5 ms·

How much copper would you need for a 4,300km 2GW subsea cable?

jillesvangurp 2y ago
That depends. It could be that they use aluminum instead. Apparently that's quite common in HVDC cables and interesting for cost reasons.
ben_w 2y ago
Aluminium is almost as good by cross section, much cheaper, and the global (mainly Chinese) production is sufficient for a global (40 megameter) multi-terrawatt power grid every 18 months or so. And yes, I did do the maths; and also yes it's really just China at the "global terrawatt" scale (they've become a dominant aluminium supplier), but a much smaller distance and power rating is probably fine even if China doesn't sell you the metal.
- bigiain 2y ago
  Interestingly, Australia is amongst the world's biggest exporters of Bauxite, which pretty much just needs electricity to turn it into Aluminium.
Tade0 2y ago
At this length aluminium would have been used instead most likely. With a typical HVDC line not exceeding 1200mm2 conductor cross section it's about 13k tonnes, so 0,025% of global aluminium production.
- raverbashing 2y ago
  > not exceeding 1200m2 conductor cross section Technically right but I think you mean 1200mm2 (a radius of approx 20mm)
  - Tade0 2y ago
    Yes! I actually had an autocorrect suggestion for "1200m2" in there for some reason.
- gdiamos 2y ago
  At $1.127/lb of aluminum that only comes out to a bit less than $30,000 for the raw materials. That's astounding.
  - thehappypm 2y ago
    I think you forgot to multiply pounds / 2000 = tons
  - Tade0 2y ago
    There's a lot more to a HVDC line than just the conductor, but ballpark estimate for a 1GW line is $1mln/km - scales accordingly with power. Still, you could wrap the world around at the equator with this for a paltry $40bln. Now scale that 10x and you have yourself a practical global grid solution for what, $400bln? That's less than half the US military budget - absolutely doable if you get enough economies on board.
55555 2y ago
~65,000 tons of copper, which would cost about $515 million dollars or maybe a billion after being turned into wires. These numbers are from ChatGPT which is good at figuring out amounts needed but useless at figuring out real industrial-scale prices.

That depends. It could be that they use aluminum instead. Apparently that's quite common in HVDC cables and interesting for cost reasons.

ben_w 2y ago

Aluminium is almost as good by cross section, much cheaper, and the global (mainly Chinese) production is sufficient for a global (40 megameter) multi-terrawatt power grid every 18 months or so. And yes, I did do the maths; and also yes it's really just China at the "global terrawatt" scale (they've become a dominant aluminium supplier), but a much smaller distance and power rating is probably fine even if China doesn't sell you the metal.

bigiain 2y ago
Interestingly, Australia is amongst the world's biggest exporters of Bauxite, which pretty much just needs electricity to turn it into Aluminium.

Tade0 2y ago

At this length aluminium would have been used instead most likely. With a typical HVDC line not exceeding 1200mm2 conductor cross section it's about 13k tonnes, so 0,025% of global aluminium production.

raverbashing 2y ago
> not exceeding 1200m2 conductor cross section Technically right but I think you mean 1200mm2 (a radius of approx 20mm)
- Tade0 2y ago
  Yes! I actually had an autocorrect suggestion for "1200m2" in there for some reason.
gdiamos 2y ago
At $1.127/lb of aluminum that only comes out to a bit less than $30,000 for the raw materials. That's astounding.
- thehappypm 2y ago
  I think you forgot to multiply pounds / 2000 = tons
- Tade0 2y ago
  There's a lot more to a HVDC line than just the conductor, but ballpark estimate for a 1GW line is $1mln/km - scales accordingly with power. Still, you could wrap the world around at the equator with this for a paltry $40bln. Now scale that 10x and you have yourself a practical global grid solution for what, $400bln? That's less than half the US military budget - absolutely doable if you get enough economies on board.

55555 2y ago

~65,000 tons of copper, which would cost about $515 million dollars or maybe a billion after being turned into wires. These numbers are from ChatGPT which is good at figuring out amounts needed but useless at figuring out real industrial-scale prices.