> If you can build giant solar arrays in GEO, you can build small ones and attach ion thrusters to them.

Well, that's just putting the cart before the horse isn't it? We have no idea how to do A, so I think you're a little premature claiming "nope"!

Maybe you should tell Don Kessler that we have it all solved. You know, the guy that they named the "Kessler syndrome" after. He's pretty adamant against SPSs: "Some of the most environmentally dangerous activities in space include [...] large structures such as those considered in the late-1970s for building solar power stations in Earth orbit" I've emailed with him recently, he hasn't changed his mind since 2009 when he wrote that. Ask him yourself.

I did the calcs on this with Don's help a few years back. I recall there being something like a 10% chance that every SPS would cause a Kessler syndrome in MEO. I'll dig up the numbers for you.

> The first one is no technological progress

So think about the *entire universe* of possible technological advances over time. Draw that as a rectangle.

Now think about all of the technical advances within that rectangle that improve solar power collection at any level. Draw a circle to cover that area. Did you do 10% of the rectangle? 20%? Excellent.

Now draw a circle representing all of the advances in the rectangle that improve solar power collection *only in space*. Draw a circle. That circle will be much smaller, and intersect the first one you drew almost entirely. In the sections that don't overlap you basically have three items, more efficient visible light lasers (more than 50% socket to light, to be exact), lightweight space structures, orders of magnitude lower launch costs.

Now throw darts at the wall. See how often you hit that section.

See the problem?

Now for fun, multiply the areas by the amount of money being spent on each of those.

> The second assumption is that any solar power station has to be photovoltaic

Read my last sentence above a second time.

> Wouldn't that make its size not grow with the size of the array of solar panels in space

Ahhh, well that depends on how high you're allowed to push the energy density of the beam.

Current international limits are 50 W/m^2. The sun at noon is 1000 W/m^2, so by that standard the rectenna is going to be very large indeed.

I seem to recall projections on beam energies around 1250 W/m^2, but now I can't seem to find anything over 40 W/m^2:

http://www.nss.org/settlement/ssp/spacepower/spacepower01.html
http://arxiv.org/abs/1401.1779

I have to be missing something here though, because I recall rectenna size projections being about 60% the size of a PV field of the same peak. Let me dig into this a bit more...

In the meantime, Do The Math digs a little deeper into that: http://physics.ucsd.edu/do-the-math/2012/03/space-based-solar-power/

> Spectrolab rates their space solar panels for 20 years at GEO

This is not the 20% degradation point I use in my article, this is the "totally dead" point. The "totally dead" point for the average solar panel on earth is unknown, because they haven't been in service long enough to know. All we know for sure is that the vast majority of panels installed in 1982 are still working just fine today (and I know examples from the 1970s). Some have suggested the totally dead point for panels on the ground is at about 100 years, although I suspect silver migration and back-rot will reduce that to the 50 year mark.

Interesting notes though: one is that the lifetime in GEO is higher than LEO which is higher than MEO. Given radiation flux is higher as you go up, one might conclude that the main damage mode is debris.

> Your comparison of operating hours neglects that in space you have 36% higher insolation

That is included in the "I" number. If I was simply day/night, then it would be 8765/2 = 4382. But the I is 2300 for the ground based panels, which is taking all of those other effects into consideration.

> cherry-picking a good location is unfair

What, like GEO? :-) Actually it's not that much of a joke, you'll have to move all the comsats out of the way, which would be fun...

But fair enough, point taken. You can use PVWatts to pick another location. Or pick from this list:

Toronto: 1,840
Calgary: 2,400 (wow!)
London: 1,302 (yikes, rainy london indeed!)
Moscow: 1,259

Looks like the break-even is around Moscow. Remember though, that's just in energy terms... in economic terms, enjoy!

> That is true if the efficiency of the panels improve

No, it's true in almost all cases.

Take a cell and put it in space and it will deliver less power. Period. It doesn't make a difference if your launchers get cheaper, or your cells get lighter, delivering less power is delivering less power. Unless your invention makes space launches cost negative dollars, you lose.

1. The panels may get much lighter, till they weigh as much as a sheet of mylar.

Does not help, less power is less power.

But that makes them lighter on the ground too, which lowers their install cost there. And thin cells like this would be perfect for putting in shingles, which makes every residential house and garage a collector.

2. Launch tech improves, and brings the cost of cargo to orbit way down.

Does not help, less power is less power.

3. The tech for transmitting and receiving power through space improves

Now this DOES help. However, this is simply a function of basic radio physics. Google up antenna factor some time. The electronics side of thing is pretty much at the efficiency limit already - I've seen solar inverters with 98% DC to AC conversion efficiency, which is pretty astounding if you think about it.

> ofc it had no budget.

Then they couldn't be intending "to test the technology in 2018", could they? You need *actual money* to build *actual hardware*.

It was a trial balloon, precisely like this one. Free press for a slow news day.

> That does not change the fact that the math is pretty solid and it would work.

Go right ahead and demonstrate the math in question. Develop from the CAPEX side through to the LCoE. Include OPEX and regulatory loads, if you care to.

Or you could save yourself the trouble and use the spreadsheet I developed. Where should I send it?

> Off the top of my head, I can't see any particularly good reason why a space-based
> system should be shorter-lived than a ground-based system.

The reasons are very clearly explained right there and I even linked to the real-world articles I took the numbers from.

>he 'd assumed a similar lifetime for the space-based system

Look at the image at the top of the page. Do you see it? That's Mir's solar panels after about *10 years*. Hubble replaced its panels twice over a period of 13 years. Space absolutely sucks for solar panels.

> Arguably, a space-based system will last less time than a ground-based system.

There's no "arguing" involved, we've had panels in space and on the ground for decades and we know very well how long these things last.

Click the links, it's not like they're going to bite you.

> You've ignored the atmospheric losses suffered by ground-based systems -- clouds, dust, the opacity of air

No, that's what the insolation number takes care of, I. There's a link right in the article to where this number came from, you can click it, type in your location, and find the number yourself. As I mentioned earlier, it definitely includes "clouds, dust, the opacity of air", as well as geometric pointing errors, day/night cycle, and even reflection off snow and dirt on the panels.

> you're also being much more generous in estimating the potential lifetime of ground-based systems

As the links at the bottom of the article note, these are real-world numbers as measured on real systems that have been in the field for decades. If you have better numbers, fine...

PROVIDE YOUR REFERENCES AND DO THE MATH YOURSELF!

> though that depends on how much we value land taken up by solar arrays

Or rectennas. You recall that SPSS's have a downlink portion, right?

> JAXA intends to test the technology in 2018

No, they don't. The project died, if it ever existed in any meaningful form, because it never had a budget.

It was a trial balloon sent up by the space industry to create demand for new rockets. That's the only reason this idea keeps getting floated, as an excuse to make more rockets or heavy launchers.

> For example I do not see why Tg is different for ground based versus space based systems

The reason should have been explained, which I now realize has not been. Basically it *should* be easier to convert constant "insolation" from the rectenna to AC power than doing the same for variable inputs from the PV panels.

But you know what, you're absolutely right. It has basically no effect on the outcome, and simply confuses matters. I'll update the article and leave a note at the bottom.

Actually I should do that anyway, because NREL *finally* updated the derate in PVWatts from 0.77, which was hopelessly outdated, to 0.86(4?) which is more in line with modern inverters.

> and why it so not eliminated as per the E term

Yes, an update is in order, it would definitely improve it in this fashion. That should have been clear to me when I was writing it. Thanks!

> he leaves out the fact that a space based PV system operates 24/7 with continuous output compared to an earth
> based system that has to deal with the vagaries of weather and that pesky thing called "night".

No, actually, I didn't. That is encoded in the insolation number. PVWatts considers day/night, clouds, reflections off snow, dirt on the panels, and all sorts of other factors in its calculations. It's totally crescent fresh, check it out for your own location:

http://pvwatts.nrel.gov

> It certainly won't happen until we get better tech, but never say "never".

You may have missed the point of the linked article. If you improve the tech of the panels, then the relative advantage of mounting them on the ground *improves*.

> But TFA is about some 93 year old retired Chinese geezer "mulling" the idea

Geez, I totally missed that.

It always is, BTW. The entire space power group is made up almost entirely of retired astronauts and rocket engineers. That and the hangers-on like the National Space Society and such. I have yet to meet a single person from the power industry that is even marginally involved.

Indeed. And invariably by someone who is a rocket guy, like this one. Wagging the dog, every time.

This stupid idea gets floated every few years. It doesn't work, even in theory. Do the math yourself.

https://matter2energy.wordpress.com/2012/03/17/the-maury-equation-redux/

> There's a reason why China has 30 nuclear plants under construction

They don't. They have 22 under construction, where "under construction" is something from "we have the signed paperwork" to "we're putting in the switchyard".

And the reason is widely recorded - they wanted to put their coal plants out of business because they're poisoning everyone. Of course a nuclear plant doesn't really compete with coal economically (few things do) so to do this the plan was giving the plants free money and cheap fuel. If this were true here, the same would be happening.

However, as the cost of wind and solar plummeted, these plans are rapidly changing. The plans used to be based on a 400+GWe nuclear buildout by 2050, but these have been scaled back to 60GW with another 30 at the outside. Meanwhile, wind power has already reached 115GW at the end of 2014, more than the nuclear plants. Current install rates for wind are far greater than the peak installation rate for nuclear would have been even at the highest end of the original projections. Since 2012, much of the planned nuclear capacity of the earlier plans has been moved to wind. Gansu alone is expected to grow to a staggering 20 GW.

Read all about it:

http://www.world-nuclear.org/info/Country-Profiles/Countries-A-F/China--Nuclear-Power/
https://en.wikipedia.org/wiki/Wind_power_in_China
http://www.bbc.com/news/science-environment-25623400

> But don't let the scientific facts get in the way of your new religion

Take your own advice.

You're quoting someone who is a professional writer and has no experience in the sciences.

His book on the topic was widely panned for taking comments out of context. It took a good 300 years for that to happen to Jesus.

So spare us your chosen savior and the holier-than-thou BS.