Hacker News Reader

Comment by DeRock | Hacker News Reader

Something that doesn't often get brought up in these energy "storage" discussions is the potential for duty cycling existing generation to time shift the duck curve. Eg. California (large solar generation) shares a grid with Washington and British Columbia (large hydro generation). When the sun is shining, hydro reservoirs can be filling, as solar is meeting the demand. When the sun goes down, hydro facilities can duty cycle on and drain the overhead. This requires some flexibility in reservoir levels, as well as the ability to stop/re-spin turbines on a daily cadence. But you eliminate all of the efficiency losses with schemes such as pumped storage (though gain some transmission losses due to the increase length of geographic arbitrage).

The same opportunity exists on the East-West axis. The sun is still shining in California after its set on the East coast.

View on HN

What you're describing sounds a lot like open-loop pumped storage [1]. You don't hear about it much because it's already quite common and has been for decades. There's something like 22GW of capacity in the US alone.

[1] https://www.energy.gov/sites/prod/files/2016/10/f33/Hydropow... (p. 183-195)

It was my understanding that this is more of a seasonal thing than on a day to day basis. That is, during the summer, southern California uses a large amount of electricity for climate control, and so power is transferred along the Pacific DC intertie in a north -> south direction.

In the winter, electricity usage in the north is devoted to heating, whereas in the south temperatures are mild. The result is that power is moved back up the intertie from south -> north. In the spring, snowpack melt refills the reservoirs in the hydroelectric system, allowing it to be ready to provide power to the south again once temperatures start to climb.

I believe it is for the DC interview, but there are also AC interties that may be for hourly fluctuations.

https://en.m.wikipedia.org/wiki/Pacific_DC_Intertie

Cause of this summer's California shortages was, in part, that it was hot up and down the west coast and Oregon didn't have anything to export.

Its my understanding that some hydro facilities do this. One issue I can think of is the losses of transmitting that much electricity from 1 area to another (1000s of miles away). the losses of transport coupled with the losses of "storage" and "generation" make it more expensive. I don't know the actual costs, just ideas I think about.

The losses of transport are relatively small, smaller than would be expected: see http://insideenergy.org/2015/11/06/lost-in-transmission-how-...

Not so much with ultra-high-voltage lines. China has been doing it for years because of the really long distances between electricity production and consumption.

https://en.m.wikipedia.org/wiki/Ultra-high-voltage_electrici...

This is true - but pumped hydro has limitations on where it can be built, meaning it becomes constrained on transmission. So, in theory it is super awesome, but in practice you need to include the cost of many years of litigation to navigate the balkanized US system for permitting power lines. Not to say it's not a good option in many cases, just to say that it's not a panacea.

Somewhat related: I cannot plug The Energy Transition Show (https://xenetwork.org/ets/) podcast enough on subjects like this. It's the only podcast I pay for. Each episode the guest will be some PhD or similar in whatever the subject is, always super in-depth. Highly recommended.

What’s most cost efficient for the duck curve at scale is single axis tracking solar systems. The added cost produces ~zero extra power at noon and significantly more power at sunrise and sunset. It doesn’t completely remove the issue, but you get about 80% of max capacity for about 90 minutes more in the morning and evening.

EX: http://ujsolar.weebly.com/uploads/2/1/0/4/21043846/9545485_o...

Currently the economics favor regular solar systems, but this will change long before large scale grid storage becomes viable.

Don't adjacent solar panels shade be each other? If not, then are the panels spaced less dense than they could be if they were fixed?

Density is lower, but daily kWh of solar power per acre stays surprisingly close. In effect flat solar panels are shading each other to some degree for most of the day. Density therefore becomes just another trade off.

Anyway, as long distance transmission is cheap panels are generally on the cheapest land you can find, often sub 1,000$ per acre. This makes minor differences in density largely meaningless. The added spacing also reduces the local environmental impact as plants can grow around the panels which can significantly reduce dust. They also make better use of limited PV manufacturing capacity.

That said, if you’re actually space limited for whatever reason, flat panels win.

as a counterpoint, nearly every grid-attached power storage system I've seen has been almost entirely described/discussed in terms of "useful for renewable power", i.e. the duck curve. though it doesn't often get mentioned by name.

the exceptions are either A) they also mention how e.g. nuclear power is very limited in its ability to change output at all, so it's also useful for day/night power imbalance in general, or B) they target short-term spikes and dips to e.g. allow gas/coal time to adjust.

Reply on news.ycombinator.com