81–90 of 112 posts
Earlier quoted context omitted.
Microsoft committed to purchase the plant's capacity for 20 years. And US electricity demand grew very slowly from 2005 to 2020. It is growing rapidly now.
At around $110/MWh, according to the article. This is about 50% higher than what utility-scale PV or wind would cost. Guess they're using OpenAI accounting.
If you were to step into the control room you’d see analog phones, tiny incandescent bulbs behind plastic covers… looks like a sci-fi set from the 60s.
The expensive part of a reactor isn’t really the reactor or tech itself, it’s the government regulation from the DOE and NRC.
I worked at Areva/Framatome/B&W and IIRC they still have the archival room where hundreds of 4 inch D ring binders held the original design docs that had to be submitted for approval.
Earlier quoted context omitted.
In most parts of the country, solar plus batteries to get through 24 hours will be cheaper than $110/MWh.
Do you have a source for that, when i googled it came up closer to $200/MWh for new york, but that was from older sources. The only thing i saw approaching this price point was if you were somewhere like las vegas. I also think you would need more than 24 hours battery. You have to prepare for freak weather events that reduce system capacity. I also wonder what time horizon we are talking. solar and batteries presuma…
Civilian nuclear reactors replace fuel gradually over 3 to 6 years typically. 20 year old solar panels work now. New solar panels are expected to work over 30 years. Utility scale lithium ion batteries are expected to last 10 to 15 years.
[1] https://www.lazard.com/media/5tlbhyla/lazards-lcoeplus-june-...
Earlier quoted context omitted.
"they have maintenance outages like anything else" not often and most importantly they are PREDICTABLE. You do understand why being able to control when a power plant is operating is a very important thing, right?
i thought the conversation was regarding utilization of capital, in which case 80% is 80%, predictability doesnt change the fact you have to let GPUs sit idle 20% of the time. I guess if I knew there would be two months with less power I might design my data center to fit into 40 foot containers so I could deploy wherever power and latency are cheapest
This is very interesting to me because a plant this old might be cheaper to operate than a new plant, but might be like the space shuttle in that replacement parts aren’t readily available and thus expensive to custom manufacture. If you were to step into the control room you’d see analog phones, tiny incandescent bulbs behind plastic covers… looks like a sci-fi set from the 60s. The expensive part of a reactor isn’t…
Not too disagree with the bulk of comment, but this sentence is not true. They're 90V neon indicator lamps, a technology that's really cool but also so inefficient that people rip it out and replace it.
Neat rabbit hole to do down.
Earlier quoted context omitted.
> There's not much to go off of on this subject in the US The main problem is that things cost more per unit if you do them less. The first new reactor in decades is going to be stupid expensive because you have new people doing it who are learning things for the first time, which often means doing them over again, which is expensive. And then we didn't even get to see if the second unit at Vogtle could improve on th…
Given the nuclear industry’s history of negative learning by doing, even during the boom years, I have a hard time seeing how ”scale” will solve it. https://www.sciencedirect.com/science/article/abs/pii/S03014...
Whereas what you want is multiple companies building these things on an assembly line, where you can plop the components out of a truck already approved and ready to be turned on.
This is very interesting to me because a plant this old might be cheaper to operate than a new plant, but might be like the space shuttle in that replacement parts aren’t readily available and thus expensive to custom manufacture. If you were to step into the control room you’d see analog phones, tiny incandescent bulbs behind plastic covers… looks like a sci-fi set from the 60s. The expensive part of a reactor isn’t…
> tiny incandescent bulbs behind plastic covers… Not too disagree with the bulk of comment, but this sentence is not true. They're 90V neon indicator lamps, a technology that's really cool but also so inefficient that people rip it out and replace it. Neat rabbit hole to do down.
>That’s cheaper than a brand-new nuclear power plant would cost, but it’s a hefty premium over wind, solar, and geothermal, according to a comparison of energy costs from Lazard. Nuclear is more expensive because there are extensive regulations. "Green" energy not only does not face so many regulations but it benefits from incentives. Also, when comparing nuclear with "green" energy, most studies don't take into acco…
Studies also don't take into account that by subsidizing solar you're subsidizing an industry that's already subsidized by the Chinese gov, it's cool in the short term but we're making the same mistake we've been making for the last 50 years Home grown nuclear programs will always be better than solar propped up by foreign entities.
OTOH, cheap energy enabled by cheap solar panels is a huge benefit to any economy.
Earlier quoted context omitted.
Maybe the loan is cheaper than using their own money.
More critically: just because a company is worth $1 trillion doesn't mean it has anywhere close to $1 billion in cash at all, let alone able to be earmarked to a given project (at the presumed exclusion of other projects). Granted, a project like this probably doesn't strictly need all $1 billion all at once, but I'd argue it's better to get whatever necessary funding upfront instead of risking having sunk a partial…
My assumption is that the real reason this is a loan from the government and not paid directly by Microsoft has to do with other factors, like Microsoft not waiting to be in the hook for the billion dollars if the partner company folds, or the potential for loan forgiveness, or other incentives that make the effectively loan cheaper than cash.
Earlier quoted context omitted.
At around $110/MWh, according to the article. This is about 50% higher than what utility-scale PV or wind would cost. Guess they're using OpenAI accounting.
Nuclear power is reliable 24/7 while wind and solar are not and handling this costs money. Microsoft has said that they have more GPUs than electricity to run them so even at $110/MWh it makes sense for them.