Sunshade

5 views
Skip to first unread message

Keith Henson

unread,
Aug 4, 2023, 8:58:44 PM8/4/23
to Power Satellite Economics, Kennedy, Robert
https://www.universetoday.com/162621/tether-a-sunshade-to-an-asteroid-to-slow-down-climate-change/

At 3.5 million tons, this is around 1/30 of the Dyson dots project.

I don't know how serious to take it.

Keith

Keith Lofstrom

unread,
Aug 4, 2023, 11:52:34 PM8/4/23
to Keith Henson, Power Satellite Economics
(Henson)

Haven't seen the "Universe Today" item, but it probably
refers to this PNAS paper by István Szapudi:

https://www.pnas.org/doi/epdf/10.1073/pnas.2307434120

The paper is quantitatively vague and poorly written
(IMHO), but the stated mass seems far too small, and
their mechanical arrangement is as silly as a flimsy
umbrella in a hurricane.

Note that a sunshield at L1 will be larger than you
might expect - it must block a 2x or more wider spot
at L1 than the projected "shadow fraction" on Earth,
because the light path from Sun to Earth is a "frustrum",
a truncated cone with the small end at Earth and the large
end at the Sun. The frustrum is twice as wide as the Earth
at Earth-Sun L1. A disk that blocks (say) 5% of the Sun's
light (for a 3 Kelvins temperature drop at Earth) will have
20% of the "disk cross section" of the Earth, 25 million
square kilometers. More than double the area of North
America.

The PNAS article mentions (but does not quantify) very
thin graphene, and assumes room temperature resistivity.
No ... the disk will get VERY hot, the resistivity will
increase, making it even hotter. Thermal runaway?

Instead, it will collapse radially inwards first.

Imagine an Asia-sized aluminum foil, blown outwards from
the Sun by light pressure, and pushed from sides to center
by the "circumferentially inward" gravitational gradient
towards the centerline. There is NOTHING to keep it flat.
It will collapse radially toward the middle, lose light
pressure, then collapse sunwards (with the counterweight
and cable "below") into a sunwards ellipse orbit.

You could configure the sail like a parachute canopy -
but you will need MANY attachment cables and MUCH
radial tensile strength in the shield disk. Again,
imagine supporting a continent scale parachute canopy
made of hot gossamer graphene (NOT perfect unobtanium).

The paper cites Colin McInnes' 1999 book "Solar Sailing",
which I don't own, nor do nearby academic libraries (If
anyone wants to sell a spare copy, let's haggle off-list).
The probable McInnes reference configuration is for an
interstellar laser-propelled light sail FAR from the hot
Sun and the nonlinear gravity gradients of planets.

I sent a longer analysis to my friend, but long analyses
may exceed the attention span of some on this list.
Tried that two weeks ago for an idea with similar physics.
TL/DR responses to the first paragraphs, sigh.

For the few of you who have read down this far, if you
ask nice I will send the longer (still rough and mostly
incomplete) email that I sent to my friend.

Keith L.

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

Keith Henson

unread,
Aug 7, 2023, 3:54:57 AM8/7/23
to Keith Lofstrom, Power Satellite Economics
"You could configure the sail like a parachute canopy -"

The definitive work on this topic is the Dyson Dots paper in the JBIS.
One of the authors is Robert Kennedy who is on this list.

I have a bootleg copy and read it, but it has been a while and I can't
remember it that well. I just skimmed it again. Not sure if the JBIS
has opened up an article from Sept 2013. It's also about the power
satellite to end power satellites, so the subject is on topic for this
list.

Abstract

No study of coping with climate change is complete without considering
geoengineering. A “Dyson Dot” is one or more large (Σarea ~700 K km2,
>200 megatonne) lightsail(s) in a radiation-levitated non-Keplerian
orbit(s) just sunward of the Sun-Earth Lagrange-1 point. The purpose
of this syncretic concept is twofold: (I) As a parasol, it would
reduce insolation on Earth by
at least one-quarter of a percent (-3.4 W m-2), same as what caused
1.5°C drop during the “Little Ice Age” (~1550-1850) and same as the
IPCC Third Report’s mid-range value for global warming by 2050. The
parasol transforms the “solar constant” to a controlled solar
variable. (II) Hosting a ~50K km2 photovoltaic power station on its
sunny side and relaying beamed power via maser to rectennas on a
circumpolar Dymaxion grid, the system could displace over 300 EJ/a
(~100 trillion kWh/yr) of fossil-fired power (total global demand for
electricity forecast by 2050), while providing USD trillions in
revenue from cheap clean energy sales (@1-3¢/kWh) to amortize the
scheme. Total system efficiency compares favorably to automobiles;
total system power density is comparable to nuclear power. This
approach — self-funding, “pay-as-you-go”, minimally intrusive,
scalable, complementary with a portfolio of other measures and above
all reversible–is not precluded by international treaty. Indeed
geoengineering may be the best “killer app” to bootstrap orbital
industry and humanity ad astra, because the terawattscale product is
comparable to the power required for interstellar travel. If Tellurian
spacefaring civilization bootstraps its
exponential growth with multi-terawatt maser beams from such
lightsails, there might eventually be enough of them to have a
detectable effect on Sol’s apparent luminosity at certain wavelengths,
as seen from far away, similar to the eponymous Dyson Sphere, hence
the moniker.

Dyson Dots & Geoengineering: The Killer App Ad Astra
Robert G. Kennedy III, Eric Hughes, Kenneth I. Roy et al.

Spinning the sails would help keep them flat. They would have to
precess 360 deg in a year to keep them oriented. I have not worked
the spin rate

There is at least one way to get hoop stress on a big area while
avoiding precession. The outside is a tube with high-velocity
vehicles in it, two tubes to zero out the precession. This should keep
the shade pointed at the sun.

I think the design details are relatively straightforward. The
problem is obtaining and moving that much mass (over 100 million tons)
to L1. I don't think rockets make sense. I don't boggle too badly at
fifty 50,000-ton power satellites taking 25,000 launches (2.5 million
tons) per year for a power satellite construction program, but 10-20
million tons per year is just over the top. (and th traffic would be
too much for the ozone.)

A possible method would be an electromagnet launcher from Earth. I
suspect this is a nonstarter because the NOx from the outgoing
payloads would totally destroy the ozone.

The proposal from the space colony days would be to mine the moon.
This could be scaled up from O'Neill's work published in 1976. This
might be a relatively lower-risk approach. A substantial mining and
processing base might be investigated. Regolith is no more than 1/3rd
metal

Last, there is mining an asteroid and sending the metal to L1. Other
than a lightweight analysis I did in 2012, there has not been a lot of
work on this subject. that I know about. But the side stream metals
might pay for the whole project

There might be another solution I missed. If you can think of one, please do.

KeithH

Arthur Woods

unread,
Aug 7, 2023, 6:02:06 AM8/7/23
to Keith Henson, Keith Lofstrom, Power Satellite Economics

There are obvious synergies between building SPS components from lunar materials as we have articulated in our GE⊕-LPS study and constructing large sunshade structures using extraterrestrial resources - Moon and asteroids.
Not only the technology for large structures but also the goal to address the energy and climate crises and the business case for doing so.

FYI: there is a  'Planetary Sunshade Foundation' founded in 2021

https://www.planetarysunshade.org

with a list of publications:

https://www.planetarysunshade.org/publications

Arthur


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

Reply all
Reply to author
Forward
0 new messages