe its air and drink its water. From dust we come, and to dust we return. The Earth is the Terrafactory.

As for automobiles, I am saying that about 20 gigafactories may be all we will ever need. So long as we find better ways to pack electrons into a fistfull of earth.
 
There are a number of implications of this theory that I think are of strategic importance.

1. The global investment in gigafactories will not be on order of $1T, but on order of $100B plus battery R&D nreed to double energy density every 10 years.

2. It will not be necessary for Tesla or competitors to add incremental gigafactories at an exponential pace to achieve exponential growth in GWh production. Rather linear growth in the number of gigafactories will suffice. Exponential growth in the technology is enough to assure exponential growth in output. (Consider that Intel is not doubling its number of factories every two years.)

3. Time is of the essence. If the planet only needs about 20 gigafactories ever, then Tesla secures a lasting advantage by moving quickly to lock in its share of the 20 gigafactories. Roughly speaking every gigafactory Tesla's builds could lock in about 5% of the future auto market. Moreover, the first gigafactories will enjoy a higher return on investment. Gigafactory 1 will enjoy a 30% reduction in manufacturing costs. This will drive down the price per kWh so subsequent gigafactories will have diminishing returns until there is no longer enough ROI to justify adding anymore factories. In fact, this is why only 20 will ever be needed. Automakers wait to long to build their own gigafactories may find it impossible ever to raise the capital to do so. 2030 may be too late.

The race to build the controlling 20 gigafactories is on. Tesla knows this, meanwhile the competition is playing compliance games with fuel cells.
 
We New Englandahs ah always dropping awe ahs. (Actually, we practice the "law of conservation of Rs." All those Rs lost when we pahk the cah are regained when we have a new idear for a sailing regattar.)

LOL! Yes, everyone likes to either drop letters (Americans drop U's like they've got clap - the U's, not the Americans) or collect them according to the time of day.