Solar Power Satellites

It distresses to me to see space advocates waste time and resources
advocating and dealing with a technological approach so completely
off the wall as solar power satellites.

Solar power satellites are not worth advocating anytime in the
foreseeable future--the balance of the century if not longer. Even
under a set of more than a dozen unrealistically optimistic as-
sumptions, a solar power satellite would produce electricity at a
cost more than 17 times that of an Earth based system--realistically,
probably hundreds of times more.

This can be demonstrated by comparing a solar power satellite system
measuring ten square kilometers with an Earth based solar power
station of the same size.

Many hoped for technological improvements will benefit both space and
Earth based systems; so many of those hoped for improvements are
irrelevant in such a comparison--for instance, the effi-ciency of
solar cells. Solar cell efficiency and cost improvements will
benefit Earth based sys-tem just as much as a space based system.
Let's assume solar cell efficiency of 30%, which is beyond anything
commercially available today, but is remotely feasible for cost
effective large scale operations in the not too distant future. But,
whether solar panel efficiency is 10% or 50% makes no difference,
since efficiency will be about the same in space and on Earth.

Another example is electricity-microwave-electricity conversion
efficiencies and costs. Increases in these efficiencies and
decreases in cost improve the economics of both power satellites and
Earth based systems. In a limiting case, where energy losses
approached zero and cost was negli-gible, power could be transported
from the most optimal locations around the Earth, and the is-sues of
diurnal, weather and seasonal variability for an Earth based system
would be eliminated.

The average delivered electrical output to a facility on Earth for
the power satellite in geostationary orbit is derived as follows:

1300 watts per square meter (solar insolation at Earth's distance
from the Sun)
times *10^7 square meters of solar panels (ten square kilometers of
solar panels)
times 30% solar panel efficiency
less 1% for station keeping
times 50% efficiency in converting electricity to a power beam
less 2% for temperature control
less 2% loss in transmission through the atmosphere
less 2% loss in dispersion and aiming errors
times 70% efficiency in converting the energy beam back into
electricity

= 1300 watts*10^7*.3*.99*.5*.98*.98*.99*.7 = 128 watts *10^7.

The average delivered electrical output for the Earth based systems
is derived as follows:

According to the US National Renewable Energy Laboratory Resource
Assessment Program, the average daily total solar insolation at the
Earth's surface using a two-axis tracking concentrator is about 10 kw
hours of electricity per day per m/2 over large areas stretching from
Southern Cali-fornia through Nevada, Utah, Arizona, New Mexico and
West Texas. Vast areas of Northern Mexico have similar, or even
greater, solar insolation. This area encompasses over a million
square kilometers, and contains the Sonoran and Chihuahuan Deserts.
Much of it is at altitudes exceeding a thousand meters; and much of
it has population densities of less than one person per square
kilometer.

The average solar insolation collectable in these areas is 10 kw
hours * 365 days per year divided by the 8766 hours in a year = 416
watts/m2. The amount of electricity generated at the ten square
kilometer Earth based station

= 416*watts*10^7*.3 = 125 watts *10^7.

The costs of these systems are summarized below:

Earth based system capital costs
Materials & Construction $1,100 million ($100 per annual megawatt
hour capacity)
Land $3 million ($1,000 per acre exclusive of mineral rights)
TOTAL $1,103 million

Power satellite capital costs
Materials and construction $5,600 million (based on 1/10 the current
cost of manned missions to orbit)
Ground Station(s) $50 million
Transport of materials and equipment to orbit $20,000 million (based
on 20,000 ton total weight and cost to geostationary orbit of $1,000
per kilogramless than a 20th of current costs)
TOTAL $25,650 million

Based on current experiences with photovoltaic systems, operations
and maintenance costs for the Earth system would be about $20 million
per year; for the power satellite, about $115 million. This assumes
the equivalent of only one shuttle mission per year, but to
geostationary orbit, and at 1/5 the current cost to low Earth orbit;
and $5 million for operations at the 11 terawatt receiving station on
Earth.

Assuming a 30 year life for the Earth based system, the cost per
kilowatt hour would be the sum of the depreciation of materials and
construction costs (you don't depreciate land) = $1.1 billion divided
by 30 = $37 million, operations and maintenance = $20 million per
year and the cost of capital per year= 5% of $1,103 million = $55
million. Total = $112 million
Divided by the total electricity produced per year (8766*125*10^4 =
about 11 terawatts) = about 1 cent per kilowatt hour.

This cost is far below both the current wholesale cost of electricity
and the current cost of gener-ating electricity from solar panels.
This is because of the optimistic assumptions about solar cell
efficiency and cost, and all the other costs that occur from the bus
forward.

Assuming a 50 year life for the space based system, the cost per
kilowatt hour would be the sum of:
The annual depreciation of materials and construction costs
including transportation to orbit= $25,650 million divided by 50 =
$513 million
Operations and maintenance = average of $115 million per year, and
The cost of capital per year = 5% of $25,650 million = $1,282
million.
Total = $1910 million divided by the total electricity produced per
year (8766*128*10^4 = about 11 terawatts) = about 17 cents per
kilowatt hour.

These estimates of output and cost for the space based system contain
more than a dozen assumptions and estimates which are barely
conceivable in this century. These include:
Cost to geostationary orbit of $1,000 per kilogramless than 1/20
the current cost;
Efficiencies of large scale energy conversions exceeding the best
laboratory results to date;
Solar panels with much higher efficiency and a tenth the cost that
is now commercially avail-able;
Weight for the solar panels (including wiring, substrates,
adhesives, spars, harnesses and other structural components) of one
kilogram per square meter, a tenth of that which is now commercially
available--even with double the efficiency and ten times lower cost.
Temperature control in space handled at negligible cost;
A 50 year life for the power satellite. The limiting factor will
probably be technological change, and it will probably manifest
itself in much sooner than 50 years;
And about ten other very optimistic assumptions and estimates.

The solar power satellite is not a system worth advocating for this
century, if ever. Even under a set of more than a dozen
unrealistically optimistic assumptions, a power satellite would pro-
duce electricity at a cost 17 times higher than that of an Earth
based system --realistically, probably hundreds of times more.