4 ms·

A wild thought just hit me: if batteries and solar get cheap enough, could the power grid as we know it become at least partially obsolete? The power grid is t

by api 2y ago

A wild thought just hit me: if batteries and solar get cheap enough, could the power grid as we know it become at least partially obsolete? The power grid is the greatest machine ever built by humans. It's gigantic, elaborate, amazingly reliable, and civilization as we know it would collapse without it. It's so ubiquitous that it's hard to imagine it not being here save in a "Mad Max" or "The Road" type collapse scenario. It makes me think of this link that was posted here a bit ago: http://cityinfrastructure.com/single.php?d=RuralOutsidePlant&t=Rural%20Outside%20Plant http://cityinfrastructure.com/single.php?d=RuralOutsidePlant... The analog "POTS" (plain old telephone system) would have seemed like the power grid to me as a kid and a teenager. It seemed that way up until the 2000s when the runaway growth of packet switching digital networks and cellular rendered it completely obsolete. It's been a while since I've even seen a POTS phone line. Could the power grid be reduced to something you only see in big cities and industrial areas where you have tons of demand? Could it eventually go away entirely, or go away "as we know it" in favor of a bunch of lateral load-sharing links between independent mini power plants? It's interesting to think about.

realusername 2y ago
For that to happen, you would need an order of magnitude better batteries than we already have. I'm not saying it's impossible but it's off the cards with the current technology.
jl6 2y ago
The copper cables that POTS ran on are still in widespread use, and their replacement is mostly wired cable/fiber, not wireless-only. Power will probably follow a similar pattern: gridless renewables for some use cases, wired power for others.
danw1979 2y ago
Microgrids have been making a lot of noise for a few years now. I think it’ll still be common to get your electricity from the grid in most towns and cities in the future, especially in northern Europe; we just don’t have the levels of insolation to generate all the energy we need from the small plots of land we live on in the UK in winter.
grecy 2y ago
Of course it can and will happen, just like our gargantuan infrastructure to extract, refine and distribute gasoline and diesel in immense quantities. It's not going to happen in 10 years, but it will slowly happen until the power grid and gas/diesel are relics from the past, much like landline phones and faxes now. They'll still exist, but only used in very odd ball cases for specific reasons. In 100 years they'll be gone, or so small as to be mostly forgotten about.
pjc50 2y ago
Cities are going to need more power than falls on their own roofs. The distributed nature of the grid is just too useful. It's like having a computer that's not connected to the Internet: it's certainly possible but inconvenient compared to the default.
- bryanlarsen 2y ago
  > Cities are going to need more power than falls on their own roofs. The US has >100,000 square miles of cities. Estimates are that ~20,000 square miles of solar panels would be needed to power the US.
  - happyopossum 2y ago
    And what percentage of those cities is actually rooftops? Looking at a few satellite images, it seems like it’s in the low single digits. Lots of parks, roads, and other uses…
himinlomax 2y ago
> if batteries and solar get cheap enough Solar is cheap enough. Batteries are never going to be cheap enough (unless, of course, some disruptive technology emerges.)
- happyopossum 2y ago
  > Solar is cheap enough Until you actually have to pay for someone to install it and connect it to the grid - real prices in a lot of markets haven’t really come down much at all over the past 10 years, the cheaper panel costs are offset by higher labor, interconnect, and other material costs.
- rjprins 2y ago
  Don't underestimate the price efficiencies of globalized industry.
- imtringued 2y ago
  Batteries have a technological learning curve similar to semiconductors and solar panels, but with a much bigger time constant. If my memory serves me well, then capacity per price (or was it by density?) is doubling every eight years.
- pfdietz 2y ago
  Batteries aren't cheap enough to use as a the sole storage modality, because they suck for very long term storage. Their cost per unit of energy storage capacity is not amortized over enough charge/discharge cycles. The answer to that is use a different (typically less efficient) way to store energy in that case. Cost savings can be dramatic at scale. The strawman argument that storage isn't feasible because batteries by themselves don't cut it is regularly seen from nuclear bros. Even MIT came out with a study doing this, a study that very carefully avoided the word "hydrogen".
elp 2y ago
> A wild thought just hit me: if batteries and solar get cheap enough, could the power grid as we know it become at least partially obsolete? No. Its definitely part of the solution but there are some issues that make it impractical as a full replacement. In summer my panels make more power than I know what to do with and my batteries are full by about 10am. In winter only about 30% of my power comes from my solar. Heavy cloud cover typically takes my solar down to about 80% of what I get on a sunny day. Falling battery prices will help to cope with rainy days but the winter drop is hard to avoid and running your own generator is VERY inefficient (i.e expensive) compared to a power station.
- pjc50 2y ago
  Yes - everyone talks about diurnal variation with solar panels, but the seasonal variation is more important, especially at high latitudes. Up here at 56N I get an annual variation that looks very dramatic: https://flatline.org.uk/daystats.html https://flatline.org.uk/daystats.html

by api 2y ago

realusername 2y ago

For that to happen, you would need an order of magnitude better batteries than we already have. I'm not saying it's impossible but it's off the cards with the current technology.

jl6 2y ago

The copper cables that POTS ran on are still in widespread use, and their replacement is mostly wired cable/fiber, not wireless-only. Power will probably follow a similar pattern: gridless renewables for some use cases, wired power for others.

danw1979 2y ago

Microgrids have been making a lot of noise for a few years now. I think it’ll still be common to get your electricity from the grid in most towns and cities in the future, especially in northern Europe; we just don’t have the levels of insolation to generate all the energy we need from the small plots of land we live on in the UK in winter.

grecy 2y ago

Of course it can and will happen, just like our gargantuan infrastructure to extract, refine and distribute gasoline and diesel in immense quantities. It's not going to happen in 10 years, but it will slowly happen until the power grid and gas/diesel are relics from the past, much like landline phones and faxes now. They'll still exist, but only used in very odd ball cases for specific reasons. In 100 years they'll be gone, or so small as to be mostly forgotten about.

pjc50 2y ago

Cities are going to need more power than falls on their own roofs. The distributed nature of the grid is just too useful. It's like having a computer that's not connected to the Internet: it's certainly possible but inconvenient compared to the default.

bryanlarsen 2y ago
> Cities are going to need more power than falls on their own roofs. The US has >100,000 square miles of cities. Estimates are that ~20,000 square miles of solar panels would be needed to power the US.
- happyopossum 2y ago
  And what percentage of those cities is actually rooftops? Looking at a few satellite images, it seems like it’s in the low single digits. Lots of parks, roads, and other uses…

himinlomax 2y ago

> if batteries and solar get cheap enough Solar is cheap enough. Batteries are never going to be cheap enough (unless, of course, some disruptive technology emerges.)

happyopossum 2y ago
> Solar is cheap enough Until you actually have to pay for someone to install it and connect it to the grid - real prices in a lot of markets haven’t really come down much at all over the past 10 years, the cheaper panel costs are offset by higher labor, interconnect, and other material costs.
rjprins 2y ago
Don't underestimate the price efficiencies of globalized industry.
imtringued 2y ago
Batteries have a technological learning curve similar to semiconductors and solar panels, but with a much bigger time constant. If my memory serves me well, then capacity per price (or was it by density?) is doubling every eight years.
pfdietz 2y ago
Batteries aren't cheap enough to use as a the sole storage modality, because they suck for very long term storage. Their cost per unit of energy storage capacity is not amortized over enough charge/discharge cycles. The answer to that is use a different (typically less efficient) way to store energy in that case. Cost savings can be dramatic at scale. The strawman argument that storage isn't feasible because batteries by themselves don't cut it is regularly seen from nuclear bros. Even MIT came out with a study doing this, a study that very carefully avoided the word "hydrogen".

elp 2y ago

> A wild thought just hit me: if batteries and solar get cheap enough, could the power grid as we know it become at least partially obsolete? No. Its definitely part of the solution but there are some issues that make it impractical as a full replacement. In summer my panels make more power than I know what to do with and my batteries are full by about 10am. In winter only about 30% of my power comes from my solar. Heavy cloud cover typically takes my solar down to about 80% of what I get on a sunny day. Falling battery prices will help to cope with rainy days but the winter drop is hard to avoid and running your own generator is VERY inefficient (i.e expensive) compared to a power station.

pjc50 2y ago
Yes - everyone talks about diurnal variation with solar panels, but the seasonal variation is more important, especially at high latitudes. Up here at 56N I get an annual variation that looks very dramatic: https://flatline.org.uk/daystats.html https://flatline.org.uk/daystats.html