over that time period are we looking at replacement issues if the lifespan is 25 yrs?

Boris
London

over that time period are we looking at replacement issues if the lifespan is 25 yrs?

Yes, and I'd actually forgotten about that in my back-of-the-envelope calculation, so thanks for pointing that out.

A 25-year lifespan means 4% yearly replacement; on 1.5M units, that's 60,000 units/yr in the final year, meaning it would dominate the installation of new units. Production would have to increase from 20,000/yr in 2030 to 80,000/yr in 2100, or an average rate of 2%/yr, which seems pretty modest.

That assumes building all-new units, though, which I suspect would be unlikely. I would imagine - and please correct me if I'm wrong - that some parts of wind turbines would wear out much faster than others; the gearing might need to be redone from scratch, for example, whereas the support tower might not have suffered much wear at all, especially for onshore installations. So I would imagine that replacing a wind turbine with an identical unit would be substantially cheaper than building the original, due to re-using and/or recycling parts, and that a wind turbine built in 2070 is likely to last longer than one built in 2007. I'd guess - guess - that the replacement costs in this scenario in 2100 would be no more than the new-unit costs.

For reference, at $2M/MW onshore, a "unit" would be $6M to build today, and 80,000 new units/yr - the maximum possible requirement - would cost $480B, or less than 1% of current world GDP, and around 5% of current world manufacturing capacity. In 2100, servicing such a large installed base of wind turbines won't be a problem; either it won't exist, or its existence will provide enough energy to keep growing the industrial base and make $480B in 2008$ an even more minor part of the world economy than it already is.

Also for reference and roughly speaking, 1.5M units @ 3MW @ 30% CF = 12,000TWh/yr = 6BT of coal @ 33% efficiency = $600B @ $100/ton, meaning that the amount of societal capacity necessary for building and maintaining the wind turbines should be roughly similar to the amount necessary for mining and providing an equivalent amount of coal at today's prices.

Towers and related electrical infrastructure (transmission lines, transformers) for on-shore projects are expected to last at least two generations for on-shore projects.

The best sites are being built out first. This will raise an interesting question, is it better to replace 2008 1 MW turbines with 2033 16 MW turbines or 1 MW turbines.

One thought is to space the new larger turbines up high (more thinly spaced) and replace 15/16s of the old 1 MW WTs with new 1 MW WTs. (Larger turbines are more thinly spaced, their wind shadow is greater).

Best Hopes for more WTs,

Alan