Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I wonder why they don't see small modular nuclear reactors as a part of the solution. Also, they mention iron-air chemistry for batteries, but don't consider them in the calculations. Their approach seems to be pragmatic; utilizing and mega-scaling tech that is available right now.


Small nuclear doesn’t seem to be ready to implement. Best estimates say 2030 earliest for commercial implementation.

https://www.energymonitor.ai/sectors/power/small-modular-rea...

I can’t seem to find anything that indicates otherwise.


All SMR companies suffer from economic constraints. If they got really investment and commitment to build many, it would be much different. It also requires a more engaged faster regulator.

And we could also just do normal nuclear until then as well.

Also 2030s isn't to late for anything.


Yes nuclear was notably missing. Also the „Li-ion (4h-8h)“ range for storage is doing a lot of work. You can’t operate a grid with intermittent energy production on a 4h buffer, 8h I‘m not sure. Unless they include nuclear I‘m assuming this compares unfavorably.


> You can’t operate a grid with intermittent energy production on a 4h buffer,

With good electricity transmission you can.

Infact, in this plan they specifically model this, and it comes out that by overbuilding solar/wind 32% you are fine.

> here is an economic tradeoff between building excess renewable generation capacity, building grid storage, or expanding transmission capability. That tradeoff may evolve as grid storage technologies mature, but with the assumptions modeled, the optimal generation and storage portfolio resulted in 32% curtailment

and the footnote:

> Currently, the transmission capacity is <1% of the combined regional peak loads (with transmission to/from Texas the lowest). Higher transmission capacities generally reduce the total generation and storage buildout, but there is an economic tradeoff between building more transmission and building more generation plus storage.


Am I misunderstanding this or does this make the whole grid a lot more vulnerable? The internet was once designed to be resilient, economic forces have made it centralized in large network exchanges and thus vulnerable. We did the same with our supply chains, and it backfired so badly during covid that we're still not fully recovered. Should we do the same with our electricity supply?


The gas generators don't evaporate. Using them for a week once a decade is a negligible amount of emissions and paying the entire staff to just mime the other 99.5% of the time on top of a renewable grid is still vastly cheaper.


What's more reliable: two massive Sun Enterprise servers, or a 100 PC's?


Nuclear is neither all that reliable nor is it dispatchable.

And yes you can with a small amount of dispatchable energy: https://www.nature.com/articles/s41467-021-26355-z


Small modular reactors don't actually seem to solve nuclear power issues to me. It just punts the regulatory regime away from the manufacturer/builder to the installer. I don't really understand the hype behind modular reactors though so perhaps there's something I'm missing. It just seems like another form of venture capital led snake oil selling. Reminds me of the hype around ridiculous ventures like Virgin Orbit/Galactic that are now failing.


I think they are going to fail if used for general power. Small reactor have uses, primarily remote locations that can't use solar or wind. Remote towns in Arctic and Anarctic that currently import and burn diesel. SMR would be be expensive but reliable option.


I suppose it makes sense in that case, but that hardly makes a business case if your customer base is small and limited.


I'd imagine because of Gates being involved in it and Elon being childish about it.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: