12 ms·

> Comparing renewables without storage with a non-intermittent source is comparing apples to oranges. Absolutely correct. There are a lot of magical hand-wavy

> Comparing renewables without storage with a non-intermittent source is comparing apples to oranges. Absolutely correct. There are a lot of magical hand-wavy arguments and false stats used when comparing solar to nuclear. My 13kW array went down to 600 W (yes, six hundred Watts) peak, not constant, during the last few weeks of rains in Los Angeles. I cannot possibly imagine an entire city relying on this for energy. At some point we have to get real. Solar isn’t the solution. Nuclear is. Solar can help, yet it is very far from being a reliable solution. I’ll post power output graphs when I get a moment.

pfdietz 4y ago
You can back up solar with hydrogen at $1/W of generating capacity for those rare prolonged outages. Because they are rare, the fuel cost is inconsiderable. At the same time, the backup generators are 1/10th (or, if you use simple cycle instead of combined cycle, 1/20th) the cost of building a new nuclear power plant, per unit of output.
- JumpCrisscross 4y ago
  > back up solar with hydrogen There are myriads of better energy-storage solutions than hydrogen, particularly at the periphery.
  - pfdietz 4y ago
    Quite possibly, but to 100% and for use cases with few charge/discharge cycles? In any case, hydrogen provides an existence proof that renewables can get to 100%, and probably more cheaply than nuclear.
- Manuel_D 4y ago
  > You can back up solar with hydrogen No, you can't, because nobody is offering hydrogen electricity storage. If you're okay with energy plans involving heretofore unused technology, then I've got a fusion plant to sell you.
  - pfdietz 4y ago
    Ah yes, your old "if no one is offering it, it cannot ever exist" argument. Yes, I am perfectly comfortable imagining the future will be using technologies that we are not currently using. Hydrogen is not much of a stretch, as it involves integrating technologies that already exist.
    - Manuel_D 4y ago
      Fusion doesn't need non-existent technologies either. It's just electromagnets and plasma. We just need to integrate these existing technologies to make the fusion process more efficient.
      - moloch-hai 4y ago
        Appealing to the immediate practicability of fusion is a desperate argument.
        pfdietz 4y ago
        His argument is apparently "if there is some technology that it is not reasonable to expect will happen, then there is no technology that it is reasonable to expect will happen." (Except maybe whatever advances are needed for nuclear fission to power the world, I'd guess.)
        moloch-hai 4y ago
        By the argument, the fission plants are all built already and just want permission to turn on.
    - robomartin 4y ago
      > Yes, I am perfectly comfortable imagining the future will be using technologies that we are not currently using. Sure. OK. The problem is that this is science fiction, not reality. I hear these kinds of arguments from people who have never done any construction project of non-trivial scale in their lives. Sure, from that perspective anything is possible. Let me tell you about reality in the US. If you want to build, say, a new instant-on hydrogen-based gigawatt-scale power generation plant, you need at least four things: - A design - A site - Environmental studies - Permits The design is tightly coupled to the site. The site is tightly coupled to the environmental studies and, of course, the permits. It could take 5 to 10 years to find a site and get it approved for a specific design. The permits could take another 5 to 10 years in the aggregate. This means you'll get some permits in a few years and others will be a battle you will have to fight for probably a decade as things are built. Finally, the construction project will likely take somewhere around 20 to 25 years. You are looking at 20 to 30 years. Just for one power plant. And I could be 100% off. It could take double that time. Here's the key: The clock starts NOW. Which means you have to design it with the technology you have NOW. Not dilithium crystals or magical hydrogen generators that do not exist. If you want the 25 year clock to start ticking today, the only way is to design with what you have, not what you wish you could have or what you think you might have. That's the problem with all of these hand-wavy arguments. They are fantasy. If we got our heads out of our collective asses we could start building modern nuclear power plants very quickly. They are not fantasy. They work. And they are far better than most, if not all, of the alternatives.
Schroedingersat 4y ago
I like how you're unable to imagine it being windy when it rains or transmitting between places that don't rain at the same time, but imagining transmission from two states over when there's a correlated outage in the local nuclear generation (which happens just as often) is fine. If we did live in this fiction where long term storage is impossible rather than simply not being the lowest hanging fruit I'd also far rather spend 5% of the next century building out renewables and then 5% of the time running fossil fuels when the alternative is: Spend 30% of the next century building reactors whilst running fossil fuels, then 30% of the next century realising the uranium ran out immediately and we have to spend another 30 years building breeders and then finally realising that nuclear needs storage and load shifting too because correlated outages aren't that rare and there's not much demand for energy at 3am.
- robertlagrant 4y ago
  > which happens just as often What happens just as often? Two nuclear power plants going offline vs it being rainy and not also windy?
  - Gwypaas 4y ago
    In the Nordics, in December we had the 3 largest reactors out of 11 out of commission combined with a cold spell. That was a nicely expensive week.
  - pfdietz 4y ago
    Nuclear stans assure us the road to success involves standardized reactor designs. Now imagine what happens when a terrible design flaw is discovered in that reactor type and all must be shut down to fix it.
    - robertlagrant 4y ago
      Are you saying that will cause as many power delivery lulls as will happen with lack of wind in wind power or lack of sun in solar? That is the context of this.
robomartin 4y ago
Here are the graphs, as promised. https://i.imgur.com/bo0s7b0.png https://i.imgur.com/bo0s7b0.png Yes, these charts are horrific. I am showing daily energy generation for the last three months and the same period 12 months ago for comparison. I have also added day charts for January 14th through the 19th of this year (the period indicated by the red arrow) for readers to get a sense of what solar reality looks like. I've done this because it is too easy to say "On January 17th we generated 42.6 kWh" and fail to understand that between 12:20 and 13:15 the system dropped from 6.672 kW to 1.632 kW (power, not energy). Intelligent readers will be able to take these charts, play with some very basic numbers and understand the significant issues facing solar. If I wanted to have a system that delivered a reliable, usable 40 kWh of energy per day it would likely grow from the 13 kWh array I have today to somewhere between 50 kWh and 100kWh (if not more). And, on top of that, I would probably need somewhere in the order of 400 kWh of batteries for storage. In other words, an unrealizable monster. I know people are going to laugh at these numbers. These are the folks who never bother to fire-up a spreadsheet, run the numbers and reason. Take a look at the daily chart for January and tell me how much energy you would need to store to be able to have a real 40 kWh per day supply and how large the array would have to be. Then tell me how you are going to charge that pack in December, because it has to be fully charged so you can use it in January. That's the reality you need to understand. And the only way you will is to do the math. And, BTW, this is living in Southern California. Almost anywhere north of this latitude --places with far more weather-- the situation is even worse.
- Schroedingersat 4y ago
  So your system in perfect weather has a performance ratio of around 80% of what an average system optimized for total power would produce on average over the month, it has terrible low light performance (almost as if it's 2010 tech or you picked a bad inverter for your use case or it's badly installed), you're pretending transmission and wind don't exist and that any suggestion of a mix involving solar be exclusively off grid, and that dispatchable backup like biomethane and hydro for a few hundred hours a year is both impossible and is never going to be involved in any nuclear based system? You're also trying to pretend weather is a 1:1 correlation with latitude and that wind and solar aren't anti-correlated. You know a vertical south facing bifacial panel a bit north of Calgary will produce just as much power in december as a flat one in singapore right? California already has about 7% hydro and biomass capacity. Include 7% dispatchable generation, then adjust your stats to match or look up the output of a real modern fixed tilt utility (or well designed off grid) system with a decent MPPT, adequate low light performance and bypass diodes, and you'll be able to see you've actually provided fairly strong evidence that 12hr storage, 7% dispatch and 100% overprovision (which you can use with your 3 day dispatchable load called an EV to avoid curtailment if you drive an average amount) on a 93% solar system is more than sufficient. Include transmission to the other side of a range to get less correlated weather and the storage and overprovision drop significantly. Add onshore wind and the requirement for dispatch, overprovision and storage plummets. Add offshore wind and HVDC and 3 hours with 30% curtailment into an electrolyser for fertilizer is overkill.

pfdietz 4y ago

You can back up solar with hydrogen at $1/W of generating capacity for those rare prolonged outages. Because they are rare, the fuel cost is inconsiderable. At the same time, the backup generators are 1/10th (or, if you use simple cycle instead of combined cycle, 1/20th) the cost of building a new nuclear power plant, per unit of output.

JumpCrisscross 4y ago
> back up solar with hydrogen There are myriads of better energy-storage solutions than hydrogen, particularly at the periphery.
- pfdietz 4y ago
  Quite possibly, but to 100% and for use cases with few charge/discharge cycles? In any case, hydrogen provides an existence proof that renewables can get to 100%, and probably more cheaply than nuclear.
Manuel_D 4y ago
> You can back up solar with hydrogen No, you can't, because nobody is offering hydrogen electricity storage. If you're okay with energy plans involving heretofore unused technology, then I've got a fusion plant to sell you.
- pfdietz 4y ago
  Ah yes, your old "if no one is offering it, it cannot ever exist" argument. Yes, I am perfectly comfortable imagining the future will be using technologies that we are not currently using. Hydrogen is not much of a stretch, as it involves integrating technologies that already exist.
  - Manuel_D 4y ago
    Fusion doesn't need non-existent technologies either. It's just electromagnets and plasma. We just need to integrate these existing technologies to make the fusion process more efficient.
    - moloch-hai 4y ago
      Appealing to the immediate practicability of fusion is a desperate argument.
      - pfdietz 4y ago
        His argument is apparently "if there is some technology that it is not reasonable to expect will happen, then there is no technology that it is reasonable to expect will happen." (Except maybe whatever advances are needed for nuclear fission to power the world, I'd guess.)
        moloch-hai 4y ago
        By the argument, the fission plants are all built already and just want permission to turn on.
  - robomartin 4y ago
    > Yes, I am perfectly comfortable imagining the future will be using technologies that we are not currently using. Sure. OK. The problem is that this is science fiction, not reality. I hear these kinds of arguments from people who have never done any construction project of non-trivial scale in their lives. Sure, from that perspective anything is possible. Let me tell you about reality in the US. If you want to build, say, a new instant-on hydrogen-based gigawatt-scale power generation plant, you need at least four things: - A design - A site - Environmental studies - Permits The design is tightly coupled to the site. The site is tightly coupled to the environmental studies and, of course, the permits. It could take 5 to 10 years to find a site and get it approved for a specific design. The permits could take another 5 to 10 years in the aggregate. This means you'll get some permits in a few years and others will be a battle you will have to fight for probably a decade as things are built. Finally, the construction project will likely take somewhere around 20 to 25 years. You are looking at 20 to 30 years. Just for one power plant. And I could be 100% off. It could take double that time. Here's the key: The clock starts NOW. Which means you have to design it with the technology you have NOW. Not dilithium crystals or magical hydrogen generators that do not exist. If you want the 25 year clock to start ticking today, the only way is to design with what you have, not what you wish you could have or what you think you might have. That's the problem with all of these hand-wavy arguments. They are fantasy. If we got our heads out of our collective asses we could start building modern nuclear power plants very quickly. They are not fantasy. They work. And they are far better than most, if not all, of the alternatives.

Schroedingersat 4y ago

I like how you're unable to imagine it being windy when it rains or transmitting between places that don't rain at the same time, but imagining transmission from two states over when there's a correlated outage in the local nuclear generation (which happens just as often) is fine. If we did live in this fiction where long term storage is impossible rather than simply not being the lowest hanging fruit I'd also far rather spend 5% of the next century building out renewables and then 5% of the time running fossil fuels when the alternative is: Spend 30% of the next century building reactors whilst running fossil fuels, then 30% of the next century realising the uranium ran out immediately and we have to spend another 30 years building breeders and then finally realising that nuclear needs storage and load shifting too because correlated outages aren't that rare and there's not much demand for energy at 3am.

robertlagrant 4y ago
> which happens just as often What happens just as often? Two nuclear power plants going offline vs it being rainy and not also windy?
- Gwypaas 4y ago
  In the Nordics, in December we had the 3 largest reactors out of 11 out of commission combined with a cold spell. That was a nicely expensive week.
- pfdietz 4y ago
  Nuclear stans assure us the road to success involves standardized reactor designs. Now imagine what happens when a terrible design flaw is discovered in that reactor type and all must be shut down to fix it.
  - robertlagrant 4y ago
    Are you saying that will cause as many power delivery lulls as will happen with lack of wind in wind power or lack of sun in solar? That is the context of this.

robomartin 4y ago

Here are the graphs, as promised. https://i.imgur.com/bo0s7b0.png https://i.imgur.com/bo0s7b0.png Yes, these charts are horrific. I am showing daily energy generation for the last three months and the same period 12 months ago for comparison. I have also added day charts for January 14th through the 19th of this year (the period indicated by the red arrow) for readers to get a sense of what solar reality looks like. I've done this because it is too easy to say "On January 17th we generated 42.6 kWh" and fail to understand that between 12:20 and 13:15 the system dropped from 6.672 kW to 1.632 kW (power, not energy). Intelligent readers will be able to take these charts, play with some very basic numbers and understand the significant issues facing solar. If I wanted to have a system that delivered a reliable, usable 40 kWh of energy per day it would likely grow from the 13 kWh array I have today to somewhere between 50 kWh and 100kWh (if not more). And, on top of that, I would probably need somewhere in the order of 400 kWh of batteries for storage. In other words, an unrealizable monster. I know people are going to laugh at these numbers. These are the folks who never bother to fire-up a spreadsheet, run the numbers and reason. Take a look at the daily chart for January and tell me how much energy you would need to store to be able to have a real 40 kWh per day supply and how large the array would have to be. Then tell me how you are going to charge that pack in December, because it has to be fully charged so you can use it in January. That's the reality you need to understand. And the only way you will is to do the math. And, BTW, this is living in Southern California. Almost anywhere north of this latitude --places with far more weather-- the situation is even worse.

Schroedingersat 4y ago
So your system in perfect weather has a performance ratio of around 80% of what an average system optimized for total power would produce on average over the month, it has terrible low light performance (almost as if it's 2010 tech or you picked a bad inverter for your use case or it's badly installed), you're pretending transmission and wind don't exist and that any suggestion of a mix involving solar be exclusively off grid, and that dispatchable backup like biomethane and hydro for a few hundred hours a year is both impossible and is never going to be involved in any nuclear based system? You're also trying to pretend weather is a 1:1 correlation with latitude and that wind and solar aren't anti-correlated. You know a vertical south facing bifacial panel a bit north of Calgary will produce just as much power in december as a flat one in singapore right? California already has about 7% hydro and biomass capacity. Include 7% dispatchable generation, then adjust your stats to match or look up the output of a real modern fixed tilt utility (or well designed off grid) system with a decent MPPT, adequate low light performance and bypass diodes, and you'll be able to see you've actually provided fairly strong evidence that 12hr storage, 7% dispatch and 100% overprovision (which you can use with your 3 day dispatchable load called an EV to avoid curtailment if you drive an average amount) on a 93% solar system is more than sufficient. Include transmission to the other side of a range to get less correlated weather and the storage and overprovision drop significantly. Add onshore wind and the requirement for dispatch, overprovision and storage plummets. Add offshore wind and HVDC and 3 hours with 30% curtailment into an electrolyser for fertilizer is overkill.