Caltech to launch space solar power technology demo into orbit in January

11 views
Skip to first unread message

k.a.c...@sympatico.ca

unread,
Jan 4, 2023, 7:16:16 AM1/4/23
to Power Satellite Economics
Caltech's SSPD experiment is now on-orbit, after yesterday's successful
launch of Momentus' Vigoride-5 satellite, which is carrying SSPD as a hosted
payload:

https://techxplore.com/news/2023-01-caltech-space-solar-power-technology.htm
l

- Kieran

Keith Lofstrom

unread,
Jan 4, 2023, 3:49:10 PM1/4/23
to Power Satellite Economics

vid.b...@fotonika-lv.eu

unread,
Jan 8, 2023, 3:50:59 AM1/8/23
to Keith Lofstrom, Power Satellite Economics
Brian Wang has an article - https://www.nextbigfuture.com/2023/01/caltech-space-solar-power-demo-launched-to-orbit-january-3rd.html#more-178949

-----Ziņojuma oriģināls-----
No: power-satell...@googlegroups.com <power-satell...@googlegroups.com> Kā vārdā Keith Lofstrom
Nosūtīts: Wednesday, January 4, 2023 10:49 PM
Kam: 'Power Satellite Economics' <power-satell...@googlegroups.com>
Tēma: Re: Caltech SSP demo; people and papers
--
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 on the web visit https://groups.google.com/d/msgid/power-satellite-economics/20230104204848.GB10735%40gate.kl-ic.com.

Keith Henson

unread,
Jan 8, 2023, 9:11:18 AM1/8/23
to vid.b...@fotonika-lv.eu, Keith Lofstrom, Power Satellite Economics
From what I know of the CalTech work, it is almost all nonsense. But
I am not concerned about the waste of resources, they are not that
large, and either the CalTech work will go nowhere or they will have
to take account of such matters as spares array, orbital mechanics,
geometry, and station keeping.

Physics rules.

Keith
> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/031b01d9233e%24544bd260%24fce37720%24%40fotonika-lv.eu.

k.a.c...@sympatico.ca

unread,
Jan 8, 2023, 12:53:22 PM1/8/23
to Power Satellite Economics
KeithH wrote:

> From what I know of the CalTech work, it is almost all nonsense. But I am not
> concerned about the waste of resources, they are not that large, and either the
> CalTech work will go nowhere or they will have to take account of such matters
> as spares array, orbital mechanics, geometry, and station keeping.
>
> Physics rules.

Once it became clear after 1980 that the US government wasn't going to provide any serious money for SPS development, the big question for the enthusiasts at the time (and those who've followed in their footsteps) has been, what can usefully be done to move the ball down the field, with what small amount of money that *can* be found? The answer has been "not a lot," other than some quite-nice small demos every few years that have served to nibble away at a few of the technical details (e.g., Nobu kaya demonstrating retro-directive power beaming on a very small scale), and keeping the torch alight.

In that context, I see the CalTech work's main significance as being that they managed to snag enough funding to pay for some flight-demo work --- good for them! That's mor than any of the rest of us have managed (the notable exception being of course Paul Jaffe at NRL).

The next-level question is, what can usefully be demonstrated in space, at a scale smaller than full-scale (which would optimistically cost $20B or more)? Several of us have our own ideas about that; my own are based on the idea that the most significant things that really need demo-ing, really need a full-scale (1 km or more in diameter) transmit aperture in GEO; the huge difficulty being the amount of funding that'd be needed to pay for that to be designed, built, flown and operated.

The CalTech guys are postulating that breaking the transmit aperture into little bits, each on its own satellite, with each of those satellites flying in formation with the others, can be made to work. I can't see it myself (yet), based on what I know about phased-array antennas and about orbital dynamics, and about satellite formation-flying control --- but maybe there's something about their concept that they haven't described in public yet, that would actually make it workable. Let's assume so just for a moment --- in that case, their concept has the virtue of having *something* useful to demo at a scale smaller than full-scale. In which case the current flight test would indeed be moving the ball down the field usefully.

Of course, maybe they haven't thought it through enough, and the basic concept is unworkable (as KeithH and I fear, due to sparse apertures, sidelobes, etc.) --- in which case their current flight demo is (from a technical perspective) a dead end. In which case it might still serve the purpose of bringing SPS to the attention of a wider audience, which *might* be parlayed into ratcheting up financial support for technical concepts that will actually work. Of course it might have the opposite effect, of 'proving the critics right"...

- Kieran



Tim Cash

unread,
Jan 8, 2023, 1:05:27 PM1/8/23
to k.a.c...@sympatico.ca, Power Satellite Economics
I do not think that a failure on their part would necessarily be a bad thing.
Nicola Tesla failed to establish a viable power beaming business, and died penniless, but it only set the field back a half century.
It has been a half century plus since the last moon landing, so yes, we have slowed down, but all of the enthusiasm of the new generation of entrepreneurs is starting to pay off.
I am against the establishment of a central establishment for space solar power, that definitely would slow things down.
The same approach as commercial space should be taken with space solar power, with perhaps additional X-Prize contests as incentives.
I am seeing recent growth in this sector, and am enthusiastic about the future of our industry.

Tim Cash

--
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.


--

Keith Henson

unread,
Jan 8, 2023, 2:12:12 PM1/8/23
to k.a.c...@sympatico.ca, Power Satellite Economics
Thanks for backing me up.

The main reason power satellites are attractive over ground solar is
that they provide baseload power.

The CalTech design has the PV on the side toward the sun and microwave
transmitters on the other side. They can provide power only when the
microwave side presents some area to the earth. Which is to say they
are intermittent like ground solar, at best providing power half the
day.

The whole thing looks like a case of premature optimization for low
mass at a cost factor of two for power half the time and another
factor of two for the checkerboard transmitter. Plus there is no
provision I know about for station keeping.

I have nothing against sandwich designs. Thermal considerations make
the combined PV and transmitter larger than the classic one km
transmitter, but that allows a top hat power beam distribution and a
more area-efficient rectenna. But you have to twist the sunlight once
a day for baseload power. That either takes big reflectors or a lot
of smaller ones as proposed for Mankin's Alpha.

Keith
> --
> 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 on the web visit https://groups.google.com/d/msgid/power-satellite-economics/30a301d9238a%2417fbd7a0%2447f386e0%24%40sympatico.ca.

Al Globus

unread,
Jan 8, 2023, 3:27:41 PM1/8/23
to Keith Henson, <k.a.carroll@sympatico.ca>, Power Satellite Economics


On Jan 8, 2023, at 11:11 AM, Keith Henson <hkeith...@gmail.com> wrote:

The CalTech design has the PV on the side toward the sun and microwave
transmitters on the other side.  They can provide power only when the
microwave side presents some area to the earth.  Which is to say they
are intermittent like ground solar, at best providing power half the
day.

Unlike ground solar and wind, however, the timing and extent of SSP deliveries to Earth can be easily calculated very reliably.

Roger Arnold

unread,
Jan 8, 2023, 8:24:46 PM1/8/23
to k.a.c...@sympatico.ca, Power Satellite Economics
I haven't read the Cal Tech proposal, so I can't comment on it directly. But I can comment on some of the comments to it. Judging from the way the proposal has been characterized here, I don't think it's as technically unsound as Keith and (with reservations) Kieran suppose.

First point is that, yes, it is entirely feasible to have a 3-D swarm of small satellites co-orbiting in a stable configuration. Note "stable", not "static". The swarm will rotate on an axis through its center, tangent to its orbit, at a rate of one revolution per orbit. Keith L got into this in connection with his "Server Sky" concept. The orbit for the swarm as a whole will be perturbed by solar radiation pressure, and will require station keeping of some sort to maintain near-circularity at GEO. I believe that the required station keeping can be effected the way Keith envisioned for Server Sky, by angling small mirrors on the periphery of the disks of individual microsats within the swarm. However I can't say that with any certainty. It's a good research topic.

Second, there's no fundamental reason that solar power microsats have to have PV cells on one side and microwave antennas on the other. Keith H's concern about the satellites only being able to transmit power in that portion of the orbit when the PV side is exposed to sunlight and the power antennas are facing Earth is, I believe, unfounded. The power antenna elements are simple electrical dipoles. They can have negligible optical cross sections. They can sit in front of the PV surface without shading it to any significant extent. The optimal design would probably have dipole elements offset from the surface on both sides of the microsat unit. The PV elements would be designed to be transparent to microwaves at the power frequency. The dipole antennas on both sides would be active simultaneously. Those on one side would radiate equally earthward and anti-earthward; those on the other would do likewise, but they'd be phased to reinforce the earthward beam and nullify the anti-earthward beam. 

Third, the "sparse aperture" issue. It's a real issue, in that a sparse aperture will end up putting unacceptable levels of power into sidelobes. Excessive sidelobe radiation is a problem for environmental and health reasons, but even if it weren't, the loss of power from the main beam would render it useless for practical power transmission. That said, what constitutes a sparse vs a filled aperture is more complicated than most of us realize. Our thinking is mostly shaped by the common 2-D case of phased radar arrays. The 3-D case is different.

In the case of a 3-D array of phased elements, we need to consider the total number of elements and the density of elements in projection. To operate as a filled aperture, the average number of dipole elements in projected beam direction must be greater than one element per square λ (where λ is the wavelength of the beam). Or maybe it's per square half-λ. Don't remember for sure. But the more the better. The distribution should be approximately uniform, but it's actually better if the locations of radiating elements are semi-random.

There's a fourth point that I'll defer for later. It has to do with the ability of a 3D holographic array to support multiple beams of varying strength in parallel. IMO, it makes for a better and more realistic business model. Allows beamed power to be used as high value backup for variable renewables.

On Sun, Jan 8, 2023 at 9:53 AM <k.a.c...@sympatico.ca> wrote:
--
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.

Keith Henson

unread,
Jan 9, 2023, 12:11:21 AM1/9/23
to Roger Arnold, k.a.c...@sympatico.ca, Power Satellite Economics
On Sun, Jan 8, 2023 at 5:24 PM Roger Arnold <silver...@gmail.com> wrote:
>
> I haven't read the Cal Tech proposal, so I can't comment on it directly. But I can comment on some of the comments to it. Judging from the way the proposal has been characterized here, I don't think it's as technically unsound as Keith and (with reservations) Kieran suppose.
>
> First point is that, yes, it is entirely feasible to have a 3-D swarm of small satellites co-orbiting in a stable configuration. Note "stable", not "static". The swarm will rotate on an axis through its center, tangent to its orbit, at a rate of one revolution per orbit. Keith L got into this in connection with his "Server Sky" concept. The orbit for the swarm as a whole will be perturbed by solar radiation pressure, and will require station keeping of some sort to maintain near-circularity at GEO. I believe that the required station keeping can be effected the way KeithL envisioned for Server Sky, by angling small mirrors on the periphery of the disks of individual microsats within the swarm. However I can't say that with any certainty. It's a good research topic.

I think it is like trying to sail upwind without a keel.

> Second, there's no fundamental reason that solar power microsats have to have PV cells on one side and microwave antennas on the other. Keith H's concern about the satellites only being able to transmit power in that portion of the orbit when the PV side is exposed to sunlight and the power antennas are facing Earth is, I believe, unfounded.

That came from the CalTech people at some point.

> The power antenna elements are simple electrical dipoles. They can have negligible optical cross sections. They can sit in front of the PV surface without shading it to any significant extent. The optimal design would probably have dipole elements offset from the surface on both sides of the microsat unit. The PV elements would be designed to be transparent to microwaves at the power frequency.

I don't think you can make PV without conductive layers that would
block or absorb microwaves. If anyone has an idea, it might also be
useful for Ian Cash's design.

> The dipole antennas on both sides would be active simultaneously. Those on one side would radiate equally earthward and anti-earthward; those on the other would do likewise, but they'd be phased to reinforce the earthward beam and nullify the anti-earthward beam.

Assuming microwaves go through PV.

> Third, the "sparse aperture" issue. It's a real issue, in that a sparse aperture will end up putting unacceptable levels of power into sidelobes. Excessive sidelobe radiation is a problem for environmental and health reasons, but even if it weren't, the loss of power from the main beam would render it useless for practical power transmission. That said, what constitutes a sparse vs a filled aperture is more complicated than most of us realize. Our thinking is mostly shaped by the common 2-D case of phased radar arrays. The 3-D case is different.

Reciprocity is fundamental to antennas. Their original checkerboard
design would lose half the power, just as if you knocked out every
other 100 meter square on the rectenna.

> In the case of a 3-D array of phased elements, we need to consider the total number of elements and the density of elements in projection. To operate as a filled aperture, the average number of dipole elements in projected beam direction must be greater than one element per square λ (where λ is the wavelength of the beam). Or maybe it's per square half-λ. Don't remember for sure. But the more the better. The distribution should be approximately uniform, but it's actually better if the locations of radiating elements are semi-random.
>
> There's a fourth point that I'll defer for later. It has to do with the ability of a 3D holographic array to support multiple beams of varying strength in parallel. IMO, it makes for a better and more realistic business model. Allows beamed power to be used as high value backup for variable renewables.

You can't split it too many ways before the cost of the rectennas
doubles the cost of power.

KeithH

> On Sun, Jan 8, 2023 at 9:53 AM <k.a.c...@sympatico.ca> wrote:
>>
>> KeithH wrote:
>>
>> > From what I know of the CalTech work, it is almost all nonsense. But I am not
>> > concerned about the waste of resources, they are not that large, and either the
>> > CalTech work will go nowhere or they will have to take account of such matters
>> > as spares array, orbital mechanics, geometry, and station keeping.
>> >
>> > Physics rules.
>>
>> Once it became clear after 1980 that the US government wasn't going to provide any serious money for SPS development, the big question for the enthusiasts at the time (and those who've followed in their footsteps) has been, what can usefully be done to move the ball down the field, with what small amount of money that *can* be found? The answer has been "not a lot," other than some quite-nice small demos every few years that have served to nibble away at a few of the technical details (e.g., Nobu kaya demonstrating retro-directive power beaming on a very small scale), and keeping the torch alight.
>>
>> In that context, I see the CalTech work's main significance as being that they managed to snag enough funding to pay for some flight-demo work --- good for them! That's mor than any of the rest of us have managed (the notable exception being of course Paul Jaffe at NRL).
>>
>> The next-level question is, what can usefully be demonstrated in space, at a scale smaller than full-scale (which would optimistically cost $20B or more)? Several of us have our own ideas about that; my own are based on the idea that the most significant things that really need demo-ing, really need a full-scale (1 km or more in diameter) transmit aperture in GEO; the huge difficulty being the amount of funding that'd be needed to pay for that to be designed, built, flown and operated.
>>
>> The CalTech guys are postulating that breaking the transmit aperture into little bits, each on its own satellite, with each of those satellites flying in formation with the others, can be made to work. I can't see it myself (yet), based on what I know about phased-array antennas and about orbital dynamics, and about satellite formation-flying control --- but maybe there's something about their concept that they haven't described in public yet, that would actually make it workable. Let's assume so just for a moment --- in that case, their concept has the virtue of having *something* useful to demo at a scale smaller than full-scale. In which case the current flight test would indeed be moving the ball down the field usefully.
>>
>> Of course, maybe they haven't thought it through enough, and the basic concept is unworkable (as KeithH and I fear, due to sparse apertures, sidelobes, etc.) --- in which case their current flight demo is (from a technical perspective) a dead end. In which case it might still serve the purpose of bringing SPS to the attention of a wider audience, which *might* be parlayed into ratcheting up financial support for technical concepts that will actually work. Of course it might have the opposite effect, of 'proving the critics right"...
>>
>> - Kieran
>>
>>
>>
>> --
>> 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 on the web visit https://groups.google.com/d/msgid/power-satellite-economics/30a301d9238a%2417fbd7a0%2447f386e0%24%40sympatico.ca.
>
> --
> 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 on the web visit https://groups.google.com/d/msgid/power-satellite-economics/CAN%3D9Pg%3DzR4ps0y1A69sQ15QVPhqN1wvcNQL%2BPdZBBN0UKk7J3A%40mail.gmail.com.

Roger Arnold

unread,
Jan 9, 2023, 4:10:07 AM1/9/23
to Keith Henson, k.a.c...@sympatico.ca, Power Satellite Economics
>> <..> I believe that the required station keeping can be effected the way KeithL envisioned for Server Sky, by angling small mirrors on the periphery of the disks of individual microsats within the swarm. However I can't say that with any certainty. It's a good research topic.


> I think it is like trying to sail upwind without a keel.

Probably not a good analogy. The swarm isn't sailing upwind. It's an oscillating system, oscillation period being the orbital period around the earth. Stationkeeping task is to counter elongation of orbit caused by solar radiation pressure. Mirrored peripheral flaps can create small drive forces normal to the axis of sunlight, and they can modulate the radiation pressure in the direction of sunlight. That seems like it should be enough to maintain stability, as long as the ratio of radiation pressure to satellite mass is below some threshold level. As I said, a good research topic.

>> The PV elements would be designed to be transparent to microwaves at the power frequency.

> I don't think you can make PV without conductive layers that would block or absorb microwaves.  If anyone has an idea, it might also be useful for Ian Cash's design.

The microwaves from the power elements are polarized. As long as the conductive lines run perpendicular to the polarization axis, they don't interfere. Lines running parallel to the polarization axis would interfere, but I think they can be tuned to be transparent. Or located far enough from the dipole elements to provide a transparent window for emitted microwaves.  It's an interesting design challenge.

>> That said, what constitutes a sparse vs a filled aperture is more complicated than most of us realize. Our thinking is mostly shaped by the common 2-D case of phased radar arrays. The 3-D case is different.

> Reciprocity is fundamental to antennas.  Their original checkerboard design would lose half the power, just as if you knocked out every other 100 meter square on the rectenna.

Reciprocity is indeed fundamental, but I'm not sure what point you're making. By "their original checkerboard design" I presume you're referring to the CalTech proposal. I suppose I should take the time to read it. I just find it hard to credit the notion that CalTech EE professors or graduate students working under them would be ignorant of antenna theory.

> You can't split it [the power beam] too many ways before the cost of the rectennas doubles the cost of power.

Hmm, a power satellite designed for delivery of a large number of beams would have a large transmitter aperture. It would be able to deliver relatively narrow beams with a relatively flat beam profile to smallish rectennas widely distributed. True, the duty cycle of the rectennas would be low, because they'd only be used for backing other energy resources. That does increase the capital cost of the collection of rectennas relative to other system elements, but rectennas should still be cheap. I'd expect that beaming variable power to distributed rectennas as needed would easily beat the cost of long distance transmission lines and energy storage capacity otherwise needed.

Tim Cash

unread,
Jan 9, 2023, 9:00:14 AM1/9/23
to Roger Arnold, Keith Henson, k.a.c...@sympatico.ca, Power Satellite Economics
Please, can anyone point me to a serious treatment of the "sparse array" or "sparse aperture" problem, mathematically or physics wise speaking?

I have done for the most part only 2D, and some 3D phased array mathematical modeling, based on the work by Arndt and Thompson, and just found a proper treatment of the Electromagnetic Compatibility of the Solar Power Satellite by K.A. Davis,
W.B. Grant, E.L. Morrison, and J.R. Juroshek, the references as follows:

Environmental Considerations for the Microwave Beam from a Solar Power Satellite
G.D. Arndt and L. Leopold
Johnson Space Center, Houston, Texas
https://ntrs.nasa.gov/citations/19790026017
https://www.semanticscholar.org/paper/Environmental-considerations-for-the-microwave-beam-Arndt-Leopold/6e563ec2fe10acd9e712e4af024c7555b8a746f7

and GD Arndt's first two references:

G.D. Arndt and L. Leopold
Microwave Transmission Characteristics of Solar Power Satellites
IEEE-MTT-S International Microwave Symposium, Ottawa, Canada, June 1978
https://ieeexplore.ieee.org/document/1123861

Solar Power Satellite - System Definition Study, Vol. IV, Part 2, NASA Contract NAS 9-15196, Boeing Aerospace Company, Seattle, Washington, Dec 1977
https://ntrs.nasa.gov/api/citations/19800012910/downloads/19800012910.pdf

DOESPS-ElectromagneticSystemsCompatibility.pdf
DOE/ER-0096
Dist. Category UC-97
Environmental Assessment for the Satellite Power System Concept Development and Evaluation Program - Electromagnetic Systems Compatibility
January 1981
K.A. Davis
Pacific Northwest Laboratory
Richland, WA 99353
and
W.B. Grant and E.L. Morrison, and J.R. Juroshek
Institute for Telecommunications Sciences
Boulder, CO 80303
Under Contract Number KD-03-8209
https://www.osti.gov/servlets/purl/6663190

The paper "Environmental Assessment for the Satellite Power System Concept Development and Evaluation Program - Electromagnetic Systems Compatibility" states:
"Electromagnetic compatibility was thus demonstrated to be an important factor in rectenna site selection. Specific recommended separation distances of from 50 to 150 km between rectennas and sensitive facilities such as nuclear reactors and radio astronomy observations have been provided for rectenna siting studies."

My own current work computes the fundamental and first twenty harmonics levels at various distances from the rectenna (1, 10, 100, 1,000 and 10,000 km) with the only significant Radio Frequency emissions being the first odd harmonic ~ 27 milliwatts.
See attached model of the Space Solar Power in Excel: I modeled the 1970's SSP and added equations for radio frequency interference at the fundamental and first twenty harmonics.
The equation does show the only substantial emissions are the odd harmonics, and only the first one of significance, the even harmonics are canceled out.
This means to me that what I thought was a 600 pound Gorilla in the room, is in fact only a pesky mosquito, however they do carry diseases, do they not?
I plan to write a book on space solar power for the Public, notice I did not say "Dummies" due to the topic being complex, non-trivial, and difficult to grasp for most of us.

I believe we can manage most technical problems, except for Physics, which has rules we must obey.
Perhaps there is a way we can "cheat" or treat sparse arrays to steal power from them?  Or just steer clear of them? 

The SSPIDR Program plans to get sunlight to the PV side of the tile while the other side remains earth facing, through some sort of contraption with mirrors and lenses?
I do not fully understand the optics, even so I believe they will get it to work after a fashion.
I agree with all of you, the simplest solution is not necessarily the lightest mass to orbit.  KISS principle, people.

Tim Cash



--
Powersat Theoretical Harmonic Content Amplitudes and Frequencies_rev_6.xlsx

Keith Henson

unread,
Jan 9, 2023, 11:11:02 AM1/9/23
to Tim Cash, Roger Arnold, k.a.c...@sympatico.ca, Power Satellite Economics
Geoffrey Landis <geoffre...@gmail.com> who is on this list wrote
the Wikipedia article on sparse arrays.

https://en.wikipedia.org/wiki/Thinned-array_curse

Keith

Keith Lofstrom

unread,
Jan 9, 2023, 4:58:42 PM1/9/23
to Roger Arnold, Keith Henson, k.a.c...@sympatico.ca, Power Satellite Economics
On Mon, Jan 09, 2023 at 01:09:53AM -0800, Roger Arnold wrote:
> >> <..> I believe that the required station keeping can be effected the way
> KeithL envisioned for Server Sky, by angling small mirrors on the periphery
> of the disks of individual microsats within the swarm. However I can't say
> that with any certainty. It's a good research topic.

Approximately correct. The corners of a 30 cm scale
"thinsat" are electrochromic mirrors, which
electrochemically change from black (1x light pressure
thrust) to reflective (2x light pressure thrust). By
changing the light pressure thrust on the corners, that
changes the thinsat light pressure thrust relative to a
constellation of thousands of other thinsats. This can
also tilt a thinsat, providing lateral thrust, though
slightly reducing radial thrust. A slow process (a turn
takes minutes, significant displacement takes hours), but
over days and weeks, this keeps the array of thinsats
together, and maintains the desired orbit. Over the
course of months, very slow accelation allows an array
to migrate to other orbits. Without expending propellant.

The purpose is not power to the ground, but computation
and communication, with heat dissipation directly to space.

As I write this, Google is expanding their enormous data
center in The Dalles, Oregon, converting almost a gigawatt
of electric power into computation and a few watts of
fiber optic photons, waste heat from the computation
warming the Columbia River. As worldwide computation
grows exponentially, power demand will increase
exponentially. That will all turn into waste heat, much
of it dumped into rivers, heating and evaporating them.

We can do wonderful things with computation, and many of
the megawatt++ applications (global atmospheric modelling,
molecular design, etc.) can tolerate minutes to hours of
communication delay. The computing chips themselves range
from milligrams down to micrograms (I've designed and
sold 0.1 microgram chips, my Hitachi partners called them
"smart dust"), so launching the chips to where power is
abundant and free, and the heat sink is the infinite
universe, makes sense for long-term giga-computation tasks.

http://server-sky.com

There are many subpages that are germane to SBSP.
The page about light-pressure-modified orbits is
particularly relevant to SBSP:

http://server-sky.com/LightOrbit

By balancing the orbit-precessing effects of the J₂ harmonic
of the oblate Earth's gravity field with the orbit-precessing
effects of light pressure, we can maintain stable elliptical
orbits for high-area low-mass satellites - like small server
sky thinsats, or giant SSPS. The key parameters to balance
are the kilograms per square meter of the satellite, and the
effects of the J₂ Earth-oblateness parameter on a slightly
elliptical orbit.

The take-home message is that a stable SSPS will also be in
a slightly elliptical orbit, with eccentricity increasing
as the mass-to-area ratio reduces. As viewed from the
rectenna, the eccentricity means that the SSPS will seem to
"orbit" in azimuth and distance over the course of a day.

As experienced by an object outside the rectenna fence,
that means the Airy pattern sidelobes from an SSPS will
also migrate over the course of a day; presuming that
there are dozens to hundreds of SSPS adding their sidelobe
contributions to any given spot, the power level and
apparent direction of the microwatt inteference will also
migrate around the sky. Note that most of the radio
devices we use today are designed for background noise
much smaller than microwatts per square meter. Learn
about "low noise amplifiers" (LNA) and "third order
intermodulation" (IM3).

Note, due to non-disclosure agreements and national
security issues, I cannot write about radiation damage
mitigations, beyond saying that pleasant surprises lurk,
which become unpleasant surprises in the wrong hands.
If I disappear after writing that, I've said too much.

James M. (Mike) Snead

unread,
Jan 9, 2023, 6:23:03 PM1/9/23
to Power Satellite Economics
Another new SSP company that I had not heard of before.

https://www.virtussolis.space/

Mike Snead
To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB7Q0SZRuaAoz9AkmPTO2-FYMB%2BiqSQUHpkMNLq2ujuYow%40mail.gmail.com.

Keith Henson

unread,
Jan 9, 2023, 10:26:56 PM1/9/23
to James M. (Mike) Snead, Power Satellite Economics
At least three of the people on their front page are members of this group.

Keith

On Mon, Jan 9, 2023 at 3:23 PM 'James M. (Mike) Snead' via Power
Satellite Economics <power-satell...@googlegroups.com>
wrote:
> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/01e901d92481%244d769b20%24e863d160%24%40aol.com.

Jay Lewis

unread,
Jan 10, 2023, 9:37:54 PM1/10/23
to Power Satellite Economics
On Sunday, January 8, 2023 at 7:11:18 AM UTC-7 hkeith...@gmail.com wrote:
From what I know of the CalTech work, it is almost all nonsense. 
OK its not just me.  My impression is they are mostly focused on the structure and assume nothing else is that hard?  
As a university student senior project its impressive. I imagine they started with something useful and looking at the budget and scheduled paired it back to the least viable system to claim its space based power.

 

Keith Lofstrom

unread,
Jan 11, 2023, 1:13:18 AM1/11/23
to Power Satellite Economics
On Sun, Jan 8, 2023 at 5:24 PM Roger Arnold <silver...@gmail.com> wrote:

> First point is that, yes, it is entirely feasible to have
> a 3-D swarm of small satellites co-orbiting in a stable
> configuration. Note "stable", not "static". The swarm
> will rotate on an axis through its center, tangent to
> its orbit, at a rate of one revolution per orbit.
>
> Keith L got into this in connection with his "Server Sky"
> concept. The orbit for the swarm as a whole will be
> perturbed by solar radiation pressure, and will require
> station keeping of some sort to maintain near-circularity
> at GEO.

Roger's explanation is pretty good, some responses to his
explanation not so much.

"Station-keeping for the swarm as a whole" ... that won't
happen for "solar sail-ish satellites", but a relaxed
version of station keeping, placing the swarm in a
slightly elliptical orbit with perigee towards the Sun,
exploits the oblate Earth's J2 gravitational harmonic
to rotate a geo-sync orbit's semimajor axis yearly.
Geosynchronous, but not quite geostationary.

See: http://server-sky.com/LightOrbit

The required perturbation grows as the "light pressure
to gravitational force" ratio grows, and can approach an
eccentricity of 0.04 if the satellite mass to Sun-facing
area ratio is as gossamer as 100 micrometers of aluminum
( 270g/m^2, typical for a postulated server sky thinsat,
far less than a comsat or postulated SSPS ).

Gravity field harmonics and orbital eccentricity are the
"keel" for this sorta-solar-sailing maneuver.

Oscillating "in and out" within the GEO plane is not
neighborly, so the elliptical radial "oscillation" should
also include some out-of-plane inclination, tilting the
orbit on the sun axis, with light pressure rotating the
inclination "vector" yearly as well. Apogee and perigee
will be in the equatorial plane but not at GEO radius.

This should be just fine for a rectenna with a wide
acceptance angle, not so much for a fixed azimuth
satellite dish.

Transmitting power beams from above or below the GEO plane
(except at noon and midnight in my scheme) will keep most
of SSPS sidelobe and harmonic power off-axis from GEO
comsat aimed-satellite dishes.

Noon and midnight? Lunch time and sleep time. SSPS
sidelobes will saturate a satellite dish LNA (front end
low noise amplifier), but they probably won't roast it.

There may be a clever way to "juggle the feeds" so that
SSPS and rectennas, comsats and satellite dishes don't
simultaneously share the same axes. Too clever for me.

Note, I stole most of "my" math from Soop's "Handbook of
Geostationary Orbits" $90 for the Springer paperback.
A good university library may have it on the shelves or
online. My unnatural acts with that math are my own
kink, and may be illegal in some southern states. :-)

Keith L.

P.S. Roger - I'd love to see the math and the simulations
for your spaceport orbital balancing act. Note that for
a high apogee launch, the spaceport capture path can be
approximately vertical, and a "pendulum" or "vertical
snake" spaceport reduces the need for bending moment on
the spaceport. That said, orbital velocity is less than
10mm/μs, plenty of time to compute which way to jig and jog
the capture path, then restore to vertical (or oscillating,
or whatever) after the rat moves up the snake.

See http://launchloop.com/CaptureRail

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

Keith Henson

unread,
Jan 11, 2023, 11:22:26 AM1/11/23
to Keith Lofstrom, Power Satellite Economics
I don't follow the physics but if you are right and GEO satellites do
not need station keeping other than light pressure, then there is an
immediate billion-dollar scale commercial application for comm sats.

KeithH
> --
> 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 on the web visit https://groups.google.com/d/msgid/power-satellite-economics/20230111060814.GC25296%40gate.kl-ic.com.

Jay Lewis

unread,
Jan 11, 2023, 8:29:41 PM1/11/23
to Keith Henson, Keith Lofstrom, Power Satellite Economics
Hi Keith and Keith, 
My experience designing the Viasat 3 GEO comsats is that they are built for station keeping. However Boeing Space was built for cost plus defense contracts.  It was a periodic battle to get them to look at actual data and reconsider ANYTHING. Most of the people there now have 30 years of experience doing what they learned the first 3 years, 10 times over. So they are very well versed on the "right" way, like a preacher knows a subset of bible passages by heart.  

But that doesn't mean they are wrong, and more often than not when I took the time to dig deeper and do my own research I found it was close enough.  Picking battles meant limiting myself to a few areas I'm enough of an expert to take out their guy that barely remembered how algebra works after a lifetime of citing tables made by his wiser ancestors.  

With a fixed price for profit mindset if we ask "how can we design this to not need station keeping" my intuition is it's a wash because accounting for the earth/moon/sun effects over time add complexity that has to be handled somehow.  But maybe there is a trade study to be done.  Perhaps something external like seasonal repair bot refueling that dock in such a way they also do this function?

Jay

You received this message because you are subscribed to a topic in the Google Groups "Power Satellite Economics" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/power-satellite-economics/rQt1zh0TjeY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to power-satellite-ec...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB7cDo9i5Em-bAa8iynwHGdafqqOg6mhZhohVuCPSmim5Q%40mail.gmail.com.

James M. (Mike) Snead

unread,
Jan 12, 2023, 11:37:05 AM1/12/23
to Power Satellite Economics
https://www.virtussolis.space/blank

From this page's illustration, they appear to intend for multiple constellations of 100,000 + satellites to sweep through LEO/MEO each day.

Keith Henson

unread,
Jan 12, 2023, 11:50:09 AM1/12/23
to James M. (Mike) Snead, Power Satellite Economics
On Thu, Jan 12, 2023 at 8:37 AM 'James M. (Mike) Snead' via Power
Satellite Economics <power-satell...@googlegroups.com>
wrote:
>
> https://www.virtussolis.space/blank
>
> From this page's illustration, they appear to intend for multiple constellations of 100,000 + satellites to sweep through LEO/MEO each day.

Looks more like Molniya orbits to me. Perhaps one of their people could say.

Keith
> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/012601d926a4%2418614460%244923cd20%24%40aol.com.

John Bucknell

unread,
Jan 12, 2023, 12:07:26 PM1/12/23
to Keith Henson, James M. (Mike) Snead, Power Satellite Economics
Keith/Mike et all,

We are pursuing highly elliptical orbits, with periods of half a sidereal day for a lot of good economic reasons.  The Molniya mission design has other implications to an architecture, but we think we have solutions.

John

John Bucknell
CEO | Founder
Virtus Solis Technologies Inc


k.a.c...@sympatico.ca

unread,
Jan 12, 2023, 12:18:43 PM1/12/23
to Power Satellite Economics

Hi John, good to hear from you.

 

One of the things I’m curious about, regarding your system architecture: you write “Satellites are grouped into massive arrays--100,000 satellites for 100MW--allowing for a highly scalable energy platform.” This sounds like each “virtual SPS” comprises a very large number of individual satellites, all flying in a formation of some sort. Since you’re on this mailing list, you will have seen the recent  discussion (and maybe also the past discussions) of the “thinned array curse,” whereby unless *all* the radiating elements are within something like half a wavelength of each other, you end up putting a lot of power into sidelobes instead of the main lobe. Myself, I’ve just been unable to conceive of how to avoid this problem, when using separate free-flying satellites --- to get that close to each other, they’d have to be touching. Is there anything you can say about that, vis-à-vis your system design?

 

- Kieran

 

 

John Bucknell

unread,
Jan 12, 2023, 12:44:46 PM1/12/23
to k.a.c...@sympatico.ca, Power Satellite Economics
Kieran,

I'm keenly aware of the thinned array curse - our analysis tools show it explicitly when we get the architecture wrong.  So yes, we have all the antenna elements within a wavelength of each other - billions of them.  The satellites are not a swarm, but a physically linked array.  And if done properly, you can get your 99.9% beam power within a couple thousandths of a degree of arc.  This is an area of proprietary design, so don't ask me how.

John

John Bucknell
CEO | Founder
Virtus Solis Technologies Inc

Gary barnhard

unread,
Jan 12, 2023, 1:33:12 PM1/12/23
to james...@aol.com, Keith Henson, power-satell...@googlegroups.com
Greetings all:

The annual Space Solar Power Symposium collocated with the National Space Society International Space Development Conference publishes the presentations at the same as electronic proceedings on the www.spacedevelopmentfoundation.org website.

Introduction to the last symposium:


Annotated Agenda with the presentations.


You can find additional information on what they are  up to and multiple other entities in their respective presentations.

Sincerely,

Gary

----- GWAVA AUTHENTICATED & SIGNED MESSAGE -----
 
Gary Pearce Barnhard
Space Development Foundation

 

>>> Keith Henson <hkeith...@gmail.com> 1/12/2023 11:49 AM >>>

James M. (Mike) Snead

unread,
Jan 12, 2023, 1:47:56 PM1/12/23
to Power Satellite Economics

Physically-linked, billions of small satellites, flying in “linked” swarms that pass through LEO/MEO twice each day.

 

This appears to be the approach being undertaken.

 

The practicality and orbital safety of this approach eludes me. The expectation of absolute perfection also is evident.

 

The world will need 50,000 GW of space solar power.

 

I am reminded of a small sign on the back of the dump truck, barely readable at 20 ft, that says stay back 200 ft and that the company is not responsible for any damage from things falling off the truck.

 

Mike Snead

--

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.

John Jossy

unread,
Jan 12, 2023, 5:41:24 PM1/12/23
to James M. (Mike) Snead, Power Satellite Economics

Roger Arnold

unread,
Jan 13, 2023, 3:04:33 AM1/13/23
to k.a.c...@sympatico.ca, Power Satellite Economics
KC > <snip> unless *all* the radiating elements are within something like half a wavelength of each other, you end up putting a lot of power into sidelobes instead of the main lobe.

Three points. Two are nits, one is important.

One nit is about that emphasized *all* in your statement. Maxwell's equations are linear, and the EM field at any given point in space is the linear sum / integral of the field contributions from all radiating elements, anywhere. So there's no abrupt nonlinear effect on the far field radiation pattern from a 2-D phased array if a few of the radiating elements in the array are missing.  

The other nit is about the half wavelength separation of elements. The EM field from any array of discrete radiating elements is always going to be an approximation of the field from a continuous surface of infinitesimal radiating elements. It's the difference between a summation and an integral. If the average separation between the discrete elements is around half a wavelength or less, the approximation will be very close. In the main beam, the two will be virtually indistinguishable. In the far field, the solutions for the discrete and continuous cases will diverge significantly -- even if the separation between radiating elements in the discrete case is less than a half wavelength. They'll be close in terms of energy density (square of the amplitude) although the discrete case will always put marginally more energy into the far field. But the finite case will generate considerable fine structure in low level distribution of amplitude and phase. That can be seen intuitively by consideration of the antenna reciprocity principle. The far field must have enough fine structure that the time reversal of the field would converge to the individual radiating elements of the phased array. 

There's nothing magic about an average separation of half a wavelength in the radiating elements of a 2-D phased array. It's the knee of a performance curve. When the separation is half a wavelength, the difference in the EM field pattern between the continuous wavefront case and the pattern from a phased array of discrete elements is small. Going to more elements with smaller separation doesn't improve the beam quality very much. But going to fewer elements with a larger separation degrades it more and more rapidly as the separation grows.

The important point is that the half-wavelength separation rule only applies for 2-D phased arrays. For a 3-D "phased cloud" of emitters, average separation can be arbitrarily large, while still producing a tight beam with very little energy in the far field. It just depends on the depth of the cloud. A good example is a low pressure gas laser. The average distance between excited atoms or molecules in the inverted population of states can be hundreds of wavelengths of light. When those excited atoms or molecules are stimulated by a pilot beam to emit coherently in phase, the resulting beam is the benchmark standard for "tight"

Bottom line: it is very much possible to produce a tight power beam from a diffuse cloud of satellites, orbiting together in a clustered set of orbits. The individual satellites just need a couple of pilot beams for reference, to determine their relative positions within the cloud. From that they can compute the proper phases and amplitudes of the radiating elements they host.




k.a.c...@sympatico.ca

unread,
Jan 13, 2023, 7:11:00 AM1/13/23
to Power Satellite Economics

Roger;

 

Very interesting points, which I don’t dispute, in principle.

 

You wrote:

 

Bottom line: it is very much possible to produce a tight power beam from a diffuse cloud of satellites, orbiting together in a clustered set of orbits.

 

You’re making a couple of major assumptions, in practice:

 

- That the individual satellites carrying each of the transmitting elements are essentially RF-transparent, so that the ones closer to Earth don’t block the path between the further-back ones and the Earth (both for the pilot beam(s) and the main transmit “beam” (which is not really a beam until you’re out of the cloud of emitters). This seems very unlikely for a constellation of individual satellites, each of which would have to carry a significant amount of equipment made of metal, to generate power, and communicate amongst themselves, and perform some amount of orbital manoeuvring.

 

- That the orbital dynamics of the cloud of satellites (and their attached emitters) is such that, as seen from the Earth, there is a sufficient areal density of emitters across the entire cloud, that there is rarely more than half a wavelength between emitters (as seen from Earth). One issue that comes to mind here, is collisions between the satellites, and the effects of those, and/ort the implications of designing them to keep collision from happening.

 

There's nothing magic about an average separation of half a wavelength in the radiating elements of a 2-D phased array. It's the knee of a performance curve. When the separation is half a wavelength, the difference in the EM field pattern between the continuous wavefront case and the pattern from a phased array of discrete elements is small. Going to more elements with smaller separation doesn't improve the beam quality very much. But going to fewer elements with a larger separation degrades it more and more rapidly as the separation grows.

 

Yes, in principle. Again in practice, though, I assume (as did the pioneering designers of the SPS systems back in the 1970s) it will be very important to design any SPS to keep the amount of power (actually, the power flux density at the ground) outside of the main lobe to very low levels, to keep the PFD outside the rectenna fence below civilian-safe levels (as specified by the local health and safety authority), and also to minimize the amount of contamination of the RF spectrum, which is being shared by many other users. The first of those two considerations was top-of-mind for the early SPS designers (my main reference here being the fairly detailed engineering design report that Owen Maynard at Raytheon wrote about the Sandwich Concept design), which did include a sensitivity study looking at the effect of transmit element failures on the PFD outside the main lobe. The answer was that it was essential to have the elements no farther apart than that “knee in the curve” distance, and some (but not many) failed transmit elements could be tolerated.

 

The individual satellites just need a couple of pilot beams for reference, to determine their relative positions within the cloud. From that they can compute the proper phases and amplitudes of the radiating elements they host.

 

I understand the math (and the mechanization in terms of RF and digital equipment and algorithms) for retro-direction down a single pilot beam, if the relative geometry of all the elements of the transmit array is known (my very first engineering job was analysis of a couple of types of phased-array radars in space, one type of which was close enough to the SPS Reference Design to make this clear). The idea of using a couple of pilot beams, to figure out that geometry, is new to me. Can you describe that, or point me to a description of how that works?

 

Thanks!

 

- Kieran

Roger Arnold

unread,
Jan 16, 2023, 6:11:09 PM1/16/23
to k.a.c...@sympatico.ca, Power Satellite Economics
Kieran,

Your comments about "in principle" vs. "in practice" for the idea of a distributed cloud of coherently emitting microsats for power beaming are valid. In particular, the problem of partial shielding and signal scattering among units in the cloud is one that has no ready solution that I can see. I don't know how much of a problem it would prove to be; it would cause some increase in power in the far field at the expense of some loss of power in the main beam, but there's no easy way I know to quantify it. It depends heavily on fine details of the microsat design. I will say, though, that RF-transparent PV cells are possible. They feature in the CalTech proposal that launched this discussion.

The idea of an orbiting cluster of microsats knit by a web of comm links and functioning as a giant multi-beam phased array goes back to Keith L's Server Sky concept. In that case, the beams were for communications, not power transmissions. Power levels were orders of magnitude lower, and noise due to partial shading and reflections off of neighboring antenna elements could be filtered. Power beaming is both a more and a less difficult proposition. Less difficult because the number of beams would be much less and the computational load for beam forming much easier. There's no need for signal extraction. More difficult because at the beam power levels involved, scattering into the far field is a major issue.

Regarding the issue of reference beams that you asked about, the situation is different than what you're thinking. This wouldn't be a system employing phase conjugation for amplified retro-reflection of a pilot beam. (I.e., the conventional arrangement envisioned for power satellites.) The reference beams are analogous to orchestra conductors. They help maintain sync among the clocks in the individual microsats and allow the satellites to periodically update their orbital parameters. The microsats are in free orbits; from their clocks and known orbital parameters, they can compute their positions with respect to the earth and to the rest of their cluster very precisely. But their orbital parameters drift slowly, due to things like uncertainties in the shape of earth's gravity field and fluctuations in solar radiation pressure. The reference beams serve to keep everybody on the same page, so to speak, for beam synthesis.

The ability to synthesize power beams and place them anywhere that software directs without requiring pilot beams from the center of earthside rectennas is a singular feature of this class of powersat designs. It's also a reason that none are likely ever to be built. They would be capable of functioning as terror super-weapons. The effective apertures of their transmitters would be very large; they would, in principle, be capable of focusing tens of gigawatts of continuous microwave power into a spot only a hundred feet or so in diameter. The intense beam could be directed to any place on earth in a satellite's view, and switched on without warning. It would be a relatively easy matter to protect stationary facilities from attack simply by shielding them under a metallic mesh. However unprotected forests and fields could be incinerated across an entire nation. Deadly firestorms comparable to the burning of Dresden by the allies near the end of WW II could be kindled in cities without warning, on a commander's whim. 

Toward the end of my career at Boeing, in the late 1970s and 1980, I spent time working out how a power satellite of this class could be designed to be proof against use as a weapon. The problem isn't actually technical. It wasn't hard for me to sketch out the software and operational protocols to ensure that beams could only be directed toward approved receiver sites, at approved power levels. A more interesting but still tractable problem was working out the firmware to ensure that the anti-weaponization features of the system couldn't be defeated through covert software updates. The real problem was (and is) geopolitics. 

The system that I sketched out relied on a mix of open source software, open source ROM firmware, public key cryptography, distributed secrets, and zero-knowledge proofs to allow parties who didn't trust one another to assure themselves that the system would operate to spec. They could be sure it was free of trap doors that would allow an enemy to break its security and use the system against them. But that meant that any single nation building it would not retain full control. The satellite would effectively become a shared asset of an international community -- presumably the UN. And I'm afraid that would be unacceptable to the power faction in charge of US foreign policy.

- Roger

Reply all
Reply to author
Forward
0 new messages