I like the joke that my house is fusion-powered; the reactor is of course located in a remote industrial area, for safety, and I rely on an array of roof-mounted receivers for long-range wireless power transfer. Conveniently enough, this reactor happens to be a deregulated and tragically undersubscribed public utility service...
Is there some name for the argument that "The future is not [a technology that currently works, if not perfectly], it's [an open research problem if it can be made to work at all]"?
The future of energy is nuclear fusion -- carried out in a tremendously massive reactor at a safe distance of 150 million km from the nearest civilian population. Technology to convert radiation from that reactor directly to electricity is already cost-competitive with other energy sources and the price is still declining rapidly. Together with secondary capture techniques that drive turbines based on air or water flows driven by the energy of that reactor, the majority of humanity's energy needs will be met by that one source in a few decades. Smaller reactors will never reach a cost scale to be competitive, even if they can eventually be made to function.
The whole point is to make it not a commodity you have to constantly buy and transport, but something everyone at all levels of infrastructure could make themselves locally.
There's a Moore's law of solar called Swanson's law which is on track for what in today's terms we'd effectively call "free energy" sometime in the nest few decades.
Another big idea is that these don't require any manual capturing, transport, and feeding of fuel. You install them, make sure they are clean and essentially that's it. Nothing more to mine or constantly load into the machine.
A robust, fault tolerant, dynamic system of community and family power is the actual future and only dramatically decentralized things like that will ultimately be politically viable and sustainable.
They will more likely survive wars and infrastructure collapse; something that a model reliant on a few multibillion dollar plants will never be able to do.
Sustainability is not only about ecology but also about decomodifying energy and not having to be constantly reliant on some large group of people maintaining an infrastructure to keep the lights on.
My off-grid friends struggle to get house insurance, because without "grid electricity" insurance companies don't consider a house worthwhile (it's actually a really, really nice house, worth close on a million dollars now).
Anyway, my friend asks the insurance assessor how often he has a power failure at his house "Oh, only a couple of times a year" says the man.
In 10 years living off grid, my friends have never had a single outage.
He has spent years and years hunting out the one or two companies that will actually cover him.
Unfortunately, he pays something like $1900/month, because he is distant from the nearest fire station, and again, city-based insurance companies see that as a big problem.
We'll get there. The decentralization is the actual revolution more than the mechanism of solar power.
Not only is local, sustainable low cost energy production getting cheaper and easier but the post-industrial household energy use is going down.
We've replaced CRTs and incandescents. Our new buildings have better weather proofing and even our kitchen appliances have dramatically improved.
Even in places like Los Angeles, the public have been clamoring for and enthusiastically voting to gasp tax themselves to build mass transit systems. The average vehicle fuel economy has been going up consistently for years. This trend towards lower cost, more efficient things is one consumers have been wildly supportive of and will continue.
The future of low energy at low cost is the stated reason that companies like PG&E are moving away from nuclear based primarily on financial forecasts.
Most people are not renewable energy enthusiasts. They don't want to install, maintain, and be responsible for their own electrical generation plant if they can just pay a utility to provide it.
People can have their own wells, or collect and store rainwater, but most don't if utility water is available.
Every home in any remotely civilized area is already wired for utility power. Utility scale renewable energy, if that can be done cost-effectively, probably makes the most sense for the most people.
Otherwise you have an expensive purchase and installation of solar panels, a transfer switch if you keep utility power connected as a backup, battery installation so you have electricity at night, etc.
It's just more complexity and expense than most people want to deal with, compared to paying a hundred dollars or so a month to the utility.
Totally agree. Unless there's a super-easy way to retrofit very few will bother. However, let's think 100 years out.
By this time, the cost threshold of things in the class of solar is perhaps a few hundred dollar fixed cost for your needs. Just like you don't need to be an energy-saving enthusiast to get a non-crt computer screen today, low-cost commodity energy generation hardware will dominate the marketplace and essentially be the only consumer-facing option.
If you are an "I don't care" consumer like most are, the cheapest and best business model would be to have either someone come by and put the thing in or to franchise it out like 7/11 so every neighborhood gets their own low-cost plant.
Either way, large power-generation infrastructure projects are likely to give way to an even larger consumer-facing power-generator infrastructure.
In the shorter term someone somewhere (maybe reading these words right now) will come along and make a version of solar or solar-equivalent tech a sexy luxury toy for wealthy people to show everyone how rich they are. It will get hyped-to-hell-and-back. Big players will slap their head and say "oh darn that was obvious" and quickly move in to snatch the market and there will be a wild ride for all the VC groups involved. It'll mint a few 22-year old billionaires from either MIT or Harvard who will be paraded as visionary geniuses for our time. So lovely.
Don't know when, don't know where, but it's coming. (ok, my real guess is geopolitics will destabilize oil prices and create a climate for adoption within the next 3 years - the cost metrics have just started working out in the past 12 months)
I bet those people also don't care about installing windows or a roof.
The thing is, a water collection system requires constant upkeep. A solar power system requires some maintenance once in several years (like your roof). Even if they want to stay connected on the grid (batteries still require maintenance), they'll probably want some solar power.
> A robust, fault tolerant, dynamic system of community and family power is the actual future and only dramatically decentralized things like that will ultimately be politically viable and sustainable.
Although I'm a big fan of distributed solar, I don't think it's by any means so clear it'll become dominant. Utility scale solar is half the cost on a leveled basis. Given existing institutions, market structure, and the very non trivial problems involved in a P2P grid I think a federated model is much more likely, with some very localized distributed stuff in the mix.
Solar does not work well in all latitudes. Wind is best placed away from homes and efficient turbines are large, so you need at least municipal installations.
Also, highrises don't have enough roof area to supply their occupants with enough energy.
And industrial energy consumers may need hundreds of megawatts for each plant. metal smelters, cement factories, ammonia synthesis, desalination, hydrogen production, transportation, ...
There are lots of non-electricity energy consumers that we want to transition off fossil fuels too.
And if we want to actively put CO2 back into the ground we might need even larger amounts of energy.
So fusion may still have a place to supply these base loads, especially in northern latitudes.
Fortunately, they seem to be growing in both efficiency and capacity/cost in a superpolynomial fashion. But they are still expensive, short-lived and require maintenance.
the centralization aspect is what puts me off fusion too. if people could buy small, safe, domestic, radiation-free fusion reactors then i'd be in favour of it
I was just reading about bubble fusion. From the info I found it seemed like the whole reason it won't work is because the researcher that discovered it was dishonest. (And it couldn't be replicated)
But it seems like one experimental design failing shouldn't rule out the entire concept? It seems like it's physically possible, no? Ie nothing in physics rules it out.
Taleyarkhan's failures alone shouldn't rule out a concept, agreed. A bigger problem is that although sonofusion does appear capable of creating small plasmas which in theory could be hot and dense enough for fusion, the amount of time during which the plasma remains at that temperature and density seems to be too small. In other words, there seems to be insufficient containment.
A common misconception about fusion is that containment is all about keeping the particles from leaving. But often, it is more about keeping the energy from leaving. A momentum-based containment design like sonofusion uses is subject to Rayleigh instabilities which carry the energy away in fast moving particles.
DotNetRocks (a popular .net podcast) does "geek outs" every so often on science/technology. They've done a couple on Nuclear Fusion. It gives a great overview of all the different companies and their approaches to nuclear fusion. Highly recommended.
I am getting to feel that fusion runs out of time. They already barely have enough time for large-scale rollout before renewables overtake everything. In 10 years from now, if someone achieves net energy positive fusion, it will be solution in search of a problem - cheap and abundant electricity will be a reality.
A large amount of very cheap energy could have uses that we haven't thought of yet. Sometimes you need a supply of ingredients before inspiration strikes. Just think what could have happened if the British government had not been so short sighted and if Babbage had not been so difficult to deal with. But then although Ada Lovelace did make some remarkably prescient suggestions no one had any real idea what really cheap computing could do.
Remember "Quantity has a quality all its own", but what that quality is isn't always easy to tell before you have it.
That is a bizarre claim to make. Energy will not be far cheaper and more abundant in 10 years. Trillions of dollars are invested in infrastructure that has lifetimes of 50+ years. It is not economically or technologically feasible to undergo an energy revolution that quickly. Even if we gathered the political and financial will, our production facilities and supply chains could not feasibly handle the load of rebuilding the world's energy infrastructure in a single decade.
Infrastructure becomes obsolete and is scrapped all the time, disappointing investors. Newcomen steam engines didn't get to their end of life or anywhere near it, they were mostly scrapped when Watt's revolution happened no matter how new they were. You're saying that investors have been caught short by present trends. I think that's true.
That's not what I am saying. I am saying that if you are competing against something that has already been built and has 50 years of fixed costs already sunk, the hurdle to surpass it is enormous, far more enormous than competing against new marginal construction. I am also saying that energy infrastructure is a thing that costs trillions. We literally do not have the capacity to replace it in a mere 10 years, even if we built 10x the solar and battery factories and ran them 24/7. Current trends extrapolated forward have no indication that this will happen.
(For the record, I am very bullish on solar power, having worked and consulted in the industry. But I am also grounded enough to realize that the revolution will be gradual, not done in 10 years. 10 years is not much time for something so capital-intensive. It's not like phones or computers.)
I regret even commenting in the first place, because I hate feeding the dynamic of giving attention to poor quality and ill-founded comments. I wish I was discussing something constructive instead.
Maybe what you want to say is that obsolete tech and plant has the "advantage" that it can now operate on marginal costs and ignore replacement costs since that won't be happening, making many previous "fixed costs" suddenly free and therefore not costs at all - something I well remember being taught in economics classes at university. This is true as far as it goes, but that path assumes relatively low marginal costs - not the case for energy. Coal isn't that economical - as for fracking, that's new tech not old; it may be a big part of the picture for a long while, and save some but not most of the old plant (large numbers of gas generators are a fairly recent phenomenon themselves.) From your selective choice of terminology, I don't think you've looked at economic theory in depth, here. In any case, all of this is perfectly consistent with what I said earlier.
When the transistor arrives, vacuum tubes are just swept aside despite the fixed costs. Ten years did that, less in fact, I was there. There's a rather extreme threshold effect, once the right price point is passed, the old tech just falls off a cliff. There's a great chart of that happening to Kodak due to digital photography, extremely suddenly, despite all the fixed costs. They were making their best profits one year, staring at death the next. This link is close to the chart I remember seeing. Note the extraordinary difference one year makes! Every mall had a very expensive photo developing plant, then not. Scrap value only.
https://macromon.wordpress.com/2012/01/19/creative-destructi...
Because solar power effectiveness varies geographically the cliff will be a bit smoother, of course.
Only in a real banana republic could you hold new tech off for a half-century, and you'd need to halt trade to do it.
I'm old, I've seen a lot of change, and tons of valuable things - including large amounts of infrastructure, become valueless very suddenly, not to mention my collection of VCR tapes.
There is such a thing as first-mover advantage, but that applies to new tech not old, so I'm assuming you don't mean to refer to that.
I regret your stooping to vituperation. It wasn't necessary, or accurate.
Yes, that's what I wanted to say. Which is why I said it.
Let's look at some facts:
* No one in the market agrees that energy will abundant and cheap in 10 years due to solar and batteries becoming far cheaper than natural gas. Natural gas companies have not been shorted to zero. Utilities are continuing to make capital expenditures. Natural gas plants are continuing to be built. Corporations are continuing to sign long-term power purchase agreements with renewable generators that lock in prices for 20 years. Maybe everyone is suffering from a collective delusion, or maybe energy transitions take decades.
* The EIA projects roughly flat energy consumption out to 2040, with natural gas and renewables taking share from coal and nuclear. Hardly abundant.
* There are strong dynamic effects that will slow the transition. As the penetration of solar grows, the non-solar consumption peak will move to night-time and the marginal value of solar falls by about 1/2. Conversely, as natural gas plants shut down, their cost of fuel will fall and the value of their spinning reserve and other ancillary services will rise.
* Building a transmission backbone for new utility-scale solar will take decades. San Diego's Sunrise PowerLink took 5 years of environmental review and construction, for instance:
* Increasing the supply of large power transformers to build out a new transmission backbone for utility-scale solar will face significant supply constraints:
* Today no energy storage solution is economical. Probably the solution with the great hope of scale-based cost reductions is batteries. How many batteries would we need? World electricity consumption is about 50,000 GWh per day. If solar displaces natural gas, then for overnight and peak backup, we need storage on that scale.
* Can we hope to even approach producing 50,000 GWh of batteries within 10 years? Global annual production of batteries is around 30 GWh a year right now. There are factories under construction now to bring that up to around 100 GWh around 2020. Given that factories take years to design and build, that means we need to scale lithium production by roughly a factor of 1,000 and scale battery production roughly a factor of 1,000 and get the plans and investment lined up in the next couple years. Is Tesla ready for 1,000 Gigafactories? Absolutely not.
No you didn't say it, there was no way to tell that you hadn't factored in threshold effects from your first post, nor that you had glossed over high marginal costs for energy. Points you then pointedly ignore when replying.
If you don't want to be asked what you're saying precisely, you just have to provide more detail earlier, or at the very least the proper rubrik. I just don't understand the hostility here mixed into factual matters.
There's little to answer in your last post. Your definition of gradual was decades; re solar power it's already been decades, but the actual transitions to new technologies - the mass adoptions - happen quickly - so you're really changing the subject. The first consumer product (radio) to appear in my city with transistors came in the early sixties, at a price of $79 - a fortune then - but only a few years after that, you couldn't buy a regular radio with tubes. The transition was very rapid. Kodak's stock took a long time to fall to earth, but the loss of revenue was sudden, and massive.
Re transmission - solar will strongly tend to be local. Batteries: you can look up discussions of Vanadium batteries (just one novel tech) elsewhere on HN, but note that individual batteries are now being built that are nearly 1GW - so 50,000 GW is very far from impossible - even if that figure turns out to be accurate. Not impossible at 5 cents a watt. Betting that this or that is impossible is usually a bad bet.
https://www.forbes.com/sites/jamesconca/2016/12/13/vanadium-...
Actually, far larger "batteries" exist - reversible dams, many built only for energy storage purposes. Analogous mass systems have also been proposed using air and salt.
I apologize if I came off as hostile. I don't feel hostile. In any case, I think it's fine to ask folks for clarification. But I think it's rude to assert: "What you're trying to say is..."
Also, even if Vanadium flow batteries achieve commercial success and drop energy storage costs down to 5 cents a kWh, that's still above wholesale electricity prices of 2-4 cents a kWh:
Lastly, the EAI disagrees with your assertion that solar will be local (which I interpret to mean rooftop). It forecasts lower costs and higher volumes for utility-scale installations, which benefit from greater economies of scale.
But fusion won't be available in 10 years either. There is a long, long way from net energy positive to mass rollout. Fusion is going to be harder than fission, and fission took 30 years from first controlled fission to 1% of electricity supply, which is the lowest which i could call 'mass rollout'. In 10 years though, renewables will be definitely making it - becoming cheaper than any new type of generation , probably almost stopping construction of any new non-relewable generation. Peak use of fossil fuels for electricity generation is now projected to happen in 2025.
Trillions of dollars are being invested in infrastructure now - and yes, it is mostly investment into renewables already.
Until someone figures out how to miniaturize it into something to fit in his chest to stop shrapnel from entering his heart... then realizes he has way more than enough power to run a super suit.
The main issue I see with fusion is a chart I saw (that I'm not sure how accurate it was though), that showed the amount of funding needed to achieve breakthroughs over time.
All funding for fusion was far below the level needed for breakthroughs to be made, so until the motivation and funding are at the level needed, I think we will probably not get fusion as soon as we need it. Eg 20 years from now instead of 10
Isn't the power production inside the sun on the order of 300 W/m3? Is it at all reasonable to think we could scale this process to 0.1-100 MW/m3, to become economically viable?
56 comments
[ 3.2 ms ] story [ 107 ms ] threadFusion, hydrogen fuel cells,...
The whole point is to make it not a commodity you have to constantly buy and transport, but something everyone at all levels of infrastructure could make themselves locally.
There's a Moore's law of solar called Swanson's law which is on track for what in today's terms we'd effectively call "free energy" sometime in the nest few decades.
Another big idea is that these don't require any manual capturing, transport, and feeding of fuel. You install them, make sure they are clean and essentially that's it. Nothing more to mine or constantly load into the machine.
A robust, fault tolerant, dynamic system of community and family power is the actual future and only dramatically decentralized things like that will ultimately be politically viable and sustainable.
They will more likely survive wars and infrastructure collapse; something that a model reliant on a few multibillion dollar plants will never be able to do.
Sustainability is not only about ecology but also about decomodifying energy and not having to be constantly reliant on some large group of people maintaining an infrastructure to keep the lights on.
Anyway, my friend asks the insurance assessor how often he has a power failure at his house "Oh, only a couple of times a year" says the man.
In 10 years living off grid, my friends have never had a single outage.
Unfortunately, he pays something like $1900/month, because he is distant from the nearest fire station, and again, city-based insurance companies see that as a big problem.
The house is in the Yukon, far out there.
This post of mine has a photo of their "cabin" that is just down the road from their house. http://theroadchoseme.com/chillin-yukon-style
I just went through my photos, I don't actually have one of their house online anywhere.
I took these on a walk from their house: http://theroadchoseme.com/northern-lights-1
Also, "off-grid house in the Yukon" is extremely common :)
Not only is local, sustainable low cost energy production getting cheaper and easier but the post-industrial household energy use is going down.
We've replaced CRTs and incandescents. Our new buildings have better weather proofing and even our kitchen appliances have dramatically improved.
Even in places like Los Angeles, the public have been clamoring for and enthusiastically voting to gasp tax themselves to build mass transit systems. The average vehicle fuel economy has been going up consistently for years. This trend towards lower cost, more efficient things is one consumers have been wildly supportive of and will continue.
The future of low energy at low cost is the stated reason that companies like PG&E are moving away from nuclear based primarily on financial forecasts.
People can have their own wells, or collect and store rainwater, but most don't if utility water is available.
Every home in any remotely civilized area is already wired for utility power. Utility scale renewable energy, if that can be done cost-effectively, probably makes the most sense for the most people.
Otherwise you have an expensive purchase and installation of solar panels, a transfer switch if you keep utility power connected as a backup, battery installation so you have electricity at night, etc.
It's just more complexity and expense than most people want to deal with, compared to paying a hundred dollars or so a month to the utility.
By this time, the cost threshold of things in the class of solar is perhaps a few hundred dollar fixed cost for your needs. Just like you don't need to be an energy-saving enthusiast to get a non-crt computer screen today, low-cost commodity energy generation hardware will dominate the marketplace and essentially be the only consumer-facing option.
If you are an "I don't care" consumer like most are, the cheapest and best business model would be to have either someone come by and put the thing in or to franchise it out like 7/11 so every neighborhood gets their own low-cost plant.
Either way, large power-generation infrastructure projects are likely to give way to an even larger consumer-facing power-generator infrastructure.
In the shorter term someone somewhere (maybe reading these words right now) will come along and make a version of solar or solar-equivalent tech a sexy luxury toy for wealthy people to show everyone how rich they are. It will get hyped-to-hell-and-back. Big players will slap their head and say "oh darn that was obvious" and quickly move in to snatch the market and there will be a wild ride for all the VC groups involved. It'll mint a few 22-year old billionaires from either MIT or Harvard who will be paraded as visionary geniuses for our time. So lovely.
Don't know when, don't know where, but it's coming. (ok, my real guess is geopolitics will destabilize oil prices and create a climate for adoption within the next 3 years - the cost metrics have just started working out in the past 12 months)
The thing is, a water collection system requires constant upkeep. A solar power system requires some maintenance once in several years (like your roof). Even if they want to stay connected on the grid (batteries still require maintenance), they'll probably want some solar power.
He has replaced the entire battery bank once, and now has more panels that track the sun.
Although I'm a big fan of distributed solar, I don't think it's by any means so clear it'll become dominant. Utility scale solar is half the cost on a leveled basis. Given existing institutions, market structure, and the very non trivial problems involved in a P2P grid I think a federated model is much more likely, with some very localized distributed stuff in the mix.
YUP
any concept of 'sustainability' that ignores politics and human complexity is silly
And industrial energy consumers may need hundreds of megawatts for each plant. metal smelters, cement factories, ammonia synthesis, desalination, hydrogen production, transportation, ...
There are lots of non-electricity energy consumers that we want to transition off fossil fuels too.
And if we want to actively put CO2 back into the ground we might need even larger amounts of energy.
So fusion may still have a place to supply these base loads, especially in northern latitudes.
Fortunately, they seem to be growing in both efficiency and capacity/cost in a superpolynomial fashion. But they are still expensive, short-lived and require maintenance.
But it seems like one experimental design failing shouldn't rule out the entire concept? It seems like it's physically possible, no? Ie nothing in physics rules it out.
A common misconception about fusion is that containment is all about keeping the particles from leaving. But often, it is more about keeping the energy from leaving. A momentum-based containment design like sonofusion uses is subject to Rayleigh instabilities which carry the energy away in fast moving particles.
https://www.dotnetrocks.com/?show=1415
Remember "Quantity has a quality all its own", but what that quality is isn't always easy to tell before you have it.
https://www.youtube.com/watch?v=5guXaWwQpe4
(For the record, I am very bullish on solar power, having worked and consulted in the industry. But I am also grounded enough to realize that the revolution will be gradual, not done in 10 years. 10 years is not much time for something so capital-intensive. It's not like phones or computers.)
I regret even commenting in the first place, because I hate feeding the dynamic of giving attention to poor quality and ill-founded comments. I wish I was discussing something constructive instead.
When the transistor arrives, vacuum tubes are just swept aside despite the fixed costs. Ten years did that, less in fact, I was there. There's a rather extreme threshold effect, once the right price point is passed, the old tech just falls off a cliff. There's a great chart of that happening to Kodak due to digital photography, extremely suddenly, despite all the fixed costs. They were making their best profits one year, staring at death the next. This link is close to the chart I remember seeing. Note the extraordinary difference one year makes! Every mall had a very expensive photo developing plant, then not. Scrap value only. https://macromon.wordpress.com/2012/01/19/creative-destructi... Because solar power effectiveness varies geographically the cliff will be a bit smoother, of course.
Only in a real banana republic could you hold new tech off for a half-century, and you'd need to halt trade to do it.
I'm old, I've seen a lot of change, and tons of valuable things - including large amounts of infrastructure, become valueless very suddenly, not to mention my collection of VCR tapes.
There is such a thing as first-mover advantage, but that applies to new tech not old, so I'm assuming you don't mean to refer to that.
I regret your stooping to vituperation. It wasn't necessary, or accurate.
Let's look at some facts:
* No one in the market agrees that energy will abundant and cheap in 10 years due to solar and batteries becoming far cheaper than natural gas. Natural gas companies have not been shorted to zero. Utilities are continuing to make capital expenditures. Natural gas plants are continuing to be built. Corporations are continuing to sign long-term power purchase agreements with renewable generators that lock in prices for 20 years. Maybe everyone is suffering from a collective delusion, or maybe energy transitions take decades.
* The EIA projects roughly flat energy consumption out to 2040, with natural gas and renewables taking share from coal and nuclear. Hardly abundant.
https://www.eia.gov/outlooks/aeo/pdf/0383(2017).pdf
* For natural gas power plants, about 1/3 of the levelized costs are capital costs.
https://www.eia.gov/outlooks/aeo/pdf/electricity_generation....
* There are strong dynamic effects that will slow the transition. As the penetration of solar grows, the non-solar consumption peak will move to night-time and the marginal value of solar falls by about 1/2. Conversely, as natural gas plants shut down, their cost of fuel will fall and the value of their spinning reserve and other ancillary services will rise.
* Building a transmission backbone for new utility-scale solar will take decades. San Diego's Sunrise PowerLink took 5 years of environmental review and construction, for instance:
https://www.irwaonline.org/eweb/upload/web_novdec_12_Sunrise...
* Increasing the supply of large power transformers to build out a new transmission backbone for utility-scale solar will face significant supply constraints:
https://energy.gov/sites/prod/files/Large%20Power%20Transfor...
* Today no energy storage solution is economical. Probably the solution with the great hope of scale-based cost reductions is batteries. How many batteries would we need? World electricity consumption is about 50,000 GWh per day. If solar displaces natural gas, then for overnight and peak backup, we need storage on that scale.
http://www.wolframalpha.com/input/?i=world+electricity+consu...
* Can we hope to even approach producing 50,000 GWh of batteries within 10 years? Global annual production of batteries is around 30 GWh a year right now. There are factories under construction now to bring that up to around 100 GWh around 2020. Given that factories take years to design and build, that means we need to scale lithium production by roughly a factor of 1,000 and scale battery production roughly a factor of 1,000 and get the plans and investment lined up in the next couple years. Is Tesla ready for 1,000 Gigafactories? Absolutely not.
http://www.visualcapitalist.com/wp-content/uploads/2015/05/l...
Regarding your transistor example - First, according to Wikipedia and Science magazine, the transition was gradual over the 50s and 60s:
https://en.wikipedia.org/wiki/V...
If you don't want to be asked what you're saying precisely, you just have to provide more detail earlier, or at the very least the proper rubrik. I just don't understand the hostility here mixed into factual matters.
There's little to answer in your last post. Your definition of gradual was decades; re solar power it's already been decades, but the actual transitions to new technologies - the mass adoptions - happen quickly - so you're really changing the subject. The first consumer product (radio) to appear in my city with transistors came in the early sixties, at a price of $79 - a fortune then - but only a few years after that, you couldn't buy a regular radio with tubes. The transition was very rapid. Kodak's stock took a long time to fall to earth, but the loss of revenue was sudden, and massive.
Re transmission - solar will strongly tend to be local. Batteries: you can look up discussions of Vanadium batteries (just one novel tech) elsewhere on HN, but note that individual batteries are now being built that are nearly 1GW - so 50,000 GW is very far from impossible - even if that figure turns out to be accurate. Not impossible at 5 cents a watt. Betting that this or that is impossible is usually a bad bet. https://www.forbes.com/sites/jamesconca/2016/12/13/vanadium-...
Actually, far larger "batteries" exist - reversible dams, many built only for energy storage purposes. Analogous mass systems have also been proposed using air and salt.
Also, even if Vanadium flow batteries achieve commercial success and drop energy storage costs down to 5 cents a kWh, that's still above wholesale electricity prices of 2-4 cents a kWh:
https://www.eia.gov/electricity/monthly/update/wholesale_mar...
Lastly, the EAI disagrees with your assertion that solar will be local (which I interpret to mean rooftop). It forecasts lower costs and higher volumes for utility-scale installations, which benefit from greater economies of scale.
Trillions of dollars are being invested in infrastructure now - and yes, it is mostly investment into renewables already.
He'll do this all in a cave of course.
All funding for fusion was far below the level needed for breakthroughs to be made, so until the motivation and funding are at the level needed, I think we will probably not get fusion as soon as we need it. Eg 20 years from now instead of 10