[thisisbrians]

Some ideas about what makes it challenging:

- A satellite will have to transmit power through the entire atmosphere from space, which will have substantial losses even in perfect conditions

- A satellite constellation in orbit would need many ground stations to transmit power to from space

- Weather will get in the way of transmission to the ground, just like it does with ordinary solar

- The cost of getting a PV array into orbit is very expensive relative to the amount of power it can generate and transmit

[JumpCrisscross]

> satellite will have to transmit power through the entire atmosphere from space, which will have substantial losses even in perfect conditions

Balancing this is at least a half an order of magnitude difference in collection efficiency.

> Weather will get in the way of transmission to the ground, just like it does with ordinary solar

True. But there are extraterrestrial atmospheres where water vapor and ozone aren’t a problem. (I’ve only seen this proposed with microwave.)

> cost of getting a PV array into orbit is very expensive relative to the amount of power it can generate and transmit

This is the killer. That said, this is a long-term research endeavour. If we contemplate such an array around the Moon or on Mars, or in a world with in-space resource extraction and manufacturing, the economics shift.

[feoren]

> half an order of magnitude

Completely off-topic, but would "half an order of magnitude" be sqrt(10)? I've just never heard someone refer to a factor of ~3 as "half an order of magnitude".

[JumpCrisscross]

> just never heard someone refer to a factor of ~3 as "half an order of magnitude"

Last time I looked, the estimates ranged from 270% to 50x. Seeing the latter, my brain went into astronomer mode and then I guess just ran with it.

[roelschroeven]

Sqrt(10) is approximately 3.162278, that's close enough to 3.

[LarryMullins]

> If we contemplate such an array around the Moon or on Mars, or in a world with in-space resource extraction and manufacturing, the economics shift.

The economics of anything on Mars or the Moon amounts to 'try to convince a government to throw tons of money at you.' A technology which only makes sense in that economic context is very limited.

[JumpCrisscross]

> technology which only makes sense in that economic context is very limited

If this were a start-up I’d be roundly criticising it. It’s not. It’s a research project. And each of its sub-projects—testing new PVs in space, a novel deployment mechanism, power transmission—has clear value outside a space-based solar context.

As a focussing mechanism, SBSP is neat because each problem needing to be solved to make it economically viable is immediately valuable on the ground. (Save for power transmission. That’s still niche.)

[LarryMullins]

> Save for power transmission

Isn't that the entire point? Putting solar panels in space has been done regularly the 1950s.

[JumpCrisscross]

I meant that I don't see a near-term use terrestrial use case for better microwave power transmission.

[ben_w]

Drones. Electric aircraft in general, recharging continuously in-flight.

Might be ill-advised for a whole mass of reasons that I, as a software engineer, know naught of; but it would be a use case.

[LarryMullins]

Maybe you could use microwave power transmission (from the ground) to power atmospheric pseudo-satellites, but practical applications seem dubious.

[wrycoder]

> cost of getting a PV array into orbit is very expensive

Musk will find a way. Another stepping stone to Mars.

[ben_w]

It would be great on Mars. Mars has planet-wide dust storms.

But for Earth, even if it was free to put the PV in Earth orbit, the ground stations need to have an incredibly low total cost to make sense.

I can't remember how low exactly.

[DennisP]

The ground stations are basically just antenna wire. For a large plant they would contribute just 0.7 cents/kWh to the total cost, according to the book The Case for Space Solar Power.

[ben_w]

Also needs a microwave frequency rectifier, plus installation costs are non-trivial as evidenced by the difference between the cost/W of a PV cell and a PV farm.

Also, cheapest ground-based PV in the world is 1.04¢/kWh (Saudi Arabia), and the cheapest in the USA is 1.50¢/kWh (New Mexico), so 0.7¢/kWh is already a large percentage.

https://commercialsolarguy.com/lowest-solar-power-prices-in-...

[DennisP]

Just to clarify, the 0.7 cents is total ground station cost, according to my source.

At ideal locations, ground solar is super cheap, but the microwave receiver works anywhere. Not needing storage is a big difference, too.

[SyzygistSix]

He's talked about the idea, how much the two aspects - solar power and space - are things that excite him, and how it just is not a viable idea.

[DennisP]

He made one comment, a decade ago, well before he started talking about $35/kg to orbit with Starship.

In response to his famous question "what's the conversion rate," the answer is about 50%, according to the book The Case for Space Solar Power. That's not bad considering a panel in geostationary collects five times as much energy in 24 hours as a panel on Earth.

[p1esk]

None of these seem like obvious show stoppers.