Power satellite problems

[Keith Henson]

I am known for pointing out problems with power satellites such as the
physics that forces large size and the number of flights needed to
construct enough to make a difference. Here I present a possible
solution to some of the daunting problems.

To make competitive power, the cost to move a power satellite to GEO
can't be much more than about $200/kg. SpaceX has probably reduced
the cost to GEO to around $100/kg. But the rest of the lift to GEO
can't be done within the cost limits using chemical fuels. This is
just a feature of the rocket equation.

It can be done with self-powered electric propulsion, spiral, or a
number of Hohmann boosts. The problem then is that such transfer
orbits take a long time and expose the power satellite to space junk
hits, typically 40 for a spiral orbit transfer. Hits are a serious
problem for a significant scale of power satellite production.
Kesstler syndrome anyone?

A possible solution occurred to me in the last day or so, but it is
well beyond what I could work out.

The idea is to calculate the transfer orbit to miss all the dead
satellites and have the active ones maneuver to miss

This is a huge traffic control problem, but the current state of
computation might be up to it.

I guess that developing the software would cost a few million. An
alternative is beamed energy bootstrapping.

I asked copilot, but it looks like it has been shut off now. Was
working yesterday.

If power satellites are constructed in LEO, moving them to GEO exposes
them to being hit by space junk and active satellites. Does it look
feasible to compute a safe path?

Copilot, 11:42 AM
Hello, if you want to continue using Microsoft Copilot, please visit
copilot.microsoft.com or download the mobile app today for Android or
iOS devices.

Keith

[Keith Henson]

If power satellites are constructed in LEO, moving them to GEO exposes
them to being hit by space junk and active satellites. Does it look
feasible to compute a safe path?

[Can't access from Skype, but the web interface worked.]

Yes, it is feasible to compute a safe trajectory for moving satellites
from Low Earth Orbit (LEO) to Geostationary Orbit (GEO) Transiting
satellites face potential collisions with both space junk and other
active satellites, so this is quite a delicate process

Transiting satellites face potential collisions with both space junk
and other active satellites, so this is quite a delicate process

Techniques involved:

Trajectory Design: Carefully planning the route to avoid known
debris fields and active satellite orbits.

Collision Avoidance Systems: Many satellites are equipped with
systems to detect and avoid collisions, automatically adjusting their
paths if necessary.

Real-time Tracking: Space agencies and companies use real-time
tracking of space debris and active satellites to make necessary
adjustments.

🛠️ Tools & Software: Several powerful tools and algorithms help
compute safe paths, considering multiple variables like the current
position of space debris, future projections, and satellite speed.
Examples include software developed by NASA and the European Space
Agency (ESA).

🚀 Challenges:

The sheer amount of space debris in LEO.

Limited propulsion capabilities for precise adjustments mid-transit.

Constantly changing environments, needing up-to-date data and
real-time adjustments.

By employing effective planning, tracking, and autonomous systems,
it's definitely possible to construct safe paths for satellites
switching orbits.

Do you think we'll see a rise in such complex missions soon?

[Is it normal for AIs to ask such questions?]

[k.a.c...@sympatico.ca]

Don't take anything that comes out of a LLM "generative AI" program as being gospel. At a fundamental level, they are parroting back to you bits and pieces of text from the various documents on which they were trained (with some bells and whistles to male the results into well-formed sentences). There is no "intelligence" involved in that parroting-back process (the AI is, AFAIK, all in the training software, using neural nets to allow much larger sets of training text to be used, and statistics for much-longer text-strings to be compiled, than you could do otherwise --- someone correct me if I'm wrong about that!), nothing remotely like "thinking." Hence the well-known characteristic of these programs, that they routinely come back with sheer nonsense --- albeit in nice, well-formed sentences, which can trick the unwary into thinking that anything that well-written must be correct.

I've played with one of these a bit (ChatGPT), and found that it tends to answer questions like "is XXX possible?" positively, even if XXX is absolutely not possible.

[Keith Henson]

On Wed, Jan 29, 2025 at 12:35 PM Doug Space <dougsp...@gmail.com> wrote:
>
> Hi Keith,
>
> In your opinion:
> - Can SPSs be launched as free-standing satellites that can beam only the amount of power that that one payload produces? In other words, can we reduce the amount of assembly done to just assembling one payload at a time or is there something in logic or physics that requires integrating between payloads?

It's not my opinion, it's diffraction-limited microwave optics that
force power satellites to be so big. This was understood from the
start. At 2.45 GHz, and a ten km footprint on the ground, the
transmitting antenna has to be a km in diameter. The math is here:
https://en.wikipedia.org/wiki/Airy_disk The smaller the transmitter,
the larger the footprint on the ground.

> - Can SPSs be designed to self-deploy at GEO so as to minimize the cross-sectional area in LEO and hence greatly reduce the risk of collision in LEO?

Of course, but that means you have to use chemical fuel or some trick
like beamed energy to get them to GEO. Chemical fuel just costs too
much.

Keith

> Thanks,
>
> Doug

>
>
> On Wed, Jan 29, 2025 at 11:47 AM Keith Henson <hkeith...@gmail.com> wrote:
>>

[jkst...@sbcglobal.net]

What about mostly automated assembly in orbit, overseen by humans on the ground?
This seems to work for other industries

John S

--
You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB6abkzrFP_wHQw6U8FLvTJq_GL_aMbOSFQukvVKE00B7Q%40mail.gmail.com.

[Keith Henson]

On Wed, Jan 29, 2025 at 3:05 PM <jkst...@sbcglobal.net> wrote:
>
> What about mostly automated assembly in orbit, overseen by humans on the ground?
> This seems to work for other industries

That's the current best concept I know about for building power
satellites. If you have to build them above most of the junk, that
puts you in the lower edge of the Van Allen belt where humans would
die of radiation in hours.

[question] If we decide to build power satellites, something will
have to be done

copilot

Absolutely, building power satellites would indeed require some
significant efforts in terms of orbital management and space debris
mitigation.

🚀 Potential solutions:

Improved Tracking Systems: Enhancing real-time tracking and prediction
of space debris movements.

Active Debris Removal: Technologies to capture and remove space debris
could help clear safe paths.

Better Coordination: International cooperation to create a more
organized and managed space traffic system.

Orbital Clean-up Policies: Enforcing stricter regulations on satellite
de-orbiting and end-of-life disposal.

🌌 It's a complex challenge, but the potential benefits of power
satellites, such as providing clean and continuous energy, could be a
game-changer for our planet. Tackling these hurdles will push the
boundaries of our current technology and may even inspire new
innovations in space exploration and safety.

What intrigues you the most about the idea of power satellites?

[question] When do you think humans might start building power satellites?

Building power satellites is an ambitious goal, and while it's
challenging, there are promising signs that we might see progress
within the next few decades

snip

While these projects and studies provide a roadmap, the actual
timeline will depend on overcoming technical, financial, and
regulatory challenges. However, the growing interest and investment in
this area suggest that we might see the construction of power
satellites within the next 20 to 30 years.

[qustion] When do you think the singularity might come about?

The concept of the singularity is intriguing and highly speculative.
Experts have various opinions on when it might occur, if at all. Here
are some of the perspectives:

Ray Kurzweil: Futurist and author of "The Singularity is Near,"
predicts that the singularity could happen around 2045. He believes
that exponential growth in computing power and artificial intelligence
will lead to this pivotal moment.

Skeptics: Some experts are more cautious, suggesting that the
singularity might be decades or even centuries away, if it ever
happens. They point out that predicting technological breakthroughs is
inherently uncertain.

Current Trends: While there have been impressive advances in AI and
technology, there are still significant challenges to overcome, such
as achieving general artificial intelligence, understanding
consciousness, and ensuring ethical considerations are met.

In essence, the timeline for the singularity remains uncertain, and
it's a topic of ongoing debate and research.

(For some reason MS has made it hard to cut and paste from copilot.)

copilot seems kind of gullible.

Keiht

> To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/0ddc01db72a2%244badc770%24e3095650%24%40sbcglobal.net.

[Tim Cash]

Consider what all of you have just discussed and intimated.  You have stated if we decide to DO space solar power the "right way", it will require all of us to innovate the "right way" of completing all the steps required to do it "the right way" and in fact will usher in Space 3.0 immediately and the "millennium" in the correct reliable engineering me manner!  Let's then get busy doing it!

Tim Cash
cash...@gmail.com

[jkst...@sbcglobal.net]

Simple -do NOT DEPLOY OR EXTEND THE ARRAYS UNTL ON STATION

JOHN s

-----Original Message-----
From: power-satell...@googlegroups.com <power-satell...@googlegroups.com> On Behalf Of Keith Henson
Sent: Wednesday, January 29, 2025 2:48 PM
To: Howard Bloom <howl...@aol.com>; Power Satellite Economics <power-satell...@googlegroups.com>
Cc: dale skran <dalels...@gmail.com>; giulio Prisco <giu...@gmail.com>; stewart atkinson <dstewa...@gmail.com>; fred becker <mac...@comcast.net>; steve kwast <kins...@icloud.com>; Don Flournoy <don.fl...@ohio.edu>; Feng Hsu <feng...@gmail.com>; jon labore <blueskies...@proton.me>; Bruce Pittman <bpit...@earthlink.net>; Richard Godwin <spac...@gmail.com>; john rutledge <jdrutl...@gmail.com>; Ryan Mcleod <ryan...@gmail.com>; Christopher Jannette <astrobi...@gmail.com>; Mark Sonter <mark....@asteroidenterprises.com>; John Strickland <jkst...@sbcglobal.net>; RICHARD ALCANTARA <ralcan...@gmail.com>; jgl...@aol.com; HENRY HALIASZ <hankh...@gmail.com>; paul werbos <pwe...@gmail.com>; Elizabeth Kennick <liz.k...@tis.org>; ELIZABETH SCHNEIDER <e...@galacticmedallion.com>; Paul Damphousse <paul.e.d...@gmail.com>; Jeroen Lapre <jeroen...@gmail.com>; 847lov...@gmail.com; Bill Gardner <william.w...@gmail.com>; kris holland <kr...@maficstudios.com>; Jason Louv <jaso...@gmail.com>; rick tumlinson <ricktu...@gmail.com>; Sam Coniglio <sam.co...@gmail.com>; James Oberg <james...@comcast.net>; Peter Garretson <budo...@gmail.com>; Gary Barnhard <barn...@barnhard.com>; Ben Shelef <b...@spaceward.org>; hicou...@aol.com; lucia chavirar <luciac...@gmail.com>; Jerome Glenn <jgl...@igc.org>; sara.a...@seds.org; Feng Hsu <feng...@gmail.com>; nancy atkinson <na...@universetoday.com>; Chris Carberry <cacar...@yahoo.com>; david ho <davi...@spacegeneration.org>; Jim Armor <james...@gmail.com>; Steven Wolfe <wol...@aol.com>; s...@etiam-engineering.com; gabriela <gabriela...@gmail.com>; Steve Nixon <yoda...@hotmail.com>; David Livingston <drs...@thespaceshow.com>; denni...@gmail.com; Greg Allison <ghal...@aol.com>; andrew lindberg <andre...@gmail.com>; dougsp...@gmail.com; Joe Rauscher <rausche...@gmail.com>; Nicola Sarzi Amade <nicola...@gmail.com>; ste...@blacksky.network; gabriela <gabriela...@nss.org>; david.c...@gmail.com; Margo Deckard <marde...@aol.com>; Kirk Woellert <kdw...@gmail.com>; Bill Gardner <analyte...@bellsouth.net>; Gerry Black <gbl...@cinci.rr.com>; Don Flournoy <flou...@ohio.edu>; Trent Waddington <trent.wa...@gmail.com>; Bruce Mackenzie <bmack...@alum.mit.edu>; MACKENZIE MCALEER <M...@mm.st>; sam spencer <sam.s...@runbox.com>; j.d...@gmail.com; rob furey <rfu...@thought.live>; Rob shapiro <robsh...@gmail.com>; David Livingston <dliv...@davidlivingston.com>; Joe Champion <joecham...@gmail.com>; John Jossy <jajos...@gmail.com>; TREVOR GARVIN <Wsht...@mail.com>; James Michael Snead <james...@aol.com>; Karen Shea <karen...@gmail.com>; leonard david <news...@aol.com>; genemey...@icloud.com; james gholston <jam...@dimensionality.com>; comp...@gmail.com; lorigor...@gmail.com; James Boyd <boyd...@newschool.edu>; ericm...@factualfiction.com; Lynne Zielinski <lziel...@comcast.net>; Jeffrey Liss <jgl...@aol.com>; Stephen Leete <stephen...@gmail.com>; rocket...@gmail.com; TOLGA ORS <tolg...@hotmail.com>; tim cash <cash...@gmail.com>; Al Globus <algl...@gmail.com>; harold...@verizon.net; Lonnie Schorer <lonnie...@aol.com>; Peter Garretson <peter.g...@us.af.mil>; JIM HURD <HURD...@gmail.com>; Manuel Pimento <ma...@virtualmoon.space>; John Spencer <jssd...@aol.com>; a.p.k...@astrox.com; rezaz zekavat <re...@mtu.edu>; bill gowan <bgo...@gmail.com>; rick zucker <rckz...@aol.com>; KEVIN RUSSELL <Kpri...@proton.me>; aggie kobrin <ag...@cecglobalevents.com>; Will Watson <willj...@gmail.com>; lauren wilson <lauren...@gmail.com>
Subject: Re: Power satellite problems

--
You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB62pMfo_uLX0fdaugVyeHM%3D6UBNxpuHbdC%3D%3D3TcwWUPow%40mail.gmail.com.

[Keith Henson]

On Wed, Jan 29, 2025 at 6:16 PM Tim Cash <cash...@gmail.com> wrote:
>
> Consider what all of you have just discussed and intimated. You have stated if we decide to DO space solar power the "right way",

The problem for me is that I don't know what the right way is. I know
a ton of ways that will not work. Maybe the AIs are right in that we
can build power satellites and thread them through the maze of objects
in orbit. I don't know. I would sure like to see a study on the
subject. The consequences of a major collision are dire.

Keith

[Keith Henson]

On Wed, Jan 29, 2025 at 6:30 PM <jkst...@sbcglobal.net> wrote:
>
> Simple -do NOT DEPLOY OR EXTEND THE ARRAYS UNTL ON STATION

OK, then how do you power the arcjet thrusters?

Keith

[jkst...@sbcglobal.net]

What % of the full array area is needed to operate them ?

John S

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB67%3DOzm18ZET_%3DTNNbDwBxv5VOJA%2BRSLFLZPB-55-3P-w%40mail.gmail.com.

[Tim Cash]

All concerned:

If we compare the effort to get a Powersat SAFELY from LEO to GEO end state against what the world space efforts are jointly, it dwarfs them.

I suggest that any/each/all of us write up what and how we recommend the orbital Powersat Constellation to be built, write up those suggestions, and create an overarching document for system reliability.

That should become THE BASELINE DOCUMENT in how anyone would design and build a Powersat Constellation.

I imagine the Chinese have already created and are NOT SHARING such documents in their own efforts to create space solar power for China.

If we are to get along on this planet and NOT destroy each other, we must share and create Global standards for space solar power, and all of the other space 3.0 technologies required.

Otherwise, forget our species' long-term survival and ever getting off this planet.

I recently created my own short list of required Space 3.0 Technologies, attached.

This document is a list of MY goals for the remainder of my life.

Tim Cash

cash...@gmail.com

--

Tim Cash

cash...@gmail.com

[Keith Henson]

On Wed, Jan 29, 2025 at 9:54 PM <jkst...@sbcglobal.net> wrote:
>
> What % of the full array area is needed to operate them ?

It is not clear, but less power takes longer which increases the
number of hits. Unless you are moving tightly packed parts, you need
to actively avoid hitting space junk.

Also, you seem to be assuming PV. PV has a lot of problems, thermal
designs are twice as efficient meaning smaller area and less station
keeping cost. Plus PV degrades quickly where a thermal system should
run for 50 years.

Keith

[Keith Henson]

Tim, it is not clear where to build the power satellites.

I have considered LEO, 2000 km and 12,000 km, in the gap between the
lower and upper Van Allen belts. They all have problems, LEO has too
much junk, 2000 km has radiation problems, 12,000 km is a long way up.
The higher you go, the more it costs in rocket fuel.

Well, if the AIs are right, power satellites are 20 to 30 years out.

Keith


Best wishes,

Keith

[James “Jim” Salsman]

Keith wrote:

... if the AIs are right, power satellites are 20 to 30 years out.

Does your AI confirm these conclusions from Gemini 2.0 Flash Thinking?

> A terrestrial solar farm covering about 12 square kilometers would typically cost around $550 million. This would provide a mean power output of about 550 MW in central Wyoming, where the capacity factor is typically 0.25.[ https://cowboystatedaily.com/2024/02/09/oregon-sheep-expert-important-to-500-million-wyoming-solar-farm/ ] Such a return on investment is reasonable for renewable energy projects, and very competitive with all forms of electricity generation, including coal and petroleum.

> For space solar power, let's just focus on the ground rectenna for a moment, again using a 12 square kilometer footprint. Just the rectenna on the ground is estimated to cost around $6 billion. This substantial expenditure is driven by the immense scale of materials and the intricate construction required to build a high-efficiency microwave rectenna of this size, which requires vast quantities of specialized materials, including conductive metals, high-performance diodes, and low-loss substrates, along with complex manufacturing and assembly processes to achieve an 85% power conversion efficiency. For that $6 billion, we'd get a mean power output of about 500 MW. So, even just comparing the ground infrastructure, terrestrial solar is better in terms of power per dollar.

> However, the $6 billion for the space solar power rectenna ignores the space-based parts of the system. To actually illuminate that 12 square kilometer rectenna from geostationary orbit and deliver that 500 MW of power, we would need a massive satellite in space. Such a satellite would require vast solar arrays spanning several square kilometers, to capture sufficient sunlight in space. These arrays would require specialized, space-qualified materials capable of withstanding the harsh space environment and efficient energy conversion. Furthermore, a kilometer-scale transmitting antenna, potentially employing advanced phased array technology for precise beam steering and control, would be necessary to direct the microwave energy towards the Earth-based rectenna. The spacecraft bus itself, encompassing power management, thermal control, attitude control, and communication systems, would need to be exceptionally robust and reliable for decades of operation in geostationary orbit. The sheer mass and volume of these components would necessitate numerous costly launches and complex on-orbit assembly procedures, potentially requiring advanced robotics and human-tended missions. Considering the cutting-edge technologies, immense scale, and inherent risks associated with space-based construction and operation at this magnitude, the cost of the satellite segment alone would undoubtedly reach into hundreds of billions of dollars, dwarfing the already substantial cost of the ground-based rectenna and rendering the overall Space Solar Power system economically prohibitive in the current technological and economic landscape.

> The estimated cost of the Space Solar Power satellite designed to deliver 0.48 GW of mean power, reaches a staggering $310 billion. This colossal figure is composed of several major cost components: Launch costs, estimated at $3 billion, account for transporting 26,000 metric tons of materials to GEO assuming $100/kg launch costs. Material costs, the largest component at $130 billion, reflect the sheer quantity and specialized nature of space-qualified materials needed for the vast solar arrays, kilometer-scale antenna, and spacecraft bus. Manufacturing and fabrication, estimated at $65 billion, cover the complex and precise construction of these advanced components. In-space assembly and astronaut labor, conservatively estimated at $10 billion, account for the unprecedented task of assembling this massive structure in orbit, even with assumed robotic assistance. Finally, non-recurring engineering, research, and development (NRE) costs, estimated at $100 billion, encompass the immense upfront investment required to design, develop, and test this novel and technologically challenging system. These individual costs, when summed, paint a clear picture of the extraordinary financial hurdle that the space segment presents for Space Solar Power.

> So, when you look at the return on capital just comparing the ground infrastructure, TSP is already more efficient. But when you factor in the absolutely enormous cost of the space satellite needed for SSP – which aren't included in that $6 billion rectenna figure – the economic case for TSP becomes overwhelmingly stronger. It amounts to a difference between $1 and $632 per Watt.

Best regards,

Jim

[Keith Henson]

On Thu, Jan 30, 2025 at 3:02 PM James “Jim” Salsman <jsal...@gmail.com> wrote:
>
> Keith wrote:
>
> ... if the AIs are right, power satellites are 20 to 30 years out.
>
> Does your AI confirm these conclusions from Gemini 2.0 Flash Thinking?
>
> > A terrestrial solar farm covering about 12 square kilometers would typically cost around $550 million. This would provide a mean power output of about 550 MW in central Wyoming, where the capacity factor is typically 0.25.[ https://cowboystatedaily.com/2024/02/09/oregon-sheep-expert-important-to-500-million-wyoming-solar-farm/ ] Such a return on investment is reasonable for renewable energy projects, and very competitive with all forms of electricity generation, including coal and petroleum.

Wait, the AI jumps to ROI without mentioning income or storage. And
generation using oil? From the numbers, a million per MW or $1000/kW.
The LCOE metric is to divide the capital cost by 80,000 to get cents
per kWh. That comes out at 1.25 cents per kWh, which is consistent
with the lowest price PV I know about which is 1.35 cents per kWh.

What is not mentioned is that the available power varies from about 2
GW to zero when the sun goes down. Making intermittent power into
steady costs 2-3 cents per kWh.

> > For space solar power, let's just focus on the ground rectenna for a moment, again using a 12 square kilometer footprint. Just the rectenna on the ground is estimated to cost around $6 billion.

Can you ask the AI where it got this number? As far as I know, I am
the only one who estimated the cost of a rectenna and I got one
billion for a 5 GW rectenna. That's $200/kW.

> >This substantial expenditure is driven by the immense scale of materials and the intricate construction required to build a high-efficiency microwave rectenna of this size, which requires vast quantities of specialized materials, including conductive metals, high-performance diodes, and low-loss substrates, along with complex manufacturing and assembly processes to achieve an 85% power conversion efficiency. For that $6 billion, we'd get a mean power output of about 500 MW. So, even just comparing the ground infrastructure, terrestrial solar is better in terms of power per dollar.

500 MW is too small by a factor of ten.

> > However, the $6 billion for the space solar power rectenna ignores the space-based parts of the system. To actually illuminate that 12 square kilometer rectenna from geostationary orbit and deliver that 500 MW of power, we would need a massive satellite in space. Such a satellite would require vast solar arrays spanning several square kilometers, to capture sufficient sunlight in space. These arrays would require specialized, space-qualified materials capable of withstanding the harsh space environment and efficient energy conversion. Furthermore, a kilometer-scale transmitting antenna, potentially employing advanced phased array technology for precise beam steering and control, would be necessary to direct the microwave energy towards the Earth-based rectenna. The spacecraft bus itself, encompassing power management, thermal control, attitude control, and communication systems, would need to be exceptionally robust and reliable for decades of operation in geostationary orbit. The sheer mass and volume of these components would necessitate numerous costly launches and complex on-orbit assembly procedures, potentially requiring advanced robotics and human-tended missions. Considering the cutting-edge technologies, immense scale, and inherent risks associated with space-based construction and operation at this magnitude, the cost of the satellite segment alone would undoubtedly reach into hundreds of billions of dollars, dwarfing the already substantial cost of the ground-based rectenna and rendering the overall Space Solar Power system economically prohibitive in the current technological and economic landscape.
>
> > The estimated cost of the Space Solar Power satellite designed to deliver 0.48 GW of mean power, reaches a staggering $310 billion. This colossal figure is composed of several major cost components: Launch costs, estimated at $3 billion, account for transporting 26,000 metric tons of materials to GEO assuming $100/kg launch costs.

I wonder where the AI got the power satellite mass? I get 32,500 tons
for 5 GW on the ground. That is 6.5 kg/kW. At half a GW, the AI is
assuming around 50 kg/kW.

$100/kg to GEO is a real trick. $100 to LEO yes, but it is a long way
from LEO to GEO. I think electric propulsion will get the cost to GEO
down to $200/kg but that's tentative.

> >Material costs, the largest component at $130 billion, reflect the sheer quantity and specialized nature of space-qualified materials needed for the vast solar arrays,

I think a real engineering study would conclude PV is a bad choice for
GW-scale power in space. Thermal is at least twice as efficient and
does not degrade from radiation. I have been using $900/kW for parts,
$130 B for half a GW is $260,000 per kW. That sounds high even for
communication satellites.

> > kilometer-scale antenna, and spacecraft bus. Manufacturing and fabrication, estimated at $65 billion, cover the complex and precise construction of these advanced components. In-space assembly and astronaut labor,

Depending on where they are being constructed, human labor does not
work. (The humans die from radiation before they can collect a
paycheck.)

> >conservatively estimated at $10 billion, account for the unprecedented task of assembling this massive structure in orbit, even with assumed robotic assistance. Finally, non-recurring engineering, research, and development (NRE) costs, estimated at $100 billion, encompass the immense upfront investment required to design, develop, and test this novel and technologically challenging system.

The AI is loading one power satellite with the entire R&D. That's not
how accounting works, you spread the R&D over a run of them. How did
the AI assume such a biased method?

> > These individual costs, when summed, paint a clear picture of the extraordinary financial hurdle that the space segment presents for Space Solar Power.
>
> > So, when you look at the return on capital just comparing the ground infrastructure, TSP is already more efficient. But when you factor in the absolutely enormous cost of the space satellite needed for SSP – which aren't included in that $6 billion rectenna figure – the economic case for TSP becomes overwhelmingly stronger. It amounts to a difference between $1 and $632 per Watt.

You sure need to watch what you get from an AI.

keith

> Best regards,
>
> Jim

>
>
> --
> You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.

> To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/e4cde822-c118-4456-9371-ad950042634cn%40googlegroups.com.

[James Salsman]

> > ... if the AIs are right, power satellites are 20 to 30 years out.
> >
> > Does your AI confirm these conclusions from Gemini 2.0 Flash Thinking?
> >
> > > A terrestrial solar farm covering about 12 square kilometers would typically cost around $550 million. This would provide a mean power output of about 550 MW in central Wyoming, where the capacity factor is typically 0.25.[ https://cowboystatedaily.com/2024/02/09/oregon-sheep-expert-important-to-500-million-wyoming-solar-farm/ ] Such a return on investment is reasonable for renewable energy projects, and very competitive with all forms of electricity generation, including coal and petroleum.
>
> Wait, the AI jumps to ROI without mentioning income or storage. And
> generation using oil? From the numbers, a million per MW or $1000/kW.
> The LCOE metric is to divide the capital cost by 80,000 to get cents
> per kWh. That comes out at 1.25 cents per kWh, which is consistent
> with the lowest price PV I know about which is 1.35 cents per kWh.
>
> What is not mentioned is that the available power varies from about 2
> GW to zero when the sun goes down. Making intermittent power into
> steady costs 2-3 cents per kWh.

The AI's reply to you:

> Okay, let's re-evaluate the economic comparison of Space Solar Power (SSP) and Terrestrial Solar Power (TSP) in prose, incorporating Keith's recommendations and assumptions, and using a more conversational tone. We'll stick to a 12.2 square kilometer ground footprint for both systems and focus on mean power output in gigawatts (GW).
>
> First, let's look at Terrestrial Solar Power. Based on Keith's input, citing a specific project in Wyoming, a 12.2 km² solar farm would cost around $559 million. This size of installation would generate a peak power output of roughly 2.24 GW. Taking into account Wyoming's typical solar capacity factor of 0.25, the mean power output comes down to about 0.56 GW. This gives us a cost of approximately $1.00 per watt. Using Keith's simplified method for calculating the Levelized Cost of Energy (LCOE), which is dividing the capital cost by 80,000, we arrive at an LCOE of 1.25 cents per kilowatt-hour (kWh). However, because solar power is intermittent, we need to add the cost of storage to provide steady, dispatchable power. Keith suggests 2-3 cents per kWh for storage, so let's use a midpoint of 2.5 cents/kWh. Adding that to the LCOE, we get a total of 3.75 cents/kWh for dispatchable power from the terrestrial solar farm.
>
> Now, let's turn our attention to Space Solar Power. Keith provides some significantly different cost estimates for the rectenna, suggesting $1 billion for a 5 GW rectenna. Scaling that down to a 0.5 GW output, which aligns proportionally with his numbers and the scale of our TSP example, we get a rectenna cost of $100 million for the 12.2 km² area. He also implies a mean power output of 5 GW for his rectenna. For our comparison, given his cost basis, we'll use that 5 GW figure as well. It is important to note that this is ten times the power output assumed in our prior analyses.
>
> The satellite cost is where things get more complex. Keith estimates a satellite mass of 32,500 tons for a 5 GW system. Using his suggested launch cost of $200/kg to GEO, the launch alone would cost $6.5 billion. For materials, he favors solar thermal over photovoltaic and suggests a cost of $900/kW for parts, totaling $4.5 billion for a 5 GW system. We'll estimate manufacturing at 50% of material costs, adding another $2.25 billion. For in-space assembly, Keith believes robotic assembly would be more feasible than using astronauts, so we'll estimate $1 billion for that, a significant reduction from our previous astronaut labor cost. Finally, for R&D, let's allocate $25 billion, which is a reduction from our previous estimate to account for amortization over a hypothetical production run.
>
> Adding up all the satellite costs, we arrive at an estimated $39.25 billion. Combining this with the $100 million rectenna cost gives us a total SSP system cost of $39.35 billion. This translates to roughly $7.87 per watt. Using Keith's simplified LCOE method, we get an LCOE of 9.84 cents/kWh.
>
> In conclusion, even when incorporating Keith's more optimistic cost assumptions, particularly for the rectenna and satellite materials, Space Solar Power remains considerably more expensive than Terrestrial Solar Power. The TSP system, with a cost of $1.00 per watt and a total LCOE of 3.75 cents/kWh for dispatchable power, is significantly more cost-competitive than the SSP system at $7.87 per watt and an LCOE of 9.84 cents/kWh. Although SSP offers the advantage of continuous, baseload power, this benefit does not outweigh the substantially higher capital costs and LCOE compared to TSP, even when using Keith's suggested cost reductions. Based on this revised analysis, TSP remains the more economically viable option for large-scale solar energy generation.

> > > For space solar power, let's just focus on the ground rectenna for a moment, again using a 12 square kilometer footprint. Just the rectenna on the ground is estimated to cost around $6 billion.
>
> Can you ask the AI where it got this number? As far as I know, I am
> the only one who estimated the cost of a rectenna and I got one
> billion for a 5 GW rectenna. That's $200/kW.

Yes, but it's just an estimate without any underlying calculations.
Maybe there are backing numbers in the chain-of-thought dialogue. I
will try it where I can see that text, but let me ask you: how did you
come up with your estimate of the rectenna and satellite in terms of
the underlying line item math? We can compare both.

Best regards,
Jim

[jkst...@sbcglobal.net]

Yes but a thermal system has MOVING PARTS.

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB7gLMSXUUm13vE0A%2BW%2BRv4qH4gJHE_2nOPccxRj2n-M3w%40mail.gmail.com.

[Keith Henson]

On Thu, Jan 30, 2025 at 8:03 PM James Salsman <jsal...@gmail.com> wrote:
>
> > > ... if the AIs are right, power satellites are 20 to 30 years out.
> > >
> > > Does your AI confirm these conclusions from Gemini 2.0 Flash Thinking?

Looking this over, I have my doubts about it coming from an AI.

It is a fairly good job except that AIs don't make typos, like hitting
9 instead of a 0.

Nor do they make an egregious error like the number of launches required.

Not sure what the point of this misinformation is. Perhaps James can
tell us, though I expect it is more likely he will vanish.

Keith

[Keith Henson]

On Thu, Jan 30, 2025 at 9:42 PM <jkst...@sbcglobal.net> wrote:
>
> Yes but a thermal system has MOVING PARTS.

I have a refrigerator that is full of moving parts. It is 25 years
old and has never failed.

It is not unusual for steam turbines to last 50 years.

[Tim Cash]

This proves the point to me that AI stands for Artificial Idiocy.

Why bother, when we have PLENTY of natural born human ID10T's around?

I find  it ludicrous to feed this conversation to an AI in the first place, the construction of a power plant is a HUMAN DECISION and should remain that way.

From what I have read, if we want a system with zero moving parts, in order to have high reliability, then go with lunar built monocrystalline PV Cells, haul them up a lunar space elevator to the lunar orbital station for Powersat construction, and then the remaining way up the tether to GEO.  I worked on that proposal a few years back, and that looks to be the best economics and reliability yet stated for space solar power.  The question to me is how long will it require to work out lunar remote construction using automation and robotics to build a remote telepresence operated factory for monocrystalline PV on the lunar surface, or any other type of remote lunar construction effort operated through telepresence?  That is the critical path for space solar power it seems to me at this point.  The difference here is that this critical path to lunar automation is managed by human and not AI efforts, which was assumed in my recent required technologies on the path to space 3.0 document.  Until and IF such a trusted AI technology knocks on my door and introduces itself properly, I will not put my trust into AI, period, end of story.  This is called common sense, people.

Tim Cash

cash...@gmail.com

--

Tim Cash

cash...@gmail.com

[k.a.c...@sympatico.ca]

This proves the point to me that AI stands for Artificial Idiocy.

Why bother, when we have PLENTY of natural born human ID10T's around?

I find  it ludicrous to feed this conversation to an AI in the first place, the construction of a power plant is a HUMAN DECISION and should remain that way.

 

Repeating my earlier point: these programs are not AI. They are fancy sentence autocomplete algorithms. They use databases that were assembled  using software that includes neural networks. If you expect them to do any “thinking,” you’ll be disappointed (or deluded).

 

 

[jkst...@sbcglobal.net]

What about the old computer chess player programs.

Their value was proved decades ago.

When focused on well defined tasks, computer AI has proved its usefulness.

What the critics are after is probably a general AI capability, where more complex decisions must be made.

It seems that some progress is happing even in this area.

.

“Stop, Dave, I’ Afraid”

 

 

John S

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAA%3Dt13z0NtEVnJuKgZJqb3V2h4VrRutcjwVwbtUsKstbAeJ6Yw%40mail.gmail.com.

[Geoffrey Landis]

"Also, you seem to be assuming PV. PV has a lot of problems, thermal designs are twice as efficient"

I don't know how many times I have to debunk this, but it is still not true.

The highest efficiency ever achieved by a concentrator solar-thermal system is 31.4% (solar in to electrical power out).  That's today's space solar cell efficiency.  So, the very best ever solar thermal system just comes up to today's commercial space PV. The very best every solar cells, on the other hand,, are 39.5% efficient for a non-concentrating cell, and 47.6% for a concentration of 665 suns.

Now, you can say "but I have a design (or, a concept for a design) that will be twice that efficient." Right. Show me your experience in thermal and optical system designs sufficient to convince me that you can make something twice as efficient as the best performance ever achieved by experts in the field.

"meaning smaller area and less station keeping cost."

Also dubious. Thermal systems need both collector area and also radiator area. Depending on how hot you run the radiators (the hotter you run them the smaller they are, but the less efficient the conversion), thermal systems can be smaller or can be larger total area than photovoltaic.

--Geoff

[Roger Arnold]

I don't have a dog in the fight over PV vs solar thermal for power satellites. I'm not a big fan of solar power satellites to be honest. Too much light pollution, and too much background RF interference from microwave power transmission for my taste. However I feel compelled to jump in on the efficiency issue.

Geoff, can you give a specific example of a concentrator solar-thermal system that achieved a 31.4% efficiency but represented the highest efficiency ever achieved for such a system? 31.4% sounds like the highest one might get from a big field of heliostats reflecting sunlight onto a boiler at the top of a central tower. But those are terrestrial systems designed for minimum capital cost and the ability to continue generating power from stored heat after sunset.  Not for efficiency. The fuel is free. The collector sits exposed at the top of the tower, white hot and radiating away half of the incident solar energy. It's a stupid design, IMHO, and nothing at all like one would want for use in space. 

For a solar-thermal power satellite, one would not be using heliostats., and one would most certainly not be shining reflected sunlight onto the outside of a boiler. What I'd use if I were designing the system would be a front-surfaced parabolic film mirror focusing an image of the sun onto a small window into an insulated cavity. Equilibrium temperature inside the cavity might be 2000 C, but withdrawal of heat for power generation would keep it lower. Power generation would be by a supercritical CO2 Brayton cycle, With the kind of low temperature thermal radiator that Keith has proposed, I'd expect a thermal efficiency of at least 60%. Combined cycle gas turbines already achieve that on earth. I can see no reason that a high temperature SCCO2 system in space couldn't match it.

Roger Arnold

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/6342a3ab-49e2-4d05-a1ea-738c4faab69dn%40googlegroups.com.

[Bryan Zetlen]

Roger what 'light pollution' from orbital microwave power arrays?

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAN%3D9Pg%3DBRDyoyjwJaHqx5iZ-SukooyjCPBYJxtZtETDxZsNBUw%40mail.gmail.com.

[jkst...@sbcglobal.net]

This issue is similar to the effort to make Starlink satellites dark colored.

Some astronomers refer to masses of satellites as “vermin of the sky”.

Reduction in visibility can be done by darker coloring, attitude and shading.

If a satellite is too dark it could overheat from the full sunlight in orbit.

 

John S

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAN%3D9Pg%3DBRDyoyjwJaHqx5iZ-SukooyjCPBYJxtZtETDxZsNBUw%40mail.gmail.com.

[Bryan Zetlen]

Thanks, that’s not an answer. Let’s drop it. 

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/017201db7467%246a253b40%243e6fb1c0%24%40sbcglobal.net.

[Keith Henson]

On Fri, Jan 31, 2025 at 6:04 PM Geoffrey Landis
<geoffre...@gmail.com> wrote:
>

> "Also, you seem to be assuming PV. PV has a lot of problems, thermal designs are twice as efficient"
>
> I don't know how many times I have to debunk this, but it is still not true.
> The highest efficiency ever achieved by a concentrator solar-thermal system is 31.4% (solar in to electrical power out).

I was not aware of any thermal systems in space, so this must be on
the ground. There are tricks you can use in space that don't work
well on the ground. I assume that a two-stage thermal system (steam
or SCCO2 and a potassium topping cycle) should do as well as a
combined cycle power plant on Earth. I have cited the patent on an
absorber that reradiates very little heat. (Light and working fluid
co-flow through the absorbers.)

> That's today's space solar cell efficiency. So, the very best ever solar thermal system just comes up to today's commercial space PV. The very best every solar cells, on the other hand,, are 39.5% efficient for a non-concentrating cell, and 47.6% for a concentration of 665 suns.

Right, I proposed to use these in the "beamed energy bootstrapping"
video and they would work fine for a few small power satellites. But
there is not enough indium to make enough power satellites to make any
difference in the energy problem.

> Now, you can say "but I have a design (or, a concept for a design) that will be twice that efficient." Right. Show me your experience in thermal and optical system designs sufficient to convince me that you can make something twice as efficient as the best performance ever achieved by experts in the field.

My experience in space radiator design goes back to a paper Eric
Drexler and I wrote in 1979. But generally, I don't think there are
any experts in GW scale radiators.

>
> "meaning smaller area and less station keeping cost."
>
> Also dubious. Thermal systems need both collector area and also radiator area.

Right, but the radiator area does not contribute to light pressure or
station-keeping problems..

> Depending on how hot you run the radiators (the hotter you run them the smaller they are, but the less efficient the conversion), thermal systems can be smaller or can be larger total area than photovoltaic.

To my surprise when I was running Excel sheets on radiator design, I
discovered that they got lighter as the radiator temperature declined.
Turns out that the vapor pressure of water falls faster than T^4 over
the liquid water range. Pressure determines the wall thickness which
is a large fraction of the radiator mass. This kind of radiator in
the GW size masses a little over a kg/kW and operates at 20 degrees C.

Keith

> To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/6342a3ab-49e2-4d05-a1ea-738c4faab69dn%40googlegroups.com.

[Roger Arnold]

Bryan,

The light pollution I referred to is from sunlight reflected / scattered from a thousand very large satellites, each with an exposed surface area of multiple square kilometers. I believe the issue was studied in the '70s when Boeing did its reference study for NASA. I was at Boeing at the time, and knew Gordon Woodcock and others who worked on that study. I don't recall specific numbers, but I recall being horrified at the magnitude of what was estimated. I believe it was on the order of a half moon equivalent. No more truly dark skies anywhere on Earth. No more camping under a dark sky and gazing up at the wonder of the Milky Way spreading across the sky. There's no easy fix for it. There are super-black materials that will absorb 99% of the light falling on them, but the solar collectors have to be exposed to sunlight.

[jkst...@sbcglobal.net]

So what kind of energy systems do you support?

I support nuclear fission, fusion research and SPS, along with continuing to use fossil fuels until we have replacement systems.

Tidal power works but needs massive infrastructure and is VERY intermittent.

Ground solar is too diffuse and intermittent to rely on as a source of base load power.

Space Solar is neither diffuse nor intermittent.

Also it is a system which has essentially no moving parts.

 

John S

 

 

 

 

From: power-satell...@googlegroups.com <power-satell...@googlegroups.com> On Behalf Of Roger Arnold
Sent: Friday, January 31, 2025 9:42 PM
To: Geoffrey Landis <geoffre...@gmail.com>
Cc: Power Satellite Economics <power-satell...@googlegroups.com>
Subject: Re: Thermal [Re: Power satellite problems]

 

I don't have a dog in the fight over PV vs solar thermal for power satellites. I'm not a big fan of solar power satellites to be honest. Too much light pollution, and too much background RF interference from microwave power transmission for my taste. However I feel compelled to jump in on the efficiency issue.

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAN%3D9Pg%3DBRDyoyjwJaHqx5iZ-SukooyjCPBYJxtZtETDxZsNBUw%40mail.gmail.com.

[Tim Cash]

Your energy solutions are identical to mine, with the addition of use of energy drilling for Geothermal energy everywhere on the planet. A recent news blurb on this new drilling method grabbed my attention.

Tim

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/01ec01db74ca%246ad4d330%24407e7990%24%40sbcglobal.net.

[Tim Cash]

As far as efficiency. Metamaterials and other tech improvements are increasing efficiency all the time.

[jkst...@sbcglobal.net]

In most locations, “hot enough” geothermal heat is too deep to be practical.

If we had faster drilling methods, we could simply drill more holes in the good geothermal areas.

Maybe drill into a volcanic area or two.

However, some forces say hands off Yellowstone, so as not to disrupt the natural geothermal features there.

 

John S

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAA%3Dt13zBk9HBAtNO5mV3qMkLWB7Xu5AuN2R_cPmWY-ujRmYsVg%40mail.gmail.com.

[Keith Henson]

BTW, Geoffrey, I would accept the results of a serious study that put
PV and thermal head to head.

Nothing of that nature has been done since the early Boeing studies.
Boeing almost discovered the fundamental diseconomy of scale for
radiators which is discussed in the 1979 paper. That is partly
avoided in condensing radiators.

Keith

[Keith Henson]

On Sat, Feb 1, 2025 at 8:57 AM <jkst...@sbcglobal.net> wrote:

> Ground solar is too diffuse and intermittent to rely on as a source of base load power.

You need storage far larger than you can get with batteries.

The cost is very attractive though. At the 2 GW scale, you can buy
power for 1.35 cents per kWh. My best efforts don't get power from
space lower than 3 cents per kWh, and that is speculative where ground
solar at this cost is real.

Storage as hydrogen doesn't work because the electrolyzers are so
expensive. However, if you heat coal, biomass, or trash in steam the
carbon and steam make hydrogen and CO. The fuel value of the gas is
3-4 times the intermittent energy input. If you have an empty gas or
oil field, you can store the gas seasonably or make it into diesel or
jet fuel.

Not a small project, though smaller than a fleet of power satellites.

Keith

> To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/01ec01db74ca%246ad4d330%24407e7990%24%40sbcglobal.net.

[Roger Arnold]

For Earth, I support advanced fission reactors, fusion reactors if any of the variants being explored ever prove practical, stratosolar PV, CO2 plume geothermal, and Allam cycle thermal power to the extent that they can be used in programs for intelligent waste management and recycling.

There's also a variant of satellite solar power that I would support. It's a variant that has received little if any public attention. It's a ring of  power satellites in sun synchronous orbit. Any site on Earth has visibility to one or more of the power satellites during a pair of one to two hour windows around local sunrise and sunset. Light pollution isn't an issue, because the satellites circle around the terminator. So it's still dusk when the ring of satellites drops below the western horizon, and the pre-dawn sky is already lightening before the ring rises above the eastern horizon. RF noise from the far field of the power beams is less of a problem for two reasons. One is that it's limited to the windows around sunrise and sunset when the satellites are visible. The other is that power is beamed over comparatively short distances never greater than about 1000 km, allowing the beams to be tight. Any given location is never exposed to the fringe field from more than a couple of passing satellites. Geosynchronous satellites, by contrast, bathe the entire earth in the far field noise from hundreds of high power satellites.

An obvious question with this form of SSP is "what about all the time that the ring of satellites is below the horizon for a given ground station"? The answer is simple: energy storage and conventional energy resources. This form of SSP is supplementary to other power sources, not a total replacement. In the worst case, 11 hours worth of energy storage is not that hard to support. This type of SSP fills in during extended periods of adverse weather -- what the Germans call "dunkelflaute". Both the satellites and the ground stations rely heavily on advanced energy storage.

The biggest challenge is probably the need for agile phased-array rectennas. The beam source is moving and must be tracked. And roughly every four minutes, the rectenna must flick to a new satellite in the ring. It's technically feasible; the harder question is whether it can be made conomically feasible.

 

- Roger

On Sat, Feb 1, 2025 at 8:57 AM <jkst...@sbcglobal.net> wrote:

[a.p.kothari astrox.com]

Yes. The easiest, cheapest (educated guess) and quickest (most likely) solution to energy needs and for net-zero (no CO2, no chemical pollutants, 1-5% of waste as current PWRs) indeed can be some of the Advanced fission reactors (Gen IV).

 

 

-------------------------------------------------------

Dr. Ajay P. Kothari

President

Astrox Corporation

 AIAA Associate Fellow

Member, AIAA Aerospace Power TC

 

Ph: 301-935-5868

Web:  www.astrox.com

Email: a.p.k...@astrox.com

-------------------------------------------------------

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAN%3D9Pgnhm0%3DFxTj1mfQdJw%3DV1%3Dy6PG4P0e4_gK12CJKASgRboQ%40mail.gmail.com.

[jkst...@sbcglobal.net]

That is why they do not use batteries.
Imagine the battery pack that can store say 5 GWH every day and discharge it as 4 GWH or so every night.
$$$$$$$$$$$$$$$$

John

-----Original Message-----
From: power-satell...@googlegroups.com <power-satell...@googlegroups.com> On Behalf Of Keith Henson

Sent: Saturday, February 1, 2025 2:33 PM
To: jkst...@sbcglobal.net

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB6WMzRfnp_095rg4-3%2BhhwLV9f%3DZ-xb2MuWqD3vS1YPaw%40mail.gmail.com.

[Keith Henson]

On Sat, Feb 1, 2025 at 12:48 PM Roger Arnold <silver...@gmail.com> wrote:
>
snip

>
> There's also a variant of satellite solar power that I would support. It's a variant that has received little if any public attention. It's a ring of power satellites in sun synchronous orbit.

I have not worked the numbers so don't take this seriously until
someone has done so. But by similarity to the problem I found with
moving power satellites from LEO to GEO my guess is that one would
take a hit per day to a hit per month from space junk. A power
satellite can't dodge and they are unavoidably big.

I think I can find the Excel sheet if someone wants to modify it.

Keith

> To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CAN%3D9Pgnhm0%3DFxTj1mfQdJw%3DV1%3Dy6PG4P0e4_gK12CJKASgRboQ%40mail.gmail.com.

[Geoffrey Landis]

> >Now, you can say "but I have a design (or, a concept for a design) that will be twice that efficient." Right. Show me your experience in thermal and optical system designs sufficient to convince me that you can make something twice as efficient as the best performance ever achieved by experts in the field.

>My experience in space radiator design goes back to a paper Eric
>Drexler and I wrote in 1979.

I read that paper. It was not an engineering design. No engineering drawings, no list of materials, not even a listing of basic fundamentals such as what the working fluid is and what the pressures were. And, more critically, did not include any analysis of losses through the efficiency chain. 

Yes, if you ignore all the losses in the system, efficiencies get pretty good. As the saying goes, “In theory, there is no difference between theory and practice. But in practice, there is.”

>But generally, I don't think there are

>any experts in GW scale radiators.

Saying "I don't think anybody else has expertise either" is not the same as having expertise.

[jkst...@sbcglobal.net]

The proof is in the (eating of the) pudding but that would be a very expensive pudding and hard to eat also!

 

John S

 

From: power-satell...@googlegroups.com <power-satell...@googlegroups.com> On Behalf Of Geoffrey Landis
Sent: Saturday, February 1, 2025 8:31 PM
To: Power Satellite Economics <power-satell...@googlegroups.com>
Subject: Re: Thermal [Re: Power satellite problems]

 

> >Now, you can say "but I have a design (or, a concept for a design) that will be twice that efficient." Right. Show me your experience in thermal and optical system designs sufficient to convince me that you can make something twice as efficient as the best performance ever achieved by experts in the field.

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/dd31d68f-1e34-40a3-a5a7-843a1180add2n%40googlegroups.com.

[Keith Henson]

On Sat, Feb 1, 2025 at 6:30 PM Geoffrey Landis
<geoffre...@gmail.com> wrote:
>
snip

>
> I read that paper. It was not an engineering design. No engineering drawings, no list of materials,

It was a concept paper. The heat reflector drawings between the tubes
are still valid. You can see them in the beamed energy bootstrapping
animation.

> not even a listing of basic fundamentals such as what the working fluid is and what the pressures were.

The title "Gas Entrained Solids-A Heat Transfer Fluid For Use In
Space"·gives away that the working fluid was rock dust and
low-pressure gas, just enough to carry the dust. If anyone wants to
read this ancient (choke!) document:

https://nss.org/wp-content/uploads/2017/07/L5-News-1979-07.pdf

https://nss.org/wp-content/uploads/2017/07/L5-News-1979-08.pdf

If you go to the sum up page 20 of the 1979 SMF conference papers, ". .
.Henson makes an impressive case for this design approach." (Credit
to Drexler, since he did much of the work.)

> And, more critically, did not include any analysis of losses through the efficiency chain.

The paper was mainly about one piece, a radiator, not a system such as
a power satellite though there was discussion about the effects of
lower temperature heat sinks on power systems. The steady state
"efficiency" of a radiator measured as heat in vs heat out is 100% by
definition.

> Yes, if you ignore all the losses in the system, efficiencies get pretty good. As the saying goes, “In theory, there is no difference between theory and practice. But in practice, there is.”
>
> >But generally, I don't think there are
> >any experts in GW scale radiators.
>
> Saying "I don't think anybody else has expertise either" is not the same as having expertise.

If you happen to know anyone who has designed or even talked about
low-temperature GW scale radiators, I would love to talk to them. I
am well aware that nobody has built anything of this scope in space
and it is possible nobody ever will. (You can't say much about the
post-singularity world.)

Keith

> To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/dd31d68f-1e34-40a3-a5a7-843a1180add2n%40googlegroups.com.

[Geoffrey Landis]

>The title "Gas Entrained Solids-A Heat Transfer Fluid For Use In

>Space"·gives away that the working fluid was rock dust and
>low-pressure gas, just enough to carry the dust. If anyone wants to
>read this ancient (choke!) document:
>
>https://nss.org/wp-content/uploads/2017/07/L5-News-1979-07.pdf
>https://nss.org/wp-content/uploads/2017/07/L5-News-1979-08.pdf

Ah, different paper. The one I was thinking of used water vapor which condensed to transfer heat.

> >And, more critically, did not include any analysis of losses through the efficiency chain.

>The paper was mainly about one piece, a radiator, not a system 

This one, as you say, is a concept for heat rejection, and not a design for a thermal conversion system at all.  I don't see any information here suggesting an approach to making a concentrating solar thermal conversion system with performance better than the world's record conversion efficiency.

[Keith Lofstrom]

On Thu, Jan 30, 2025 at 9:42 PM <jkst...@sbcglobal.net> wrote:
> Yes but a thermal system has MOVING PARTS.

On Thu, Jan 30, 2025 at 10:37:29PM -0800, Keith Henson wrote:
> I have a refrigerator that is full of moving parts. It is 25 years
> old and has never failed.
>
> It is not unusual for steam turbines to last 50 years.
>
> Keith

Just scrolled down to this two-week-old email.

My 15yo kitchen refrigerator is somewhat smaller than a
thermal SSPS, which will be VASTLY "fuller" of moving parts.

My refrigerator sits in a 20C room. Not 1366W/m² sunlight,
black body temperature hotter than 120C, nor the occasional
spring/fall Earth shadow, quickly plunging towards cryogenic
temperatures if the structure has minimal thermal mass.

A terrestrial steam turbine can slowly leak steam over the
years; there's plenty more water where that came from.
Lost coolant from an SSPS must be replenished from Earth.
The Earth is far less delta V than any other source of
coolant in the solar system; leave that off the table.

An orbital steam system with square kilometers of
surface may have a zillion times more pinholes in
(presumably) some very thin weight-limited surfaces.
I don't have much of a micrometeoroid problem with
my refrigerator.

My refrigerator doesn't have a flywheel or a motorized
counterweight to counteract Earth gravity gradient and
keep it perpendicular to the Sun. It won't get torqued
to shreds if the orientation pivot seizes. Note that
MOST photovoltaic SPSS designs will have the same
orientation hazard, but they won't have a large mass of
working fluid adding to the emergency control problem.

All that said ... I have FIVE refrigerators: automobile
aircon, kitchen, freezer, office snacks, and also a
medical supplies fridge ... plus 3 failed "office cube"
refrigerators which I haven't recycled yet. I've
repaired them, but I don't have the proper tools for
changing coolant. If I extrapolate their surface area
and failure rate to a city-sized SSPS, we will need a
city of repair robots to keep it running.

I will gladly donate my three cube refrigerators to the
team developing thermal SSPS repair robots. Refrigerator
repair robots might be a lucrative product here on Earth,
if sidelined appliance repair shop owners don't sabotage
the robot factory. If the repair robots NEVER leak
coolant during terrestrial repairs, that will keep a lot
of super-greenhouse gases out of the Earth's atmosphere.

Keith L.

P.S. The only PV I've deployed is a square foot, which
keeps my riding mower battery charged. I'd love to get
rid of the mower and replace TOO MUCH GRASS with a
quarter acre of PV, but huge trees block direct sunlight
to my 45N latitude lawn area. True Oregonians compost
politicians before they get rid of trees.

--
Keith Lofstrom kei...@keithl.com

[Bryan Zetlen]

American corporate trade with China has eliminated the concepts of durability, reliability and warranty protection for consumers. 

--
You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/20250214043947.GA31057%40gate.kl-ic.com.

[Keith Henson]

On Thu, Feb 13, 2025 at 8:39 PM Keith Lofstrom <kei...@keithl.com> wrote:
>

> On Thu, Jan 30, 2025 at 9:42 PM <jkst...@sbcglobal.net> wrote:
> > Yes but a thermal system has MOVING PARTS.
>
> On Thu, Jan 30, 2025 at 10:37:29PM -0800, Keith Henson wrote:
> > I have a refrigerator that is full of moving parts. It is 25 years
> > old and has never failed.
> >
> > It is not unusual for steam turbines to last 50 years.
> >
> > Keith
>
> Just scrolled down to this two-week-old email.
>
> My 15yo kitchen refrigerator is somewhat smaller than a
> thermal SSPS, which will be VASTLY "fuller" of moving parts.
>

Let me preface this with the statement that I don't think power
satellites will be constructed on this side of the singularity. It is
not that they can't be, but there just isn't time. Starting today it
would take at least 20 years, and that most likely put it beyond the
singularity. On the other side, who knows?

> My refrigerator sits in a 20C room. Not 1366W/m² sunlight,
> black body temperature hotter than 120C, nor the occasional
> spring/fall Earth shadow, quickly plunging towards cryogenic
> temperatures if the structure has minimal thermal mass.

You put your finger on an unsolved problem. Most of the structure
area is light concentration reflectors. No particular reason to think
they would be hurt by a cryogenic plunge. But the radiators are full
of low-pressure steam. They will freeze out. Again, that might not
hurt them, but I don't know how to restart them. Ideas welcome.

> A terrestrial steam turbine can slowly leak steam over the
> years; there's plenty more water where that came from.
> Lost coolant from an SSPS must be replenished from Earth.
> The Earth is far less delta V than any other source of
> coolant in the solar system; leave that off the table.

It is not conceptually difficult to capture turbine seal leakage, the
harder problem is the radiator tubes. See
https://www.youtube.com/watch?v=VEkZkINrJaA Beamed energy
bootstrapping

> An orbital steam system with square kilometers of
> surface may have a zillion times more pinholes in
> (presumably) some very thin weight-limited surfaces.
> I don't have much of a micrometeoroid problem with
> my refrigerator.
>
> My refrigerator doesn't have a flywheel or a motorized
> counterweight to counteract Earth gravity gradient and
> keep it perpendicular to the Sun. It won't get torqued
> to shreds if the orientation pivot seizes. Note that
> MOST photovoltaic SPSS designs will have the same
> orientation hazard, but they won't have a large mass of
> working fluid adding to the emergency control problem.

Something has to twist. If the light concentrators' bearings freeze
up, it will not hurt anything if they are relatively light. Steam
radiators are relatively light, a little over a kg per kW, but that
said, they are enormous.

The big and unsolved problem for PV is transmitting a lot of power to
the transmitter. They have arcover problems with comm sats at the kW
level.

snip

I have a similar problem with a completely mundane project to make
trash and intermittent PV (at GW scale) into syngas and from there
into diesel or jet fuel. It might be done in a panic if people get
concerned about their brains getting full of plastic fragments. No
idea what damage it is doing, but 7 gm of plastic in your brain can't
be doing any good.

KeithH

> --
> Keith Lofstrom kei...@keithl.com

[Paul Werbos]

Refrigerators are important -- but I have gotten a lot deeper into longevity issues with mammals and cars. More mammals than cars since I retired, and have less contact with the auto industry.

It is fascinating how extreme the differences are in longevity in different species, and what they connect to. For SOME of us, 

the longevity issues cannot be ignored. I remember calculating: if use of 5% of my free time and energy might double the amount of 

such time left, it should be included in the portfolio. AND THEN, when Biden and Trump both demonstrated just how extreme the costs of age related cognitive decline (ARCD) could be, I looked a bit further. But here, as with low-cost access to space, science DOES give us powerful

new options, while politics and vested interests mostly prevent it.

[Claudio Cioffi]

I agree, Paul. And the mathematical theory of systems reliability offers a powerful albeit mostly unappreciated scientific perspective on all kinds of aging processes. A truly interdisciplinary field since it has applications across numerous disciplines, including the natural, the engineering, and the social sciences.

–––
Claudio Cioffi-Revilla, Ph.D., D.Sc.Pol.
University Distinguished Professor Emeritus, George Mason University
Jefferson Science Fellow of the U.S.A. National Academy of Sciences
Elected Fellow of the  American Association for the Advancement of Science AAAS
        


Research Gate: https://www.researchgate.net/profile/Claudio_Cioffi
Academic page: https://socialcomplexity.gmu.edu/faculty/csc-faculty-dr-cioffi/
National Academies page:   http://sites.nationalacademies.org/PGA/Jefferson/PGA_046486
ORCID https://orcid.org/0000-0001-6445-9433

SSRN https://papers.ssrn.com/sol3/results.cfm

On Feb 14, 2025, at 7:32 AM, Paul Werbos <paul....@gmail.com> wrote:

External Message: Use Caution

To view this discussion visit https://groups.google.com/d/msgid/power-satellite-economics/CACLqmgdzGzQGptZXU7TzmXJbNV1vQ4W4tuGYo5G-hcfTXG3gpA%40mail.gmail.com.