3 ms·

Not following how this would work. So overnight demand of the NW and CAL looks to be ~60GW. Assuming you can get 20GW of hydro between the two (which is a bit a

by erentz 5y ago

Not following how this would work. So overnight demand of the NW and CAL looks to be ~60GW. Assuming you can get 20GW of hydro between the two (which is a bit above current peaks) as your supplement. So you'd need enough solar to on the cloudiest of days to both keep things running with daytime demand in NW and CAL while still providing enough excess to (picking some numbers) charge up ~30% of the batteries, while the hydro runs to supplement charging the other ~30% of charge, so that overnight the 60% of charge can be drained in conjunction with the 20GW of hydro continuing to run. Why even rely on hydro if solar can still do all this, just build even more solar. Seems shaky and fragile. And this is before you even get to solving the rest of North America.

Retric 5y ago
You need to expand the maximum capacity of hydroelectric generation not the average capacity which means zero new dams. As a sanity check if hydro can average 7% of US electricity 100% of the time that’s equivalent to 70% of US electricity 10% of the time. To greatly simplify if daytime demand is 200GW and nighttime demand is 100GW. You build enough solar to in theory provide 600GW in optimal conditions and 60GW in the worst case accepting most of that is wasted. Still 90% of the time they provide 300+GW in the day which can charge batteries to provide 100GW at night. Which means 90% of days hydro is producing ~0 electricity. The other 10% of the time solar is producing between 300GW and 60GW in the day. If it’s 299GW then hydro needs to supplement 1GW at night. If it’s only 60GW then hydro needs to supplement 140GW in the day and 100GW that night. Now in the real world things aren’t that simple. Hydro isn’t a light switch, and wind is part of the equation etc, but as a first order approximation it works especially as you can use hydro to charge the same batteries that solar or wind would.
- erentz 5y ago
  You say expand to 70% for 10% of the time. You mean you want to multiply several times over the number of generator halls, penstocks, turbines, generators, and water outflow from existing hydro schemes? There isn't the space for all that. And the cost would be astronomical. Plus the environmental impact of having several times the outflow from most dams at full power would be incredible. You could maybe at a stretch 1.5x it across the US on average. But most plans to expand hydro scheme output around the developed world get met with incredible resistance.
  - Retric 5y ago
    Increasing peak flow from dams is actually great for the environment as flooding is part of the natural ecosystem, this actually doesn’t go far enough to simulate natural variability. Also, dams don’t operate anywhere close to 100% 24/7, what I am describing is generally closer to a 2x increase in maximum generating capacity than 10x which isn’t free but the value of electricity produced would be vastly high more than making up for these costs. I don’t get objections like this: There isn't the space for all that. That’s not an inherent limitation, people built these dams in the first place, we can remove them and build a new structure if that’s what it takes. Don’t flood properties downstream is a perfectly reasonable limitation, don’t build any new infrastructure isn’t. 2019 we produced 100x as much solar power as in 2009, change happens.

by erentz 5y ago

Retric 5y ago

You need to expand the maximum capacity of hydroelectric generation not the average capacity which means zero new dams. As a sanity check if hydro can average 7% of US electricity 100% of the time that’s equivalent to 70% of US electricity 10% of the time. To greatly simplify if daytime demand is 200GW and nighttime demand is 100GW. You build enough solar to in theory provide 600GW in optimal conditions and 60GW in the worst case accepting most of that is wasted. Still 90% of the time they provide 300+GW in the day which can charge batteries to provide 100GW at night. Which means 90% of days hydro is producing ~0 electricity. The other 10% of the time solar is producing between 300GW and 60GW in the day. If it’s 299GW then hydro needs to supplement 1GW at night. If it’s only 60GW then hydro needs to supplement 140GW in the day and 100GW that night. Now in the real world things aren’t that simple. Hydro isn’t a light switch, and wind is part of the equation etc, but as a first order approximation it works especially as you can use hydro to charge the same batteries that solar or wind would.

erentz 5y ago
You say expand to 70% for 10% of the time. You mean you want to multiply several times over the number of generator halls, penstocks, turbines, generators, and water outflow from existing hydro schemes? There isn't the space for all that. And the cost would be astronomical. Plus the environmental impact of having several times the outflow from most dams at full power would be incredible. You could maybe at a stretch 1.5x it across the US on average. But most plans to expand hydro scheme output around the developed world get met with incredible resistance.
- Retric 5y ago
  Increasing peak flow from dams is actually great for the environment as flooding is part of the natural ecosystem, this actually doesn’t go far enough to simulate natural variability. Also, dams don’t operate anywhere close to 100% 24/7, what I am describing is generally closer to a 2x increase in maximum generating capacity than 10x which isn’t free but the value of electricity produced would be vastly high more than making up for these costs. I don’t get objections like this: There isn't the space for all that. That’s not an inherent limitation, people built these dams in the first place, we can remove them and build a new structure if that’s what it takes. Don’t flood properties downstream is a perfectly reasonable limitation, don’t build any new infrastructure isn’t. 2019 we produced 100x as much solar power as in 2009, change happens.