4 ms·

> There's a non-storage solution to the duck curve "problem" which is just to waste some solar power. Can you explain this? How does “wasting some solar” solve

by bryan0 5y ago

> There's a non-storage solution to the duck curve "problem" which is just to waste some solar power. Can you explain this? How does “wasting some solar” solve the problem when your max solar output is 0 at peak demand?

a9h74j 5y ago
Can't speak for the poster, but I have seen the argument that if solar and wind can both be over-provisioned by a factor of 3, then storage will not be much of an issue. Fits with a "solar and wind will be too cheap to meter [the amount of cost and materials going into manufacture]" paradigm.
- hn_throwaway_99 5y ago
  > I have seen the argument that if solar and wind can both be over-provisioned by a factor of 3, then storage will not be much of an issue. That makes 0 sense. If you have 3x needed capacity at midday but 0x capacity at midnight, you could over-provision by 100x - zero times a hundred is still zero.
  - lostapathy 5y ago
    Wind capacity at midnight isn't going to be zero over a large enough connected grid.
ZeroGravitas 5y ago
So, the duck curve problem although it caught people's imaginations, no-one really read the original source, so most people dont know what the duck curve problem is. https://www.vox.com/energy-and-environment/2018/3/20/17128478/solar-duck-curve-nrel-researcher https://www.vox.com/energy-and-environment/2018/3/20/1712847... This features some slides from the original use of the term. Note the repeated reference to "Ramp" since that's what the problem was. > We have to be really careful when we talk about something breaking. The only thing that’s going to break is the economics of solar. The lights are gonna stay on. All it comes down to is more curtailment. One of the other solutions was asking the gas plants to ramp faster. The specs said the equipment could do it, it was just institutional inertia as they hadn't needed to do it before and so weren't prepared for it.
- bryan0 5y ago
  Ok I see. The problem that the duck curve is really about, is not about the lack of solar at certain times of the day, but really that other sources of power need to ramp up and down efficiently around solar.
  - ZeroGravitas 5y ago
    Yep, the duck curve was a problem they anticipated they'd hit when trying to get to 15% solar in California, not a general problem for getting to 100% renewable. Even then it was one of those "good problems to have" as it meant you had an unimaginably high (at that time) level of solar. Some places arent as lucky as California when it comes to solar provision. The fix for the totally seperate "nighttime" problem is some mix of oversupply of wind, hydro, demand management, transmission lines, batteries etc. but we only need to continue with the next low hanging fruit at the moment as we've not even maxed out solar pv (fallacy #1 in the article is expecting instant perfection and having solutions to all the problems we haven't actually hit yet)
- hn_throwaway_99 5y ago
  > The only thing that’s going to break is the economics of solar. Yes, that is the exact point. Unless storage capacity is introduced to the system: 1. It becomes uneconomical to add any more solar plants, because the time when the solar plants can produce already has demand fully met by other, existing solar/wind plants. 2. Since you reach peak daily demand when solar and wind approaches 0 output, it means you need just as much fossil fuel plant generation capacity as before - you can't decommission any of those plants (or, at present, perhaps you just convert them to natural gas). I'm not sure how this thread went off the rails. All I was originally responding to was the sentiment that "you don't need to worry about storage capacity until you already have the vast majority of your energy met by renewables", and that statement is easily provably false because there are many areas of the world that still have relatively low levels of total power generation from renewables, but until the "time shifting" storage problem is solved it will be uneconomical to add more solar to the system.
  - 5y ago
tedsanders 5y ago
It's a good question. The short answer is that although the daily peak in California tends to be around 7pm, the annual peak is more like 3pm-5pm due to air conditioning on hot summer days. So solar, at our current margins, actually does supply peak demand and thereby reduce capex needed for gas peakers. (And annual peak demand is the correct timeframe to use, since our system targets reliability on the scale of years not days.) Today's peak: http://www.caiso.com/TodaysOutlook/Pages/index.html?fbclid=IwAR1rodt4fciRKkUDvTkcVoFmwYyhvh26uIb9E4cv13tWDQWcivFXfYFwWtY http://www.caiso.com/TodaysOutlook/Pages/index.html?fbclid=I... Annual peaks: https://www.caiso.com/documents/californiaisopeakloadhistory.pdf?fbclid=IwAR3lVViQ0LmD4pYZW7OOkJJ8vW0Nd96qOQ0a_eoTzhuGB-d4pDBy1ZVHsV0 https://www.caiso.com/documents/californiaisopeakloadhistory... I wrote a detailed version of this comment later in another thread: https://news.ycombinator.com/item?id=28436432 https://news.ycombinator.com/item?id=28436432

by bryan0 5y ago

a9h74j 5y ago

Can't speak for the poster, but I have seen the argument that if solar and wind can both be over-provisioned by a factor of 3, then storage will not be much of an issue. Fits with a "solar and wind will be too cheap to meter [the amount of cost and materials going into manufacture]" paradigm.

hn_throwaway_99 5y ago
> I have seen the argument that if solar and wind can both be over-provisioned by a factor of 3, then storage will not be much of an issue. That makes 0 sense. If you have 3x needed capacity at midday but 0x capacity at midnight, you could over-provision by 100x - zero times a hundred is still zero.
- lostapathy 5y ago
  Wind capacity at midnight isn't going to be zero over a large enough connected grid.

ZeroGravitas 5y ago

So, the duck curve problem although it caught people's imaginations, no-one really read the original source, so most people dont know what the duck curve problem is. https://www.vox.com/energy-and-environment/2018/3/20/17128478/solar-duck-curve-nrel-researcher https://www.vox.com/energy-and-environment/2018/3/20/1712847... This features some slides from the original use of the term. Note the repeated reference to "Ramp" since that's what the problem was. > We have to be really careful when we talk about something breaking. The only thing that’s going to break is the economics of solar. The lights are gonna stay on. All it comes down to is more curtailment. One of the other solutions was asking the gas plants to ramp faster. The specs said the equipment could do it, it was just institutional inertia as they hadn't needed to do it before and so weren't prepared for it.

bryan0 5y ago
Ok I see. The problem that the duck curve is really about, is not about the lack of solar at certain times of the day, but really that other sources of power need to ramp up and down efficiently around solar.
- ZeroGravitas 5y ago
  Yep, the duck curve was a problem they anticipated they'd hit when trying to get to 15% solar in California, not a general problem for getting to 100% renewable. Even then it was one of those "good problems to have" as it meant you had an unimaginably high (at that time) level of solar. Some places arent as lucky as California when it comes to solar provision. The fix for the totally seperate "nighttime" problem is some mix of oversupply of wind, hydro, demand management, transmission lines, batteries etc. but we only need to continue with the next low hanging fruit at the moment as we've not even maxed out solar pv (fallacy #1 in the article is expecting instant perfection and having solutions to all the problems we haven't actually hit yet)
hn_throwaway_99 5y ago
> The only thing that’s going to break is the economics of solar. Yes, that is the exact point. Unless storage capacity is introduced to the system: 1. It becomes uneconomical to add any more solar plants, because the time when the solar plants can produce already has demand fully met by other, existing solar/wind plants. 2. Since you reach peak daily demand when solar and wind approaches 0 output, it means you need just as much fossil fuel plant generation capacity as before - you can't decommission any of those plants (or, at present, perhaps you just convert them to natural gas). I'm not sure how this thread went off the rails. All I was originally responding to was the sentiment that "you don't need to worry about storage capacity until you already have the vast majority of your energy met by renewables", and that statement is easily provably false because there are many areas of the world that still have relatively low levels of total power generation from renewables, but until the "time shifting" storage problem is solved it will be uneconomical to add more solar to the system.
- 5y ago

tedsanders 5y ago

It's a good question. The short answer is that although the daily peak in California tends to be around 7pm, the annual peak is more like 3pm-5pm due to air conditioning on hot summer days. So solar, at our current margins, actually does supply peak demand and thereby reduce capex needed for gas peakers. (And annual peak demand is the correct timeframe to use, since our system targets reliability on the scale of years not days.) Today's peak: http://www.caiso.com/TodaysOutlook/Pages/index.html?fbclid=IwAR1rodt4fciRKkUDvTkcVoFmwYyhvh26uIb9E4cv13tWDQWcivFXfYFwWtY http://www.caiso.com/TodaysOutlook/Pages/index.html?fbclid=I... Annual peaks: https://www.caiso.com/documents/californiaisopeakloadhistory.pdf?fbclid=IwAR3lVViQ0LmD4pYZW7OOkJJ8vW0Nd96qOQ0a_eoTzhuGB-d4pDBy1ZVHsV0 https://www.caiso.com/documents/californiaisopeakloadhistory... I wrote a detailed version of this comment later in another thread: https://news.ycombinator.com/item?id=28436432 https://news.ycombinator.com/item?id=28436432