Future 2040 Solar Will Be Five Times Cheaper Than Fossil Fuel Electricity

The best predictions of the solar energy cost and solar energy adoption have come from Ramez Naam. There have been hundreds of forecasts of the future costs of solar power and the projected share solar power would have in the future. Ramez has been an optimist around solar energy and solar energy development but actual development has even exceeded his previous forecasts in 2011 and 2015.

Energy Storage is Needed to Add to Solar for a Large Scale Energy Transition

This potential low cost of solar energy will need to be marked up to account for the needed large scale battery. Battery prices will also drop a lot. I believe Tesla and CATL and others will be able to drop prices down to $30-50 per kwh 2025-2030. I foresee the Tesla battery day goals to get achieved in 2030. Batteries of some form will get to $10/kwh in 2030-2040.

Previous Nextbigfuture articles covered the future of energy storage. Solar energy at one-fifth the cost would end up around half the price of fossil fuel when very cheap battery storage is included to actually compare more apples to apples with baseload storage.

The actual cost of solar-battery depends upon the percentage of energy that is used in a particular area. MIT has some analysis. It boils down to the higher percentage of renewable and battery usage ends up costing a lot more. This will take longer to go into detail. Getting up to 10% renewables is pretty easy and can be stable. Getting renewables and super-cheap batteries, you can affordably and stably get up around 50% and maybe 60-80%. It would be better to have 20%-40% of reliable baseload with nuclear or hydro.

Ramez Naam’s Case for Projecting Ultra-cheap Solar

Ramez Naam has new forecasts that use something like Wright’s law. Wright’s law has been used by Ark Investments to predict the cost reductions in Tesla electric cars and batteries.

T.P. Wright saw that doubling of scale led to a constant percentage reduction in cost in 1936 while studying the production costs of airplanes. This relationship between cumulative production and cost has been found in other areas like smartphones, computers and other products.

Solar Energy costs have dropped by a factor of 5 since 2010. Utility-scale solar energy is around 4-5 cents per kilowatt-hour. The lowest cost solar project bids are at costs of 1.35 cents per kwh in Abu Dhabi or 1.6 cents per kwh in Portugal. Solar in 2020 is:

– Less than half the price Ramez forecast in 2011.
– Less than a quarter of the price the IEA forecast in 2010.

Wright’s Law applies to the cost of solar modules. The price of solar modules per watt of power drops by somewhere around 25% for every doubling of cumulative manufacturing. Solar modules are about one-third of the cost of solar electricity. The rest comes from related equipment (inverters, trackers, cabling, mounting systems), land, labor, and other non-module costs.

In 2015, Ramez confirmed a very consistent learning rate for electricity produced from utility-scale solar.

At the end of 2019, solar produced just over 2% of global electricity. After two more doublings when 2,400 GW of solar are producing roughly 8% of current electricity demand, solar costs (of the most recently built and operational projects) will have dropped in half from today’s levels.

Cumulative Solar Energy deployment is doubling every 3-4 years. The IEA forecasts doubling every 4.5 years. Even with the IEA estimation, the world should reach 2400 GW of solar in 2027. We should reach 4800 GW of solar by 2032 and 9600 GW of solar in 2036. Global energy demand will grow and electricity will replace fossil fuels in transportation. This will mean a doubling of electricity demand by 2040 just from the electrification of transportation. There will also be a doubling or two from economic growth. The world solar industry is growing installations at about 23-25% each year in 2020-2021.

Natural gas has reached 43% of US electricity production. It was able to displace coal. Coal had a 50% share of US electricity generation and coal is now 24%. It is tougher to displace the last part of different forms of energy because the last ones are the lowest cost, most efficient and strongest operators.

Also, fossil fuels and nuclear are better able to generate direct heat and steam which are needed for industrial applications.

SOURCES- Ramez Naam
Written By Brian Wang, Nextbigfuture.com

197 thoughts on “Future 2040 Solar Will Be Five Times Cheaper Than Fossil Fuel Electricity”

  1. You're skirting the issue. Solar is just as beholden to supply and demand as any other potential source of energy.

  2. So, do you see a difference between sunlight and coal, which alone is pretty useless too? For example, sunlight can be concentrated with mirrors for great amount of direct heat, or made into electricity. With two possible products, what is this useless stuff that is not free? It can't be the electricity or the heat, because they are not the same thing. I know! it is energy! Free and virtually infinite as sunlight. That is what I mean.

  3. Do you recognize the difference between free cattle, all you want, and cattle you have to pay for? The fact that I pay a chef to cook it does mean it was not free! very basic stuff here.

  4. Smh, you don't think clearly. In fact, you're not thinking at all.

    Sunlight alone WILL NOT power a darned thing. You need a system for converting photons into electrons, and then storing that energy for use. That system will supply a certain wattage, and if electrical demand exceeds that certain wattage, an additional system must be put in place to meet the demand that exceeds the first system, thus there is still a supply and demand paradigm.

  5. Supply of sunlight in Space in the solar system? You use words incorrectly. edit: it is not the processed resource that is infinite, it is the resource, the energy, the sunlight. How much do you pay for sunlight in Space? Compare to anything from Earth, you would run out, not free.

  6. Surely you jest.

    Even if you had 24 hours of sun light, you're still limited by supply and demand, because there isn't an unlimited supply of solar panels.

  7. Sunlight has infinite supply for our purposes. "Definition of supply and demand : the amount of goods and services that are available for people to buy compared to the amount of goods and services that people want to buy If less of a product than the public wants"

  8. According to Kehvan fossil fuels plants emit pixie dust and not co2 , nor cause 4 million premature deaths a year from their pollution.

  9. solar and windfarms are terrible eyesores it says….hilarious….
    so what about the 4 million people who die every year prematurely from fossil fuel pollution.

  10. So far there has not been the big wide-spread adoption of solar power onto single-family homes. I don't see them on apartment complexes either, where they might make even more sense.

    Once solar power reaches its "iPhone Moment" and everyone is racing to get solar panels on their roofs, the market forces will kick in more strongly. Prices will drop and efficiency/innovation will rise.

    Unless your model predicts that, it won't be accurate.

  11. Yes, called plate tectonics and continental drift is about where they came from and where they are going. 

    There is a good chance there was never any planet there. The stuff we see could all be debris from other things crashing together in the past. Thousands of past collisions that resulted in the planets we see. There are almost certainly pieces of the primeval Earth. There is a lot of science to do out there. It could tell us a lot about the planets at various times in the past. They are snapshots of some place and time.

  12. Again, this isn't so much about reducing GH emissions, though it has that effect. It's about a trend that we're already seeing, where cheap solar is displacing other forms of energy production during the day. And about the outcomes of that if it continues.

    Yes, EV owners might generally prefer the lower cost to charge during the day – a good example of a change from current conditions. But if the cost difference isn't great, and they find it convenient to charge in a garage overnight, they may still do it. The inefficiency will only matter to them in terms of the higher cost – they won't otherwise much care whether the electricity is from nuclear or gas or solar energy stored in batteries of whatever type.

  13. Thanx!! When I was a frosh in college Scientific American's Sept "one topic" issue was about plates, but they may have called it "continental drift". Probably the fastest change in a big theory I've ever seen.

    I've always thought that the asteroid belt was much larger in the past, and could have had at least some biggies with gravity sep before things got blasted apart. Then, most stuff gets slung away by Jupiter. My hope is that the very small amounts we need will be present amidst the lunar craters' or NEOs' astronomical amounts.

  14. The non-optimal roof facings are covered with inactive glass tiles. The active PV tiles are only placed where they receive sufficient insolation (ie, south-ish exposures).

  15. Yes, trickle down has been demonstrated as false. Beyond a certain level of wealth, they spend no more, and if they do, it is money spent out of the country.

  16. There are ways you can reduce the size of the array you need.
    High efficiency refrigerators/freezers, double pane windows, insulation, heat pump water heaters, heat pump air heater/cooler, LED lighting (I like low K value ones…just like incandescent), laptop rather than desktop, hand wash your dishes, cook more with microwave oven and induction stove, ear phones rather than speakers, smaller OLED TV…
    And if you have an electric dryer, be sure to avoid over drying. A little damp is ok. You can just set the clothes on things in the house and they will dry fairly quickly as long as they can breathe. For comforters, pillows and blankets…there is no shortcut.

    Random air heat pump: https://www.alpinehomeair.com/viewproduct.cfm?productid=453064230
    Less random water heater heat pump: https://www.homedepot.com/p/Rheem-Performance-Platinum-65-Gal-10-Year-Hybrid-High-Efficiency-Smart-Tank-Electric-Water-Heater-XE65T10H45U0/312741511
    I plan to order this in the next few days. My gas water heater has sprung a leak.

    Do only what makes sense to you, obviously. At $100/month there is not a whole lot of fat to trim. 
    Are you air conditioning in the summer…or just enduring the heat?

  17. I had a reply but it vanished. 
    The problem is electricity could be free to the utility, but don't expect your rates to change…they have a monopoly.
    Self generation is really the only way around that.
    And that is not my downvote. I very rarely downvote. Though I seem to receive a lot, from ideological nuts.

  18. We block entry to all sorts of products into the country. That is not a tax. It is a tax that you can't legally buy cocaine, or elephant tusks?
    And large corporations do not typically flagrantly violate regulatory rules. There are no Mom & Pop GPU foundries/companies.
    Performance will not be cut in half just because power is. Maybe a temporary 25% hit. The next generation you will get all that performance back. People will not suddenly decide they don't like video games any more. The companies will still make and sell lots of GPUs.

  19. Criswell picture does not have he standard phased array, so perhaps he knows something! edit: you certainly know that all Space Solar uses a beacon.

  20. check it out! convince them that they are incorrect in supporting Space Solar, it is far more important than Criswell Lunar Solar Power being impossible. You will be famous!

  21. Simple cycle gas turbines cost $0.70 per watt all-in – if you convert a coal plant like $0.25/watt. $1.20 for combined cycle. That's it. Even if it was a requirement that 1 full watt of gas turbine to be bought/built per nameplate watt of solar or wind after a certain % of wind/solar energy penetration of the total (and you'd need less than 1 full watt for certain), those wind and solar options would STILL be cheapest and with only intraday storage required. Windless/sunless day? Run your gas turbines at full tilt and store 100% not used into the battery arrays, spread it out during the day. The USA uses 11 TWh per day, so you'd need less than half of that in batteries if it was all gas/solar/wind. Like 5.5 TWh. As there is existing hydro, nuclear, and geo you're down to about 3TWh. In a realistic scenario you're off by a factor of 17.
    And the gas turbines would only generate around 13% of the total energy during a typical year, so fuel prices wouldn't be a big deal. Re: GHG plant trees to offset. So consider modular reactors if you can't feasibly site a gas plant, perhaps but unlikely. You could run your turbines on biofuel and it would still be cheaper in most (87%+ of the time) instances than building new nuclear.
    And there are now operating Allam cycle gas plants in demo (real, not powerpoint), so anywhere you can pipe CO2 for a reasonable cost small modular reactors could compete there never.

  22. If you're trying to reduce greenhouse emissions, having more expensive electricity at night means electric vehicles would be charged during the day, presumably by solar. Since many of them are used during the day, you'd need to put the solar into one battery, then decant it later into the car, losing ~15% each time and doubling battery cost. Allowing for cloudy days, more battery cost. A lithium ion battery has 100 times less energy density than gasoline, and uranium has two million times more energy density than gasoline.

  23. "nss.org"

    Is that supposed to mean anything? Like, NSS has spent a billion dollars investigating space solar power?

  24. Yes, solar would eventually price-out all competition during solar-energy hours.

    But the price during 'daylight' hours would at that point be so low that – assuming the lower price is passed on to consumers during those hours – consumers would be incentivized to find ways to shift energy use to those hours.

    Meanwhile, nearly all the fixed costs of energy production systems like gas power plants used during non-daylight hours would be applied to the electric rates during those hours, with those higher rates further pushing consumers to avoid consuming electricity then.

    None of that depends on subsidies. Though if batteries never get cheap enough to compete with natural gas to cover much of night time demand, it might make sense to subsidize nuclear to avoid totally depending on natural gas at night. Maybe just enough nuclear to cover the minimum overnight demand level, with night time rates high enough to subsidize daytime nuclear prices to be competitive with solar.

  25. So you WOULD have fines (confiscation of graphic cards in your example) for anyone violating your preferred wattages.

    A 50W (or even 5W) graphics system might be fine for some, but not for others. Attempting to press everyone into the same mold is very much the same mistake made by communist systems. If you want energy conservation for some reason, just charge more for electricity.

    Personally, I'd rather see plentiful electricity made without CO2 – so I prefer a carbon tax, with revenues used to accelerate building out non-CO2 energy sources (including nuclear) and storage systems. Long term it might be necessary to shift the remaining carbon tax revenues over to subsidize maintenance of mostly idle natural gas power plants and stores of gas, so they're ready when needed to supplement intermittant renewables.

  26. Industrial solar and wind have a HUGE environmental impact for the amount of energy they produce. I'm all for personal solar on roofs as that space is already taking up a footprint. I am not for converting 20,000 square miles of pristine desert ecosystems into giant solar parking lots.

  27. Deep learning AI is quickly approaching bird-poop-recognition parity with humans. By 2027 we'll have drones that can clean poop from panels for under $1/hr.

    However, new material science will provide poop-repellant solar panel coatings that cause bird droppings to simply roll off. Then it'll be a race to the bottom in pricing between drones versus coatings. Ultimately, I expect coatings to win.


  28. The impact on congestion should be minimal, probably insignificant.

    An EV typically won't need to go charge itself more than once a week even for those with fairly long commutes, and far less often for those with short commutes (i.e. lots of city apartment dwellers) or those who can charge at home. SD charging would also be spread over the whole day, not just during high congestion periods. SD cars will probably schedule their charging times online, arriving just in time rather than forming long queues.

    City streets (where most SD cars would be driving themselves to get charged) have low speed limits – driving slower won't save much. And by the time SD EVs are common, charging stations will likely become so common that cars won't need to drive very far to charge – often just to a station in the same parking garage.

    The key transition point will be when there's enough SD EVs to justify having human (or eventually robotic) attendants at most charging stations.

  29. And to answer your question about space solar…

    Geosynchronous orbit is at a height of about 36000 kilometers, about 1/10:th of the distance to the moon.

    From the equations above, we understand that the cone angle scales as 1/distance and that the one dimensional number of emitters is proportional to the distance. The phase requirement scales as ~distance^2.

    So, putting it together, we get N~180 000 (one dimension), or number of emitters ~40 billion. The cone angle would be ~0.12 degrees and the phase requirement ~6*10^-19 seconds.

    Still impossible, Dan.

    Now, if you think I am wrong, Dan, please counter with your own estimate or cite a paper that estimates the phase requirement. Who knows, maybe I've made a mistake somewhere in my estimate? It was kind of late… But I must be wrong by a great number of orders for it to matter…

  30. Dear Dan, a few things.

    Has there been 1 billion USD spent on space solar power? Please cite a source. All I see is a very shallow power point and a proponent that has not even performed a rudimentary sanity check on the feasibility. I can't imagine that costing one billion dollars. I can imagine that there is a sufficient amount of pork money to keep a few enthusiasts producing power-points, though…

    Second, what does the suposed one "billion" dollars of research say about the phased array requirements on beam divergence, the phase aligment of the individual elements etc? If you can link to a source that contradicts my claims, please do so.

    Third, what do you yourself think about the phase requirements and the number of arrray elements? How many elements do you think are required and what is the phase requirement on each element, according to your estimate?

  31. Left to itself, solar would hit a point where it was competing with any new solar coming on stream, and new solar would not be built until solar plus storage was cheaper than the competition. It hasn't been left to itself – places like California and Germany have renewables mandates, must-take rules at set prices no matter what demand is, tax breaks, etc. The result is that prices go negative – customers are paid to take power. That's why the management of California's last nuclear power plant is pulling the plug – they can make power 24/7 cheaply enough, and it's as carbon free as wind and cleaner than solar, but they have to be able to sell the product. Eventually they'll reach the stage where even gas and coal, whose price is dependant on fuel cost, not the cost of the plant, will need subsidies. Blacking out the cities is not an option. https://www.utilitydive.com/news/prognosis-negative-how-california-is-dealing-with-below-zero-power-market/44

  32. Reality can get complicated. The water in the Earth is not all available. Most of it is locked up in hydrates: With heating it can escape the lattice but it will often be locked away in the crystal lattice when things cool a bit. Then it can't do any nifty sorting. Also, on anything smaller than Mars, water will likely be blown away by the solar wind. On Earth, subduction zones don't just ingest rock, they ingest a far amount of water as well. This is how most of this cool sorting takes place on Earth. This subducted plate with the water sinks but the less dense silicates, water and other less dense minerals rise and make volcanic chains. This is where you find quartz veins and gold and other minerals.
    Hoping to get the same process in space, may be wishful thinking.

  33. Having a temperature gradient is a factor, but to get circulation you need gravity.
    Everything on Earth is likely present in asteroids, but unseperated, some stuff will be very sparse. Thing is, when stuff is sparse, it is very hard to concentrate, taking a great deal of time and energy.
    The crust is the solid part. Below that it is "plastic", that is to say, it oozes a bit but mostly stays together though large volumes move not unlike liquid. Below that it is just liquid, until you get to the center where more iron, nickle, and heavy elements are but gravity is high enough to jam the atoms together making a solid even though it is above melting temp. Most molecules/elements are sold when compressed…water is an exception. That is why you can skate on ice. The compression liquefies the top of the ice. 
    But this is with a planet that has liquefied and then cooled somewhat like the Earth. With smaller stuff, they solidify much faster because they have more surface area relative to volume. I would expect Ceres to be completely solid.
    The Earth has reached an equilibrium. The heat generated within due to radioactive decay equals the heat lost at the surface. The original heat from the friction of all the impacting planetesimals was lost. It is estimated that it took the Earth 30 million years to cool and reach equilibrium. The Moon probably took much less. Maybe a million or two. Something 1/10 the mass of the Moon probably less than 100k years. So not much time to stratify.

  34. Thanx! Did not realize *water sorting* required gravity. Is it possible that on small bodies the water process could happen deeper, below the watery crust we have? (edit: I'm wondering how something could be *more* on the crust than deeper, in the asteroid parent, as these are heavy atoms.?) But there is still the more crude gravity sorting. Then, even if the rares are not concentrated, we may be melting and centrifuging everything anyway, and will have plenty of *most* things. We should go see what is there!

  35. The cost to produce solar does matter, because the price of electricity during the day will fall to the point where solar is just profitable, eventually driving out most everything else when it is producing.

    Non-solar-hour rates then rise somewhat to keep solar alternatives (including battery storage of solar generated electricity) profitable. Slow ramping producers like coal and nuclear suffer compared to fast-ramping producers like natural gas and battery storage.

    And as discussed elsewhere in this topic, the skewed pricing creates incentives for consumers to shift their consumption to solar production hours, leading to further increases in solar electricity demand, further skewing of day-night rates, and so on.

  36. Either your neighbor is using a LOT more power than you, or they got ripped off.

    Tesla's system recommendation site, for $100/mo, suggests a ~8kW roof, installed, for $16500, ~$12000 after incentives. Tesla recommends 2 PowerWall batteries for about $22K total after incentives, saving net of ~$12K after 25 years.

  37. The crust of a planet is a very small fraction of the mass. When you break it up…if, in fact, that is where the asteroids came from which is far from proven…only a very small percentage will be former crust. And the total mass of the asteroid belt is approximately 4% the mass of the Moon. That would not make much of a planet.
    Also, early in the Solar System there was a lot more residual heat from impacts and more radiation (stuff has moved down the decay chains over the millions and billions of years since then) and that heat made a lot of stuff molten. That can help sort some things, but if there is no crust on your hypothetical planet, those things which are concentrated by water won't be. Further diminishing the odds is that fact that the sorting by water is density differential driven by gravity…similar to what scientists do with centrifuges. With 1/10 the gravity of the Moon that would be a profoundly slow process…and given that hypothetical planet's short existence, one would have to really enjoy squandering money to look for rare earths in the asteroid belt. Maybe some small chance on the 4 big asteroids. But I would think you would have a better chance on the Moon…and that would not be a great spot to try either.
    I expect there are a lot of goodies on Mercury, likely sorted by processes uncommon on Earth. Europa, Ganymede, Enceladus, and Triton are interesting places to prospect as they have water to dissolve and concentrate minerals.

  38. "And the conclusion is that it is technically impossible to build the beam that you would require for moon space solar power." So, it is also technically impossible to build the beam that you would require for space solar power. Note I removed the word *moon*. LSP is a form of Space Solar, which "is based on >1B$ of space power & lunar studies– Profitable with 1980s technology". You have single handedly overturned over $1B in research! If you are correct. LSP is same stuff as well known GEO solar power sats, ~10 times further away. Go to the big game, and attack GEO sats!

  39. Again, the discussion at hand is whether pricing incentives could or would change exactly the issue you're talking about. I.e. the problem you are "raising" is part of the implicit background to the discussion.

    "Reptilian model" is a very good term for the proposal to heavily shift energy consumption to daytime hours, btw.

    Please note that I'm not advocating the reptilian model – I'm trying to explore its feasibility/desirability.

    However, if solar keeps getting cheaper and more common, this model might to some degree force itself on us, as the cost of electricity falls during daylight hours.

  40. OK, and now for the real kicker…

    How much can "phi" change before the emission ange changes by 0.012 degrees, and the beam misses the target rectenna field altogether? Let us use the second portion of the original equation:

    N*phi/2 = 0.012 degrees
    phi/2 = 0.012 degrees / N
    phi = 5.5*10^-9 degrees

    Let us consider this phase difference. The frequency of the beam is 2.5 GHz. The period of the wave is 1/(2.5*10^9) seconds, or 400 ps. How much is 5.5*10^-9 degrees of that in the time domain? Well, we have that 360 degrees correspond to:

    400 ps / 360 degrees = 1.1 ps / degree

    Here we have 5.5*10^-9 degrees, which corresponds to:

    1.1 ps * 5.5*10^-9 = 6.05 * 10^-21 seconds

    So you need to synchronize ~4*10^12 emitters by 6*10^-21 seconds, just to get your desired characteristics of the beam divergence.

    Dan, there is no way, let me repeat, no way, so synchronize anything to the level of 6*10^-21 seconds. Not even two emitters, let alone 4*10^12 (!). We are more than a thousand billion times off from what is technically possible for two objects. And the conclusion is that it is technically impossible to build the beam that you would require for moon space solar power.

    Don't believe me? I urge you to make your estimation before you waste more of your life pushing for something that is literally impossible to build.

  41. OK Dan, I did some googling, and here is the equation of a line array [1]. The math of a rectangular array is similar, you only have a separate equation of X and Y [2].

    Let's stick with the first source for it's simplicity. Let us define the central angle as where the amplitude is 1/2 of maximum. We then obtain:

    sin(N*pi/4*sin(theta)+N*phi/2) = 1/sqrt(2)
    (N*pi/4*sin(theta)+N*phi/2) = 45 degrees
    only the first term depends on theta

    theta = emission angle
    phi = phase shift from each of the N emitters
    N = number of emitters

    Our goal is that changing theta by 0.012/2 degrees will make the amplitude half. At maximum, the whole expression within the paranthesis was 90 degrees. So, we would like the term


    …to change the angle by 45 degrees. Here, theta (original) is 90 degrees. The new theta is 89.94 degrees.

    (sin(90)-sin(89.94))*N*pi/4 = pi/4 (radians)

    I.e. you need almost 2 million emitters in the X direction. By combining it with [2], we see that the total number of required emitters to have the N^2, or about 4*10^12 emitters.


  42. Was dispersion a typo too? Given that Criswell determined that it was cost effective in the 80s, and that it can work at 2% solar beam intensity, I would guess about 10 times cheaper than it needs to be by now.

    I took em radiation in my Physics major at Michigan in the late sixties. There was far more *known physics* about radar than you want to hear, at the time. Stuff like the math leading to the horn precise shape, basic stuff. Look at any OTHER Space Solar plan that uses power beaming for detailed info. It is all the same basic stuff. Or, find recent Goat Guy description. Your chance of success is low, but very important if you do find something that is a *show stopper*. Enjoy the education!

  43. Similar principle as the GOP's trickle down theory : Poor people will be better off if we give rich people more money.

  44. I am not a believer in 100% solar. I think wind and hydro also have their place. And during an emergency, gas turbine units have their place. I also believe that high voltage transmission lines can reduce the need for storage. I also believe that storage will get a lot cheaper in the next two decades.

    Distribution utilizes look at the weather forecast and determine the generation and purchase mix for the day. If the weather is going to be cloudy, they buy additional power or have fossil fuel power plants spin up their generators. Yes, they will keep a few as backup.

  45. "diveregence" I assume divergence

    Yes Dan, you caught a typo. Good for you.

    About falling costs of power transistors. Power transistors do not follow Mores law, since they do not benefit from scaling to smaller sizes. On the contrary, the power of a transistor scales with area and volume. But Dan, how far off is todays costs from what you need to make space solar theoretically cost competitive? A factor of two? A factor thousand? Ten thousand?

    Could you post the link to the paper that describes the "known physics" about the beam? I got a haunch that the math in it will allow me to show that it's not technically feasible.

  46. You are asking me to read you Criswell. "Plus, if your average power is 100 GW, what is the peak power of the system?" Well, the min design is 20 TWe, but up to 200 TWe is contemplated. Check it out! Then, we can talk about it meaningfully, without spoon feeding. For example, from the abstract: "The LSP System can quickly grow to provide ≥ 2 kWe/person to 10 billion people, ≥ 20 TWe, and enable GWP ≥ 800 T$/y. "

  47. "diveregence" I assume divergence.
    "What beam dispersion have been demonstrated" from before.
    "If by dispersion you mean diffraction, it is thru space, so standard formula applies" my earlier reply. There is no dispersion in Space, if you mean the refraction stuff.

    So, I'm guessing by divergence you mean diffraction??? In that case, Goat Guy will show you the details (standard formula), as he has me many times, of the diffraction limit equation, which I first learned in my youth. Criswell's sizes are correct. He has the large arrays you say are needed, so I won't jump in. The Physics is well understood, as the military is very interested. Electronics costs are going down continuously. Please note that most of your objections are to all Space Solar that uses power beaming of this frequency, not just Criswell LSP.

  48. So Dan, you would need a large number of antenna elements, fantastic phase control, enormous power at 2.5 GHz and low inductance local buffering. And this is just scratching the surface by a rube.

    Plus, if your average power is 100 GW, what is the peak power of the system? When you do your "napkin math" estimation, perhaps you have to assume a safety margin of 10, i.e. you should have 1000 GW (not 100 GW) of 2.5 GHz amplifiers on your "shopping list"? Yes? No?

    Now Dan, you have looked into every aspect of space (and moon) solar power before becoming a proselyte. To me, these aspect as obvious potential problems that any undergrad engineering student can identify by thinking on the subject for half an hour. And perhaps all of these possible road blocks have already been considered and have been solved.

    But then Dan, would you be so kind as to share the evidence that ultra low divergence power beaming is feasible and has a reasonable cost per installed W, or at least has the potential to be reasonably priced per W?

    Note that if you cannot go below two USD per Watt, then there is not way space solar could become competitive because you would already be more expensive than the complete system cost of energy systems on earth, just by making the power beaming part of the system.

  49. If I believe you would depend on your replies. Above, DrPat convinced me that I was wrong about volume heating being dangerous. Long term exposure may be of course be a different question..

    So let's get back to the beam diveregence and cost. You have checked every aspect of space solar. What has been demonstrated in terms of diverergence for a phased array at 2.5 GHz? Nothing? There should be experiments for radar, at least?

    And cost. Have you checked how expensive/cheap 100 GW of switching transistors would be? Today, high speed transistors – for GHz – are expensive per unit power. We are talking about more than 1000 USD per Watt.

    ..On the other hand…

    You are bound to reduce the cost per unit of power as you increase the power. But, you are probably not going to be able just adding a lot of the cheapest transistor in parallel. To get the narrow beam, you must add – remembering undergrad physics – a lot of elements in your phased array. Furthermore, in order to get your incredibly non-divergent beam (0.012 degrees central angle), you will most likely have to control the phase of each element very precisely. To get high power, you probably – please feel free to jump in and correct me at any time – have to add large swaths of antennas. Plus, you have to buffer your power locally to get good phase characteristics and good control.

  50. Bezos understands O'Neill, so he is developing a Space/lunar development and industrialization company, Blue Origin, to do that himself. The first big project will be Space Solar of some kind, am not aware that Bezos understands Criswell however. He has announced that he will buy much launch for Kuiper, but not from Musk necessarily. As I have pointed out, O'Neill plans avoid launch, no matter how cheap. Does Musk have anything like Blue Moon?

  51. Using drones and robots for maintenance? Inspection, I suppose. A flying drone with a screwdriver, I have not seen.
    Are you talking blowing dust off the panels? Bird poop removal with a squirt gun and a spatula? If it is not autonomous, this is going to be a big waste of time and money.
    Well, I suppose you could pay people in India to clean your panels using drones operated via the Internet for $2/hour.

  52. So, why is Bezos not just getting his space solar power up by using SpaceX rockets? Buy as a service? Appart from the fact that Bezos does not like Elon Musk, there doesn't seem to be much stopping him, right?

  53. The Tesla Solar Roof looks better than traditional solar arrays on roofs, but it has some drawbacks. A traditional instillation is done on the south facing part of the roof in the northern hemisphere because it will generate more power on that angle vs the slope on the north side. By installing it on both you get a system that not only performs more poorly compared with its Watt rating, but one that has a sharper peak of power at the 10AM-2PM and much less power produced in the early and late hours of sunshine.
    If this type of roof became popular it would exacerbate the issues of solar, increasing the need for supplemental energy storage beyond the need from the same level of increase from traditional roof arrays.
    Better than either, by far, is an array on a pole with 2-axis tracking.
    The Tesla Solar Roof will not even average to flat roof level of performance.
    It is possible to build a house with this kind of solar system in mind. It would look pretty strange. The roof would likely be a section of a sphere pointed 37.8 degrees southward (in S.F, Ca.). I suspect the center of the sphere would probably be about 2-3x distant the width of the roof.
    Probably some messy math or computer simulations to find the exact optimal for a stationary roof.

  54. The problem with self-driving cars either driving to a charging station or back home to get a charge, is that you nearly double the driving congestion, creating far more waste. And it drives unnecessary distance, using more electricity.
    And if the only energy it used was from home to work, it can't top that off by driving home and back to pick you up. It would be at about the same level. 
    Charging stations would be overwhelmed with robot cars topping off making it difficult for others who need to get somewhere.
    And the real zinger: selfish people would have it drive very slow conserving its energy saving a few cents when they are not in it, making traffic far worse.

  55. Used or old inventory is not what I am talking about. There are only 2 companies making the chips for graphic cards, I doubt either would risk having their cards impounded.

  56. Wiki: "The record conversion efficiency for a rectenna is 90.6% for 2.45 GHz"

    Don't block the beam! Even if it is safe, unless you are trying to get warm, in which case you should pay for the electricity.

  57. Thank you. At 2.5 GHz, the penetration depth seems to be [1] 3cm for bone and about 6 cm for fat (~brain). Pretty ideal for penetrating the skull and heating the brain.

    Now, it may be that volume heating is not dangerous, as per DrPat's answer above. Of course, there is a SAR-safety level of maximum 2 W/kg for hand held devices. Given that the head has a cross section of bout 2 dm2 from above, this would equate to 4W of heating in the brain of 1.4 kg, i.e. about 3W/kg. Below the threshold.

    Now, if the rectenna efficiency is less or equal to 50% the transmission would be double, and so would the absorbed power be. I.e. 6 W/kg, or above the SARS threshold.


  58. Since solar has a capacity factor of about twenty percent most places, the 'alternative' power source would actually be the main power source – or modern industrial societies, which rely on 24/7 power, would have to revert to a more reptilian model. A glance at the daily power graph for California and France ( which have similar sized economies) will show very clearly that part time solar is far less effective, if you want to reduce emissions, than full time nuclear. https://www.electricitymap.org/zone/FR?wind=false&solar=false

  59. I have checked every single issue you mentioned, starting in 1977, with Glaser design, and 1990, when I first heard of Criswell. YOU check on YOUR questions, and verify. YOU won't believe me, will you?

  60. Interesting stuff.

    You have convinced me that volume heating of the human body is not an issue, at least not until above several hundred watts. Of course, the harmful levels noted for radio waves – 4W/kg – is about 300 W for a 75 kg person, so that could be true at the same time as you assessment of several hundred watts of volume heating being safe.

    One last thought. Would the skull not prevent a lot of vibrations from entering the brain cavity? The skull is basically a closed box with a few thin entrances for the spine and for the optical nerves. I intue that these small holes would constitute "obstacles" for the vibrations, and that the skull and brain would vibrate in unison. I.e., the heating would take place outside of the brain. But this may be an incorrect assumption. Did your sources say anything about the volume heating in the brain due to the vibrations?

    Also, what is the effect of prolonged – thousands of hours – of exposure to vibrations?

  61. So you are a strong propent of having solar cells on the moon, and in all these years you did not bother to check if anyone has demonstrated phased array power beaming with a sufficiently concentrated beam?

    That's taking an awful lot on "belief".

    What other things have you not bothered to check? Did you make any sanity check on the costs? Power electronics, mechanics of the solar cells..? About placements of the (necessary) multiple power beaming stations, the required power lines on the moon, availability of raw materials for building the different parts, the effect of micro meterorites on the solar installation?

    Please don't tell me that you have read the paper cited above, and then been spamming the internet about space solar for years, without making some kind of check your self except for that very paper?

  62. Wow! Really? Who would have thought…? Clouds! huh… Just never occurred to me that clouds might interefere with solar power…

    So, let's consider how is this "cloud" issue is relevant to a discussion of setting incentives to modify consumer behavior and thus the demand side of the energy provision equation, to reduce the need for large scale battery storage so sunny day solar power could usually meet most energy demand.

    I presume that on these "cloudy days" (is that the right technical term?) some alternative form of power production would have to kick in. To maintain the incentive structure, the cost of that cloud-induced-non-solar production would need to be spread over other periods.

    Ideally the alternative power system would have relatively low capital costs (compared to adding truly massive quantities of long-duration batteries to handle nearly all 'cloud' interference), so that the cost of creating and maintaining it could be merged into the grid connectivity fee or at worst averaged into the per kWh charges.

  63. What happens if someone tries to sell a graphics card that uses 2x the limit? What if someone buys that card?

    In such cases the main alternative to fines of some form would seem to be jail time.

    Self driving EVs could help a lot to enable daytime charging, even if a human attendant is necessary to plug them in.

  64. Outside of a desert, even sunny climes have cloudy days, and cloud knocks solar output right back, especially in winter, or during the monsoon in tropical countries.

  65. And not noted for steady winds and sun either. Now they are burning a lot of other peoples forests and call it renewable.

  66. Interesting. People would need to be certain that the free daytime electricity is permanent for the incentive to generate changes.

    A hot/cold storage and exchange system could provide heating/AC and hot/chilled water – probably the largest expense and biggest barrier to converting to this way of living. Smart freezers might make and store ice during the day to keep their chill through the night – we'd go back to calling them "ice boxes".

    Very little battery capacity would be needed for plentiful lighting through the evening. And many electronic gadgets already have their own batteries built in.

    Big screen TVs are kind of power hungry and commonly used at night, but maybe AR/VR goggles will displace those? Gaming rigs are also power hungry, but could be paired with batteries adequate to the system and amount a particular person plays.

    Food could be cooked during the day (or purchased pre-cooked) and just re-heated for evening meals. A bit inconvenient…

    Of course, you could always pay the premium for night-time electricity for anything not otherwise handled. Assuming night-time electricity is still partly generated by nuclear, there'll be times at night when the price may be a bit lower, for things like charging EVs at home.

    Might work. Would take a LOT of change-over of appliances, with AC/heating being the worst of it.

  67. I think energy will be free in daytime hours due to a surplus. With that people will change behavior and the need for batteries are greatly exaggerated

  68. Reminds me of the guy who was trying to take stress off of his back by putting springs on his backpack, and found a bounce that would actually save energy as he walked.

    Also, please note that Space Solar power beams are not meant to be body heaters, but would be safe. The general idea is to not block them.

  69. Curiously, the 70s' plan for these things is fairly independent of launch costs. Think of a ratio of launching product v launching factories for that product. The factories would use Space resources that are thus NOT launched, which drives the whole idea. At some amount of product launch, you will wish you had launched the factories instead. Esp if these are factories that make factories. Criswell pg 8 for example of why: exponential growth with factories, but linear launch growth, never gets better. So, if the launch costs go down, launch cheaper and bigger factories!!!

  70. Brett has it 100% right below. It’s on old game of the Left: “We can make our’s cheaper by making their’s more expensive…” Meanwhile, “We” get poorer – and a select few of them get rich beyond their wildest dreams!
    P.T. Barnum, “There’s a sucker born every day…”
    “W. C. Fields, “A bunch of fool’s and their money will soon be parted,,,”
    Vladimir Lenin, “They west is ripe with Useful Idiots…”

  71. Yeah, they said the same thing in 2010… And in 2000 they told us we’d be out of fossil fuels by 2024, and in 1975 they said an ice age was rapidly approaching!
    Lions, Tigers, and Bears, oh my!

  72. " Standard formulas apply"

    I didn't ask what the theory states, but what has been demonstrated. So are you going to answer the question or are you going to evade it once more?

  73. Yes, for Earth use. I was challenging the notion that the *panels* were to be on Earth, not the rare earths' mines for Earth use. So, I'm assuming you agree with Space mining for Space use! Eventually, to Earth delivery too based upon what is found and cost comparison. (edit: probably WAY easier to get rare es from stuff you are melting anyway, the first step in Space, where Sunlight is free. Sep in a fuge.)

    On Earth, most rare earths were processed from material otherwise processed for *normal* (more abundant) stuff, so the digging was already being done. Otherwise, too expensive. This may be changing to escape China market control.

    "even assuming that some non-hydrological process has concentrated them somehow" This brings up an interesting point. Why would there have been no such (edit: hydrologic) process before the planetoid was broken up to make the asteroid? Plus, if there was the normal gravity sep that steals everything from our surface, bingo!

  74. I love the idea of space solar, but right now launch costs are too high. Maybe when the Spacex "Starship" is in regular service it will become viable. Should get launch costs down more than 90%.

  75. from the absract: is Lunar Solar Power is converted on the Moon into to electric power and then to beams of ~2.5 GHz microwaves to which Earth's atmosphere is transparent.

  76. Please see ppg 10-11 for rectennae info. There are very many of them, not just one big one. If by dispersion you mean diffraction, it is thru space, so standard formula applies. If you mean dispersion by the atmos, very little if the correct frequency is chosen.

  77. How did you arrive at "fines"? I suggested no fines. You are against efficiency? The Fed had a program to make efficient lighting. It did not cost much: https://www.ncsl.org/research/energy/energy-efficient-lighting.aspx
    We can just as easily set standards for graphics cards CPUs, power supplies, computer monitors, TVs, refrigerators, washing machines, and dish washers. The downside is small. We already have an "Energy Star" thing, so people can be informed and chose efficient appliances. I am just saying we can insist all the makers comply rather than just a few.

    I am suggesting actions that make it less of an issue to switch to more solar. For supply and demand to mesh well, we need power from solar from dawn to dusk and near the same levels hour by hour. That is achieved by using solar with tracking. And there is a big surge in demand in the evening. If nothing is done to reduce that, solar is of limited utility. And you will need a lot more storage or keep the fossil. And keeping the fossil defeats the purpose. 

    If everyone just charges their car in the evening when there is no solar going into the grid, the cars are powered by fossil or expensive stored energy. The car is at work or at shopping centers in the day…so that is when and where they need charged. But it needs to be at rates identical or very close to those one would pay charging at home, so people use it.

  78. Because in 2020, and for decades to come, it's going to be much, much, cheaper to search for, mine, and process rare elements where we can drive a Toyota HiLux to the mine site.

    Arguably we can identify huge lumps of nickel iron, or water, by telescope and send out some probe to grab chunks. And those bulk materials can be processed in fairly simple processing plants that (in theory, nobody has done this that I'm aware of) can be operated completely remotely, in zero g, hard vacuum, and space like temperature extremes, without needing on site workers.

    But tracking down rare elements, even assuming that some non-hydrological process has concentrated them somehow, is another matter. And even then we are usually looking at fractions of a % ore concentrations, so the processing facility needs to be complex, which means much more difficulty in getting the thing to work in space.

  79. In the last 15 years, road bicycles have undergone a significant design change as people understood that vibration absorption by the human body is a great way to turn mechanical energy to a hot rider.
    Unlike the motorised armoured vehicles, this has two problems.

    1. The rider gets hotter than they would otherwise, which makes them more fatigued.
    2. That mechanical energy… was generated by the rider in the first place. So they get even MORE fatigued.

    So there has been a major push to prevent rough road surfaces from vibrating the rider of bikes.
    At one extreme this has taken the place of ever more elaborate and fancy suspension systems on the mountain bike type vehicles, which have trickled down so that even vehicles designed purely for road use can now come with sprung and damped suspension.
    At the other end of the range, even the lightweight and aerodynamic road and racing bikes are now using much softer, fatter tyres than they did in say 2001. Super hard tyres 18 mm wide are great on polished track surfaces, but on even a good road they transmit all this vibration up into the bike, and thence the rider, where it is absorbed as heat. And like the tank drivers, this can be measured to be as bad as 100W. When your "engine" only has a few 100W continuous output, losing 100W to making yourself hotter is clearly a bad thing.

    tl;dr humans regularly absorb heat at the 100W scale into their bodies. It isn't good.

  80. (…continued)
    So, a couple of hundred watts (at the high end studied) of heat appearing in the muscle, fat, organs etc of anyone in a vibrating vehicle.
    And the body then has to get rid of this heat. Fortunately we have a fluid circulating through our body (blood) which carries this heat to the skin which has cooling mechanisms of radiation, evaporation of sweat, etc.
    Which is how the body dumps all the heat from our metabolism, muscle energy use, etc. So it doesn't injure you to ride in a tank.
    Hundreds of watts is a similar level to the heat generated in hard physical activity.
    But it does require the body to deal with this heat, which uses a lot of energy and (unconscious) effort, and the result was that after hours of sitting on a hard metal seat in a vibrating vehicle you are absolutely exhausted, even though it doesn't seem that you've done much.
    Put in seating designed to not transmit this vibration to the people and then test their ability to do complex tasks after hours of travel and you find that a "comfortable seat" gives a huge performance improvement.
    Now I don't have a tank personally (yet… growth mindset) and so I encountered this story in the next stage. What other vehicle has just about zero damping in a rigid structure that develops lots of vibration and where the only soft component is the human?

  81. Internal volume heating of humans has actually come up in a completely different application I have interest in.
    And you have foolishly brought up the subject so now I'm going to infodump with impunity.
    This was actually identified in a US military study on tanks and the seats therein. Tanks (and other armoured vehicles), as you might suspect, have a lot of vibration in them. They are big, rigid, metal boxes with huge engines and transmissions running at high (ish) speeds over very rough terrain…. vibration.
    And soldiers inside these tanks find it uncomfortable and so there are projects to have softly sprung and damped seats so the soldiers don't get vibrated. And what it was found was that it isn't just more comfortable, it also drastically improved their performance.
    Long story short (curse you word limit!) in a rigid metal structure the vibration is just stored and transmitted back and forth until it reaches something that will actually dissipate the mechanical energy, something with damping. And in a traditional armoured vehicle, the only material with damping is… human flesh. All the metal mass and elastic material would vibrate with quite low viscous loss and only the human flesh would absorb that vibration and turn it into… heat. (See where this is going?)
    So, if you are subject to high levels of vibration (frequency obviously plays a part) you can be absorbing up to a couple of hundred watts of energy that is just dumped into all your flesh as heat.

  82. What beam dispersion have been demonstrated with phased arrays? From your linked power-point, the goal would be to beam 100 GW (at least) onto a "spot" with 80 km diameter. That would correspond to roughly 80km/380000km radians of "cone" angle for the beam, or 0.012 degrees. To me, that seems like a very sharply defined beam.

    So my question to you is, what kind of beam dispersion have been demonstrated with phased array radio beams?

  83. I don't disagree about the disco dating of the paper…

    There is a difference between surface heating and volume heating. Sunlight doesn't penetrate very deep in the body and it is definitely blocked by the human skull. This means that sunlight is unlikely to increase the temperature of the brain. If the skin and upper muscle tissue is slightly warmer, that's OK. Also, if the skin is gets hotter, a bit of sweating will cool the skin.

    Volume heating would presumably heat the brain (and testicles) directly, and you could not dissipate any heat from the brain through sweating. Your only chance would be to transport the heat away with the blood passing through the brain. Would that suffice? You tell me; I would absolutely not bet my brain on it, anyway.

    The cellphone limit of Europe is max 2 W/kg, which seems kind of small considering the power levels that are suggested in the paper above (200 We/m2 of the rectenna).

  84. You only need 4W/kg of absorbed power for the radiation to be dangerous. The absorption depends on the wavelength of the radiation. What frequency radio waves are you suggesting, Dan Lantz? Tell me that and I can tell you how much heating your radiation would cause in a body at a power level of more than 200 W/m2.

    And about 8 USD per watt. How much does a phased array system cost per watt on earth? 1000 USD? 100 0000 USD? And won't it be more expensive on the moon?

  85. My neighbor and I live in NW Florida and both of us have south facing roofs. I also like the idea of energy security if it make economic sense. I believe that there will be huge breakthroughs in energy technology in the next few years.

  86. I agree. If the rectenna field on earth would be 80 km in diameter (100 GWe of power), then the beam from the phased array would have to be theta ~80 km/384000 km = 0.012 degrees. Is it even possible to make such a narrow beam with a phased array? Not theoretically, but practically?

  87. I agree…we should have a capacity or similar payment to recognize the service provided by fossil and nuclear plants.

    On the other hand, we should also have a charge representing the cost of fossil emissions.

  88. Based on your detailed map, you miss the point of Brian's piece…rapid and steep decline in the cost of solar. The map to which you link is dated 2016. It is based on data that is at least 5 years old. The cost of solar has dropped dramatically and will likely continue to drop. Therefore, the amount of area "in the red" economically justified zone on your map is likely to have increased substantially, and will likely continue to do so.

    I look forward to your updated map.

  89. "take stock of the 'problem with beaming'. It is real, and it is not easy
    to work round. We have certainly parried this before, you and I. Yet it
    remains… " We have always decided in the past that Criswell's diffraction numbers work. You seem to hope that somehow I will "see the light" and realize that power beaming is impossible when presented with equations that are the same as they were when I was 11, and support Criswell's dimensions. Are Criswell's calculations correct? Do GEO sats work similarly, with smaller radars?

  90. The main thing is building the rectennae, so I would say that Earth to Earth power beaming is the way to start. Then, launched GEO, then lunar ISRU for whatever you want.

  91. The NSSO did a study on Space Solar ~2007. The Col who ran the blog, Coyote Smith, sent me a thank you for participating, heavily if you can imagine. They included putting the panels on the Moon as a possible plan, in the appendix, which means I did have an impact.

    By Alan Boyle Science editor msnbc.com updated 10/12/2007 12:06:42 PM ET—"I think we have found the killer application that we have been looking for to tie everything together that we're doing in space," Air Force Col. Michael V. "Coyote" Smith, who initiated the study for the Defense Department's National Security Space Office, told msnbc.com on Thursday.

    Space Force will test solar power transmission during X-37B space plane’s next mystery flight Alan Boyle Geek Wire May 6, 2020, 7:33 PM CDT—When a Boeing-built X-37B space plane is sent into orbit this month for the test program’s sixth flight, it will try out a technology that’s been more than a decade in the making: space-based solar power.An experiment designed by the U.S. Naval Research Laboratory will transform solar power into a microwave beam for transmission down to the ground. If such a power-beaming system could be perfected, concentrated microwave energy from space could conceivably be converted to electricity for far-flung military outposts.Back in 2007, the Pentagon issued a report saying the U.S. military could be an “anchor tenant customer” for space-based power generation systems. That report piggybacked on a NASA study that was written a de

  92. I continue to think geosynch is superior to Criswell, due to the smaller minimum viable product, and lack of need for redirector satellites and their losses.

    The satellite system just can start working and paying investors at a much smaller scale, with uninterrupted power delivered to a specific base station.

  93. That was basically just a science faire project for their academy. That's all. I don't think it was intended to actually learn anything, as such.

  94. You can check that on electricitymap.org. It's easiest to see in places like California, southern Italy, or Kyushu, where there's enough solar to plonk an orange bell curve in the middle of the generation graph. Demand peak is almost always in the evening, though it's often more of a continuing plateau. Price peak is also in the evening, so eventually all the power in the middle of the day will be solar – but at negative prices, which will cover the subsidy needed to keep the fossils burning for the rest of the day. https://www.electricitymap.org/zone/US-CAL-CISO?wind=false&solar=false

  95. He has developed and sold a better gas rocket than Raptor, an H rocket for second and already existing 3rd stage of New Glenn, and a deep throttle small H rocket for lunar lander and tug. He is testing an advanced landing system, for unprepared sites, and is in charge of crewed lunar landing effort, altho he does not do the crewed pieces, only the lander. When New Glenn tests next year, it will be far better than FH, crew rated. 25 reuse first design iteration. And he is into Space Solar. "Bezos cites the increased energy demands of better living standards for people as a contributor to his off-world goals. “We will run out of energy,” Bezos said during the presentation. “This is just arithmetic. It’s going to happen. Do we want stasis and rationing or do we want dynamism and growth? This is an easy choice. We know what we want. We just have to get busy.”"

  96. $100/month = $1200/year = 3%.

    This year you might be able to get a home building/renovating loan for less than that. So you might be ahead of the costs. Now if you can then sell your home for $40k more because "fully solar mate, no power bill!" then it is sort of a viable investment.

  97. We already have a series of fines for people who use electricity. They have to pay per kW.h and at the cost of the electricity at the time they use it. This is judged by the electricity supplier to cover the cost of providing that electricity, plus a profit.

  98. Demand DOES peak in the middle of a sunny day in any reasonable location. Though if you're living near the arctic circle like in Iceland or the USA this is probably different.

  99. Death ray of 200W/m2? Direct sunlight is 5 to 6 times as intense. You aren't killing anyone with that.

    You might fry their mobile phones, but for many people that would be a net improvement.

    His USD $8/W numbers are calculated using technology from the 1970s.

    WHich is why I'm not taking the whole project too seriously if nobody has bothered to update this paper since the invention of Disco.

  100. That's easy then. We go back to cheap coal and gas and intend to liquify the CO2 and dispose of it somehow.
    Some guy (I suggest myself) is paid millions of dollars per year to investigate how to do this, with a report due as soon as it becomes feasible, and everyone is happy.

  101. SPS can be part of the solution. But I think we will need a robust manufacturing base in space first. Probably 50 years or more imo.

  102. I think that you've got this backwards.

    Paying people to do nothing doesn't prevent the more energetic segment of the population from simply finding other ways to make money on the side. The inevitable result is that the added production of the "free enterprise" people increases the supply of money which leads to inflation and in the long term reduces the purchasing power of the basic income.

    More production, with the same money supply, leads to deflation, not inflation.
    Now if people were making currency, that would produce inflation, but making goods and services would not.

    The real problem with UBI is that people calculate that by 2100 if everything goes perfectly according to optimistic forecasts then a UBI would be perfectly affordable.
    So they figure they can introduce it now.

    There is a serious flaw in that thinking.

    Like someone in 1820 working out that application of industrial technology to farming means that by 2020 only 2% of the population as farm workers would be needed to feed everyone. So they start moving everyone to the city in 1821.

    We haven't got those ridiculous levels of automated production yet, so UBI can't be done.

    So people sit down to do the math and conclude it can be done providing you redefine UBI so it only applies to a small fraction of the population and /or isn't enough to live on. And then call for UBI because the math works!

  103. 30% of world production of Tellurium is used in solar panels.

    Although Tellurium is NOT a "rare earth" element, it is often listed as such in non-technical publications. It is rare, but that's not what rare earth means.

    There are other, experimental solar cells that do use rare earth elements, but as far as I can tell only Te is used in any commercial quantity.

    Neodymium IS a rare Earth element, and is heavily used in the permanent magnets in many wind power generators.

    Throw in a couple of battery designs with a tiny level of rare earths needed to tweak the behaviour of the anodes or something.

    These facts are often mixed together in non-technical reports as "rare earths are used in renewable power" and after being filtered through 5 layers of non-technical reporting copying off earlier reports, ends up as "rare earths are used in solar power".

  104. You mean, there's no way the PV system will last long enough to pay for itself, even if your neighbour does.

  105. I'm not so sure about robots replacing human labor completely. Augmenting yes. Improving productivity, yes. But completely, probably not. Robots cost money, and they have to be maintained and powered, which also cost money. They have to be managed, and they have to be programmed or have to have an AI capable of human level intelligence. The challenge there is the Isaac Asimov problem – they revolt and kill us all. "I, Robot" is the natural result.

    Basic income is a communist system. No such system has ever worked. Even the pilgrims tried and abandoned it. Communism essentially stifles human enterprise and free market economies. Paying people to do nothing doesn't prevent the more energetic segment of the population from simply finding other ways to make money on the side. The inevitable result is that the added production of the "free enterprise" people increases the supply of money which leads to inflation and in the long term reduces the purchasing power of the basic income. Over the long term, the basic income isn't sufficient to sustain the masses and eventually the system collapses.

  106. It is mainly the utility scale solar that electrifies a large part of world energy. We still need the other options like nuclear to generate the heat

  107. Home installation cost just has to be less than the cost of the roof. I will be writing about the Tesla Solar roof shortly and I will likely be getting one. 100 million homes in the US with 10 kw of solar. 1TW of solar. 4X for world. This is relevant but does not electrify everything.

  108. Space based solar can work. Should be affordable with fully reusable spacex super heavy. Plus we can put giant mirrors in space to light up the large solar farms.

  109. "Bezos cites the increased energy demands of better living standards for people as a contributor to his off-world goals. “We will run out of energy,” Bezos said during the presentation. “This is just arithmetic. It’s going to happen. Do we want stasis and rationing or do we want dynamism and growth? This is an easy choice. We know what we want. We just have to get busy.”"

  110. No. Robots will replace labour. Universal basic income needs to replace such wages.

    Remember a time when humans had to hunt or forage for food? That is what earning a living would feel like in a couple of generations.

  111. Rare earth minerals are mined to provide rare earth elements. Which of the REEs are used in solar panels? The REEs are the lanthanides plus scandium and yttrium.

  112. So are you as tired of hearing about Mars as I am, or global heating, or overpopulation, or do you just feel uncomfortable with an actual solution that is rarely mentioned, altho Bezos may be about to change that. You will certainly be happy to hear that I have stopped my 4 decades of complaining that we should go to the Moon before Mars. Because we are. Now to Space, not the Moon, other than for resource exploitation.

  113. You made your own Dr. Goebbels' principle that
    propaganda is strengthened by repetition.
    You have reached the point that you should be
    put in charge of a re-education camp for taikonauts.

  114. "RF rectennas are used for several forms of wireless power transfer. In
    the microwave range, experimental devices have reached a power
    conversion efficiency of 85-90%. The record conversion efficiency for a rectenna is 90.6% for 2.45 GHz, with lower efficiency of about 82% achieved 5.82 GHz." But these are definite problems, not solved by citations. As for the focus stuff, the LSP radars have to be replicated as the power delivery size goes beyond the diffraction reqmt, so it gets easier to just have more stations. The point is, you can easily make fewer and larger if needed. Think of Criswell as similar to O'Neill, he is presenting Physics based possibilities, with much engineering to be done. All of the advances in launch cost (soon) and electronics in general may make Criswell 2009 details *old hat*, while making his overall plan even more attractive. The overlap with orbiting Space Solar, esp L5, is almost complete, so 'problem with beaming' is very important to understand well. Global heating and all that. Criswell sez 80's tech economical, but better works fine too.

  115. It depends where you live and what kind of deal you get, and a few other things like how much power you use, your service level (ours has to be upgraded), and the shape and orientation of your roof. I live in San Diego County, so quite a bit of sun, and our rates are abusive. Thing is, my roof is not lasting as long as it was supposed to. I got some used panels cheap, and I have some old forklift batteries. I am going to look for loopholes to avoid permits. I may be able to put one on the roof of an old trailer I am using for storage. Small off grid stuff. Power my big freezer and a little fridge. I like the idea of having that secure from power outages. And maybe I could wire up a few lights in the house that are just used during blackouts and maybe an outlet for a fan. Last outage it was very hot and no breeze.

  116. Dan, instead of perpetually referencing Criswell's paper, do take stock of the 'problem with beaming'. It is real, and it is not easy to work round. We have certainly parried this before, you and I. Yet it remains… 

    EFFICIENT beaming of power is a technological and physics basket of feral cats. Mad cats. With claws and teeth. 

    Basically the transmitter must be frequency (wavelength) well-matched to the the receiver. Not much of a problem, that. But it is one cog in the wheel. After that, one has to choose a wavelength for power beaming. Since the kinds of DC-to-EMF-beam technologies (semiconductors and TWTs) both work more efficiently at lower (longer) wavelengths, ideally one might want to choose the longest wavelength that'd practically work-the-distance-gap. 

    However, therein is the hard tradeoff: the product of transmitter and receiver 'diameter' directly predicts the longest wavelength (for a given distance) one need deploy. Longer the distance, smaller the wavelength, all other being equal. 

    Lastly, even using exotic AC-to-DC synchronous active rectification, the received power is quite a bit lower than that transmitted. 

    It stings when only 50% of the produced power … is received and useable for 'the grid'. Hurts a lot. 

    ⋅-⋅-⋅ Just saying, ⋅-⋅-⋅
    ⋅-=≡ GoatGuy ✓ ≡=-⋅

  117. You have made an excellent case for Space Solar and the related power beaming over Earth solar alone. I like nukes for H and heat, but Space Solar for H and grid. Can't get past boiling water as a waste of time to make elect.

  118. In this case, the cost does not go up with increased usage. You have no looming shortage of Sunlight, in Space, so the *energy* itself is free for the taking. Some other mass production ideas lead to shortages and new costs with scale up. (edit: whale oil, for example, as primary source of lighting). Not Space Solar, at any relevant scale whatsoever.

  119. My neighbor spent $40,000 for a solar voltaic system for his home. My electric bill averages around $100 per month. There's no way that my neighbor will live long enough for his savings to pay for the system.

  120. Earth to Earth power beaming uses the same orbiting re director sats as Lunar Solar Power, so is mostly in Space, altho twice thru the atmos. Please don't lie about my being a liar.

  121. I think the analysis ignores energy storage costs (no sun at night, wind varies) and the need for oncall generation capacity. If you want reliable energy available when needed, you must have a huge increase in storage or the current huge nuclear, hydro and fossil fuel sources. Also, solar and wind farms kill animals and are a terrible eyesore. So nope, we are not that close to solar overtaking efficient fossil. It is decades away, if ever.

  122. And, please note that the cost of fracked natural gas and oil has fallen much faster than the price of solar energy during the last 10 years.

  123. Beaming things through atmosphere incurs an energy loss due to dissipation, please don't lie about such things as it doesn't benefit your position.

  124. Production isn't where the problems are for solar; storage and distribution are. If those costs aren't dropping, it doesn't matter how far panels themselves drop.

  125. Solar generation goes to zero every night and stays there for hours. That isn't a small scale change in production. Solar also peaks around midday, which is not when demand peaks. Generation also falls off a cliff on cloudy days and when snow covers the panels. Solar is hugely variable so it needs huge amounts of battery backup. If USA were using solar for 50% of our energy use we would probably need roughly 50 TW/h of storage at the minimum. Even with 50 TW/h I wouldn't feel great about it, but it would be a decent level of storage at least. Producing that many batteries would take centuries at current rates. Even with productivity improvements and new factories, it will take decades to get to 50% renewable energy.

    Better to just build small modular reactors.

  126. Power beaming solves the intermittancy problem. see ppg 12-13 of the above Criswell paper for the clear and obvious solution.

  127. Of course, the wind blows in the desert, so it is FAR harder to do sheets of solar collector material 1/10th as thick as tissue paper there than on the Moon, or in Space, as recommended by O'Neill in the '70s.

  128. "The lowest cost solar project bids are at costs of 1.35 cents per kwh in Abu Dhabi" And that is without power beaming!

  129. If it was that cheap or even feasible to set up unmanned solar panel production in the middle of a desolate wasteland and have them stamp out solar panels they would be doing it in every desert on earth.

  130. Homes need roofs anyway, so rooftop solar has a future, as the structure/land cost is mostly there already. Then power beaming from deserts and Space takes over. Laying a very thin sheet on the Moon surface is very cheap, once you are there.

  131. "transmission constraints" probably need to get careful with the word "transmission" as power beaming uses a frequency that only very heavy rain will impact at all, so inherently has no constraints. Many call "conduction" land lines "transmission" lines, for some reason.

  132. His cost estimates are undoubtedly high, using expensive launch, pre-Musk. Otherwise, things are cheaper and easier in Space than on Earth, right?

    Beams follow pilot signals sent to request energy, and the miracle of phased array allows exactly what you describe as to multiple (thousands) of rectennae.

    "Death ray" of 200 We per m^2? Tell the birsds that nest in the radio towers to get out!

    Your objections are spurious and uninformed. And very OLD and answered.

  133. Ironic… considering you need to grind rocks to dust in order to get the rare earth minerals necessary to make solar panels.

  134. Solar operators need incentives to have horizon to horizon tracking systems. Maybe pay them more for the power they make near dawn and dusk. Dusk especially, as that is where most of the shortfall is. 
    We also need more efficient appliances that are typically used in the evenings. TVs/large monitors need to be particularly efficient but also routers, phone charging systems, microwave ovens, electric ranges, refrigeration, washing machines, dish washers, lights (LEDs/OLEDs are good, we just need to ban the sale of the other inefficient stuff), computer processors need to be limited to 50W, game consoles to 40W, graphics cards to 40W.
    I would not presume to say when and how people can use these products in their homes, this would all be on the manufacturers shoulders.
    We could also encourage single piece VR systems (or ones that work with smartphones) that use say 25W or less. The more people use that, the less will be using big TVs and large monitors.
    It should be noted that electric cars are a very poor fit with solar as they are being charged at night typically. Unless we can get charging at typical destinations (not necessarily fast charging, just say 220V), this mismatch will be an increasing burden and slow the phaseout of fossil fuel.
    Of course, if we just get the big nuclear plants built, solar can just help with the air conditioning surges.

  135. There is no intermittency problem. People forget that demand is intermittent. Every time an elevator moves demand increases. The power distribution system is built to handle small scale changes in production and demand. Something simple as dropping voltage. Longer term power demand is calculated based on the weather forecast and generation asset and power purchases are scheduled to fill the demand. In case of an emergency gas turbine generators are brought online quickly. A note to the fans of solar intermittency problem, generators do tripped, transmission and distribution feeders do fail, and thunder storms do happen.

  136. Couple of things people don't understand about generation and solar. First solar operates during the day when demand is at its peak and cost for power is twice as high. Power is dispatched as the next cheapest $/MW. This $/MW is calculated from fuel cost and maintenance cost not financing cost. Since solar uses no fuel, it wins all the time. If there is no transmission constraints every MW of solar available will sell. The same can't be said for fossil power. To put in plain term, every GW of solar built and installed will replace a GW of fossil power. First it will be coal and then it will be gas.

  137. I took the trouble of reading through your source, and now I regret it.

    It assumes that you can make a complete system – solar generation, transmission and at least two gigantic directional (phased array) antennas for 8 USD per Watt. On the moon. And most of the material and production equipment needs to be build on the moon.

    Also, it assumes that you will actively direct a beam of at least 100 GW to discrete locations on earth while the two planets are rotating, having a power density of about 200 We/m2 on earth, i.e. covering an area of about 500 km2. So there would be a 80 km diameter beam with more than 200 W/m2 power (efficiency of rectenna< 100%). The beam would have to switch locations several times per day due to the rotation and track the rectenna fields when "locked" to one field.

    Imagine you direct the beam towards New York by mistake. Would you kill everyone in New York? How about airplanes, would you shoot them down with the death ray? And what if wild birds or animals would wonder into the beam? Slow heating death? You only need about 4 W/kg to damage an animal/human.

    The whole thing is a power-point-worthy idea, nothing else.

  138. We've been through this many times. It's a given. Mature and mass produced tech-anything dives in cost. End of story.

  139. They had to build coal plants to replace the nuclear they dumped because of Fukushima fears. (Germany's not exactly noted for their earthquakes or tsunamis, oddly enough.)

  140. Still, innovation in installation, logistics, etc etc are also still being made.
    For instance:

    • Using drones and robots for maintenance
    • designing modules so that they precisely fit inside containers. This would result in a shipping container housing 50% more solar panels. This would reduce shipping cost alone by a factor of 2 for the shipment of modules.


    Keeping improving on all fronts is what get's that 30% learning curve.

  141. Actually, the Space Solar is cheaper for the sunlight intensity and duration factors, AND ALSO any other advances that make the projections for Earth solar so cheap. The power beaming used in all Space Solar removes most need for load leveling, even if only Earth to Earth at first. Fossil fuels are artificially cheapened by lack of C fee, which would ideally be used for:
    which also opens Space. Glad to help!

  142. Intermittency is indeed a problem that will have to be solved. Personally, I’m optimistic on that front. When the cost of generating electricity gets low enough, many energy storage technologies become viable. There’s going to be significant progress in this area.

    But I’m confused how the referenced article debunks Brian’s analysis. The article says that the increase in German emissions was due to decommissioning nuclear and replacing it with coal and natural gas generation. Can you elaborate? Or provide a more relevant article?

  143. Rooftop solar can be as expensive as nuclear per unit.

    Fission is too hard for Americans, the more they build the more inept they become at it.



  144. Well, the cost for a large installations also include installation cost, so even if this is not an apples-for-apples comparison, some of that dramatic price drop should rub of on consumer installations.

    In addition, I don't believe that the installation cannot be made simpler. You could add auto detection of parts so that you only have to connect them and the system would know how to be configured, for starters. And you could do all connections with weather-proof external contacts so that it's just a question of "clicking things into slots". It should be possible to make it so that the only actual wire-to-wire connections would be the power connection from the solar panels to AC-power of the house. The already installed electrical system is unlikely to have an AC-power input contact…

    Moverover, considering that a 16 kW Tesla system costs about 32 000 USD to install [1], the installation part of that should be negligible. Lets say two men over two days, totaling 32 hours. 50 USD per hour, meaning about 1800 USD. If you half the equipment cost from, say, 30 000 USD to 15 000 USD, the final price including installation would be 17 000 USD, i.e. pretty close to half the original price.

    Of course, installation procedures will be simplified if the market expands…


  145. In 2040 the sun will still only shine during the day, of this I am sure. If solar is cheaper than fossil fuels, it will only be because fossil fuel plants are compelled to be unnaturally expensive by forcing them to provide uncompensated load leveling for solar.

    Even the current pricing relies on uncompensated load leveling from fossil fuels, after all. Where are battery costs in your projections?

    Now, maybe by 2040 we'll be building SPS's, for which the Sun only shining during the day is irrelevant. But I doubt they'll be FIVE times cheaper, unless the cost of fossil fuels has been artificially inflated.

  146. "Solar modules are about one-third of the cost of solar electricity. The rest comes from related equipment (inverters, trackers, cabling, mounting systems), land, labor, and other non-module costs."
    Permits and inspections for solar are already much easier to get than for any other industry.

  147. No innovation in installation? The maze of permits, inspections, sales, etc. can't be made more efficient?

  148. While solar panel price might drop to cents per kilowatt, installation price will not decrease. So for home installation there will always be a highish cost

Comments are closed.