I get where you're coming from, but let me correct some of your numbers for other readers:
> you replace bricks every couple hundred years, but solar panels every 10–20, right?
Old panels perhaps, but modern solar panels come with performance warranties that guarantee they will be producing >85% of their initial output after 30 yrs.
> 2000W solar panel is like $300 or something and it's half the size of a door
2000W solar panels generally don't exist, so I assume that's a typo for 200W? Modern utility scale panels top out at ~700W with dimensions of 2.4 m x 1.3 m, however rooftop panels for commercial buildings are in the 500W range and ~ 2 x 1 m (so yeah about a door). International wholesale prices for these from Tier 1 manufacturers are now < $0.10 USD / W (although from what I understand more expensive in the USA).
Thanks for the work you've done with ShadeMap - I used this extensively when I we were looking for somewhere to rent, as living in hilly city some areas lose the sun quite quickly. Happy to say we are now living in a place that gets plenty of sun, and this summer has yielded a lot of tomatoes in a city where that can be difficult.
Yeah when it comes to these discussions, I think most people are unaware just how much the cost of solar & BESS (at the large scale) has fallen in the past couple years.
The battery installed along solar farms is usually on the order of hours of energy delivery though. Like 5 hours. It’s fine, but not everyone will be able to ornaffors to buy electricity on when it is cloudy and still for a few days.
I love how your response conveniently ignores any mention of energy storage technologies, several types of which are having rapid technological advancement.
You’re right that there has been huge advancements but there’s still a pretty brutal economic cost with it. I generally use the Tesla Megapacks as an easy example because prices and specs are easy to find on Wikipedia (and my local utility is testing them, so it’s relevant to me).
A Megapack 2XL can output ~2MW and has a capacity of ~4MWh for $1.39M.
A GE BWRX-300 is rated for 300MW and an 18-24 month refueling cycle and allegedly costs ~$1B.
You can build 150x Megapacks for $208M to match that 300MW output, but there is only enough energy stored to provide that output for two hours. If you want to provide 12 hours capacity (to run through the night), you need 900 units at a cost of $1.25B. That’s just for the storage though, you still need the source of electricity to charge the packs, overprovisioned to deal with the capacity factor issues that solar and wind have.
Will the nuclear plant go over budget? Almost certainly. Will it then provide a long-term baseline source of power? Also yes.
I’m pro-renewables and pro-storage, but there’s a mix needed here. Even with storage tech, there needs to be something else that can just sit there and run and produce reliable and controllable power output long-term.
Right, so that means that to produce 300MW reliably from the solar or wind farms you'd need about 900MW nameplate capacity. I'd be really curious about the solar side of it too and whether that 30% is overall or just during daylight. Either way, you end up having to overprovision the unreliable sources such that you have enough capacity to both charge your battery pack and provide power to the grid.
30% overall, including night time. Honestly seems high to me and makes me wonder if they have overprovisioned panels vs the maximum allowable injected power to get a better capacity factor.
I keep wondering what the long-game (if any) of LLMs is... to make the world dependent on various models then jack the rates up to cover the costs? The gravy-train of SV funding has to end eventually... right?
Yep I second this, I usually find out that I've been invited to an event when someone makes a post in the event, and often not from the event invite itself.
reply