Hacker News Reader

> 1. Long distance transmission lines.

Those are really expensive. They're part of the toolbox, but they're not tool #1.

> 2. Some type of "clean, firm, dispatchable" power. Examples include: Nuclear fission, fusion power, deep geothermal, and space based solar power.

If you're relying on that to supply power during those winter weeks without sun & wind then it has to scale up to 100% of power needs. And if it can do that, why build anything else?

To get to 100% carbon free with > 99.99% reliability for under $1T, your primary tool is modelling.

Then you reach for:

- source diversity. Wind is more expensive than solar, but it tends to be highest at dawn/dusk so is a great complement. - overprovisioning. Enough solar to supply needs on a cloudy winter day - storage. - long distance interconnect. There's never been an hour in recorded history where there's no sun or wind somewhere in the continental US.

https://mitpress.mit.edu/9780262545044/electrify/

> There's never been an hour in recorded history where there's no sun or wind somewhere in the continental US.

But is that sufficient to handle the full load across the entire continental US? And how do you do that without the really expensive long distance high voltage transmission lines?

Where I live, bad winters can see us go for weeks of full cloud cover and little wind in January. If we really get away from fossil fuels and run heat pumps, that means electrical use in winter will rival that in summer.

No it isn't. That's why I said that modelling is tool #1. The whole US might not go an hour without sun and wind, but your area might go 3 weeks. But the combination of your area and a neighboring area might max at 3 days. So thus instead of building a continent wide interconnect and no storage, you build a regional one and 3 days of storage.

the sun shines and the wind blows in the winter. Plus, batteries. Giant redox flow batteries are coming online now, sodium batteries, it's not like there aren't options for storage people are working on.

> Clean, firm, dispatchable power

Besides the examples you listed, there's also synthetic fuels. I don't know if they'll pan out, but the concept is intriguing.

Essentially, the argument goes that there's a critical solar price point at which synthesizing methane from atmospheric gas capture becomes cheaper than drilling. Said methane can be burned for power in existing plants (forming a closed cycle) or refined into heavier liquid hydrocarbons for vehicles and polymers.

The advantage here is that you don't need batteries or inverters - just dirt cheap panels - and the synthesis plants can be engineered to be productive despite only operating during the day.

I know one company is working on this with industrial scale in mind (Terraform Industries), and I believe SpaceX is also pursuing it on-site for Starship (which consumes ~1000 T of methane per launch, all of which currently has to be trucked in at great expense.)

I wonder if this explains why Prometheus Fuels decided to do methane…

Probably. Methane is easier to synthesize than fuel alcohols, and has better synergy with existing infrastructure (dead simple, highly responsive gas plants don't care where their fuel comes from, after all - they'll just take the cheapest option.)

> So you don't have to go beyond 75% renewables to start with.

I think the 75% aggregate over some period. If 25% of your total capacity is nuclear/hydro you will still have extreme shortages during peak times if there is no sun/wind.

That why it has to be gas/etc. which can be scaled up and down very rapidly (unfortunately you can’t “overload” a nuclear reactor to make it generate more power for a few hours on a regular basis..)

>(unfortunately you can’t “overload” a nuclear reactor to make it generate more power for a few hours on a regular basis..

You could throw excess power away from an oversized reactor and not throw it away when it's needed. Financially not very smart, but technologically feasible

Bitcoin mining is doing exactly in some cases. E.g. there's a fair number of small, remote, hydroelectric plants whose construction and/or renovation were financed by Bitcoin mining, with the amount of mining declining over time as the community using them grew and ramped up consumption. Also in Canada there's a company working on making finance deals for small scale nuclear energy plants in the far north where the excess capacity will be temporarily used for Bitcoin mining while the community grows and/or industrial uses like mines ramp up production.

Re: the nuclear version, good chance none of it happens due to anti-nuclear sentiment of course. So far exactly zero of these small scale nuclear plants have been built.

Too little, too late. We should've been switching to solar in the 80s. Even if we could switch to be carbon free tomorrow, the amount of CO2 already in the atmosphere is predicted to cause breadbasket collapses within the next 20 years.

If it makes folks feel better, there's a good chance you probably had no control/influence over this outcome if you were born after 1980.

https://www.sciencealert.com/researchers-weve-underestimated...

We were trying to switch to solar in the 80s, but it was infeasible. The technology just wasn't there. Now it is and we're adopting it en masse.

We will spray the atmosphere to buy more time.

People will haggle over it because of the unknowns, but when imminent social chaos becomes obvious, we'll be forced to pull the trigger on it.

It's not hopeless. The risk of crop failures may be higher than it would be if we were going to experience less warming, but having a bad harvest year isn't existential. We'll work to mitigate things.

This is why I think we need to roll the dice on geoengineering. We can try to tilt the odds in our favor, but it's still a crapshoot. From what I've read, iron fertilization would be one of the better paths to go. A potentially better path would be the creation of synthetic whale poop.

https://www.smithsonianmag.com/innovation/scientists-are-cra...