Hacker News new | ask | show | jobs
by Retric 832 days ago
You double dipped, use outdated numbers "in 2020", hand waved, and still didn't get orders of magnitude difference.

Pairing batteries with solar provides actual useful power from solar for most of the day. You don't need both 24h of batteries AND redundant solar farms to get 24h of energy. Extremely redundant solar costs less per kWh. Therefore a combined system will cost less, provide more useful power, and charge the batteries at zero additional cost. I'll stick with batteries alone, but include that zero charge cost as part of a cheaper overall system.

Construction costs vary by country so you need a country with both battery systems and nuclear to get an apples to apples comparison. For the US "In 2023 costs had increased to $34 billion, with work still to be completed on Vogtle 4.", with work still to be completed on Vogtle 4, and that's just construction on two 1.1GW reactors. https://en.wikipedia.org/wiki/Vogtle_Electric_Generating_Pla...

400MW * 6 = 2.4 GW, 4h * 6 = 24h. 36 * 560m = 20 billion so well under that 34+B even before considering the cost savings from solar. We are comparing with Nuclear at 2.2GW which again goes offline for long periods, but we care about orders of magnitude so redundancy is a non issue.

Now you need to consider operating costs for both systems. In constant dollars without subsidies construction works out to roughly 1/3 of nuclear total lifetime costs ~1000 workers + fuel + insurance + new equipment + decommissioning adds up. I'll be conservative and double that 34 billion instead so 68B.

Actual studies look into foretasted costs. However for orders of magnitude 5%/year of install costs per year in maintenance would be equivalent to creating redundant facilities to cover 50% degradation in 10 years, so that's conservative as swapping out batteries would lower costs here. Lifetime costs should therefore be below 20B + 20B * 5% * 50 years = 70B ie roughly the same as Nuclear per kWh and far more flexible.

Now, you can easily quibble about these numbers but your not getting an orders of magnitude difference here.

1 comments

> You double dipped, use outdated numbers, hand waved, and still didn't get orders of magnitude difference.

Feel free to provide better numbers. Those are the two largest most recent projects. That those numbers disagree with what studies say they should be aren't a problem with the numbers, but problems with the studies.

Also you should read up on the difference between power and energy since you're confusing the two in pretty much every line of your post. Watts aren't joules and joules aren't watts.

> Feel free to provide better numbers.

I did provide a US example for an apples to apples comparison. Providing lower numbers is irrelevant when higher numbers already prove the point.

> Also you should read up on the difference between power and energy since you're confusing the two in pretty much every line of your post. Watts aren't joules and joules aren't watts.

Don't hand wave, pick a single example.

If you don't understand my notation you can simply compare 2 * 1.1 GW nuclear reactors * 24h = 52.8 GWh per day. The battery example you gave was 400MW * 4h = 1.6GWh per day. 52.8 GWh / 1.6 GWh= 33x. I used 36x for a total of 20.16 Billion dollars.