jump to navigation

Here we go again with the SPSs. September 1, 2009

Posted by Maury Markowitz in solar power satellites.

The SPS’s just won’t die. In spite of being utterly uneconomical, and doomed to being destroyed in space even if you press ahead anyway, the announcements just keep rolling in. Nothing like a little unobtanium to lighten up the “credit crunch”.

The latest is a Bloomberg story about how Mitsubishi and IHI have joined an effort to build a 2 trillion yen ($21 billion USD) SPS that generates 1 GW of power. At $12,000 a pound those numbers seem a bit fishy to me, so I read on. When noting the high cost of the space launches needed to put this stuff up there, quickly-contacted talking head Hiroshi Yoshida said “These expenses need to be lowered to a hundredth of current estimates”. A hundredth eh? And how are they going to do that? “Humankind will some day need this technology, but it will take a long time before we use it”.

Ahhh, they’re going to wish it into existence! Excellent. And when they do, it will consist of a cardboard box full of monkeys that says “Peekaboo Perfect Space Explore!” on the side (sorry, can’t resist, NSFW page if you want a laugh).

Once again I’m left scratching my head: if you’re going to wish for some super-tech to save your SPS, why is that tech always a better rocket? Why not just hope for a better solar cell? After all, that seems much more likely than a better rocket, given the Japanese track record in that department. But I know why; if you make a better solar cell, you’re still better off leaving it here on planet Earth. And this isn’t really about SPS’s, it’s about excuses to build cool rockets. Funny that Mitsubishi makes JAXA’s  H-2 rocket… but that’s just a coincidence, I’m sure.

Since my last diatribe, a reader pointed me in the direction of this excellent report. So don’t take my word for it, read it and weep.


1. John C. Mankins - September 1, 2009

Dear Maury,

It seems fairly evident from your rather histrionic criticism of the idea of solar energy from space that you don’t actually know anything about the topic. If you have some interest in information, please let me know. If not, then please go ahead as you might wish…

Best regards,

– John

Maury Markowitz - September 2, 2009

II’ll happily take you up on your offer – but you didn’t mention how I should contact you.


Maury Markowitz - September 2, 2009

Opps, ignore than, still learning about this software.

2. Joel Gilmore - September 2, 2009

Thanks for all your posts on this topic, it’s been useful.

I certainly agree that, at the moment, it’s not at all competitive with conventional power.

I can’t help, though, but think you’re being a little too gleeful about your results! Obviously these guys are not morons and think that, in time, this could be a viable option. A factor of 10 decrease in launch costs in the medium to long term seems very believable to me with competition from Virgin Galactic, SpaceX, etc, etc. A factor of 2-3 on solar cells also seems very believable. Another factor of 2-3 somewhere (e.g. a doubling of power prices, certainly conceivable under an emissions trading scheme over the next 20-30 years, and you’re home and hosed.

I’m not saying it’s possible right now, but I would *never* state that a technology is impossible or uneconomical – particularly when technology is advancing at an exponential rate and the space market is really taking off. Bill Gates might not have actually said “640kb should be enough for anyone”, but plenty of people were thinking it!

Discussion is great, but your posts would be even more useful/believable if you gave a more balanced view – even if you disgaree with them, they clearly have arguments of some merit.

Thanks for posting, though!

Maury Markowitz - September 2, 2009

> think you’re being a little too gleeful about your results!

Well I was trying to be sarcastic, but that apparently failed. From now on I promise to leave that to the professionals at Seanbaby. But to the meat of it:

>A factor of 10 decrease in launch costs in the medium to
> long term seems very believable to me

I’ve been watching the space launch business for a long time now. Between Saturn V and Shuttle heavy launch for loads over 25,000 lbs went up in price. Medium launch, 10k to 25k, has gone down about 15 to 20%. Light launch has gone up and down, all over the place, and ended up pretty much where it started.

During that period there were an endless number of proposals for new low-cost launchers. The recent com-sat inspired boom-and-bust is simply the latest in a chain of these going back to the 1960s, and not even the outline has changed – it’s still Big Dumb Booster vs. flyback vs. SSTO. None of these demonstrated any ability to lower costs in the real world.

We’ve had whole new countries come in too. Even with the dollar factors being artificially skewed, the Russians and Chinese can’t offer rates that are more than 1/2 better than what we have now.

So that’s the last 40 years in a nutshell. What part of that history suggests that GEO access is going to fall by a factor of 10 in the foreseeable future? None of it. You can dream, but in the meantime we have real problems to solve, and dreaming isn’t going to solve them.

> factor of 2-3 on solar cells also seems very believable.

More than believable, I fully expect this to occur within the next two years (and I know much more about this side of the equation that you might imagine).

But then you run into the other side of the problem. I posted this equation before, but I’ll do so again here:

1) calculate the payback from the amount of energy a solar panel will produce over its lifetime if you place it in the Nevada desert.
2) calculate the figure that the same panel will produce in space
3) calculate the cost of launching that panel into space.

Unless the difference between (1) and (2) is greater than three, you are better of building in the desert, and using the money you would have spent on (3) on buying more panels.

Right now, (2) is about four to five times (1). However, (3) is about 10 times (2). That places the overall economics of by way more than one order of magnitude.

The problem here is that (1) and (2) are linear with the price of the panel. So no matter what happens to the price of panels, the only consideration in terms of space vs. ground is the launch costs. They need to go down about 100 times before the math works out in your favour (don’t take my word for it, take the SPS booster’s own comments in the article above).


Mark Wonnacott - November 9, 2009

Those calculations are all very well, but does Japan have anything equivalent in size or light intensity as the Nevada desert? Is it more feasible for them to transmit power from Nevada to Japan than from space to Japan?

3. Tom Andersen - September 19, 2009

The nice thing about putting solar in the Arizona desert (and other sunny hot spots) is that output peaks with the A/C demand peak every day.
One thing to watch out for is subsidies. Right now with huge subsidies, there is actually less incentive to bring the panel prices down, especially for companies with big investments and profits from current technology. See this article for example. As subsidies in Spain dropped, all of a sudden people started caring about how much the panels cost to make!

4. Peter Stuyvesant - November 9, 2009

Nice calculation to get to the 100 factor but do not forget that Japan has no Nevada. I think they want to be independent of a foreign country with regard to their energy supply.

Maury Markowitz - November 9, 2009

Too true. But by 2030 there will be a whole host of energy systems that will make this one superfluous anyway. If we’re going to assume a two-orders of magnitude reduction in infrastructure costs due to improved economies of scale, then one must consider the fact that with the same reductions, offshore wind power is about 0.5 cents a kWh, considerably less than this solar effort. And I would consider a 100x reduction in wind power costs to be much, much more likely than the same reduction in launch costs.

5. Fer - November 9, 2009

Dude, you’ve got an attitude…

6. YetAnotherBob - November 9, 2009

I see a few problems with your approach.

First, you are ignoring the problems with Wind Power. There are some serious ecological effects already apparent. There are also effects that are beginning to be seen that are not recognized yet. There is a reason why serious ecologists don’t like large wind power installations. Wind power will not be able to provide more than a fraction of our total energy needs for the long term.

Second, you assume that there will be no ‘breakthroughs’ in construction technology in the future. Serious advocates of Solar Power Satellites do not assume that the raw materials for construction of anything other than demonstrators will be lifted from earth. It’s just too expensive with chemical rockets. It will require advances in processing of materials, and remote control robotics. The major materials will probably come from the Moon. Long term, probably asteroids. It is still too soon to try to build one today. Except for the small prototype the Military wants to put up to provide power to remote locations. That would be in the neighborhood of a 250 KW unit. A good demonstrator, but not EVER economical. For the intended use, economics is not an issue.

Third, you vastly overstate the inefficiencies of the system. As currently envisioned, the efficiencies will be on the order of 16%. That sounds bad until you realize that your car has an efficiency of 35% (if you drive a Prius), or a power plant with an efficiency of about 45%, with a transmission efficiency of around 50%, giving an overall efficiency (for what we NOW use of around 20%. Suddenly, the space effeciency doesn’t sound so bad.

Fourth, you greatly exaggerate the danger from space debris. To be effective, an SPS will need to be in geosynchronous orbit. Most of the space trash is in low earth orbit, below 600 KM. The geosynchronous orbit is almost an order of magnitude higher. The greatest source of damage will be from micrometeorites. And yes, we will need to think about that in design. That involves a degree of redundancy, and system replacement every 20 to 40 years. Same as with all other power plants.

I will readily admit that we aren’t ready or able to build a realistic system today, but in about 10 years, a realistic research program could be ready to begin serious production. Realistic projects I have read about are all looking at 20 years out. (20 years out is federal speak for ‘It’s never going to happen”, but some of them do pan out, and this one is going forward. It’s going forward faster than the Space Elevator. (That AC Clark pegged at 50 years out!)

NASA has had demonstrators that have beamed power for over 20 miles to power a 100 watt light bulb for years now. The planned transmitter would not be a microwave radio antenna, as you seem to think. (Please tell me if you are thinking differently.) Alll the serious proposals involve masers (microwave lasers) and have a beam diameter on the ground of around 100 Meters. Total power density on the ground then is, in the array area, only enough to raise body temperature a degree or so. This has been found in laboratory experiments not to damage birds or insects. Total microwave exposure outside of the array, and underneath it are below FCC exposure safety levels.

A question I have for you is, Are you really a power engineer? I am. A lot of your objections seem to be based on non-factual assumptions.

7. Maury Markowitz - November 9, 2009

> First, you are ignoring the problems with Wind Power

I’m not suggesting that Japan build wind farms. I’m suggesting that any technological advances that lower the cost of one system are just as likely to lower the costs of other systems too. So if you pick any technological development that you believe will lower the cost of space power, it’s highly likely it will also lower the price of ground power too.

> Second, you assume that there will be no ‘breakthroughs’ in construction technology in the future.

I am assuming there WILL be breakthroughs. But any breakthrough is just as likely (or much more likely) to lower the cost of earth-based generation even more.

Lets say, for instance, someone perfects a way to make really cheap carbon fiber structures, 1/100th the price that they cost to make now. That will significantly reduce the cost of building an SPS. But it will also reduce the cost of building a wind turbine just as much – they’re dominated by shipping costs.

With a single exception, the cost of launching rockets, any advance you might suggest is likely to lift all boats equally. So SPS’s become an argument about launch costs – as it is now.

> Third, you vastly overstate the inefficiencies of the system

One of us has missed the point here. I don’t care about absolute efficiencies of these systems. The only effect that has is the launch mass. if you double the cell efficiency you half the launch cost – but you also half the amount of land you need to cover if you leave it on Earth.

> Fourth, you greatly exaggerate the danger from space debris.

No, sorry, I’m not. As you’ll see…

> Most of the space trash is in low earth orbit, below 600 KM

No, actually it’s between 800 km and 1500 km. Yes, that is also below GEO, but any satellite that gets to GEO has to go through this band to get there. And that’s what I’m worried about.

If my calculations are correct, every 100 MW baseline NASA design will have a 4% chance of being hit by debris on the way up to GEO. That doesn’t sound too bad, and if one of them does get hit, it’s only one launch out of dozens, so you simply send up another.

Buuuut, if one of them does get hit, it will cause thousands of new debris to be released in this band of space. Those fragments will collide with the other satellites already there, spalling off more debris, and so one.

So the way to look at this isn’t to worry about the SPS, but everything else: every 100 MW of SPS has a 4% chance of wiping out everything in this band of space. And this is the band of space that contains all the GPS satellites, a considerable amount of our communications gear, many of our ground mapping satellites, and basically the vast majority of everything else you can think of (that’s why the debris is there, that’s where we put everything).

So can you imagine NASA or the Air Force allowing anyone to try this? I can’t.

> Realistic projects I have read about are all looking at 20 years out.

Which ones are those, exactly? I’ve read all of the published reports, and they’re all the same pie-in-the-sky garbage about 100 times cheaper rockets and all sorts of other non-existent technologies. One of them actually proposed putting a 100 by 300 meter sat right at 1100 km! Har!

> lot of your objections seem to be based on non-factual assumptions.

Unnamed ones, I see. Really, you should be very careful before accusing me of not knowing what I’m talking about from a single short blog post.


Mark Wonnacott - November 9, 2009

There are many conceivable technological advances that wouldn’t lift both boats equally: advances in rocket propellant for one.

Maury Markowitz - November 10, 2009

> advances in rocket propellant for one.

For sure! And that’s what this is really about.

SPS’s are not about generating power. If they were, you might expect to see representatives of the power generation firms and groups at the meetings. SPS’s are really an excuse to develop a new rocketry infrastructure. That’s why the meetings consist almost entirely of ex-NASA engineers and hangers-on.

It’s a sci-fi “con” in disguise. Instead of Klingons and phasors, these guys are into “real science”, huge space structures and moon bases.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s