[mschuster91]

> You could also technically use this as a grid-battery, taking in excess grid energy when it is cheap and converting it into natural gas that can be run back through a gas peaking plant that spins up to meet peak demand.

The loss in such a cycle is abysmal, alone from thermal loss (not to mention the loss during compression and decompression) - even straight fuel cells are at 60% round-trip, compared to batteries with >>90% efficiency.

[pydry]

Batteries are good at smoothing out daily or even weekly variations but do not make any sense whatsoever for seasonal power storage.

It's ridiculously cost ineffective to charge a battery in July only to discharge it in December.

60% roundtrip is cost effective if you're synthesizing when the sun is blazing and the wind is blowing hard and burning it when wind, solar and batteries have all tapped out.

Thats especially so if the equipment has low capex which it seems like this does. Unlike batteries that makes it cost effective to overbuild and idle it most of the year.

[ben_w]

> It's ridiculously cost ineffective to charge a battery in July only to discharge it in December

Indeed.

This is why the cheapest solutions in most places are a mix of a mere few days off storage plus a target production level that is a little higher than you need on an average day in winter.

While this doesn't work above the arctic circle (you could do it with a power line somewhere sunnier or a synthetic fuel, and possibly also geothermal or nuclear etc., devil is in the details for all options) overproduction + 35-90 hours of batteries is sufficient for most people and places:

https://tonyseba.com/wp-content/uploads/2020/11/RethinkingEn...

[cbsmith]

Is that more effective than the traditional, "store the energy as kinetic potential energy by pushing water uphill, so you can let it go down the hill later" approach.

[pydry]

For storing energy cheaply for months at a time and transporting it across large distances, yes.

Pumped storage has ~90% roundtrip efficiency, good at storing energy for days or weeks but maxes out easily. The energy density of water pushed uphill is very low.

I think we should be pushing a lot more water uphill, but I see it as an alternative to or competitor to grid-scale batteries and a complement to syngas.

Syngas production will probably be most useful if built next door to a wind or solar farm and used to siphon off energy which is currently curtailed when the grid is maxed out.

It can then be easily stored in enormous quantities and easily transported by ship to anywhere in the world that needs it.

[danmaz74]

It's not more effective where you have the right conditions for pumped storage, but those conditions aren't very common around the world.

[pydry]

I'm curious about where you read this. I see this idea that pumped storage geography is rare pop up a lot on Hacker News but I don't know where it's coming from and it rarely seems to come with citations.

If you look at this map, you'll see that unlike, say, dam-appropriate geography, it's actually extremely common:

https://www.energytransitionpartnership.org/uploads/2023/05/...

[mschuster91]

The problem is, the potential for buildouts of pumped hydro isn't that large any more. In Europe, most usable areas have been built out, and new projects are likely to be denied because anything involving creating dams or bodies of water with rapid differences in water level is incredibly devastating on nature and wildlife.

[cbsmith]

More or less devastating than extracting heavy metals and setting up a battery farm?

[danmaz74]

I'll admit that I wasn't talking about potential green-field sites not linked to rivers or existing reservoirs, as described in your link. How many of the sites they identified as viable with their algorithm would actually be economically, socially and environmentally viable is a big question though. Not saying some of these sites can come to fruition, but for sure the capex and lead time for this kind of projects is huge.

[pydry]

I guess that answers my question. People are getting pumped storage and river dams mixed up. It seems mschuster91 also mixed them up.

Yes, the capex and lead times on one of these things can be huge, but it's comes out ahead of nuclear power on those fronts.

[genman]

Figure 2 doesn't agree with you. It's common but not geographically equally distributed.