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

Well first of all energy storage, is not production, but as you asked:

1) Reduced need for energy storage. Right to repair for everything, write code that is efficient and lower power, distributed systems which don't require complex centralized systems to run. Taxes on unused compute cycles to help create incentives for this, perhaps.

2) For actual energy storage, something like the sand heat system recently put into use in Scandinavia, or the mechanical earth dams (store energy in potential energy mass, less dangerous than an actual water dam, a lot easier to build). For immediate electric storage at scale you can do e.g. saline water storage tanks which hold mild electric charge, who knows maybe there is some inert chemistry which could be devised for a safer transportable version of a lithium ion battery...

3) For energy production, I am a long time advocate of geothermal. There's no real downside, besides digging holes and I guess maybe a well collapse, but you're limited to loss of whatever is in the whole/immediate surrounding in the case of a cave in, there are no engineering problems to solve except pumping water around, which is a well known task. Solar/wind for ships, airplanes, space vehicles, electric/hydrogen for storage/consumption scenarios where the grid is not accessible (remote locations e.g. the poles, alaska, siberia, African/Asian planes)



> mechanical earth dams (store energy in potential energy mass, less dangerous than an actual water dam, a lot easier to build)

a lot easier to build?

This claim seems wrong, given that pumped-storage hydroelectricity is in actual use, when this is purely theoretical.

"mechanical earth dams" gives info about water dams - is it existing even as a theoretic design?


https://www.energyvault.com/gravity

Its a company with as I understand it proven designs.

Besides, the concept is trivial and applicable and replicable by almost anyone. I don't understand why there is so much skepticism around these things...


> Besides, the concept is trivial and applicable and replicable by almost anyone.

Has anyone made one operating at major scale? And comparable to say serious pumped storage? At their page I see only "demonstration unit".

I suspect that either it is not solved, expensive or not trivial. And not replicable or there is nothing to replicate or not worth replicating.

For comparison https://pl.wikipedia.org/wiki/Elektrownia_Por%C4%85bka-%C5%B... can operate at 500MW for 4 hours, was build in 1979 and is a normal pumped storage battery (large enough to be classified, operated and treated as a power plant).


Nuclear energy as a concept is trivial. There are a lot of details between "concept is trivial" and "it's practical at scale"


> write code that is efficient and lower power, distributed systems which don't require complex centralized systems to run.

huh?


This is a really lazy response, but given the possibility that you genuinely don't understand the gp: Writing better software and reducing the (currently massive) waste of energy on enterprise bullshit and cryptomining would have a meaningful impact on international energy usage, and so help climate change.


It's marginal compared to the other power expenses to be sure, but computing is another rising power cost. Besides my AC, and cooking, I don't have any regular power expenditures other than digital devices, so it seems reasonable to me to want to optimize power expenditure there.

Regarding the distributed vs centralized, the reasoning is large data centers are inefficient and could be replaced mostly with local, low power systems which are barely on at all, versus constant-on, constant-ready server rack systems.




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

Search: