Author Topic: Solar Power Satellites  (Read 32517 times)

Offline SICA Design

  • Full Member
  • *
  • Posts: 179
  • UK
  • Liked: 42
  • Likes Given: 51
Re: Solar Power Satellites
« Reply #20 on: 06/30/2014 07:36 PM »
Peak beam intensity for most proposals is kept below 300W/m2 (about 1/4 to 1/3 mid-day sun)

This may be a problem, think about it, if the power density is less than that of the sun, why not just use the sun light directly? Yes I know there're nights, weather issues with terrestrial solar power, but they seem to be much easier to solve than the problems facing SPS.

If you have access to a desert wasteland then why not use it? India announced "The World's Largest Solar Power Plant" at 4GW: http://www.scientificamerican.com/article/india-to-build-worlds-largest-solar-power-plant/

When you look deeper, that 4GW is actually "6.4 billion kilowatt-hours per year", in other words an average of 730MW. At a construction cost of $4.4 billion, that's $6B/GW (the build cost for a typical nuclear power station is $8B/GW) - and that doesn't include any provision for energy storage; a 17.5 million kilowatt-hour battery would probably cost many times the plant itself - and would supply an average of 730MW for...     one day.

I'm sure all these cost and storage issues can be resolved (and will be) over time.

The solar plant covers an area of 77 square kilometres - which arranged as a disc would be nearly 10km across, or nearly 5 times greater land area than a 4.5km diameter rectenna. The land under solar panels is shaded and unsuitable for farming. The land under the rectenna receives just 5% of the stray microwave beam (max 15W/m2 for 300W/m2 peak beam intensity), but more than 90% of the sunlight.

The mature launch price for a 1GW HESPeruS is $3.2B (for all costs recovered, according to REL data).

[EDIT: to be fair, you need 3 HESPeruS to supply 2GW total for 24/365 shared between UK & Japan, so that's $4.8B/GW]

« Last Edit: 06/30/2014 07:56 PM by SICA Design »

Offline gospacex

  • Senior Member
  • *****
  • Posts: 3028
  • Liked: 537
  • Likes Given: 604
Re: Solar Power Satellites
« Reply #21 on: 06/30/2014 09:02 PM »
The land under solar panels is shaded and unsuitable for farming.

There are forest plants which *require* shade to grow. Some of them have edible berries.

Offline SICA Design

  • Full Member
  • *
  • Posts: 179
  • UK
  • Liked: 42
  • Likes Given: 51
Re: Solar Power Satellites
« Reply #22 on: 06/30/2014 09:50 PM »
The land under solar panels is shaded and unsuitable for farming.

There are forest plants which *require* shade to grow. Some of them have edible berries.

Thanks for the info

Offline bad_astra

  • Full Member
  • ****
  • Posts: 1862
  • Liked: 252
  • Likes Given: 306
Re: Solar Power Satellites
« Reply #23 on: 06/30/2014 10:14 PM »
Why not just pump a fraction of the vast amount of money this project would cost into a commercial fusion reactor?

It was a nice idea in the 70's when shuttles were going to fly once a week, fusion was 50 years away, and O'neal needed  a reason to design his wondrous colonies. This idea makes little sense now.
"Contact Light" -Buzz Aldrin

Offline dror

  • Full Member
  • ****
  • Posts: 584
  • Israel
  • Liked: 138
  • Likes Given: 373
Re: Solar Power Satellites
« Reply #24 on: 06/30/2014 10:26 PM »
85% efficient over what distance?
Good question!

That is the efficiency for converting the incident beam power falling within the confines of the rectenna.
...
84% of the beam power is confined within a circular region called the "Airy disk" (George Biddell Airy), which dictates the minimum size of the rectenna and transmit antenna for a given frequency and beam path length.
...
Other losses are at the satellite-end of the beam and include:

PV efficiency - 40% expected from multi-junction concentrated cells (44.7% demonstrated)
                    -  20% --> >50% expected from future thin-film (e.g. perovskite research, 19% demonstrated)
                    -  bear in-mind that earth-orbit solar flux is 57% greater ...
...
                     - overall conversion efficiency (electrical --> microwave) could achieve >70%

So we add to that 2.5 longer time (24 hours instead of ~10) and we get:

Solar satelite energy = solar power plant energy × 0.85 × 0.84 × 0.7 × 1.57 × 2.5 = ~1.96×solar_power_plant_energy

So the satelite produces about twice as much energy per unit area, but requiers launch, structure and transmiter/antena/rectena infrastructure and 4.5 km radius on the ground.
The land solar plant uses about 5 times more erea of some desert..
I seriously dought the economics for that to succeed.

On top of that, any beamed energy from GEO is aditional heat for earth and will increase global warming as a fact.
 
"If we crave some cosmic purpose, then let us find ourselves a worthy goal. "
Carl Sagan, Pale Blue Dot

Offline SICA Design

  • Full Member
  • *
  • Posts: 179
  • UK
  • Liked: 42
  • Likes Given: 51
Re: Solar Power Satellites
« Reply #25 on: 06/30/2014 10:33 PM »
Why not just pump a fraction of the vast amount of money this project would cost into a commercial fusion reactor?

It was a nice idea in the 70's when shuttles were going to fly once a week, fusion was 50 years away, and O'neal needed  a reason to design his wondrous colonies. This idea makes little sense now.

I'm all for fusion, but the closest thing to a commercial reactor is ITER, predicted to produce 500MW of HEAT (zero electricity) at a development cost "expected to be in excess of €15 billion" (Wikipedia).

That's $20B, or $40B/GW, or $67B/GW if you assume 60% thermal-->electrical efficiency.

Meanwhile, we have access to a fully functioning fusion reactor just 150 million km away.

We KNOW how to beam microwaves from space; we do it all the time from comms sats and GPS, it's a matter of scaling up and lowering launch costs.

Offline QuantumG

  • Senior Member
  • *****
  • Posts: 8653
  • Australia
  • Liked: 3586
  • Likes Given: 843
Re: Solar Power Satellites
« Reply #26 on: 06/30/2014 10:34 PM »
It was a nice idea in the 70's when shuttles were going to fly once a week, fusion was 50 years away, and O'neal needed  a reason to design his wondrous colonies. This idea makes little sense now.

Fusion is still 50 years away, reusable rockets are almost here (well, first stage reuse anyway, if SpaceX can actually launch again), the good professor's name was O'Neill, his colonies are still a good idea, and space solar power does indeed make more sense if you're looking at it from the perspective of people in space trying to produce something to trade with people on Earth.
I hear those things are awfully loud. It glides as softly as a cloud. What's it called? Monowhale!

Offline SICA Design

  • Full Member
  • *
  • Posts: 179
  • UK
  • Liked: 42
  • Likes Given: 51
Re: Solar Power Satellites
« Reply #27 on: 06/30/2014 10:55 PM »
So we add to that 2.5 longer time (24 hours instead of ~10) and we get:

Solar satelite energy = solar power plant energy × 0.85 × 0.84 × 0.7 × 1.57 × 2.5 = ~1.96×solar_power_plant_energy

So the satelite produces about twice as much energy per unit area, but requiers launch, structure and transmiter/antena/rectena infrastructure and 4.5 km radius on the ground.
The land solar plant uses about 5 times more erea of some desert..
I seriously dought the economics for that to succeed.

As I said, if you have access to a handy desert, use it - just use it more effectively than India or Ivanpah (Ivanpah... ask me, please  :D).

http://www.contemporaryenergy.co.uk/picts/solarm1.gif

The UK averages about 2.5kWh/m2 insolation per day, or 104W/m2. In-space average is 1367W

I make that 13 times better, not "twice as much"

Quote
On top of that, any beamed energy from GEO is aditional heat for earth and will increase global warming as a fact.

An 85% efficient rectenna reflects about 5% beam power back into space, so 10% is direct waste heat.
For 1GW to the grid, that's <120MW into the environment.

The best gas power stations have about 60% thermal efficiency, i.e. 40% direct waste heat into the environment.

For 1GW to the grid, that's 667MW into the environment.

Replacing 1GW of gas-fired power with equivalent SBSP reduces Earth's heat load by 547MW (fact).

In practice, the waste heat is negligible compared with the Earth's total insolation and the effect of C02.

Offline Asteroza

  • Full Member
  • ****
  • Posts: 644
  • Liked: 83
  • Likes Given: 2
Re: Solar Power Satellites
« Reply #28 on: 06/30/2014 11:52 PM »
All the more reason why laser SPS has it's ups and downs. Small sizes are easier to demo overall. Tight spot size is great for reducing receiver area, but then it functionally approaches weapons grade beam densities rather quickly. Though I like to roll out that most space things are weapons already (most if not all satellites potentially can become kinetic impactor weapons from orbit).

Very true!

Quote
The old SPS studies chose microwave because laser wasn't ready, and they chose the density to appeal to the average man who subjectively didn't relish the idea of being microwaved from orbit. As mentioned above, the physics says it needs to be 1Km or more for the transmitting antenna, at around 1GW based on those constraints. That fundamentally makes GEO demos hard. The japanese intend to do a LEO microwave demo as a proof-of-concept though, which is easier due to the distances involved at the expense of antenna tracking.

Microwaves between 1-10GHz were also chosen for their ability to pass through thick cloud and rain; the 5.8GHz ISM band is near optimal.

I have a concept for a 1 gigawatt SPS called HESPeruS - Highly Elliptical Solar Power Satellite - which traverses a Molniya orbit, a highly-elliptical orbit first used by the Russians. Like GEO, this is also a synchronous orbit, in-that the SPS loiters over the same two northern hemisphere locations alternately every 12 hours. There are several benefits:

* HESPeruS is of rigid construction, built from multiple hexagonal modules into an essentially flat structure capable of collecting sunlight and beaming to any latitude  above 45 degrees, without the complex rotating optics as needed for the 2nd Japanese option (and ALL GEO satellites).
* HESPeruS is complementary to GEO space solar power proposals, which suffer additional losses when beaming to high latitudes; 30% of Earth's ice-free land and a $7 trillion economy is located above 45 degrees latitude.
* It's 5.8GHz antenna is distributed over the full 4.4km diameter, with individual elements vacuum-spaced by <25mm, drawing power (10s of milliwatts) from local PV cells which point directly at the sun throughout the orbit - so negligible distribution losses.
* The large antenna to PV ratio means it can operate within the diffraction limit at one-tenth scale (100MW), meaning a pilot system can be launched at lower cost, and then expanded later (Molniya is a servicable orbit!).
* It costs less than half as much (principally propellant mass) to place a satellite on a Molniya orbit than it does to GEO.
* Just 3 HESPeruS satellites can supply 1GW continuous (24/365) power to two rectenna sites, such as the UK and Japan.

Quote
I have a personal belief (not necessarily backed with facts mind you) that laser (either laser thermal or hybrid laser PV/thermal) would be a better choice for the receiver architecture considering the process heat applications and not just electrical use. That, and it would enable a favorite pet idea, laser assisted Skylons.

That's Keith Henson's idea (he comments here) - hopefully his paper will be published around the same time as mine!

(Edited for typos)

How does HESPeruS handle local sunrise/sunset orientation with respect to downlink antenna tracking? You appear to describe a single flat panel, but at local sunrise it will be edge-on with respect to the ground so the phased array aspect of the distributed aperture won't do you any favors there.

Offline su27k

  • Full Member
  • ****
  • Posts: 1141
  • Liked: 845
  • Likes Given: 75
Re: Solar Power Satellites
« Reply #29 on: 07/01/2014 04:14 AM »
The mature launch price for a 1GW HESPeruS is $3.2B (for all costs recovered, according to REL data).

[EDIT: to be fair, you need 3 HESPeruS to supply 2GW total for 24/365 shared between UK & Japan, so that's $4.8B/GW]

Thanks for the numbers, they're interesting to look at. For the launch price, I assume this is using Skylon? Also what about the cost of the satellite themselves and the receivers? Japanese's plan seems to have an artificial island as receiver, that can't be cheap.

Offline dror

  • Full Member
  • ****
  • Posts: 584
  • Israel
  • Liked: 138
  • Likes Given: 373
Re: Solar Power Satellites
« Reply #30 on: 07/01/2014 07:10 AM »
So we add to that 2.5 longer time (24 hours instead of ~10) and we get:

Solar satelite energy = solar power plant energy × 0.85 × 0.84 × 0.7 × 1.57 × 2.5 = ~1.96×solar_power_plant_energy

So the satelite produces about twice as much energy per unit area, but requiers launch, structure and transmiter/antena/rectena infrastructure and 4.5 km radius on the ground.
The land solar plant uses about 5 times more erea of some desert..
I seriously dought the economics for that to succeed.

As I said, if you have access to a handy desert, use it - just use it more effectively than India or Ivanpah (Ivanpah... ask me, please  :D).

http://www.contemporaryenergy.co.uk/picts/solarm1.gif

The UK averages about 2.5kWh/m2 insolation per day, or 104W/m2. In-space average is 1367W

I make that 13 times better, not "twice as much"
But satelites has about 50% convertion panelty (0.84×0.85×0.7) that ground based facility dont have so it is about 7.5 times better for England and "about twice as much" for low altitude ereas.
Quote
Quote
On top of that, any beamed energy from GEO is aditional heat for earth and will increase global warming as a fact.

An 85% efficient rectenna reflects about 5% beam power back into space, so 10% is direct waste heat.
For 1GW to the grid, that's <120MW into the environment.

The best gas power stations have about 60% thermal efficiency, i.e. 40% direct waste heat into the environment.

For 1GW to the grid, that's 667MW into the environment.

Replacing 1GW of gas-fired power with equivalent SBSP reduces Earth's heat load by 547MW (fact).

In practice, the waste heat is negligible compared with the Earth's total insolation and the effect of C02.
All of the energy ends up as heat either way. 1 GW heat of microwave beam equals 1 GW heat of burnt coal or degraded plutonium.
When comparing to another non sustainable alternative, of course both are as bad.
I compared it to solar energy which uses the heat flux that is already coming to earth so it does not add to global warming (disregarding albedo difrences caused by the ground based panels) to show that space based solar energy cant be counted as sustainable, in opose to the general opinion.
"If we crave some cosmic purpose, then let us find ourselves a worthy goal. "
Carl Sagan, Pale Blue Dot

Offline SICA Design

  • Full Member
  • *
  • Posts: 179
  • UK
  • Liked: 42
  • Likes Given: 51
Re: Solar Power Satellites
« Reply #31 on: 07/01/2014 07:58 AM »
How does HESPeruS handle local sunrise/sunset orientation with respect to downlink antenna tracking? You appear to describe a single flat panel, but at local sunrise it will be edge-on with respect to the ground so the phased array aspect of the distributed aperture won't do you any favors there.

HESPeruS modules comprise a 3:1 stepped planar array, with fully integrated elements (PV + PMAD + Tx) at the half-wavelength scale (~25mm). The PV is at 90 degrees to the nominal boresight ( = rotation axis, normal to ecliptic), with the whole SPS rotating once-per-year to maintain PV directly sun-facing. The diagram shows the polar plot and local radiation pattern modelled for 22 elements.

HESPeruS travels along a highly elliptical Molniya orbit, loitering at northern apogee (40,000km) with a high-speed pass of the southern hemisphere at 450km perigee ("Molniya" == "Lightning", IIRC for its speed). I've defined "sunrise" as an SPS elevation of 20 degrees seen at the rectenna (northern latitudes above 45 degrees), which is sized accordingly (12km diameter).

Typical beam times are 9 hours 29 minutes for each 11 hour 58 minute orbit, satisfying elevation (>20 deg), beam sweep (+/- 50 deg) and diffraction limits (using effective apertures).

Offline Alf Fass

  • Full Member
  • ****
  • Posts: 451
  • The Abyss
  • Liked: 89
  • Likes Given: 83
Re: Solar Power Satellites
« Reply #32 on: 07/01/2014 08:07 AM »


As I said, if you have access to a handy desert, use it - just use it more effectively than India or Ivanpah (Ivanpah... ask me, please  :D).

http://www.contemporaryenergy.co.uk/picts/solarm1.gif

The UK averages about 2.5kWh/m2 insolation per day, or 104W/m2. In-space average is 1367W

I make that 13 times better, not "twice as much"


that 104W/m2 is for an unsteered surface flat to the ground.
When my information changes, I alter my conclusions. What do you do, sir?
John Maynard Keynes

Offline SICA Design

  • Full Member
  • *
  • Posts: 179
  • UK
  • Liked: 42
  • Likes Given: 51
Re: Solar Power Satellites
« Reply #33 on: 07/01/2014 08:13 AM »
The UK averages about 2.5kWh/m2 insolation per day, or 104W/m2. In-space average is 1367W

I make that 13 times better, not "twice as much"
But satelites has about 50% convertion panelty (0.84×0.85×0.7) that ground based facility dont have so it is about 7.5 times better for England and "about twice as much" for low altitude ereas.
Quote
Quote
On top of that, any beamed energy from GEO is aditional heat for earth and will increase global warming as a fact.

An 85% efficient rectenna reflects about 5% beam power back into space, so 10% is direct waste heat.
For 1GW to the grid, that's <120MW into the environment.

The best gas power stations have about 60% thermal efficiency, i.e. 40% direct waste heat into the environment.

For 1GW to the grid, that's 667MW into the environment.

Replacing 1GW of gas-fired power with equivalent SBSP reduces Earth's heat load by 547MW (fact).

In practice, the waste heat is negligible compared with the Earth's total insolation and the effect of C02.
All of the energy ends up as heat either way. 1 GW heat of microwave beam equals 1 GW heat of burnt coal or degraded plutonium.
When comparing to another non sustainable alternative, of course both are as bad.
I compared it to solar energy which uses the heat flux that is already coming to earth so it does not add to global warming (disregarding albedo difrences caused by the ground based panels) to show that space based solar energy cant be counted as sustainable, in opose to the general opinion.
Yes, SBSP is typically 7.5 times better than non-tracking terrestrial PV (including all losses) for northern latitudes. How much benefit do you associate with predictable baseload power? How much cost do you associate with storage to overcome intermittency?

The best terrestrial PV (concentrated) is still about 40% efficient, with low albedo, so 40%-60% waste heat rejected to the environment. I'm happy to disregard albedo; all rejected and useful power ends up as heat - it's still negligible compared with solar influx and heat trapping due to C02.

Offline SICA Design

  • Full Member
  • *
  • Posts: 179
  • UK
  • Liked: 42
  • Likes Given: 51
Re: Solar Power Satellites
« Reply #34 on: 07/01/2014 08:27 AM »
The mature launch price for a 1GW HESPeruS is $3.2B (for all costs recovered, according to REL data).

[EDIT: to be fair, you need 3 HESPeruS to supply 2GW total for 24/365 shared between UK & Japan, so that's $4.8B/GW]

Thanks for the numbers, they're interesting to look at. For the launch price, I assume this is using Skylon? Also what about the cost of the satellite themselves and the receivers? Japanese's plan seems to have an artificial island as receiver, that can't be cheap.

Yes, launch price is based on mature Skylon operations, including servicing missions for Fluyt. Data was kindly supplied by REL specifically for launching 10 tonnes payload (HESPeruS + Fluyt propellant) to 300km orbit at 63.4deg inclination, with construction via Fluyt carrying 49 tonnes of HESPeruS to Molniya before returning to 300km.

From memory, JAXA was assuming much higher beam intensity - which would have to be away from population centres.

I've focussed on launch price, which has traditionally been the dominant factor. I don't yet have useful materials costs, which will depend on mass production.

Fortunately, wireless comms (incl 5.8GHz WiFi) and terrestrial solar are both driving research and mass production in the right direction.
« Last Edit: 07/01/2014 08:32 AM by SICA Design »

Offline Asteroza

  • Full Member
  • ****
  • Posts: 644
  • Liked: 83
  • Likes Given: 2
Re: Solar Power Satellites
« Reply #35 on: 07/01/2014 08:59 AM »
Japanese construction companies (the big general contractors) are addicted to large civil works projects like a crack fiend. That, and the general japanese government view that large civil works projects are the preferred form of economic stimulus (via construction worker salary trickle down economics effects).

That said, rather than a true artificial island, they may go for a megafloat modular concrete barge island. There's currently on the table a potential project for a large offshore megafloat based heliport structure to service offshore oil/gas rigs which are too far to service by conventional helicopter easily from japan. Though at the scales involved, a PSP type module for the megafloat rather than a conventional rectangular barge may be a better option.

Offline Asteroza

  • Full Member
  • ****
  • Posts: 644
  • Liked: 83
  • Likes Given: 2
Re: Solar Power Satellites
« Reply #36 on: 07/01/2014 09:16 AM »
How does HESPeruS handle local sunrise/sunset orientation with respect to downlink antenna tracking? You appear to describe a single flat panel, but at local sunrise it will be edge-on with respect to the ground so the phased array aspect of the distributed aperture won't do you any favors there.

HESPeruS modules comprise a 3:1 stepped planar array, with fully integrated elements (PV + PMAD + Tx) at the half-wavelength scale (~25mm). The PV is at 90 degrees to the nominal boresight ( = rotation axis, normal to ecliptic), with the whole SPS rotating once-per-year to maintain PV directly sun-facing. The diagram shows the polar plot and local radiation pattern modelled for 22 elements.

HESPeruS travels along a highly elliptical Molniya orbit, loitering at northern apogee (40,000km) with a high-speed pass of the southern hemisphere at 450km perigee ("Molniya" == "Lightning", IIRC for its speed). I've defined "sunrise" as an SPS elevation of 20 degrees seen at the rectenna (northern latitudes above 45 degrees), which is sized accordingly (12km diameter).

Typical beam times are 9 hours 29 minutes for each 11 hour 58 minute orbit, satisfying elevation (>20 deg), beam sweep (+/- 50 deg) and diffraction limits (using effective apertures).

Maybe I'm not understanding correctly, but what happens when the SPS lines up with the day/night terminator line while near the top of the Molniya loiter then, unless you phase orbits to avoid that scenario?

Offline gospacex

  • Senior Member
  • *****
  • Posts: 3028
  • Liked: 537
  • Likes Given: 604
Re: Solar Power Satellites
« Reply #37 on: 07/01/2014 10:16 AM »
The best terrestrial PV (concentrated) is still about 40% efficient, with low albedo, so 40%-60% waste heat rejected to the environment. I'm happy to disregard albedo; all rejected and useful power ends up as heat - it's still negligible compared with solar influx and heat trapping due to C02.

You didn't understand the argument. With terrestrial solar panels, you capture as useful energy some energy which *would otherwise end up absorbed by Earth anyway*. With SBSP, the energy you capture and send to Earth *would otherwise fly past the planet*.

Offline Alf Fass

  • Full Member
  • ****
  • Posts: 451
  • The Abyss
  • Liked: 89
  • Likes Given: 83
Re: Solar Power Satellites
« Reply #38 on: 07/01/2014 10:21 AM »
The best terrestrial PV (concentrated) is still about 40% efficient, with low albedo, so 40%-60% waste heat rejected to the environment. I'm happy to disregard albedo; all rejected and useful power ends up as heat - it's still negligible compared with solar influx and heat trapping due to C02.

You didn't understand the argument. With terrestrial solar panels, you capture as useful energy some energy which *would otherwise end up absorbed by Earth anyway*. With SBSP, the energy you capture and send to Earth *would otherwise fly past the planet*.

A pointless argument, whether space or surface based the difference it makes to the amount of energy the Earth receives is truly negligible.
When my information changes, I alter my conclusions. What do you do, sir?
John Maynard Keynes

Offline SICA Design

  • Full Member
  • *
  • Posts: 179
  • UK
  • Liked: 42
  • Likes Given: 51
Re: Solar Power Satellites
« Reply #39 on: 07/01/2014 10:33 AM »
Maybe I'm not understanding correctly, but what happens when the SPS lines up with the day/night terminator line while near the top of the Molniya loiter then, unless you phase orbits to avoid that scenario?

HESPeruS is never in Earth's shadow when other beam constraints are met. A Molniya orbit is inclined at 63.4deg taking it WAY above the northern hemisphere. The diagram is not to scale (obviously), but the solar angle (shown from Earth's tilted perspective) and orbit proportions are correct. The "SPS" represents a small section of one module - the tiered arrangement is at ~25mm scale.

Were you considering something similar to 70 minute GEO outtages during Spring/Vernal equinoxes?