Built on Facts

In theory I like solar power a lot. There’s a lot to be said for energy that falls from the sky for free for as long as we care to use it. But actually getting that energy to do useful work in an economical way. Solar panels are expensive and not terribly environmentally friendly to produce, and sunlight is only available during the day and only to the extent that the sky is clear. Getting energy to use during less sunny periods requires either lots of excess capacity during the day along with storage facilities, or alternate generation methods. The first is wildly uneconomical, the second at least partially defeats the purpose.

Still, it’s not a bad technology and with time it’s likely to become more and more practical at least for areas with high insolation. (Insolation being the snazzy word for the solar power shining on a particular location, typically averaged over a year to take into account day/night/seasonal/weather variations.) The average insolation per square meter at the earth’s surface is this, according to Wikipedia:

White corresponds to about 200 watts per square meter, so for instance your computer will probably take about 1 square meter worth of solar panel to operate on average. During a sunny day it will take a lot less, but during the night you’ll have needed the extra area to store energy to use while the sun is down.

There are alternatives. You could put a solar panel in space and beam the power back down to earth. Leaving aside all the practicality considerations, it’s not a bad way around some of the problems of solar. There’s no night and no weather, so you’ll get the full ~1300 watts per square meter of sunlight 24 hours per day, 7 says per week. Which is pretty nice. You’d still need a lot of solar panels – the US eats about 3.8 trillion kilowatt hours per year, which works out to about 130 square miles of solar panels each getting the full blast of constant sunlight. Still, it’s not something that’s entirely inconceivable. You’d need many times that to get the same amount of energy from panels at the earth’s surface.

What about getting the power from space to the ground? Running an extension cord is out of the question. The easiest way is to turn the energy back into light and shine it down to collection stations. Clearly just using visible light wouldn’t help, you want something easier to collect. Microwaves are an easy solution. Set up a field of cheap metal antennas oriented in the appropriate direction and you can capture the beamed down radiation cheaply and easily.

The Japanese space program is planning to give it a try:

The concept of solar panels beaming down energy from space has long been pondered–and long been dismissed as too costly and impractical. But in Japan the seemingly far-fetched scheme has received renewed attention amid the current global energy crisis and concerns about the environment. Last year researchers at the Institute for Laser Technology in Osaka produced up to 180 watts of laser power from sunlight. In February scientists in Hokkaido began ground tests of a power transmission system designed to send energy in microwave form to Earth.

The laser and microwave research projects are two halves of a bold plan for a space solar power system (SSPS) under the aegis of Japan’s space agency, the Japan Aerospace Exploration Agency (JAXA). Specifically, by 2030 the agency aims to put into geostationary orbit a solar-power generator that will transmit one gigawatt of energy to Earth, equivalent to the output of a large nuclear power plant. The energy would be sent to the surface in microwave or laser form, where it would be converted into electricity for commercial power grids or stored in the form of hydrogen.

It will be crazy expensive. To get a payload to geostationary orbit is something like $10,000 a pound, and space-hardened solar panels and microwave transmitters are not exactly featherweight. The power generated will have to be very cheap and reliable indeed to make up for the truly massive infrastructure costs.

To make matters worse, those of you who used to play Sim City might remember a disaster where just this technology would get knocked out of alignment and set your city on fire. But fortunately this wouldn’t happen in the real world. The power densities at the ground are relatively small; in the proposed eventual Japanese implementation they work out to about 200 watts per square meter. Now this wouldn’t leave death and flames in its wake, but it is about 4 times the occupational limit for microwave exposure. Stringent safeguards would have to be in place.

I’d love to see this technology becoming commonplace, but I don’t honestly see it happening until we find a cheap way to get to orbit. Maybe my other pet sci-fi dream of the space elevator will eventually come to pass and bring space solar along for the ride.

Comments

#1 Mike Olson
September 30, 2009

I really enjoyed this entry. To me it simply has always seemed that what looks ridiculous now, might one day be a possibility. Sci Fi, offers some interesting examples: The b/w labs of Dr. Frankenstien, the original Star Trek…on the one hand you’ve got “transporters” on the other no LED displays. Folks making predictions get just as goofy: we were all going to have nuclear cars and jet packs by the year 2000. But, no one thought we’d have women scientists or ship captains(watch the episode of Star Trek with the psychotic woman who takes Kirk’s body). No one ever guessed we’d have computers in nearly every home, or use this advanced computer knowledge to unravel the human genome or have the advances in medical care we’ve got. Of course we’re not all wearing skin tight white clothing, with no body or cranial hair either. Solar power might come around…we also might find something nearly as cool.
#2 Winter Toad
September 30, 2009

Actually, in the classic design there’s zero chance of the beam going off-target. The satellite doesn’t have the hardware to send a beam to any place at all. Instead, the ground station sends up a pilot beam that strikes the phased array on orbit. Electronics there phase-reverse the inbound signal at each antenna element and amplify it from the power-generation system. The net effect is that the beam isn’t “sent” anywhere, it’s merely returned to its origin, but at higher intensity. Lose the pilot beam and the antenna decoheres, spraying its energy harmlessly.

Because of the physics of the phased array, the transmitter and receiver sizes, and the wavelengths involved, if some miscreant wanted to redirect the beam elsewhere he would have to send up the pilot beam from a similarly sized antenna array. It’s unlikely that somebody could stealthily build a football-field-sized phased array microwave antenna in a populated area.
#3 Dan Nielsen
September 30, 2009

Hi, does the beam not get partially scattered and absorbed by the atmosphere to any large degree?
#4 Chris
September 30, 2009

How does putting a solar panel in space solve the problem of downtime at night? If it’s in geosynchronous orbit, then the Earth will still be between it and the sun for several hours a day. And satellite TV all but stops working during a storm, would this system be similarly affected by weather?
#5 Winter Toad
September 30, 2009

Chris: Actually, a satellite in geosynchronous orbit is almost always in sunlight. A geosynchronous satellite enters “eclipse season” twice a year. Each interval is about 6 weeks long, at the time of the equinoxes, and an eclipse can last as long as 70 minutes at the height of the season. This, in fact, is a significant constraint on commercial satellite design. The batteries have to be designed to handle sunlight outages of more than an hour in two clusters each year. Because batteries are heavy, this is a major cost in the design. There have been companies that have been investigating the possibility of providing artificial sunlight with ground-based lasers for customer satellites in these circumstances, relieving the customers of the need to loft those big batteries into space to handle the relatively rare eclipses.

Lower Earth orbits have more severe eclipse problems, as well as the problem of the pilot beam having to track a moving target.

Note one more thing, however. When a satellite is in eclipse, it’s local night. The demand for electricity reduces at night, so it may not be a big deal that the satellites sometimes don’t provide power.
#6 ppnl
September 30, 2009

If solar is not competitive on earth even in daylight at peak prices then it seems unlikely to be competitive after spending the money to lift square miles of cells into orbit.

How do you make such large structures safe anyway. Even something as small as the shuttle or space station has to worry about collisions. Square miles of cells that has to function for many decades? Get real.

Space based solar would seem to be incompatible with geosynchronous communication satellites. I think communication satellites are kinda space limited as it is.

The technology to make it possible would also make it irrelevant. This is well beyond the technological horizon of fusion power let alone safer fission power.
#7 Josh
September 30, 2009

One major issue with solar here on earth is water. The panels need to be washed regularly or they lose efficiency. Since the areas that are ideal for solar are often deserts, problems ensue. Does anyone know what kind of maintenance panels in space would need to ensure continued efficiency? Sending crews up to service them would add dramatically to the expense.
#8 Omega Centauri
September 30, 2009

We should contrast this with another “easier” solar technology. Concentrated photovoltaics (CPV). In CPV some sort of optics is used to focus sunlight on a small area of photovoltaic material. Typical concentration factors are several hundred to a few thousand times. For either CPV, or space based power the cost per unit area of photovoltaics has to be high enough to justify all the extra infrastructure to optimally use the limited PV material. Of course CPV has the usual “won’t work at night -or with cloudy weather” problem, but in principle it might work out to be cheap.

A minor thing in favor of these schemes, which go to great lengths to make optimal use of a small amount of active PV material, is that higher quality (higher efficency) cells can be used than would be economical for simple ground ased fixed panels. But this efficiency factor is likely no more than a factor of two.

I think CPV makes good sense, in terms of the risk/reward for investment, whereas space based requires a very high investment. Both carry the risk, that if bulk PV becomes cheap enough per unit area, that the cost of either concentration (or space basing) isn’t worth it.
#9 TBRP
September 30, 2009

Chris:
To add to what Winter Toad said, if you care more about having the satellite in the sun than having it constantly beaming down power, it could be put in a sun-synchronos orbit. Then it will never be in shade except for the rare case of a lunar eclipse. You would need a constellation in order to perpetually get power to a single spot on earth, but you’d get the bonus of being able to sell power to anywhere on the planet if they want to build the ground stations.
#10 Winter Toad
September 30, 2009

TBRP: Quite correct, a sun-synchronous orbit would keep the satellite in the sun. Unfortunately, you’d have a ruinous duty cycle. Those orbits are usually about 800 km up, so they don’t have a very large ground-visible footprint. The area of the Earth that sees the satellite at least 30 degrees above the horizon is about 0.1% of the total surface area of the Earth. You’d need 1000 ground stations, evenly distributed over the solid, damp, and icy parts of the planet to get continuous use our of your satellite, and you can’t store power for later transmission, it’s “beam it or lose it”.
#11 doug l
September 30, 2009

I’ve long been an ardent supporter of developing space based solar and have no illusions that it will be easy or even cheap at the beginning, but it should be noted that in doing so we will be creating an industry and engineering know-how to create the platforms from which the further exploration and exploitation of the features and resouces that are there beyond low earth orbit.
Of course the expense now is because the cost to lift mass into orbit is so expensive but that is largely because we have been using a somewhat inappropriate and technlogically archaic legacy of when space exploration was a byproduct of the cold war’s nuclear balistic missile program. A man in a can atop a tube filled with explosives is indeed possible but surely there are better ways. Thinking beyond that and looking at alternatives such as the use of electromagnetic mass drivers, nuclear propelled lifting bodies and a few other possible technologies have not been fully explored due to apprehensions in the case of nuclear lifting craft, and lack of money due to the incredible expense that the operation of the shuttle and those darn ballistic missiles pose.
A suggestion that has shown some promise and is gaining some momentum is to offer industry significant prize money to accomplish specific goals. No payment until targets are achieved and once they are achieved it’s hard to imagine there would not be applications that would stagger the current imagination. NASA is a tremendous organization but in its last couple of decades it has become focused on perfecting a system that was meant to be only a step into something far more practical. But to save a penny we’ve spent dollars…or billions of dollars actually and jeopardized the lives of crewmen and the very existence of the program in which so much as been invested without gain.
To do what it has done has required an enormous amount of effort, expenditure of resources and organized manpower and those together have created a kind of inertial that may be harder than the planet’s gravity to overcome, unless we re-ignite the drive that began when a relatively small but very smart group of engineeers were told they were going to the moon in ten year, so get moving.
Humans and delicate packages can be sent by the equivalent of a taxi service while massive, high-g capable components and bulk materials like fuel or water or shielding can be shot into space at forces humans cannot endure for far less money..some have speculated that a 100kg cannister might be lifted to low earth orbit for a few hundred dollars without emissions and repeatedly every few minutes.
Our capacity with remotely operating equipement and robots will further enable us..and when I say us, I’m not too sure I actually mean the US anymore. There’s gold in them thar hills, or as it has been sometimes said “it’s raining soup and we’re getting pretty hungry”.
#12 Geoff
September 30, 2009

Omega: Why not take the best of both worlds and make a CPV array in space? You’d still need an array of the same size, but it could be lightweight lenses instead of heavy solar panels. This would at least reduce the installation costs (hopefully).
#13 Nathan Myers
September 30, 2009

Has no one noticed that Pacific Gas and Electric of California has contracted for a 200 MW space-based solar array? The contractor is Solaren. It’s supposed to be providing power in 2016. It would be followed by a series of GW-scale systems.

http://www.grist.org/article/2009-04-16-solera-space-solar/

For a cynical take, this is hoisted from the comments there:

Hard not to laugh at this. I started by career post-grad school with Arthur D. Little, a company that was on the cutting edge of a huge array of technologies, going back to the turn of the 20th century – we had a role in the development of everything from the Xerox to the Stairmaster, and not a few pretty novel R&D plays in the energy space. (One of my early projects was working with the team to allow fuel cells to work on gasoline.)Anyway, any company with a 100 year history in R&D comes up with it’s share of duds – as well as a legacy of figuring out how to extract large sums of $ from excited investors for continued consulting revenue in support of those duds. (My personal dud-based source of steady consulting revenue was the hydrogen economy. Knew it was bonkers, but also knew how to extract consulting fees out of continued investigations.) Anyway, whenever us worker-bees would get onto conversations about our more dubious consulting assignments, one of the old-timers would inevitably raise microwave-beaming of solar space power, which had generated huge consulting fees for us through the years and was the perfect dud consulting gig. Never likely to work, but always likely to attract $ from naive, wealthy idealists (think DOE, EPRI, DARPA, etc.). Better yet, it could get rolled out again every time there was an energy crisis, or an increase in NASA budgets, or a build-up at the Pentagon. Damned near recession proof, and pure consulting genius. But the big money isn’t in turning those duds into consulting revenue… it’s in figuring out how to turn it into a big IPO. Good luck, Solaren!
#14 Whomever1
September 30, 2009

I imagine that the cost of building 20 geosynchronous solar power stations would be much less than 20 times the cost of building one. But we might have build some infrastructure in space first.
#15 Michael Varney
September 30, 2009

Reminds me of an old Usenet post I made several years ago working a quick little Fermi problem on solar panels on the moon. A much better idea in my opinion than blowing the moon up.

Also, a local CU blogger, Science Geek Girl, did her thesis on higher efficiency photovoltaic systems.
http://blog.sciencegeekgirl.com/

She may have some interesting info on the topic.
#16 Uncle Al
October 1, 2009

How big is an economic sunsat? Ground facilities are typically 1-5 GW. To pay back your fabrication and installation costs MTBF you will need more than 100X that output baseline. No staff on board, either: Outside both the atmosphere and the magnetosphere they well be thoroughly fried by radiation:

http://science.nasa.gov/headlines/y2009/29sep_cosmicrays.htm
NASA’s 50-year data model projection is mirror-image wrong.

Human skin area is 1.8 m^2, 4(pi)steradians on a sphere, 70 kg asstronaught. [8x10^(-4)][(270+450)/2](22.6)[3000] = 19,500 hits/second or 8.8 billion hits/kg-year whole body.

Tell us how to collimate a 500 GW microwave beam over a 23,500 mile path – something about numerical aperture, wavelength vs. emitter span, comes to mind. Then, side lobes and atmospheric scattering. The ionosphere will not like you no matter what. A meteor zooms in to ionize a conductive path through the upper atmosphere and… EMP.
#17 Abby Normal
October 1, 2009

Would a space elevator put “running an extension cord” back into question?
#18 Margaret
October 1, 2009

Coincidentially, I listened to this today. (Scroll down to “Space Solar Power”.) Mp3 file here of a debate on pros and cons of developing space solar power, with an overview of some of the technical and economic issues that would have to be solved if it is to become viable.
#19 CCPhysicist
October 1, 2009

Solar electricity has all of the problems you mention, but solar hot water does not. It is a lot easier to store hot water than electricity!
#20 uber
October 1, 2009

I find it amusing to read people’s thoughts on solar power. The very same people that are constantly touting (and/or relying on) the elegance of the internet with it’s multi-node, distributed computing power approach; put down solar power as not ‘the answer’ to our power needs because of the items you mention in your post.

Why is it that when the thought of using solar power comes up, we revert back to ‘main frame’ thinking and find ways to argue that solar power is not the ‘holy grail’ to our power requirements.

Then you argue, if we were to rely on another type of power generation *GASP* it would “partially defeats the purpose.”

Here’s an idea, use solar power when it’s available, use another renewable power source(s) when it’s/they are available and (sham-wow!) we’ve covered our energy needs!

Why do you all insist on thinking in terms of ‘mainframes’ and single point computing power and then applying it to the ‘solution’ of renewable energy?!?!
#21 Michael Varney
October 1, 2009

Just go with nuclear. *shrug*
Oh, wait… enviro-whiners… =/
#22 Furkan
October 2, 2009

Really we have solar power. We must decrease our oil inveteracy. Deserts are suitable for solar power. But Sands actions will close top ground. Also Solar power productivity ise very low.
#23 Taylen Peterson
October 2, 2009

It may seem crazy, but so did landing on the moon at one time. And the world is flat, right? Who knows what the future has in store. Great post, though! Thanks for diving into the possibility of space solar, always interested in the solar world.
#24 A. lagrangian
October 4, 2009

Great great post. I have been reading your blog for awhile, and I really enjoy it.

thanks matt
#25 zuggernaut
October 15, 2009

There is a book called Sunstroke written by author David Kagan that clearly details the biohazards of power-beaming concentrated microwaves from space to the ground. Sunstroke also describes the real dangers if a malfunction develops in the so-called “fail-safe pilot beam” mechanism that will supposedly defocus the high-intensity microwave transmission in the event of an emergency. This “pilot beam” can also be mimicked from other clandestine sources, with absolutely disastrous results to populated areas and to the environment in general.

Sunstroke spells out quite clearly that such a space solar power station employing high-intensity microwave beams will be used as a devastating multi-pronged weapon: it can supply electricity to the military, be used to fry enemy ground troops, aircraft, ocean-going vessels, and also be used to disrupt enemy communications as well as destroy their agricultural capabilities. All in the guise of being a “purely civilian alternative energy project”.
#26 Jim Thomerson
October 18, 2009

Have you read O’Neil’s “The High Frontier”? He proposed large peopled satellites at the Lagrange points; doing manufacturing is space and beaming power down. This was back in the 70′s.