Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

SPS Questions....

10 views
Skip to first unread message

Tom Moore

unread,
Jul 13, 2000, 3:00:00 AM7/13/00
to sci-spa...@moderators.isc.org
Ok... heres an SPS question that has been plaguing me... well its not
unique to SPS, but it was SPS that raised the question.

A lot of SPS designs call for microwave transmission of the power to a
surface station. A very lowlevel description of how the microwaves work
is by exciting molecules, which causes them to generate thermal energy.
Now plain air isnt affected much by microwave energy (as evidenced by
the cool air in my microwave oven after just finishing heating my
dinner) but water is reactive ( which is why I use the microwave to
heat water for tea etc). There is a great deal of water in the
atmosphere. What kind of effect will the high intensity of the SPS
microwave transmission have on heating that water, and subsequently
affecting atmospheric warming, weather etc.

Now as to why this isnt unique to SPS systems, every radar, and every
microwave communications system is pouring energy waves into the
atmosphere, passing through the water vapor etc. One has to wonder how
much heat all that energy has been transfering to our atmosphere over
the past 60 years.

Tom


Tom Moore

unread,
Jul 13, 2000, 3:00:00 AM7/13/00
to

cfr...@my-deja.com

unread,
Jul 15, 2000, 3:00:00 AM7/15/00
to
In article <A.396E53A...@bellsouth.net>,

Tom Moore <t_m...@bellsouth.net> wrote:
> Ok... heres an SPS question that has been plaguing me... well its not
> unique to SPS, but it was SPS that raised the question.
>
> A lot of SPS designs call for microwave transmission of the power to a
> surface station. A very lowlevel description of how the microwaves
work
> is by exciting molecules, which causes them to generate thermal
energy.
> Now plain air isnt affected much by microwave energy (as evidenced by
> the cool air in my microwave oven after just finishing heating my
> dinner) but water is reactive ( which is why I use the microwave to
> heat water for tea etc). There is a great deal of water in the
> atmosphere. What kind of effect will the high intensity of the SPS
> microwave transmission have on heating that water, and subsequently
> affecting atmospheric warming, weather etc.
>

No, only a few special frequencies of microwaves are absorbed by
water. Microwave ovens specifically are tuned to those frequencies.

The frequencies used by SPS will be different, and will be tuned toi
those which are not absorbed by water, which is the majority of the
band.

> Now as to why this isnt unique to SPS systems, every radar, and every
> microwave communications system is pouring energy waves into the
> atmosphere, passing through the water vapor etc. One has to wonder
how
> much heat all that energy has been transfering to our atmosphere over
> the past 60 years.
>

Practically nill.

The worst culprit will be leakage from microwave ovens, and weather
monitoring radars, since those frequencies are intentionally absorbed
by water.

The frequencies used by most radar and telecommunications are
specifically chosen to avoid water absorbption, since that will
adversely affect their system performance.

Sent via Deja.com http://www.deja.com/
Before you buy.

gbaikie

unread,
Jul 15, 2000, 3:00:00 AM7/15/00
to
In article <A.396E53A...@bellsouth.net>,
Tom Moore <t_m...@bellsouth.net> wrote:
> Ok... heres an SPS question that has been plaguing me... well its not
> unique to SPS, but it was SPS that raised the question.
>
> A lot of SPS designs call for microwave transmission of the power to a
> surface station. A very lowlevel description of how the microwaves
work
> is by exciting molecules, which causes them to generate thermal
energy.
> Now plain air isnt affected much by microwave energy (as evidenced by
> the cool air in my microwave oven after just finishing heating my
> dinner) but water is reactive ( which is why I use the microwave to
> heat water for tea etc). There is a great deal of water in the
> atmosphere. What kind of effect will the high intensity of the SPS
> microwave transmission have on heating that water, and subsequently
> affecting atmospheric warming, weather etc.
>
> Now as to why this isnt unique to SPS systems, every radar, and every
> microwave communications system is pouring energy waves into the
> atmosphere, passing through the water vapor etc. One has to wonder
how
> much heat all that energy has been transfering to our atmosphere over
> the past 60 years.
>
> Tom
Some guesses:
First microwave energy in your oven is at a wavelength that efficently
heats water, it not necessarily true that communication microwave is at
the same frequency- so it might still heat water, but imagine to save
energy and have better signal they would use a frequency the most
transparent to atmosphere and water vapor. The other side of this is
how much energy is lost in microwave transmission- if it's inefficent
enough that it's heating up any significant amount of water vapor- they
are going to have quite a power bill. There is a lot of water in our
atmosphere- at least enough to cover the Earth in a inch or so of
water, the air in atmosphere if liquified would be tens of feet deep.
It would take a huge amount of energy to heat up the atmosphere by 1
degree even for a single day, even if you design some system that does
in the most efficent way. Compare this with the energy from the sun,
which something like 1300 watts per sq meter- that's a lot energy on a
global scale. So even if you had microwave footprint of a mile square
and it had wavelength and intensity of a microwave oven, it's affect on
global temperature would be utterly insignificant. A mile sq black
asphalt parking lots would probably add more heat (during the daytime).

Gord Deinstadt

unread,
Jul 15, 2000, 3:00:00 AM7/15/00
to

Tom Moore wrote:
> A lot of SPS designs call for microwave transmission of the power to a
> surface station. A very lowlevel description of how the microwaves work
> is by exciting molecules, which causes them to generate thermal energy.
> Now plain air isnt affected much by microwave energy (as evidenced by
> the cool air in my microwave oven after just finishing heating my
> dinner) but water is reactive ( which is why I use the microwave to
> heat water for tea etc). There is a great deal of water in the
> atmosphere. What kind of effect will the high intensity of the SPS
> microwave transmission have on heating that water, and subsequently
> affecting atmospheric warming, weather etc.

Microwave ovens use one of three or four specific frequencies at which
water
molecules resonate. At most other frequencies water vapor is pretty
much
transparent to microwaves. Clouds can absorb (or reflect) radio waves
if the
droplet size is comparable to the wavelength of the microwaves, but that
only
affects the higher frequency bands.



> Now as to why this isnt unique to SPS systems, every radar, and every
> microwave communications system is pouring energy waves into the
> atmosphere, passing through the water vapor etc. One has to wonder how
> much heat all that energy has been transfering to our atmosphere over
> the past 60 years.

Until recently communications and radar systems stayed clear of the
"oven"
frequencies to avoid interference and because they wanted a signal that
would penetrate the atmosphere, not be absorbed by it. However recently
advances in digital radio have made it possible to use these bands
for short-range communications. Furthermore, for license-free
operation it is an advantage that signals are absorbed by the
atmosphere,
because it reduces the worry that a transmitter will cause interference
with
other electronic equipment. You don't have to worry about such systems
warming
up the atmosphere, though, because they transmit at power levels of a
fraction
of a watt.
--
Gord Deinstadt

Ryan Healey

unread,
Jul 15, 2000, 3:00:00 AM7/15/00
to
Well if the microwaves aren't at one of the resonance frequencies of water
then they wont heat it, that wouldn't be very hard to do as we can figure
out what are aren't resonance.

Even if it did there isn't enough energy dumped into the water (not dense in
atmosphere) to cause problems, in fact you would cause more damage (to
whether and earth's surface) by building the solar panels on the earths
surface.

Tom Moore wrote:

> Ok... heres an SPS question that has been plaguing me... well its not
> unique to SPS, but it was SPS that raised the question.
>

> A lot of SPS designs call for microwave transmission of the power to a
> surface station. A very lowlevel description of how the microwaves work
> is by exciting molecules, which causes them to generate thermal energy.
> Now plain air isnt affected much by microwave energy (as evidenced by
> the cool air in my microwave oven after just finishing heating my
> dinner) but water is reactive ( which is why I use the microwave to
> heat water for tea etc). There is a great deal of water in the
> atmosphere. What kind of effect will the high intensity of the SPS
> microwave transmission have on heating that water, and subsequently
> affecting atmospheric warming, weather etc.
>

> Now as to why this isnt unique to SPS systems, every radar, and every
> microwave communications system is pouring energy waves into the
> atmosphere, passing through the water vapor etc. One has to wonder how
> much heat all that energy has been transfering to our atmosphere over
> the past 60 years.
>

> Tom


jti...@my-deja.com

unread,
Jul 17, 2000, 3:00:00 AM7/17/00
to
In article <396E53A7...@bellsouth.net>,

Tom Moore <t_m...@bellsouth.net> wrote:
> Ok... heres an SPS question that has been plaguing me... well its not
> unique to SPS, but it was SPS that raised the question.
>
> A lot of SPS designs call for microwave transmission of the power to a
> surface station. A very lowlevel description of how the microwaves
work
> is by exciting molecules, which causes them to generate thermal
energy.

Only one specific frequency of microwaves excite water molecules. If
your microwave transmitter is _not_ tuned to that frequency, moist air
is very transparent to the microwaves.

Regards,
Jack Tingle

wm...@my-deja.com

unread,
Jul 18, 2000, 3:00:00 AM7/18/00
to


In article <8kusj0$idp$1...@nnrp1.deja.com>,

That's right. Also, radio waves, microwaves, infrared energy, light
energy, ultraviolet energy, x-rays, are all forms of ELECTROMAGNETIC
RADIATION! The only difference is the frequency or color.

Electromagnetic radiation arises whenever charged particles are
accelerated. In low frequency radiations the acceleration can be
slow. In radios, this occurs when electrons move through circuits. In
microwave circuits electrons are accelerated in magnetic fields -
vibrating sort of like a gas in a whistle. Infrared is radiation is
caused by the vibration of atoms when heated. In visible radiation
electrons jump from one level in an atom to another - accelerating
them. X-rays can occur when fast moving electrons hit a wall.

Microwaves are easily produced at high efficiency (>85%) in KLYSTRON
TUBES. These tubes are easily and cheaply produced and can be found in
microwave ovens.

Jon J Thaler

unread,
Jul 18, 2000, 3:00:00 AM7/18/00
to
jti...@my-deja.com wrote:

> Only one specific frequency of microwaves excite water molecules. If
> your microwave transmitter is _not_ tuned to that frequency, moist
> air is very transparent to the microwaves.

I think this isn't the case. See Jackson "Classical Electrodynamics"
second edition, page 290. There is an absorption peak in the near
infrared (10^14 Hz, not relevant here). In the microwave region, the
absorption is smoothly varying, and increases with frequency. The
absorption is stronger than 1/cm for all microwave wavelengths
shorter than 10 cm.

MA Lloyd

unread,
Jul 18, 2000, 3:00:00 AM7/18/00
to

It's not. The sharp microwave lines are for rotational energy levels,
you won't see them except in that gas phase - and low pressure gas
phase at that, in a condensed phase quantitized energy levels all smear
out. Your microwave oven isn't tuned, the heating is purely classical
dielectric heating, not quantitized. On the other hand, there are windows
in which moist air is fairly transparent, and you'd naturally choose
those for an SPS beam, or an all weather radar/lidar/whatever sensor for
that matter.

--
-- MA Lloyd (mall...@io.com)

wm...@my-deja.com

unread,
Jul 21, 2000, 3:00:00 AM7/21/00
to
In article <8l2rd0$cbj$1...@eris.io.com>,
There's a formula available in any good engineering textbook (which I
don't have handy right now) that tells you how much energy available
from the Sun is absorbed by the atmosphere. Any loss of microwave
energy must be better than this absorption for microwaves to be more
efficient.

At the latitudes of North America the Sun during high noon at summer
radiates about 600 watts per square meter incident radiation normal to
a surface oriented to the sun. In space there is about 1,370 watts per
square meter. So, more than half is lost. The space value varies plus
or minus 15 watts per square meter primarily in the UV region due to
changes on the surface of the Sun.

Of course in winter months and at sunrise and sunset more energy is
absorbed by the atmosphere, reducing the amount of power available per
unit area by half again as much. (only 250 watts/square meter getting
through!)

And at night and on overcast days, things are worse yet.

In most parts of the US there are between 1500 and 2000 hours of
sunlight a year. There are 8766 hours in a year. So, about 5/6 to 4/5
of the time the sun is unsuable. Then add the atmospheric losses and
you find that 19/24 to 9/10 of the sunlight available per unit area is
actually available on the Earth's surface.

Tracking the sun in space is much easier than tracking the sun on the
ground. So, on the ground you don't do that. This limits your
conversion efficiency so your collector area increases for a given
pwoer output by an order of 3 or 4. Also, ground based collectors must
withstand gravity and weather. This triples costs again! So, overall
a square meter in space can be equivalent to 288 square meters on the
ground worst case, or 30 square meters best case.

Then, you've got the added inefficiencies and costs of energy storage.
To take energy and make fuel is not 100% efficient. And to take fuel
and make energy is not 100% efficient. Furthermore, the equipment to
do this conversion and storage is not free. Finally, the conversion
equipment must be sized for peak solar output, so is not as economic as
it might otherwise be. Assuming 30% efficiencies for each step and a
doubling again of the capital costs to make fuels of solar power, we
have another 28.89 fold increase (1/.3 x 1/.3 x 2) meaning solar based
power requires 832 times the cost on Earth for collector, and energy
management than an equivalent power in space.

BrianF5070

unread,
Jul 23, 2000, 3:00:00 AM7/23/00
to
wm...@my-deja.com writes:

Lots and lots of snipped stuff concluding with:

>meaning solar based
>power requires 832 times the cost on Earth for collector, and energy
>management than an equivalent power in space.
>

Or you simply take the CAPEX of the solar panels for the latest generation of
GEO sats ($50 - $100 MM) and divide by the expected power output (15 kW at
start of life; 9 kW at end of life) integrated over the 15 year rated life,
ignore OPEX (which is small relative to CAPEX), and find that current space
solar power costs at least $30 - $60 per kW-hr.

Then you compare with a current terrestrial solar system, which is going for
around $0.20 to $0.30 per kW-hr, depending on location, assumptions, etc.

Hmmm. It's not really that difficult a comparison after all, is it?


Rand Simberg

unread,
Jul 23, 2000, 3:00:00 AM7/23/00
to
BrianF5070 wrote:

> Or you simply take the CAPEX of the solar panels for the latest generation of
> GEO sats ($50 - $100 MM) and divide by the expected power output (15 kW at
> start of life; 9 kW at end of life) integrated over the 15 year rated life,
> ignore OPEX (which is small relative to CAPEX), and find that current space
> solar power costs at least $30 - $60 per kW-hr.

It's not possible to usefully extrapolate current costs of space solar panels to
potential cost of mass-produced space solar panels for power satellites, any
more than one can infer the costs of a well-designed, high-flight-rate reusable
launch system from Shuttle costs.

--
************************************************************************
sim...@interglobal.org * 310 372-7963 (CA) 307 739-1296 (Jackson Hole)
interglobal space lines * 307 733-1391 (Fax) http://www.interglobal.org

"Extraordinary launch vehicles require extraordinary markets..."


wm...@my-deja.com

unread,
Jul 24, 2000, 3:00:00 AM7/24/00
to
In article <397AA2DE...@interglobal.org>,

Rand Simberg <sim...@interglobal.org> wrote:
> BrianF5070 wrote:
>
> > Or you simply take the CAPEX of the solar panels for the latest
generation of
> > GEO sats ($50 - $100 MM) and divide by the expected power output
(15 kW at
> > start of life; 9 kW at end of life) integrated over the 15 year
rated life,
> > ignore OPEX (which is small relative to CAPEX), and find that
current space
> > solar power costs at least $30 - $60 per kW-hr.
>
> It's not possible to usefully extrapolate current costs of space
solar panels to
> potential cost of mass-produced space solar panels for power
satellites, any
> more than one can infer the costs of a well-designed, high-flight-
rate reusable
> launch system from Shuttle costs.
>


That's right! And besides, Brian didn't compare earth based solar to
space based solar.

But Briand did have a point about costs. Maybe I shouldn't have used
the word 'cost' in my original posting. But its hard to find a
suitable replacement.

One can talk about the costs of things in many ways.

Cost can be related to the infrastructure in the market to make a
thing based on historical patterns of investment. That's what Brian
looked at.

One can also look at fundamentals to attain ultimate outcomes. That's
what I was speaking of.

You need some 832 times the collector area in North America to reliably
produce a kwhr of electricity in a year than you do in orbit. If
launch and deployment costs can be kept in line, then we can see from
this fundamental advantage of space based solar, what it will take to
make SPS competitive.

These are the approaches I take in my proposed program;

1) reduce mass to about 5 to 25 tons per square km of collector area;
2) reduce launch costs to about 1/10th comparable launch costs;
3) reduce deployment and support on orbit by launching a complete self
deploying satellite built on Earth;
4) reduce transmission costs by using moderately high energy density
laser beams rather than very low energy density microwave beams;
5) increase revenues by providing power to markets that value power
highly, namely the peak power user.

Peak power users in the US pay about $300 million per TWhr energy and
$350 million per GW of power generating capacity.

A 2GWatt SPS power plant is thus worth $700 million and can generate
17.532 TWhr per year worth another $5.26 billion.

The NPV of this income stream with typical discount rates is over $100
billion. $100 billion is enough to build the first station.

Recurring costs of about $5 billion per SPS implies that once one
station is operational, numbers of stations can double every year until
the market is saturated.

Market saturation for stationary power will occur in about 1,000
stations, in about 10 years after the launch of the first station.

Total capital cost of $4 trillion valued at about $20 trillion. Once
the US utility market is served this way, other markets can be added to
maintain growth.

At about 20,000 stations all utility demand for the entire world can be
met. This will occur in about 15 years following the first profitable
SPS orbited. This will have a cost of $80 trillion and a value of $400
trillion. The entire productive capacity of the world will be about 11
times greater than that found today.

Beyond this penetration into the mobile market, will fuel additional
growth. This includes providing power to laser powered rockets and
jets operating on Earth and off.

Throughout all of this SPS will produce power for less than ground
based fuel burning plants after the first $100 billion is spent.

BrianF5070

unread,
Jul 29, 2000, 3:00:00 AM7/29/00
to
Rand Simberg wrote:

>> Or you simply take the CAPEX of the solar panels for the latest generation
>of
>> GEO sats ($50 - $100 MM) and divide by the expected power output (15 kW at
>> start of life; 9 kW at end of life) integrated over the 15 year rated life,
>> ignore OPEX (which is small relative to CAPEX), and find that current space
>> solar power costs at least $30 - $60 per kW-hr.
>
>It's not possible to usefully extrapolate current costs of space solar panels
>to
>potential cost of mass-produced space solar panels for power satellites, any
>more than one can infer the costs of a well-designed, high-flight-rate
>reusable
>launch system from Shuttle costs.

Of course it is. For both of these - solar panels and RLVs.

I work in the petroleum industry. Revolutionary technoloies in our industry
will drop the technical cost of a process by 30% in a generation. Evolutionary
technologies, which are far more common, will drop technical process costs by
10 - 15%. The same sort of technology improvement cost curves apply in *every*
mature technical industry, from semiconductors to chemicals to petroleum to
automobile manufacture.

The launch market is a very mature industry. The chances of a technology
breakthrough dropping costs from current launch costs by an order of magnitude
are negligible.

Current ELV costs are order of 10,000 per kg. EELV will drop this by 15%.
This is believable. Beal is looking to drop it by another 30% or so. I might
even believe that.

The current semi-RLV market is about $20,000 per kg. The chances of the very
next RLV being 30% less than this are low; being 50% less are negligible. The
next RLV generation with Shuttle-type capabilities will cost $10,000 - $15,000
per kg to LEO. Count on it. (Of course, you may claim that the next RLV will
have fewer capabilities than Shuttle, which will trim costs; i.e., I think
Pioneer Rocketplane was targeting around $5000/kg. This may be possible. But
I can guarantee that if V* is ever operational, it will be more than
$10,000/kg).

Likewise, the next generation of space solar panels (which is also a very
mature industry) will be order of $25/kW-hr. It will come down from present
costs. I have no doubt. It will continue down in the future. But it will be
a long time before it reaches ground based panel cost.

Like it or not, this is the way the world works. Claims of rabid RLV advocates
notwithstanding.

Brian

"If it sounds to good to be true - it is."


Terrell Miller

unread,
Jul 29, 2000, 3:00:00 AM7/29/00
to
God, this guy is persistent, give him that! Every time someone shoots down
his numbers, he changes the subject and comes up with a whole bunch of other
numbers.

Looks like he's either an overeducated idiot, or he's aggressively looking
for investors who are. My advice: don't give this guy any of your money!

--
Terrell Miller, Ordo Pantheris
terrel...@mindspring.com

"Faith is love at its most intense and selfless."
-Linda Nagata

<wm...@my-deja.com> wrote in message news:8lg15a$kad$1...@nnrp1.deja.com...


> In article <397AA2DE...@interglobal.org>,
> Rand Simberg <sim...@interglobal.org> wrote:

> > BrianF5070 wrote:
> >
> > > Or you simply take the CAPEX of the solar panels for the latest
> generation of
> > > GEO sats ($50 - $100 MM) and divide by the expected power output
> (15 kW at
> > > start of life; 9 kW at end of life) integrated over the 15 year
> rated life,
> > > ignore OPEX (which is small relative to CAPEX), and find that
> current space
> > > solar power costs at least $30 - $60 per kW-hr.
> >
> > It's not possible to usefully extrapolate current costs of space
> solar panels to
> > potential cost of mass-produced space solar panels for power
> satellites, any
> > more than one can infer the costs of a well-designed, high-flight-

> rate reusable


> > launch system from Shuttle costs.
> >
>
>

Bill Bonde

unread,
Jul 29, 2000, 3:00:00 AM7/29/00
to

BrianF5070 wrote:


>
> Rand Simberg wrote:
>
> >> Or you simply take the CAPEX of the solar panels for the latest generation
> >of
> >> GEO sats ($50 - $100 MM) and divide by the expected power output (15 kW at
> >> start of life; 9 kW at end of life) integrated over the 15 year rated life,
> >> ignore OPEX (which is small relative to CAPEX), and find that current space
> >> solar power costs at least $30 - $60 per kW-hr.
> >
> >It's not possible to usefully extrapolate current costs of space solar panels
> >to
> >potential cost of mass-produced space solar panels for power satellites, any
> >more than one can infer the costs of a well-designed, high-flight-rate
> >reusable
> >launch system from Shuttle costs.
>

> Of course it is. For both of these - solar panels and RLVs.
>
> I work in the petroleum industry. Revolutionary technoloies in our industry
> will drop the technical cost of a process by 30% in a generation. Evolutionary
> technologies, which are far more common, will drop technical process costs by
> 10 - 15%. The same sort of technology improvement cost curves apply in *every*
> mature technical industry, from semiconductors to chemicals to petroleum to
> automobile manufacture.
>
> The launch market is a very mature industry. The chances of a technology
> breakthrough dropping costs from current launch costs by an order of magnitude
> are negligible.
>

Why do you say this?

> Current ELV costs are order of 10,000 per kg. EELV will drop this by 15%.
> This is believable. Beal is looking to drop it by another 30% or so. I might
> even believe that.
>
> The current semi-RLV market is about $20,000 per kg. The chances of the very
> next RLV being 30% less than this are low; being 50% less are negligible.
>

Where is that $20K per kilogram going? It isn't to fuel. Imagine how
much it would cost for an airline ticket if you needed a new airplane
after every flight. Or if you took the engines and much of the rest of
the plane apart after every flight?

> Likewise, the next generation of space solar panels (which is also a very
> mature industry) will be order of $25/kW-hr. It will come down from present
> costs. I have no doubt. It will continue down in the future. But it will be
> a long time before it reaches ground based panel cost.
>

Panels used on the ground can work in Space, can't they? Is there some
vacuum or radiation or something else issue? I thought the reason space
rated panels were so expensive was because they needed to be as
efficient as possible. In any case, if you can make solar cells on the
Moon directly onsite from lunar regolith, potentially your costs could
go way down.

Phil Fraering

unread,
Jul 29, 2000, 3:00:00 AM7/29/00
to
brian...@aol.com (BrianF5070) writes:

> Examples of hundreds of technologies through all industries show the same
> recurring patterns. The launch induustry is in no way unique.
>
> Indeed, it should be the rabid advocates of cheap RLV technology who
> should be defending their claims that all the lessons of history and
> economics will be repealed in their specific case.

Following simple analysis of the fact that a year and a half ago the
price of oil was $ 10 per barrel, and it's now $ 30 per barrel, more
or less, seems to imply that it will be $ 50 per barrel in another
18 months, and 65 per barrel in 3 years... just draw the line through
the points on the graph.


--
Phil Fraering "One day, Pinky, A MOUSE shall rule, and it is the
p...@globalreach.net humans who will be forced to endure these humiliating
/Will work for tape/ diversions!"
"You mean like Orlando, Brain?"

Bill Bonde

unread,
Jul 29, 2000, 3:00:00 AM7/29/00
to

BrianF5070 wrote:
>
> >Where is that $20K per kilogram going? It isn't to fuel. Imagine how
> >much it would cost for an airline ticket if you needed a new airplane
> >after every flight. Or if you took the engines and much of the rest of
> >the plane apart after every flight?
>

> An oft-used analogy and completely irrelevant. Why use expendable robots for
> Mars exploration instead of recoverable and reusable ones? The point is, which
> is cheapest? ELV's are used because they are cheaper than the existing RLV.
>
ELVs are cheaper because the demands on the system are so high, parts
don't last very long, so why waste time taking the system apart after
every use? Just use it once. But if you can reach a point where you
don't have to fix mostly everything after every launch costs have to
come way down.

> When the next RLV is designed and built, it will only have to compete with
> existing ELV's. Thus, the designing firm will only build it such that total
> per flight cost will be lower than the competition. If current launch costs
> are $10,000 per kg to LEO, and they think that they can get sufficient market
> share for a large IRR at $7000/kg, then this is what they will design it for.
> This cuts vehicle CAPEX, R&D expenditures, time to market, and program risk.
>
> Nobody with money to invest, who is in their right mind, would risk it on
> trying to cut costs to $1000/kg, when there is no business need to do so, and
> valid reasons against.
>
OK, so we get two RLVs and they compete. Or two companies both buy
examples of a single RLV and compete. Two airlines manage to compete
with each other even though they might both buy only Boeing planes.
Also, a new system is going to be unproved. Coming in at half the launch
cost of competitors is a good way to take business. Do I risk a 50
million dollar satellite on something with no track record? Can I get
insurance even?

> >> Likewise, the next generation of space solar panels (which is also a very
> >> mature industry) will be order of $25/kW-hr. It will come down from
> >present
> >> costs. I have no doubt. It will continue down in the future. But it will
> >be
> >> a long time before it reaches ground based panel cost.
> >>
> >Panels used on the ground can work in Space, can't they? Is there some
> >vacuum or radiation or something else issue? I thought the reason space
> >rated panels were so expensive was because they needed to be as
> >efficient as possible. In any case, if you can make solar cells on the
> >Moon directly onsite from lunar regolith, potentially your costs could
> >go way down.
>

> No, ground panels cannot work in space, unmodified. The radiation deteriorates
> panels without protection, they need charge buildup protection, they need
> automated deployment, and so on.
>
They don't NEED automated deployment if you are talking about an SPS.
You have people on orbit helping things out. Or Lunar SPS where the
cells are manufactured right on the regolith.


> Space panels are not so expensive because they need to be efficient - they are
> efficient because they are so expensive. Efficiency makes deployment easier,
> and reduces launch weight and cost. But an equally efficient GaAs cell on
> Earth does not cost nearly as much.
>
> The fact is, space panels are designed very carefully to minimize total system
> cost. Again, it is all tied together with dozens of other factors, from market
> size, to reliability requirements, to bus integration. And the fact is that
> the cost of the panels won't come down faster than the typical technology curve
> predicts.
>
But in an SPS system, the market is huge by definition. And the
manufacture is best with space resources, so a lot of the reasons for
the current higher cost are gone. In Lunar SPS, this includes even the
issue of efficiency.

Rand Simberg

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
On 29 Jul 2000 23:50:58 GMT, in a place far, far away,
brian...@aol.com (BrianF5070) made the phosphor on my monitor glow
in such a way as to indicate that:

>>It's not possible to usefully extrapolate current costs of space solar panels
>>to
>>potential cost of mass-produced space solar panels for power satellites, any
>>more than one can infer the costs of a well-designed, high-flight-rate
>>reusable
>>launch system from Shuttle costs.
>
>Of course it is. For both of these - solar panels and RLVs.
>
>I work in the petroleum industry. Revolutionary technoloies in our industry
>will drop the technical cost of a process by 30% in a generation. Evolutionary
>technologies, which are far more common, will drop technical process costs by
>10 - 15%. The same sort of technology improvement cost curves apply in *every*
>mature technical industry, from semiconductors to chemicals to petroleum to
>automobile manufacture.

None of this is relevant to the launch industry.

>The launch market is a very mature industry.

No, it's not. The expendable launch vehicle industry is a very mature
industry. The launch market is in its infancy.

>The chances of a technology

>breakthrough dropping costs from current launch costs by an order of magnitude
>are negligible.

The chances become essentially a hundred percent as soon as someone
makes an investment to do so.

>Current ELV costs are order of 10,000 per kg. EELV will drop this by 15%.
>This is believable. Beal is looking to drop it by another 30% or so. I might
>even believe that.
>
>The current semi-RLV market is about $20,000 per kg. The chances of the very

>next RLV being 30% less than this are low; being 50% less are negligible. The
>next RLV generation with Shuttle-type capabilities will cost $10,000 - $15,000
>per kg to LEO. Count on it. (Of course, you may claim that the next RLV will
>have fewer capabilities than Shuttle, which will trim costs; i.e., I think
>Pioneer Rocketplane was targeting around $5000/kg. This may be possible. But
>I can guarantee that if V* is ever operational, it will be more than
>$10,000/kg).

You clearly don't understand the cause of high launch costs. It has
very little to do with vehicle design, or technology. It has almost
everything to do with flight rate.

>Likewise, the next generation of space solar panels (which is also a very
>mature industry) will be order of $25/kW-hr. It will come down from present
>costs. I have no doubt. It will continue down in the future. But it will be
>a long time before it reaches ground based panel cost.
>

>Like it or not, this is the way the world works. Claims of rabid RLV advocates
>notwithstanding.

And if you built as few ground-based solar panels as space-based ones,
they would cost almost as much. *That's* how the world works.

************************************************************************
simberg.interglobal.org * 310 372-7963 (CA) 307 739-1296 (Jackson Hole)

interglobal space lines * 307 733-1715 (Fax) http://www.interglobal.org

"Extraordinary launch vehicles require extraordinary markets..."

Replace first . with @ and throw out the "@trash." to email me.
Here's my email address for autospammers: postm...@fbi.gov

gbaikie

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to

> > Current ELV costs are order of 10,000 per kg. EELV will drop this
by 15%.
> > This is believable. Beal is looking to drop it by another 30% or
so. I might
> > even believe that.
> >
> > The current semi-RLV market is about $20,000 per kg. The chances of
the very
> > next RLV being 30% less than this are low; being 50% less are
negligible.
> >

> Where is that $20K per kilogram going? It isn't to fuel. Imagine how
> much it would cost for an airline ticket if you needed a new airplane
> after every flight. Or if you took the engines and much of the rest of
> the plane apart after every flight?

The semi-RLV "market" is the Shuttle, which isn't an market.
Sea launch's Zenit gets a payload to LEO for 4367 per kg
http://www.spaceandtech.com/spacedata/elvs/zenit_specs.shtml
and another Russian made rocket can do better than this, Proton:
http://www.spaceandtech.com/spacedata/elvs/proton_specs.shtml
The American BA-2 is promising even less- 2941 per kg
http://www.spaceandtech.com/spacedata/elvs/beal_ba2_specs.shtml
Though it won't launch for at least a couple of years, meanwhile
the US government has stopped imposing a quota of Russian launches
and has stopped requiring the the launch price be within 7 percent
of other American launches. Therefore we are finally getting to a place
where this market is less interfered with by US government.
It's fairly safe bet that it would be possible to now get a Russian
Proton for about 2/3 of the list price (or less) about 3,000 per kg.
Considering the price MirCorp got for it's launches, this price could
quite bit higher than actual price. With a free market we will also
"magically" get lower American launch prices.


>
> > Likewise, the next generation of space solar panels (which is also a
very
> > mature industry) will be order of $25/kW-hr. It will come down from
present
> > costs. I have no doubt. It will continue down in the future. But
it will be
> > a long time before it reaches ground based panel cost.
> >

> Panels used on the ground can work in Space, can't they? Is there some
> vacuum or radiation or something else issue? I thought the reason
space
> rated panels were so expensive was because they needed to be as
> efficient as possible. In any case, if you can make solar cells on the
> Moon directly onsite from lunar regolith, potentially your costs could
> go way down.
>

I think the $25/kW-hr is referring to cost/price of solar panel launched
into orbit. A "ground based panel" is about 25 cent per kW-hr. When the
total price gets to around 10 cent a kW-hr, people will buy them instead
of paying for power bills, and then prices will drop even further due to
market forces.
Anyway, 25/kW-hr in orbit is probably a bit on the low side, but if you
could get electrical power at this price it's quite a bargain- for a
couple hundred buck per kg you could convert water into rocket fuel.

BrianF5070

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
>>I work in the petroleum industry. Revolutionary technoloies in our industry
>>will drop the technical cost of a process by 30% in a generation.
>Evolutionary
>>technologies, which are far more common, will drop technical process costs
>by
>>10 - 15%. The same sort of technology improvement cost curves apply in
>*every*
>>mature technical industry, from semiconductors to chemicals to petroleum to
>>automobile manufacture.
>
>None of this is relevant to the launch industry.

It's relevant to the launch industry because it is relevant to *all*
industries. All industries show these sorts of cost curves. No exceptions.
An immature industry has rates that can be higher than this, but the launch
industry is in no way immature.


>>The launch market is a very mature industry.
>
>No, it's not. The expendable launch vehicle industry is a very mature
>industry. The launch market is in its infancy.

The launch market has been around for, what, about 40 years? When was the
first Intelsat launched? So, to put it in perspective, it has been around as
long as the commercial ammonia/urea industry and 10 years longer than the
liquefied natural gas industry (which is where I spend a lot of my time). I'm
real familiar with technology cost curves.

>>The chances of a technology
>>breakthrough dropping costs from current launch costs by an order of
>magnitude
>>are negligible.
>
>The chances become essentially a hundred percent as soon as someone
>makes an investment to do so.

Disproved by past experience. Let's see, "someone" has put investments in:
Shuttle
V*
Roton
Beal
Titan
Delta
Atlas
Ariane
Long March
Proton
Pioneer Rocketplane
Kelly Spaceplane

And how many dozens of others? I know, I know - none of these were
"breakthrough." Or if they were, they were underfunded. Blah, blah, blah.
The excuse doesn't matter - the fact is that launch costs are continuing down
the same old utterly predictable cost curves. Someone investing in launch
technology doen't stand a 100% chance of an order of magnitude reduction - they
stand a 0% chance.

>You clearly don't understand the cause of high launch costs. It has
>very little to do with vehicle design, or technology. It has almost
>everything to do with flight rate.

I assure, I clearly understand this. You seem to be the one who doesn't.
Let's put it real clearly - no matter what the technology (ELV, RLV, laser
launchers, rubber bands, I don't care), the launch costs will not be
significantly lower until the market can support it. Yet you made the claim a
couple sentences ago that as soon as someone invests in breakthrough launch
technology, they are 100% assured of an order of magnitude drop in costs.

Technology cost curves and market curves march in lockstep. They provide
feedback for each other, both negative and positive. You can't force either
one, and if you try you will fail. If someone, tomorrow, developed a launch
technology optimized for 10,000 launches per year, and tried to market it, then
they would go broke. Real fast. Examples of this litter the history through
dozens of industries.

The result of all of this is that launch costs, and market usage, are
completely predictable in the long term.

>>Like it or not, this is the way the world works. Claims of rabid RLV
>advocates
>>notwithstanding.
>
>And if you built as few ground-based solar panels as space-based ones,
>they would cost almost as much. *That's* how the world works.

Perhaps. But the fact remains that the market and the technology are coupled,
and you can't treat them separately. Perhaps if you built as few ground
panels, they would have a comparable cost to space panels. But we don't.
Perhaps in some happy theoretical world with massive space panel production,
costs would approach ground based. But this world is not the world we live in.
And it won't be. Not before a time that can be easily and comfortably
predicted from existing growth rate and technology cost curves.


BrianF5070

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
>> The launch market is a very mature industry. The chances of a technology

>> breakthrough dropping costs from current launch costs by an order of
>magnitude
>> are negligible.
>>
>Why do you say this?

There are an infinite number of reasons. The technologies don't exist.
Development will take longer and be more expensive than predicted (it always
is). Operations will be more expensive than predicted (they always are). The
market won't support it. Even if someone had this technology that could drop
costs that much, they only have to compete with existing technology, so they
would only drp cost by a small amount and generate huge margins.

Examples of hundreds of technologies through all industries show the same
recurring patterns. The launch induustry is in no way unique.

Indeed, it should be the rabid advocates of cheap RLV technology who should be
defending their claims that all the lessons of history and economics will be
repealed in their specific case.

>> The current semi-RLV market is about $20,000 per kg. The chances of the


>very
>> next RLV being 30% less than this are low; being 50% less are negligible.
>>
>Where is that $20K per kilogram going? It isn't to fuel. Imagine how
>much it would cost for an airline ticket if you needed a new airplane
>after every flight. Or if you took the engines and much of the rest of
>the plane apart after every flight?

An oft-used analogy and completely irrelevant. Why use expendable robots for


Mars exploration instead of recoverable and reusable ones? The point is, which

is cheapest? ELV's are used because they are cheaper than the existing RLV.

When the next RLV is designed and built, it will only have to compete with
existing ELV's. Thus, the designing firm will only build it such that total
per flight cost will be lower than the competition. If current launch costs
are $10,000 per kg to LEO, and they think that they can get sufficient market
share for a large IRR at $7000/kg, then this is what they will design it for.
This cuts vehicle CAPEX, R&D expenditures, time to market, and program risk.

Nobody with money to invest, who is in their right mind, would risk it on
trying to cut costs to $1000/kg, when there is no business need to do so, and
valid reasons against.

>> Likewise, the next generation of space solar panels (which is also a very


>> mature industry) will be order of $25/kW-hr. It will come down from
>present
>> costs. I have no doubt. It will continue down in the future. But it will
>be
>> a long time before it reaches ground based panel cost.
>>
>Panels used on the ground can work in Space, can't they? Is there some
>vacuum or radiation or something else issue? I thought the reason space
>rated panels were so expensive was because they needed to be as
>efficient as possible. In any case, if you can make solar cells on the
>Moon directly onsite from lunar regolith, potentially your costs could
>go way down.

No, ground panels cannot work in space, unmodified. The radiation deteriorates


panels without protection, they need charge buildup protection, they need
automated deployment, and so on.

Space panels are not so expensive because they need to be efficient - they are

gbaikie

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
In article <3983CC8C...@mail.com>,
Bill Bonde <std...@mail.com> wrote:

>
>
> BrianF5070 wrote:
> >
> > >Where is that $20K per kilogram going? It isn't to fuel. Imagine
how
> > >much it would cost for an airline ticket if you needed a new
airplane
> > >after every flight. Or if you took the engines and much of the rest
of
> > >the plane apart after every flight?
> >
> > An oft-used analogy and completely irrelevant. Why use expendable
robots for
> > Mars exploration instead of recoverable and reusable ones? The
point is, which
> > is cheapest? ELV's are used because they are cheaper than the
existing RLV.
> >
> ELVs are cheaper because the demands on the system are so high, parts
> don't last very long, so why waste time taking the system apart after
> every use? Just use it once. But if you can reach a point where you
> don't have to fix mostly everything after every launch costs have to
> come way down.

I think the main difference is that a ELV uses stages, at least 2
stages, lower atmosphere and upper atmosphere. This allows nozzles
to be more optimized, and you can use different fuels- denser fuel to
get the rocket going as fast as possible. And upper stage a fuel such as
LH&LOX to increase payload size. Plus you can get rid of part the rocket
mass: the big engine, tanks, stronger structure of lower stage.
Now, a RLV can use stages but it needs some mechanism to recover, such
as a parachute. Now, unless the stage has wings which can fly back, you
need to send crew to recover. Depending on terrian (or ocean) there will
be certain problems you could encounter. Also how soon the stage is
dropped will also be a factor- a solid stage burns quickly, therefore
will drop at lower attitude and slower speed, but you also aren't saving
much of value. If using flybacks or any other elabrorate means you will
be added a percentage of deadweight (as far as your purpose of getting
something into orbit) plus you are added to the cost of the total cost
of building the stage and maintaining it. Plus there's chance of failure
of this stage safely landing.

Rand Simberg

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
On 30 Jul 2000 04:20:54 GMT, in a place far, far away,

brian...@aol.com (BrianF5070) made the phosphor on my monitor glow
in such a way as to indicate that:

>>None of this is relevant to the launch industry.


>
>It's relevant to the launch industry because it is relevant to *all*
>industries. All industries show these sorts of cost curves. No exceptions.
>An immature industry has rates that can be higher than this, but the launch
>industry is in no way immature.

Yes, it is.

>>>The launch market is a very mature industry.
>>

>>No, it's not. The expendable launch vehicle industry is a very mature
>>industry. The launch market is in its infancy.
>
>The launch market has been around for, what, about 40 years? When was the
>first Intelsat launched? So, to put it in perspective, it has been around as
>long as the commercial ammonia/urea industry and 10 years longer than the
>liquefied natural gas industry (which is where I spend a lot of my time). I'm
>real familiar with technology cost curves.

Maturity of an industry does not come from the mere passage of
time--it comes from developing it to the point that economies of scale
have been fully employed, and the output product has been
commoditized. The launch industry is at the very bottom of the curve
in this respect.

>>>The chances of a technology

>>>breakthrough dropping costs from current launch costs by an order of
>>magnitude


>>>are negligible.
>>
>>The chances become essentially a hundred percent as soon as someone
>>makes an investment to do so.
>

>Disproved by past experience. Let's see, "someone" has put investments in:

>Shuttle

Thirty years ago, when technology actually was somewhat of a problem,
and the goal was less to reduce launch costs than to keep NASA busy
after Apollo.

>V*

No significant investment has been made.

>Roton

Only a small percentage of the investment was made.

>Beal

Jury out.

>Titan

No goal to reduce costs by order of magnitude.

>Delta

Ditto.

>Atlas

Ditto.

>Ariane

Ditto.

>Long March

Ditto.

>Proton

Ditto.

>Pioneer Rocketplane
>Kelly Spaceplane

No significant funding.

>And how many dozens of others? I know, I know - none of these were
>"breakthrough." Or if they were, they were underfunded. Blah, blah, blah.

If you know, then why waste bandwidth posting a bunch of irrelevant
examples?

>The excuse doesn't matter - the fact is that launch costs are continuing down
>the same old utterly predictable cost curves. Someone investing in launch
>technology doen't stand a 100% chance of an order of magnitude reduction - they
>stand a 0% chance.

Data?

>>You clearly don't understand the cause of high launch costs. It has
>>very little to do with vehicle design, or technology. It has almost
>>everything to do with flight rate.
>
>I assure, I clearly understand this. You seem to be the one who doesn't.
>Let's put it real clearly - no matter what the technology (ELV, RLV, laser
>launchers, rubber bands, I don't care), the launch costs will not be
>significantly lower until the market can support it. Yet you made the claim a
>couple sentences ago that as soon as someone invests in breakthrough launch
>technology, they are 100% assured of an order of magnitude drop in costs.

Then you're saying that it's *not* a technology problem (i.e., you're
changing your story)? Then how can you claim that Shuttle is
representative of future launch vehicle costs, or that ISS solar cell
costs are representative of future space solar cell costs?

>Technology cost curves and market curves march in lockstep. They provide
>feedback for each other, both negative and positive. You can't force either
>one, and if you try you will fail. If someone, tomorrow, developed a launch
>technology optimized for 10,000 launches per year, and tried to market it, then
>they would go broke. Real fast. Examples of this litter the history through
>dozens of industries.

Data?

>>And if you built as few ground-based solar panels as space-based ones,
>>they would cost almost as much. *That's* how the world works.
>
>Perhaps. But the fact remains that the market and the technology are coupled,
>and you can't treat them separately. Perhaps if you built as few ground
>panels, they would have a comparable cost to space panels. But we don't.
>Perhaps in some happy theoretical world with massive space panel production,
>costs would approach ground based.

That is the world that is being proposed, hence your claim that
current solar cell costs have anything to do with costs in that world
are nonsense.

>But this world is not the world we live in.
> And it won't be. Not before a time that can be easily and comfortably
>predicted from existing growth rate and technology cost curves.

That was the time that was being predicted.

earthf...@my-deja.com

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
In article <20000723014453...@ng-df1.aol.com>,

brian...@aol.com (BrianF5070) wrote:
> wm...@my-deja.com writes:
>
> Lots and lots of snipped stuff concluding with:
>
> >meaning solar based
> >power requires 832 times the cost on Earth for collector, and energy
> >management than an equivalent power in space.
> >
>
> Or you simply take the CAPEX of the solar panels for the latest
generation of
> GEO sats ($50 - $100 MM) and divide by the expected power output (15
kW at
> start of life; 9 kW at end of life) integrated over the 15 year rated
life,
> ignore OPEX (which is small relative to CAPEX), and find that current
space
> solar power costs at least $30 - $60 per kW-hr.
>
> Then you compare with a current terrestrial solar system, which is
going for
> around $0.20 to $0.30 per kW-hr, depending on location, assumptions,
etc.
>
> Hmmm. It's not really that difficult a comparison after all, is it?
>
>
Excellent analysis.
What is the current cost per kilowatt-hour on earth for natural gas
powered gas turbines? Wind power? Hydro-power? Nuclear power?

Ian St. John

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
<earthf...@my-deja.com> wrote in message
news:8m1p9e$avd$1...@nnrp1.deja.com...
<snip>
> Excellent analysis.

Stupid analysis. It ignores economies of scale, improvements in technology,
and costs for unmanned heavy lift vs manned missions. Not to mention that
when you make a satellite solar panels they are higly customized and
engineered for *no* service. SPS design will be much more based on modular
simple pieces with some contingency planning.

But you just wanted to ridicule, so I guess the analysis is 'excellent' in
your terms.

Another my-deja twit. Ugh.


James A Davis

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
Rand Simberg wrote:

> >The chances of a technology
> >breakthrough dropping costs from current launch costs by an order of magnitude
> >are negligible.
>
> The chances become essentially a hundred percent as soon as someone
> makes an investment to do so.

Rand, it seems to me that you place yourself in an untenable position by
advancing this argument. Your premises seem to be:

1. Launch costs can be reduced by an order of magnitude with a large
enough investment.

2. No one has made such an investment.

The question becomes: given (1), why (2)?

Possible answers:

1. The financial reward for lowering launch costs are inadequate for the
investment required.

You've made it clear that you don't believe this.

2. There is a conspiracy afoot (a la William Mook) to ensure that such
investment is not made, either by direct intimidation or by spreading
fear, uncertainty, and doubt about various plans.

No recent (last 15 years) evidence of this although NASA and the US
government did try to eliminate potential shuttle competitors in the
1975-1985 period. You've indicated in the past that you think
NASA/Boeing/LM stifle investment by their very existence but haven't
given any specific examples (Joe Doe was going to invest $15 million in
Rotary but his consultant Bob Smith, who works for NASA, told him Boeing
was thinking about doing something like Roton).

3. Premise 1, although true, is honestly believed to be false by
virtually all potential investors.

If this is the case then you would be in the uncomfortable position of
explaining why Rand Simberg is right about premise 1 and all the money
in the world is wrong.

Are there other explanations I've overlooked?

Jim Davis

Scott Lowther

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
Rand Simberg wrote:
>
> Here's what
> John Schilling said the most recent time that you did it, a couple of
> months ago.
>
> "
> If you ever see someone building, or even proposing, an actual solar
> power satellite with an antenna big enough to serve as a space-based
> death ray weapon, fell free to cry foul, scream bloody murder, and
> whatnot. If you've done your math right, odds are a lot of us here
> will join you."

Foul. Bloody murder. Whatnot.

Take a look at William Mook's proposals for 200+ solar intensity laser
SPS systems. Sure, not quite instant death, but a dandy weapon
nonetheless.

Now, if the original statment had been: "If you ever see someone
RATIONAL building, or even proposing..."

George William Herbert

unread,
Jul 30, 2000, 3:00:00 AM7/30/00
to
Michael Kagalenko <mkag...@lynx.dac.neu.edu> wrote:
> I have addressed this falsehood a few times over. As the microwave
>beaming can't be economical if there is too much land area taken
>by the receiving array, a practical SPS will be useable as
>a weapon. And, of course, beams from several arrays
>can be combined to increase the weapon potential.

Uneconomical at twenty cents a square meter? That's $200,000 per
square kilometer; a 20 km diameter array only would cost $60 million,
and at a very very *very* low end 10 watts/m^2 could bring in around
3 gigawatts per array. If we assume power prices drop insanely to
one cent per kWh, that's worth an income stream of $30,000/hr,
$720,000/day, $262 million/year.

Tell us again how land costs are going to be unreasonable.

> And, as before, the second point , -danger to planes and
>satellites,- is conspicuously ignored, as there exists not even
>a fallacious excuse like the one by Schilling to address that.

At reasonable power levels... tens to perhaps a hundred watts per
square meter... the *power* coming down is a fraction of sunlight's
1.35 kW/m^2 noon peak. It has some electromagnetic interferance
effects with electronics, but shielding isn't going to be all that
difficult if the frequency range is narrow, and establishing an
air traffic control exclusion zone around the array should reduce
the risk acceptably.

Sure, someone could fly into it and get zapped if nobody turned
off the juice and their plane was particularly badly unshielded.
Someone could get zapped flying into a missile test range if
the range control didn't notice them there, too. That hasn't
happened so far in the history of aviation, to my recollection.
I, and society at large, should be able to buy into that level
of risk.


-george william herbert
gher...@retro.com


Rand Simberg

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
On Sun, 30 Jul 2000 18:50:21 -0500, in a place far, far away, James A
Davis <jimd...@primary.net> made the phosphor on my monitor glow in

such a way as to indicate that:

>> >The chances of a technology


>> >breakthrough dropping costs from current launch costs by an order of magnitude
>> >are negligible.
>>
>> The chances become essentially a hundred percent as soon as someone
>> makes an investment to do so.
>
>Rand, it seems to me that you place yourself in an untenable position by
>advancing this argument.

Gee, I don't think so... :-)

>Your premises seem to be:
>
>1. Launch costs can be reduced by an order of magnitude with a large
>enough investment.

Yes.

>2. No one has made such an investment.

Also yes.

>The question becomes: given (1), why (2)?
>
>Possible answers:
>
>1. The financial reward for lowering launch costs are inadequate for the
>investment required.

>You've made it clear that you don't believe this.

No, I don't believe it, but I don't have the money. The correct
answer is that the perceived financial reward for lowering launch
costs are inadequate for the investment and market risk required.

And note that I didn't say that there was a hundred percent chance of
making a profit--just of reducing costs.

My point is that cost reduction is achievable on a per pound basis by
building a system that can launch lots of pounds and then using it to
launch lots of pounds. The risk is not that the cost target won't be
hit, but rather that there will not be a sufficient customer base who
is willing to pay more than your costs.

>2. There is a conspiracy afoot (a la William Mook) to ensure that such
>investment is not made, either by direct intimidation or by spreading
>fear, uncertainty, and doubt about various plans.

No, no conspiracies, but institutional incentives can have the same
effect.

>No recent (last 15 years) evidence of this although NASA and the US
>government did try to eliminate potential shuttle competitors in the
>1975-1985 period. You've indicated in the past that you think
>NASA/Boeing/LM stifle investment by their very existence but haven't
>given any specific examples (Joe Doe was going to invest $15 million in
>Rotary but his consultant Bob Smith, who works for NASA, told him Boeing
>was thinking about doing something like Roton).

American Rocket can provide some real examples.

>Are there other explanations I've overlooked?

Just the one provided above.

Michael Kagalenko

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
Ian St. John (ist...@spamcop.net) wrote
]<earthf...@my-deja.com> wrote in message

The two biggest problems with the space solar arrays beaming power on
Earth are that they are a potential weapon and a hazard for planes
and satellites.

Rand Simberg

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
On 31 Jul 2000 01:21:22 GMT, in a place far, far away,
mkag...@lynx01.dac.neu.edu (Michael Kagalenko) made the phosphor on

my monitor glow in such a way as to indicate that:

> The two biggest problems with the space solar arrays beaming power on


> Earth are that they are a potential weapon and a hazard for planes
> and satellites.

Michael, you never seem to tire of posting this idiocy. Here's what


John Schilling said the most recent time that you did it, a couple of
months ago.

"As has been pointed out to you many times, no actual SPS proposal has
an antenna of sufficient size to serve as a lethal weapon. No actual
SPS proposal has an antenna within even an order of magnitude of
sufficient size to serve as a lethal weapon. An antenna of sufficient
size to serve as a lethal weapon, would be a horrendously expensive
construction of limited utility to an actual SPS system, and will not
be built.

If you ever see someone building, or even proposing, an actual solar
power satellite with an antenna big enough to serve as a space-based
death ray weapon, fell free to cry foul, scream bloody murder, and
whatnot. If you've done your math right, odds are a lot of us here
will join you.

Until then, your time would be better spent studying than screaming,
as you've got a lot to learn on the subject."


You obviously have been avoiding your studies again...

Michael Kagalenko

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
Rand Simberg (simberg.i...@trash.org) wrote
]On 31 Jul 2000 01:21:22 GMT, in a place far, far away,

]mkag...@lynx01.dac.neu.edu (Michael Kagalenko) made the phosphor on
]my monitor glow in such a way as to indicate that:
]
]> The two biggest problems with the space solar arrays beaming power on
]> Earth are that they are a potential weapon and a hazard for planes
]> and satellites.
]
]Michael, you never seem to tire of posting this idiocy. Here's what
]John Schilling said the most recent time that you did it, a couple of
]months ago.
]
]"As has been pointed out to you many times, no actual SPS proposal has
]an antenna of sufficient size to serve as a lethal weapon. No actual
]SPS proposal has an antenna within even an order of magnitude of
]sufficient size to serve as a lethal weapon. An antenna of sufficient
]size to serve as a lethal weapon, would be a horrendously expensive
]construction of limited utility to an actual SPS system, and will not
]be built.
]
]If you ever see someone building, or even proposing, an actual solar
]power satellite with an antenna big enough to serve as a space-based
]death ray weapon, fell free to cry foul, scream bloody murder, and
]whatnot. If you've done your math right, odds are a lot of us here
]will join you.
]
]Until then, your time would be better spent studying than screaming,
]as you've got a lot to learn on the subject."
]
]
]You obviously have been avoiding your studies again...

I have addressed this falsehood a few times over. As the microwave


beaming can't be economical if there is too much land area taken
by the receiving array, a practical SPS will be useable as
a weapon. And, of course, beams from several arrays
can be combined to increase the weapon potential.

And, as before, the second point , -danger to planes and

wm...@my-deja.com

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
Your speculations about my motivations are off the mark.

Your advice is pointless.

With respect to your assertion that my numbers have been 'shot down' or
that my numbers are somehow inconsistent, you are in error.

I you would like to discuss your error meaningfully, please provide the
factual basis of your misperception and I will attempt to explain your
error to you.

William Mook
Orbatek, Inc.


In article <8lvqpp$3u1$1...@slb3.atl.mindspring.net>,


"Terrell Miller" <terrel...@mindspring.com> wrote:
> God, this guy is persistent, give him that! Every time someone shoots
down
> his numbers, he changes the subject and comes up with a whole bunch
of other
> numbers.
>
> Looks like he's either an overeducated idiot, or he's aggressively
looking
> for investors who are. My advice: don't give this guy any of your
money!
>
> --
> Terrell Miller, Ordo Pantheris
> terrel...@mindspring.com
>
> "Faith is love at its most intense and selfless."
> -Linda Nagata
>
> <wm...@my-deja.com> wrote in message
news:8lg15a$kad$1...@nnrp1.deja.com...
> > In article <397AA2DE...@interglobal.org>,
> > Rand Simberg <sim...@interglobal.org> wrote:
> > > BrianF5070 wrote:
> > >

> > > > Or you simply take the CAPEX of the solar panels for the latest
> > generation of
> > > > GEO sats ($50 - $100 MM) and divide by the expected power output
> > (15 kW at
> > > > start of life; 9 kW at end of life) integrated over the 15 year
> > rated life,
> > > > ignore OPEX (which is small relative to CAPEX), and find that
> > current space
> > > > solar power costs at least $30 - $60 per kW-hr.
> > >

> > > It's not possible to usefully extrapolate current costs of space
> > solar panels to
> > > potential cost of mass-produced space solar panels for power
> > satellites, any
> > > more than one can infer the costs of a well-designed, high-flight-

> > rate reusable


> > > launch system from Shuttle costs.
> > >
> >
> >

> > That's right! And besides, Brian didn't compare earth based solar
to
> > space based solar.
> >
> > But Briand did have a point about costs. Maybe I shouldn't have
used
> > the word 'cost' in my original posting. But its hard to find a
> > suitable replacement.
> >
> > One can talk about the costs of things in many ways.
> >
> > Cost can be related to the infrastructure in the market to make a
> > thing based on historical patterns of investment. That's what
Brian
> > looked at.
> >
> > One can also look at fundamentals to attain ultimate outcomes.
That's
> > what I was speaking of.
> >

> > You need some 832 times the collector area in North America to

Rand Simberg

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
On 31 Jul 2000 02:54:11 GMT, in a place far, far away,

mkag...@lynx01.dac.neu.edu (Michael Kagalenko) made the phosphor on
my monitor glow in such a way as to indicate that:

>]You obviously have been avoiding your studies again...


>
> I have addressed this falsehood a few times over. As the microwave
>beaming can't be economical if there is too much land area taken
>by the receiving array,

Data?

>a practical SPS will be useable as
>a weapon. And, of course, beams from several arrays
>can be combined to increase the weapon potential.

This of course ignores John's admonition to point out to us whenever
someone actually proposes such a system. No one has, to date, other
than possibly Mr. Mook.

> And, as before, the second point , -danger to planes and
>satellites,- is conspicuously ignored, as there exists not even
>a fallacious excuse like the one by Schilling to address that.

And just what is the danger to vehicles that are enclosed in an
aluminum faraday shield?

wm...@my-deja.com

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
In article <20000730002054...@ng-fk1.aol.com>,

brian...@aol.com (BrianF5070) wrote:
> >>I work in the petroleum industry. Revolutionary technoloies in our
industry
> >>will drop the technical cost of a process by 30% in a generation.
> >Evolutionary
> >>technologies, which are far more common, will drop technical
process costs
> >by
> >>10 - 15%. The same sort of technology improvement cost curves
apply in
> >*every*
> >>mature technical industry, from semiconductors to chemicals to
petroleum to
> >>automobile manufacture.
> >
> >None of this is relevant to the launch industry.
>
> It's relevant to the launch industry because it is relevant to *all*
> industries. All industries show these sorts of cost curves. No
exceptions.
> An immature industry has rates that can be higher than this, but the
launch
> industry is in no way immature.


This is bogus. Lets talk specifics rather than hard to dissect
generalities. The launch market is immature because minor investments
can make big differences in the market. The European Space Agency for
example spent something like $4 billion to build an ELV using then
current technology for the specific purpose of capturing the commercial
space launch market. As a result within 18 months of achieving flight
readiness, ESU captured over half the commercial launch market and
still retains a large fraction today!

Its not time or even investments, since much of the investments made by
government are politically motivated, and not motivated by launch
efficiency. The difficulties engineers have in reducing costs in an
environment that pays on a cost plus basis are legion. Government
inefficiency is legend. So, quit pretending the launch vehicle market
is mature just because its 50 years old and has had some $500 billion
spent on it worldwide. Very little of that money has been spent to
reduce costs or increase operational efficiencies, and a strong bias
against these factors created by cost plus contracting keeps the status
quo going.


>
> >>The launch market is a very mature industry.
> >
> >No, it's not. The expendable launch vehicle industry is a very
mature
> >industry. The launch market is in its infancy.
>
> The launch market has been around for, what, about 40 years? When
was the
> first Intelsat launched? So, to put it in perspective, it has been
around as
> long as the commercial ammonia/urea industry and 10 years longer than
the
> liquefied natural gas industry (which is where I spend a lot of my
time). I'm
> real familiar with technology cost curves.

If it were a simple learning curve problem, then your experience would
be germaine. But it is not. Quit pretending it is. We've ascended
several learning curves and the nature of how funding is awarded at
NASA labs ensures that we keep ascending them. Witness the billions
spent on NASP. Its not the role of government labs to increase
efficiencies but to develop new technologies. Its not the role of the
astronautics sector of the aerospace business to create operational
efficiencies since doing so *reduces* profits, and does not increase
them.

> >>The chances of a technology
> >>breakthrough dropping costs from current launch costs by an order of
> >magnitude
> >>are negligible.
> >
> >The chances become essentially a hundred percent as soon as someone
> >makes an investment to do so.
>

> Disproved by past experience.

Nonesense. You are claiming we can't achieve 10 fold reductions in
launch costs even with future technology! Hell, we could achieve a ten
fold reduction with 20 year old technologies if someone with sufficient
resources and political clout would put them together in the right mix
and airframe! Any rocket scientist knows this, which is why so many of
them have gone off half-cocked and undercapitalized and attempted to
build a multi-stage reusable vehicle using one or another variant of
existing technologies mix.

Let's see, "someone" has put investments in:
> Shuttle

> V*
> Roton
> Beal
> Titan
> Delta
> Atlas
> Ariane
> Long March
> Proton
> Pioneer Rocketplane
> Kelly Spaceplane

The experiences you cite are not germaine, the reasons are as long as
your list. The single instance of ESU capturing a 50% market share in
18 months with a measely $4 billion investment is a credible
counterpoint to your list. When the investments has been sufficient
and correctly focused with no countervaling political concerns, results
are easily forthcoming. Your claim that no foreseeable technology can
wring out a 10 fold improvement in launch efficiencies doesn't stand.
Your defenses of this laughable position are knuckleheaded and lame.

I would like to say a word about Shuttle since its on your list.
Shuttle was buggered up in committee after the folks in Huntsville came
up with a viable design. If we would have stopped at an H2 variant for
the SSME rather than pushing the performance of the SSME way beyond
where it needed to be; and put those savings in a fully reusable first
stage; and accepted a slightly higher areal mass for the shuttle tile;
and flown the Shuttle at the rates projected for it in the late 60s;
the Shuttle would have achieved many of the promises made for it. But
the Shuttle was killed because having that many high mass missions
meant that NASA would need to take lots of risks on orbit with their
astronauts, and that is something NASA was not set up to do. With a
moonbase, an orbiting hotel, outposts on Mars, and three manned flights
to orbit a week, Apollo 13 type disasters would happen once or twice a
year. A crew in transit to Mars might be killed by a solar flare once
or twice a decade. And those deaths wouldn't be fast, they'd be slow,
and painful, and ugly, and televised around the world. NASA opted for
the safest route possible. Namely, doing nothing, and blame as their
reason the very very high costs of space launch. In reality, manned
space exploration is risky dangerous business. Doing variations of
Mercury flights year in and year out is not.


> And how many dozens of others? I know, I know - none of these were
> "breakthrough."

I'm not surprised guy. I doubt you would recognize a breakthrough
technology if it bit you in the arse.

>Or if they were, they were underfunded. Blah, blah, blah.

Billions can be spent on programs, and programs that look good going in
can have their performance sapped by mismanagement and shifting
performance goals. This is clearly true in cases like NASP or
Shuttle.

> The excuse

Not excuses fact. When Europe can steal half the commercial launch
market it shows just how little we actually do with the money we do
spend on space and just how big a target we've become. This will be
confirmed again when Japan takes most of the rest of the commercial
launch market and builds most of the worlds satellites, and takes over
the broadband space based telecom business the US has now abandoned.
When Japan solves her energy problems with viable SPS, we'll see it
confirmed again. When China flies to mars on a nuclear pulse rocket we
could've built 50 years ago, maybe you'll understand what we're talking
about here. But by then, it will be too late.

> doesn't matter - the fact is that launch costs are continuing down
> the same old utterly predictable cost curves.

Please cite a reference to this for commercial launch costs, with all
external funding removed!

Even if true, its beside the point because, its also utterly
predictable that if NASA switched from cost plus to fixed price
reimbursement the US aerospace business could halve launch costs within
a year, making hash of your curves.

Its also utterly predictable that an unpiloted fully reusable
multistage rocket could launch Atlas class payloads for about $2
million each after investing less than $2 billion using completely off
the shelf technologies. Thats a reduction to about 50:1 - again making
hash of the curves you

> Someone investing in launch
> technology doen't stand a 100% chance of an order of magnitude
reduction - they
> stand a 0% chance.

And you are 100% wrong because you are misapplying what you think you
know in the oil and gas business inappropriately to the space launch
business. If you knew the first thing about rocket design, you'd
understand how overpriced our current ELV and manned fleets are. If
you knew the first things about the history and politics of space in
the US you'd understand why.

> >You clearly don't understand the cause of high launch costs. It has
> >very little to do with vehicle design, or technology. It has almost
> >everything to do with flight rate.
>
> I assure, I clearly understand this. You seem to be the one who
doesn't.

This is pointless bickering. You clearly understand something relating
to cost analysis in the oil and gas business as you claim. However,
your clear misapplication of what you know also makes it clear you
don't know the first thing about rocket design, and why we use
inefficient rockets rather than efficient ones, and why things are that
way.

> Let's put it real clearly - no matter what the technology (ELV, RLV,
laser
> launchers, rubber bands, I don't care),

This covers your technical ignorance quite nicely doesn't it? I mean,
if you could take a knowing look at existing ELV allowable expenditures
you could come up with many ways to save costs. But try to get it off
the drafting table and into production. This isn't the same kind of
business as the oil and gas business. You don't make more money
cutting costs. You make less. That's why much better and more
experienced rocket scientists than you or I are stymied. Degarabedian
and his Big Dumb Booster concepts, are a case in point. Heck, it was
hell getting a laptop on board the Shuttle.

Fact is, we could cut costs to half their current levels immediately,
and reduce them to 2% of current costs in 18 months with 20 year old
technology. The problem with that is the political battle that would
ensue in fighting over the savings. That's not the only one. A bigger
issue is if we have several flights to orbit a week and bases across
the inner solar system, at current prices or higher, we'd have
challenger type disasters as often as we have aircrashes, and Apollo 13
type disasters once or twice a year, and worse less frequently. Whole
villages get wiped out if a volcano burps and suffocates everyone.
What would the consequences be for NASA if a solar flare or a gas leak,
or a technical glitch kills a village of folks on the moon? NASA's not
ready for this. They've made their decision, and they've blamed high
launch costs.

> the launch costs will not be
> significantly lower until the market can support it. Yet you made
the claim a
> couple sentences ago that as soon as someone invests in breakthrough
launch
> technology, they are 100% assured of an order of magnitude drop in
costs.

I'm going one further. If someone invests a piddling $2 billion in 20
year old technology they'd be 100% assured of two orders of magnitude
drop in costs! That's what inspires all those rocket scientists to
start up a rocket business on a shoestring.

> Technology cost curves and market curves march in lockstep. They
provide
> feedback for each other, both negative and positive. You can't force
either
> one, and if you try you will fail.

Can you cite references for this important finding?

> If someone, tomorrow, developed a launch
> technology optimized for 10,000 launches per year, and tried to
market it, then
> they would go broke. Real fast.

Yes precisely. The shuttle is cheap at 100 launches per year (and with
proper investments to improve ground support and launch logistics using
well known technologies that have deteriorated since Apollo days).

>Examples of this litter the history through
> dozens of industries.

Yes. But counterexamples exist. Railroads in America during the 19th
century for example. They needed volume to make them pay. There was
an extensive period of development where many railroads came and went.
But then, once all the details of developing the MARKET as well as the
train were fully worked out, great transcontinental powerhouses arose.
Space faring has all the earmarks that it is overdue for such a
dramatic transformation, and that's what I'm talking about.

> The result of all of this is that launch costs, and market usage, are
> completely predictable in the long term.

Geez. Your argument reminds me of the fella that proved the bigger
your CPU the less costly your calculation. This lead IBM precisely in
the wrong direction in the 1970s. When the costs of circuits were
dropping they were spending millions trying to assemble them into
larger and larger mainframes. Meanwhile, two hired hands working at an
apple orchard created the personal computing industry - despite all the
lofty pronouncements of the experts. Continued reliance on these
misapplications of knowledge ensured IBMs failure to enter the market
in a timely or decisive way. Leading them to farm out the OS
development for their PC when it did come, to a small Washington based
company Microsoft. While they were frying their bigger mainframe fish
they failed to negotiate strongly with this little startup writing
software for a little computer that could never compete in real
business - and the course of the next 30 years of computing was set,
all to the detriment of the big players.

Comparable misanalysis worked against Xerox - despite their investments
in PARC - whose important results were discounted by the experts -
leading to a situation where another small cap firm Hewlett Packard now
dominates the computer printer market, and through that door, enter the
scanner and copying business to Xerox's detriment.

> >>Like it or not, this is the way the world works.

A look at the history of how the big computing and information
companies lost the central computing and information markets, resulting
in the creation of trillions of dollars of brandnew wealth, and how
computing was never really regulated by government in the first place
leading to a rebirth of libertine practice in America, indicates how
the world works. Your misanalysis doesn't withstand the test of
reality. Sorry.

>>Claims of rabid RLV
> >advocates
> >>notwithstanding.
> >

> >And if you built as few ground-based solar panels as space-based
ones,
> >they would cost almost as much. *That's* how the world works.
>
> Perhaps.

No perhaps about it. You already agreed with this statement above.
Sheez.

> But the fact remains that the market and the technology are coupled,

True.

> and you can't treat them separately.

True.

> Perhaps if you built as few ground
> panels, they would have a comparable cost to space panels. But we
don't.

Again, quit the pussyfooting. Fact is, you *can* figure out the
economic volumes needed to meet demands of investors and customers for
a wide range of existing technologies. Fact is, we already know how to
build vastly cheaper space vehicles and know what markets they will
serve and the launch rates that would ensue.

> Perhaps in some happy theoretical world with massive space panel
production,
> costs would approach ground based.

Not perhaps. For sure. For sure if the investments are made to reduce
costs the volume would increase to support the investments. That's the
whole point.

> But this world is not the world we live in.
> And it won't be.

Nonesense. The real world used to have people like E.F.Hutton (the man
whose company bears his name). He was an economic advisor to
Carnegie. Before Carnegie the steel industry was highly fragmented,
and suffered from technological innovation and was not very popular.
The market was limited and anyone with a few dollars could build a
small steel mill and push prices lower.

E.F. Hutton saw there would be economies in large scale production. He
also saw that the whole world could begin using steel in innovative
ways. Carnegie was not well established before he met Hutton.
Carnegie was an uneducated latecomer to the steel business, trading in
steel and such among better capitalized and better educated and more
stable competitors. Carnegie bought Hutton's idealistic vision of a
steel based economy - and made it happen. He integrated steel
production in large efficient plants, and then flooded the market with
far more steel than the 'experts' thought could be supported. As a
result steel displaced the use of wood in buildings, carriages, crates,
boxes, glass jars, furnishings, you name it, a steel version was
created and agressively marketed. This in addition to steel rails and
steel for engines and train cars.

This is the real world. This is the world that says loudly - the space
frontier is here to solve all our world's needs - all we have to do is
reach out and take the bounty that is offered us. But experts now,
like experts then, will tell you the past must always predict the
future - no variation can exist. These experts are fools. Deep down
they know it!

> Not before a time that can be easily and comfortably
> predicted from existing growth rate and technology cost curves.

You ought to read some of the correspondence between Hutton and other
steel producers before he met Carnegie! They said the same damn
thing. And while they were waiting, Carnegie created US Steel and
swamped them with steel so low in cost they had to shut their doors.

There was a period of gradual increase in steel production, but there
is a discontinuity before Carnegie/Hutton and after. This
discontinuity exists in space travel technologies and markets. Ignore
it at your own risk.

Those who plan seriously the future of the United States and seek to
make strategy based on those plans take note! We are dooming our
nation to long term second tier status with our current policies. Open
the space frontier now and share the riches with the rest of the
world. We'll all end up equal eventually, but by doing this the US
will remain first among equals, and that will be enough!

Rand Simberg

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
On 30 Jul 2000 21:43:14 -0700, in a place far, far away,
gher...@gw.retro.com (George William Herbert) made the phosphor on my

monitor glow in such a way as to indicate that:

>Uneconomical at twenty cents a square meter? That's $200,000 per


>square kilometer; a 20 km diameter array only would cost $60 million,
>and at a very very *very* low end 10 watts/m^2 could bring in around
>3 gigawatts per array. If we assume power prices drop insanely to
>one cent per kWh, that's worth an income stream of $30,000/hr,
>$720,000/day, $262 million/year.
>
>Tell us again how land costs are going to be unreasonable.
>

>> And, as before, the second point , -danger to planes and
>>satellites,- is conspicuously ignored, as there exists not even
>>a fallacious excuse like the one by Schilling to address that.
>

>At reasonable power levels... tens to perhaps a hundred watts per
>square meter... the *power* coming down is a fraction of sunlight's
>1.35 kW/m^2 noon peak. It has some electromagnetic interferance
>effects with electronics, but shielding isn't going to be all that
>difficult if the frequency range is narrow, and establishing an
>air traffic control exclusion zone around the array should reduce
>the risk acceptably.
>
>Sure, someone could fly into it and get zapped if nobody turned
>off the juice and their plane was particularly badly unshielded.
>Someone could get zapped flying into a missile test range if
>the range control didn't notice them there, too. That hasn't
>happened so far in the history of aviation, to my recollection.
>I, and society at large, should be able to buy into that level
>of risk.

George, don't confuse poor Mr. Kagelenko with facts--he's posted this
nonsense so often, and in the face of irrefutable refutations, that
it's obvious that he's impervious.

BrianF5070

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
>Excellent analysis.
>What is the current cost per kilowatt-hour on earth for natural gas
>powered gas turbines? Wind power? Hydro-power? Nuclear power?

Each project is highly site specific. That being said, the tendencies and
broad outlines are clear.

Natural gas is currently running on order of $500 per installed kW. OPEX
depends on natural gas prices, which are currently about $4/ mcf and are
forecasted at around $3.2/mcf for the next five years. Overall, you can figure
on $0.03 to $0.04/kW-hr for a large combined cycle in baseload duty.

Wind power has been averaging about $3000 per installed kW, but the price of
turbines is continuing down. Total cost is highly sensitive to capital and
interest rates since operating costs are pretty low. Figure $0.12/kW-hr all
told.

Hydropower in the US is running at about $0.03/kW-hr, but this is pure
operating cost since the capital base is completely installed. Almost no one
is able to put in new hydro in the US anymore. In South America, hydro is
going in very quickly and is typically running around $0.08/kW-hr all up.

Some of the new nuclear designs should be able to go in for order of
$1000/installed kW, but the price of uranium is increasing quite rapidly, and
environmental constraints increase the installation costs considerably above
the technical cost. If you want to install nuclear, figure on at least $0.10
to $0.15 per kW-hr to get a reasonable return. More in the US.


BrianF5070

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
>Maturity of an industry does not come from the mere passage of
>time--it comes from developing it to the point that economies of scale
>have been fully employed, and the output product has been
>commoditized. The launch industry is at the very bottom of the curve
>in this respect.

Of for crying out loud. Economies of scale are *never* fully employed. Output
products are *never* fully commoditized. In large part, these factors are what
continues to drive down the cost of technologies in mature markets. All
technologies, including natural gas, automobiles, computers, and rockets. The
launch industry is middling near the bottom compared to some markets, but
nowhere near enough the bottom to be classed as immature.

If you ever have the chance to walk the floor at LockMart's Titan/Atlas
assembly facility, you'll see the degree to which the vehicles are
commoditized. You'll see how economies of scale are employed. You'll realize
that there is no way this manufacturing facility (or industry) is immature.

>>And how many dozens of others? I know, I know - none of these were
>>"breakthrough." Or if they were, they were underfunded. Blah, blah, blah.
>
>If you know, then why waste bandwidth posting a bunch of irrelevant
>examples?

I know what you will say. I also don't agree with it. There have been dozens
of programs. They all have received investment. They have *all,* without
exception, failed to drive down costs by an order of magnitude from the
immediate previous generation. All the ones that have actually brought
vehicles to market have followed the nice typical cost curves that one expects
to see.

I could not care less what excuses you make for past investment results. This
one is a special exception, that one is a special exception, yeah, yeah,
whatever. When is the exception the rule? My point is only that new
investments DON'T get orders of magnitude reduction in costs. There are market
reasons for this, technical reasons, etc. Doesn't matter. You WON'T get the
cheap launch costs you want without going through the whole process predicted
by your typical cost curves.

>>The excuse doesn't matter - the fact is that launch costs are continuing
>down
>>the same old utterly predictable cost curves. Someone investing in launch
>>technology doen't stand a 100% chance of an order of magnitude reduction -
>they
>>stand a 0% chance.
>
>Data?

Every single past launch cost reduction program. Every one. No exceptions.

>Then you're saying that it's *not* a technology problem (i.e., you're
>changing your story)? Then how can you claim that Shuttle is
>representative of future launch vehicle costs, or that ISS solar cell
>costs are representative of future space solar cell costs?

I may be completely hallucinating, but I don't think I ever said that it would
be impossible to lower launch costs by an order of magnitude because technology
would prevent it. Technology is certainly a portion of the barrier, but is a
far cry from being the only stopping point.

Still doesn't matter. Shuttle and current ELV's are perfectly legitimate
points to use on a cost curve for future vehicles. And current space solar
panels are perfectly legitimate points to use to cost future systems.
Estimating tools use past data like this to predict future costs, and they are
wonderfully accurate.

>>Perhaps. But the fact remains that the market and the technology are
>coupled,
>>and you can't treat them separately. Perhaps if you built as few ground
>>panels, they would have a comparable cost to space panels. But we don't.
>>Perhaps in some happy theoretical world with massive space panel production,
>>costs would approach ground based.
>
>That is the world that is being proposed, hence your claim that
>current solar cell costs have anything to do with costs in that world
>are nonsense.

OK, I guess you are saying: "In the world where solar cells cost what we say
they cost, then they will cost what we say they cost." That's a tautology I
can agree with.

But in the REAL world, solar cells cost what they currently do. And current
solar cell costs have everything to do with that. Next year, they'll cost
maybe 15% less, if we are fortunate. The same with the year after that, and
after that. It makes it very predictable to find what costs in a future world
will be based on current costs.

>>But this world is not the world we live in.
>> And it won't be. Not before a time that can be easily and comfortably
>>predicted from existing growth rate and technology cost curves.
>
>That was the time that was being predicted.

No, that wasn't the time that was being predicted by the original poster to
this thread, who said that costs are immediately competitive or can be
competitive within some given time frame (2010, I think). Anyone looking at
the situation knows that space solar power is not competitive right now. Nor
will it be in the near future.


Rand Simberg

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to
On 31 Jul 2000 06:05:53 GMT, in a place far, far away,

brian...@aol.com (BrianF5070) made the phosphor on my monitor glow
in such a way as to indicate that:

>>Maturity of an industry does not come from the mere passage of


>>time--it comes from developing it to the point that economies of scale
>>have been fully employed, and the output product has been
>>commoditized. The launch industry is at the very bottom of the curve
>>in this respect.

>Of for crying out loud. Economies of scale are *never* fully employed. Output


>products are *never* fully commoditized.

Yes, they are. Sugar is fully commoditized. Pork bellies are fully
commoditized. Gold is fully commoditized. RAM chips are fully
commoditized. Electricity is fully commoditized. Even airline
tickets are commoditized to a large degree.

>In large part, these factors are what
>continues to drive down the cost of technologies in mature markets. All
>technologies, including natural gas, automobiles, computers, and rockets. The
>launch industry is middling near the bottom compared to some markets, but
>nowhere near enough the bottom to be classed as immature.

It is nowhere near a commodity.

>If you ever have the chance to walk the floor at LockMart's Titan/Atlas
>assembly facility, you'll see the degree to which the vehicles are
>commoditized.

You obviously don't understand the meaning of commoditized.

You mean that I can buy Titan futures on the Chicago Exchange? Can I
buy generic Titans?

>You'll see how economies of scale are employed. You'll realize
>that there is no way this manufacturing facility (or industry) is immature.

I've already said that ELV's are mature. This has nothing to do with
the future price of access. Access to space will be commoditized, and
the industry mature, when I can buy a ticket on Priceline.

>>>And how many dozens of others? I know, I know - none of these were
>>>"breakthrough." Or if they were, they were underfunded. Blah, blah, blah.
>>
>>If you know, then why waste bandwidth posting a bunch of irrelevant
>>examples?
>

>I know what you will say. I also don't agree with it. There have been dozens
>of programs. They all have received investment.

Many of your examples, particularly the ones that had some hope of
significant cost reductions, received little or no investment. None
of your examples were relevant to my claim.

>They have *all,* without
>exception, failed to drive down costs by an order of magnitude from the
>immediate previous generation.

The ones that were actually invested in weren't invested in with that
goal, other than possibly the Shuttle, but that goal was subsumed in
the "keeping NASA busy" goal. I don't understand why I have to repeat
this.

>All the ones that have actually brought


>vehicles to market have followed the nice typical cost curves that one expects
>to see.

>I could not care less what excuses you make for past investment results. This
>one is a special exception, that one is a special exception, yeah, yeah,
>whatever. When is the exception the rule?

Never. None of the vehicles that you listed (with the possible
exception of the Shuttle, though I don't think that it meets it
either) satisfied my rule. I'll repeat it.

When someone makes an investment (that is, a full investment) in a
vehicle with the intent of reducing costs an order of magnitude (or
more), it will happen. You still haven't provided any examples in the
last two decades where anyone has even made the attempt, other than
Beal, for whom the jury remains out (though I notice that you left out
Kistler, another vehicle on whom the jury is out).

>My point is only that new
>investments DON'T get orders of magnitude reduction in costs. There are market
>reasons for this, technical reasons, etc. Doesn't matter. You WON'T get the
>cheap launch costs you want without going through the whole process predicted
>by your typical cost curves.

You do if you take a new approach. It's called jumping the curve.

>>>The excuse doesn't matter - the fact is that launch costs are continuing
>>down
>>>the same old utterly predictable cost curves. Someone investing in launch
>>>technology doen't stand a 100% chance of an order of magnitude reduction -
>>they
>>>stand a 0% chance.
>>
>>Data?
>

>Every single past launch cost reduction program. Every one. No exceptions.

None of them were attempting major reductions. They were all
evolutionary investments in existing designs and approaches.

>Still doesn't matter. Shuttle and current ELV's are perfectly legitimate
>points to use on a cost curve for future vehicles. And current space solar
>panels are perfectly legitimate points to use to cost future systems.

Not when you simply assume, as you did, that the current costs *are*
the future costs.

>Estimating tools use past data like this to predict future costs, and they are
>wonderfully accurate.

Not when you simply assume, as you did, that the current costs *are*
the future costs.

>>>Perhaps. But the fact remains that the market and the technology are
>>coupled,
>>>and you can't treat them separately. Perhaps if you built as few ground
>>>panels, they would have a comparable cost to space panels. But we don't.
>>>Perhaps in some happy theoretical world with massive space panel production,
>>>costs would approach ground based.
>>
>>That is the world that is being proposed, hence your claim that
>>current solar cell costs have anything to do with costs in that world
>>are nonsense.
>

>OK, I guess you are saying: "In the world where solar cells cost what we say
>they cost, then they will cost what we say they cost." That's a tautology I
>can agree with.

No, what I'm saying is, in the world where space solar cells are being
mass produced (as they would have to be for SPS production), instead
of built in a boutique and in limited quantities (as they are now),
they will be orders of magnitude cheaper than they are now, and there
is no way to reliably infer what they will cost from the cost of
today's solar cells.

>But in the REAL world, solar cells cost what they currently do. And current
>solar cell costs have everything to do with that. Next year, they'll cost
>maybe 15% less, if we are fortunate. The same with the year after that, and
>after that. It makes it very predictable to find what costs in a future world
>will be based on current costs.

Yes, for terrestrial cells. Space cells are so far from being a
commodity that you cannot predict their future cost from their current
cost. It is overwhelmed by amortization of their cost of development,
certification and production. If you'd used the marginal cost, then
your number might have some credibility, but your simplistic analysis
has no relevance to a discussion of the future costs of space solar
cells.

>>That was the time that was being predicted.
>

>No, that wasn't the time that was being predicted by the original poster to
>this thread, who said that costs are immediately competitive or can be
>competitive within some given time frame (2010, I think). Anyone looking at
>the situation knows that space solar power is not competitive right now.

Yes.

>Nor will it be in the near future.

It is not a matter of simple time frame, or even technology
advancement--it is simply a matter of mass production. We could
easily have low-cost solar cells for space in a decade (or less) if
someone decided to make in investment in the necessary plant.

Bill Bonde

unread,
Jul 31, 2000, 3:00:00 AM7/31/00
to

BrianF5070 wrote:
>
>
> >No, what I'm saying is, in the world where space solar cells are being
> >mass produced (as they would have to be for SPS production), instead
> >of built in a boutique and in limited quantities (as they are now),
> >they will be orders of magnitude cheaper than they are now, and there
> >is no way to reliably infer what they will cost from the cost of
> >today's solar cells.
>

> Sigh. Well, we won't agree here. I know, with complete certainty, that you
> can infer what space solar cells will cost ten years from now, based solely on
> the cost of today's cells and recent trends. Why don't we get back together in
> 2010? If space solar cells cost less than $10/kW-hr on satellites launched at
> that time, I will publicly abase myself to your superior intellect, experience,
> and all around good-fellowness.
>
> But they won't.
>
Do a price vs. performance analysis on x86 compatible CPUs using only
information and trends from before AMD got its act together.

wm...@my-deja.com

unread,
Aug 1, 2000, 3:00:00 AM8/1/00
to
In article <3984EF...@ix.netcom.com>,
Scott Lowther <lex...@ix.netcom.com> wrote:

> Rand Simberg wrote:
> >
> > Here's what
> > John Schilling said the most recent time that you did it, a couple
of
> > months ago.
> >
> > "
> > If you ever see someone building, or even proposing, an actual solar
> > power satellite with an antenna big enough to serve as a space-based
> > death ray weapon, fell free to cry foul, scream bloody murder, and
> > whatnot. If you've done your math right, odds are a lot of us here
> > will join you."
>
> Foul. Bloody murder. Whatnot.
>
> Take a look at William Mook's proposals for 200+ solar intensity laser
> SPS systems. Sure, not quite instant death, but a dandy weapon
> nonetheless.

Not a weapon Scott. But we've had this conversation before. Which is
why I guess you put that caveat "not quite instant death" in your
statement.

One way to think about it more clearly is to imagine that SDI research
produced a solar pumped laser with a measly 200 solar intensity laser.
That's about the limit you can achieve with present nonlinear optical
systems anyway.

In this instance critics of SDI would decry the 200 solar laser as
useless as a weapon. Incapable of deflecting a nuclear attack or any
other kind of attack, not cost effective, and too easy to knock out.
And they'd be right!

You need at least thousands of times my power beam intensity from a
hardened satellite to even begin having a useful weapon system.

The outcome of SDI was to prove how difficult it is to create a useful
weapon with space based assets. And my SPS is not a weapon by any
meaningful definition of the term.

Recall, Teller initially proposed SDI to Reagan as a nuclear bomb
pumped X-ray laser that could pop up from hardened sites on Earth, and
knock out scores of targets with each firing. A nuclear powered xray
flashbulb with fancy optics.

These lasers here were millions of times as intense as my power lasers
and not permanently in orbit. These could have made useful terminal
phase defense if the details could be worked out. And even these were
useless against ground troops and such, due to strong atmospheric
absorption.

Later versions had hardened satellites holding dozens of nuclear
power 'shots' that were easily separated from their bus and fired over
enemy territory to get the enemy missiles in their boost and cruise
phases.

Again, these were millions of times brighter than my modest lasers.
These indeed are instant bolts of death from the blue down to about 25
km altitude. But my modest power lasers - especially when used in
conjunction with my beam control technology - are in now way weaponlike.

Later, Reagan, said no nuclear powered weapons in space.

So, SDI retreated from orbiting lasers to battle mirrors. Now you
switched from X-rays to UV/Vis, which decreased efficiency, and made it
vastly easier to defend your targets because you needed to use mirrors,
and x-ray mirrors don't really exist as they're needed for battle
mirrors. Mirrors had to be deployable from hardened satellites, or else
they could be destroyed before they could come into use. My power
beams use IR/Vis range by comparison and are easily countered against
using any old type of mirror or even a white cloth!

Then when complaints arose from citizens who didn't want exploding
nukes powering UV/Vis lasers in their neighborhood, SDI was revised
again first to chemical lasers, but the vast quantities of exotic fuels
were hard to defend against preemptive strike, and folks still worried
about all those weird chemicals.

So, SDI was revised again to brilliant pebbles. KKVs from orbit could
have a destructive effect on a missile during the launch phase! But
because KKV range is limited, and boost phase defense is time critical,
the number of battle stations increased dramatically, along with cost.

The targeting technology for KKV was vastly different, and less
reliable.

Although brilliant pebbles could be adapted to area defense and systems
like Patriot arose from the technology. Imagine a slew of Patriot type
missiles on orbit ready to deorbit and smack into a missile at launch,
and you get an idea of what SDI was reduced to.

> Now, if the original statment had been: "If you ever see someone
> RATIONAL building, or even proposing..."

My system is rational. Sadly I can't say the same of your criticisms.
Sorry.

William Mook
Orbatek, Inc.

BrianF5070

unread,
Aug 1, 2000, 3:00:00 AM8/1/00
to
>>If you ever have the chance to walk the floor at LockMart's Titan/Atlas
>>assembly facility, you'll see the degree to which the vehicles are
>>commoditized.
>
>You obviously don't understand the meaning of commoditized.
>
>You mean that I can buy Titan futures on the Chicago Exchange? Can I
>buy generic Titans?

I don't know if you can by Titan futures, but I don't believe there is any
reason why someone couldn't create that market if desired. And yes, you can
buy generic Titans. Go to Lock/Mart, say that you want this model Titan, with
such and such an upper stage, and the manager will pull out his little book, go
bing, bang, say this is how much it will cost, this is when it will be ready,
then you enter the order and X months later it rolls off the assembly line.

Better yet, go to ILS and order a Proton.

This is a product that has been commoditized.

Rockets are commoditized in much the same way as automobiles. Or airplane
tickets. There is a certain amount of haggling and arguing and customization,
but by and large both buyer and seller know the cost and schedule and come
rather quickly to agreement. These markets are less mature than pork bellies,
as you eloquently state, but they are pretty mature.

Chemical plants are a different story. If you want an ethylene plant, you go
to two or more engineering and construction firms. You give them the location
and environmental parameters. They take 3 - 6 months to develop a design, cost
estimate, and contract (for Mr. Mook's benefit - about 3/4ths of all chemical
plant contracts are reimbursable. Yet cost effective). Then you run it
through your in house design teams and business teams, make a go/no-go
decision, arrange financing, select a contractor, and start construction.

This is a product that has *not* been commoditized.

>>You'll see how economies of scale are employed. You'll realize
>>that there is no way this manufacturing facility (or industry) is immature.
>
>I've already said that ELV's are mature. This has nothing to do with
>the future price of access. Access to space will be commoditized, and
>the industry mature, when I can buy a ticket on Priceline.

I think this is a rather narrow definition. Is it not possible to commoditize
cargo delivery to LEO before it is possible for human delivery there?

>>I know what you will say. I also don't agree with it. There have been
>dozens
>>of programs. They all have received investment.
>
>Many of your examples, particularly the ones that had some hope of
>significant cost reductions, received little or no investment. None
>of your examples were relevant to my claim.

But just to note, your originally claim was that as soon as someone made an
investment in reducing launch costs by an order of magnitude, they were 100%
guaranteed of results. You didn't include qualifiers such as, "a sufficient
investment."

>>They have *all,* without
>>exception, failed to drive down costs by an order of magnitude from the
>>immediate previous generation.
>
>The ones that were actually invested in weren't invested in with that
>goal, other than possibly the Shuttle, but that goal was subsumed in
>the "keeping NASA busy" goal. I don't understand why I have to repeat
>this.

Part of the reason you are repeating this is that we are talking past each
other. You say that the vehicles weren't invested in with the goal of reducing
launch costs by an order of magnitude. I say, "No kidding!" Investors DON'T
invest for goals like reducing launch costs. They invest to make a profit. To
make a profit, they only have to be cheaper than their competitors. They
normally tend to choose the least expensive, lowest risk means of accomplishing
this.

>When someone makes an investment (that is, a full investment) in a
>vehicle with the intent of reducing costs an order of magnitude (or
>more), it will happen. You still haven't provided any examples in the
>last two decades where anyone has even made the attempt, other than
>Beal, for whom the jury remains out (though I notice that you left out
>Kistler, another vehicle on whom the jury is out).

At the bottom of the list I noted that I have left out dozens of other
projects. Beal is attempting to drop launch costs by around 30% or so. As is
Kistler (though I was under the impression their project was close to
moribund). These are real projects following tried and true rules of economics
and engineering development. Personally, I'd love to see either succeed. But
you won't get order of magnitude here.

You can throw in Sea Launch if you want. They had some neat sort of ideas and
OPEX is pretty low, but the CAPEX I think will sink the project (no pun
intended). I doubt if they can even keep up with interest payments based on
their margin and launch rate.

Energiya-M? Now there is a concept that could have really driven down launch
costs. But a bit ahead of its time.

>>My point is only that new
>>investments DON'T get orders of magnitude reduction in costs. There are
>market
>>reasons for this, technical reasons, etc. Doesn't matter. You WON'T get
>the
>>cheap launch costs you want without going through the whole process
>predicted
>>by your typical cost curves.
>
>You do if you take a new approach. It's called jumping the curve.

Well, jumping the curve is talked about a lot. But it is far harder to do than
you seem to think. And, in reality, it usually ends up being more of a
curve-acceleration rather than a true discontinuity.

Energiya-M had a real chance to be a discontinuity in the launch cost curve.
It went broke, as is normally the case.

>>Still doesn't matter. Shuttle and current ELV's are perfectly legitimate
>>points to use on a cost curve for future vehicles. And current space solar
>>panels are perfectly legitimate points to use to cost future systems.
>
>Not when you simply assume, as you did, that the current costs *are*
>the future costs.

I never assumed this. I assumed that future costs can be predicted from
current ones. Rather accurately for the most part.

>No, what I'm saying is, in the world where space solar cells are being
>mass produced (as they would have to be for SPS production), instead
>of built in a boutique and in limited quantities (as they are now),
>they will be orders of magnitude cheaper than they are now, and there
>is no way to reliably infer what they will cost from the cost of
>today's solar cells.

Sigh. Well, we won't agree here. I know, with complete certainty, that you

Rand Simberg

unread,
Aug 1, 2000, 3:00:00 AM8/1/00
to
On 01 Aug 2000 05:34:51 GMT, in a place far, far away,

brian...@aol.com (BrianF5070) made the phosphor on my monitor glow
in such a way as to indicate that:

>>>If you ever have the chance to walk the floor at LockMart's Titan/Atlas


>>>assembly facility, you'll see the degree to which the vehicles are
>>>commoditized.
>>
>>You obviously don't understand the meaning of commoditized.
>>
>>You mean that I can buy Titan futures on the Chicago Exchange? Can I
>>buy generic Titans?
>
>I don't know if you can by Titan futures, but I don't believe there is any
>reason why someone couldn't create that market if desired.

Then let me inform you that you cannot buy Titan futures. The reason
is that there is insufficient volume in the market. It is not a
commodity.

>And yes, you can buy generic Titans.

No, you can't. They don't exist, in the sense that commodities do.

>Go to Lock/Mart, say that you want this model Titan, with
>such and such an upper stage, and the manager will pull out his little book, go
>bing, bang, say this is how much it will cost, this is when it will be ready,
>then you enter the order and X months later it rolls off the assembly line.

That's not a commodity. You continue to demonstrate a profound lack
of understanding of the term "commodity."

>Better yet, go to ILS and order a Proton.
>
>This is a product that has been commoditized.

No, it's not.

>Rockets are commoditized in much the same way as automobiles.

Automobiles are not a commodity.

>Or airplane
>tickets. There is a certain amount of haggling and arguing and customization,
>but by and large both buyer and seller know the cost and schedule and come
>rather quickly to agreement.

Airplane tickets are commoditized, but not as fully as the other items
mentioned.

>These markets are less mature than pork bellies,
>as you eloquently state, but they are pretty mature.

Yes, and rockets and access to LEO are nowhere near that state.

>Chemical plants are a different story. If you want an ethylene plant, you go
>to two or more engineering and construction firms. You give them the location
>and environmental parameters. They take 3 - 6 months to develop a design, cost
>estimate, and contract (for Mr. Mook's benefit - about 3/4ths of all chemical
>plant contracts are reimbursable. Yet cost effective). Then you run it
>through your in house design teams and business teams, make a go/no-go
>decision, arrange financing, select a contractor, and start construction.

Chemical plants are not commoditized, but chemicals (their outputs)
are.

>This is a product that has *not* been commoditized.

Never claimed otherwise. Nor has a rocket plant.

>>>You'll see how economies of scale are employed. You'll realize
>>>that there is no way this manufacturing facility (or industry) is immature.
>>
>>I've already said that ELV's are mature. This has nothing to do with
>>the future price of access. Access to space will be commoditized, and
>>the industry mature, when I can buy a ticket on Priceline.
>
>I think this is a rather narrow definition. Is it not possible to commoditize
>cargo delivery to LEO before it is possible for human delivery there?

Yes, but I cannot buy cargo to LEO on Priceline either, nor is there a
futures market for it.

>>>I know what you will say. I also don't agree with it. There have been
>>dozens
>>>of programs. They all have received investment.
>>
>>Many of your examples, particularly the ones that had some hope of
>>significant cost reductions, received little or no investment. None
>>of your examples were relevant to my claim.
>
>But just to note, your originally claim was that as soon as someone made an
>investment in reducing launch costs by an order of magnitude, they were 100%
>guaranteed of results. You didn't include qualifiers such as, "a sufficient
>investment."

Well, excuse me. I thought that this would be obvious.

Did you really think that I meant that it would happen as soon as
someone put up one percent, or ten percent of the money? Do you really
want people in the newsgroup to think that you are that dim?

>Part of the reason you are repeating this is that we are talking past each
>other. You say that the vehicles weren't invested in with the goal of reducing
>launch costs by an order of magnitude. I say, "No kidding!" Investors DON'T
>invest for goals like reducing launch costs. They invest to make a profit. To
>make a profit, they only have to be cheaper than their competitors. They
>normally tend to choose the least expensive, lowest risk means of accomplishing
>this.

I have never claimed otherwise. All I claimed was that if someone
made such an investment, that the costs would be reduced. Whether or
not that is an economically sensible thing to do is an entirely
different matter, as I explained to James Davis.

>>When someone makes an investment (that is, a full investment) in a
>>vehicle with the intent of reducing costs an order of magnitude (or
>>more), it will happen. You still haven't provided any examples in the
>>last two decades where anyone has even made the attempt, other than
>>Beal, for whom the jury remains out (though I notice that you left out
>>Kistler, another vehicle on whom the jury is out).
>
>At the bottom of the list I noted that I have left out dozens of other
>projects. Beal is attempting to drop launch costs by around 30% or so.

No, they're attempting to drop prices by that amount. If they want to
make a good profit to amortize their investment, they'll drop costs
quite a bit more.

>As is
>Kistler (though I was under the impression their project was close to
>moribund). These are real projects following tried and true rules of economics
>and engineering development. Personally, I'd love to see either succeed. But
>you won't get order of magnitude here.

You continue to confuse cost and price.

>>>My point is only that new
>>>investments DON'T get orders of magnitude reduction in costs. There are
>>market
>>>reasons for this, technical reasons, etc. Doesn't matter. You WON'T get
>>the
>>>cheap launch costs you want without going through the whole process
>>predicted
>>>by your typical cost curves.
>>
>>You do if you take a new approach. It's called jumping the curve.
>
>Well, jumping the curve is talked about a lot. But it is far harder to do than
>you seem to think. And, in reality, it usually ends up being more of a
>curve-acceleration rather than a true discontinuity.

If it's harder to do, it's because of the uncertainty of financial
return--not because it's intrinsically difficult to do.

>Energiya-M had a real chance to be a discontinuity in the launch cost curve.
>It went broke, as is normally the case.

Why? That is, why did Energiya-M have a real chance?

>>>Still doesn't matter. Shuttle and current ELV's are perfectly legitimate
>>>points to use on a cost curve for future vehicles. And current space solar
>>>panels are perfectly legitimate points to use to cost future systems.
>>
>>Not when you simply assume, as you did, that the current costs *are*
>>the future costs.
>
>I never assumed this. I assumed that future costs can be predicted from
>current ones. Rather accurately for the most part.

Not at all. You didn't even postulate what you thought that future
costs would be. You simply stated present costs of space solar cells,
and compared them to present terrestrial solar cell costs, as though
there was no other analysis necessary. It was, in the blunt (but as
this discussion goes on, apparently appropriate) words of another
poster, a "stupid analysis."

>>No, what I'm saying is, in the world where space solar cells are being
>>mass produced (as they would have to be for SPS production), instead
>>of built in a boutique and in limited quantities (as they are now),
>>they will be orders of magnitude cheaper than they are now, and there
>>is no way to reliably infer what they will cost from the cost of
>>today's solar cells.
>
>Sigh. Well, we won't agree here. I know, with complete certainty, that you
>can infer what space solar cells will cost ten years from now, based solely on
>the cost of today's cells and recent trends.

Well, aren't you smart. Why ain't you rich?

>Why don't we get back together in
>2010? If space solar cells cost less than $10/kW-hr on satellites launched at
>that time, I will publicly abase myself to your superior intellect, experience,
>and all around good-fellowness.

I have no expectation that space solar cell costs will be
significantly lower in 2010, because I don't expect anyone to make the
investment necessary to cause that to happen. My only point is that
it could happen if it were required or desired, and that *this* is the
most important criterion, not what they cost today.

Scott Lowther

unread,
Aug 1, 2000, 3:00:00 AM8/1/00
to

Much irrelevant SDI blahblahblah snipped.

And here's where your thought process shows its lack of flexibility. An
M-1 Abrams CANNOT SHOOT DOWN AN ICBM. By your reasoning, that makes it
"not a weapon." An aerosol can of anthrax CANNOT SHOOT DOWN AN ICBM.
This also classifies it as a "William Mook certified non-weapon." The
Boston Strangler couldn't shoot down an ICBM. How'd you like him to show
up at your house some night?

Look, a 200 solar laser spot 30-50 centimeters wide is a dandy weapon.
It'd do a great job setting forest fires, melting asphalt roads,
blinding people, burning people, burning down wooden structures
(wouldn't LeMay have LOVED to have such a thing), damaging delicate
structures (radar setups and the like), confusing IR sensors, setting
petroleum plants on fire, causing stampedes, altering weather,
evaporating ponds, destroying crops, dazzling pilots on final approach,
lighting up enemy forces at night, psychological warfare ("Booga booga,
we control the Sun"), etc. Such a system would be useful in any of a
number of foreseeable military situations. And this assumes that you
don't overlap lasers... wouldn't take too long to pile up 200 solar
lasers to make something capable of damaging tanks and sinking ships.

> > Now, if the original statment had been: "If you ever see someone
> > RATIONAL building, or even proposing..."
>
> My system is rational. Sadly I can't say the same of your criticisms.
> Sorry.

Well, it's obviously because "rational" in your universe only applies to
nuclear war. Well, sorry, that's a minor concern. Wars will continue to
be fought with things other than ICBM's and nukes.

wm...@my-deja.com

unread,
Aug 1, 2000, 3:00:00 AM8/1/00
to
In article <398680...@ix.netcom.com>,

Bull. You clipped it because it made a strong and cogent argument as
to why no one would consider my solar pumped laser powersat a military
threat! Sheez.

> And here's where your thought process shows its lack of flexibility.

Not.

>An
> M-1 Abrams CANNOT SHOOT DOWN AN ICBM.

Yes it can. If its sitting right at the launch center and fires
somewhere very close to ignition.

> By your reasoning, that makes it
> "not a weapon."

By your reasoning anything that's dangerous is a weapon. That's
extraordinarly foolish. By this standard a pack of matches, a cutlery
set, and a baseball bat constitute meaningful weapons. Fact is, in the
sense of modern military uses, they don't.

> An aerosol can of anthrax CANNOT SHOOT DOWN AN ICBM.
> This also classifies it as a "William Mook certified non-weapon."

Bull. A can of anthrax doesn't cost $60 billion, and besides, if
deployed in the launch center well before launch, it can inactivate a
large number of ICBMs.

>The
> Boston Strangler couldn't shoot down an ICBM. How'd you like him to
show
> up at your house some night?

There are any number of dangerous and horrendous things I would rather
not have happen. Whats it to do with your argument.

If you believe my powersat system is a useful and effective military
device giving its owners supreme control over the Earth (so much so
you've said previously that the moment it was deployed it would become
a universally abhorred target!) - then why don't you quit all this
offensive and meaningless falderah and tell us all just precisely in
your mind makes it such a gawdawful weapon?

You've yet to provide a clear and meaningful and cogent reply to this
question, and until you do, we can all safely write your postings off
as that of a meaningless crank!

> Look, a 200 solar laser spot 30-50 centimeters wide is a dandy weapon.

By this standard a 230 V three phase and a set of bare wires make a
dandy weapon. I have yet to see them as standard issue in any military
force.


> It'd do a great job setting forest fires, melting asphalt roads,
> blinding people, burning people, burning down wooden structures

So does 230 V three phase, if things get out of control. So what?
Electrical distribution can be dangerous if improperly handled. That
doesn't make an electrical distribution system a weapon.

I freely admit lasers can be dangerous if improperly handled. That
doesn't make my power distribution system using lasers a weapons
system.

You have consistently shown a total ignorance and lack of understanding
of four wave mixing, nonlinear optics, and all the other critical
factors that go into making my powersat work. Until you come to
understand these factors why don't you bite your tongue, reign in your
stupid and foolish comments, and try to recover some modicum of respect
here? You really don't know what you're talking about on this one
Scott. I'm sorry to see you play the rabid fool so often.

> (wouldn't LeMay have LOVED to have such a thing), damaging delicate
> structures (radar setups and the like), confusing IR sensors, setting
> petroleum plants on fire, causing stampedes, altering weather,
> evaporating ponds, destroying crops, dazzling pilots on final
approach,
> lighting up enemy forces at night, psychological warfare ("Booga
booga,
> we control the Sun"), etc. Such a system would be useful in any of a
> number of foreseeable military situations.

<shrug> There may be a role a big laser in the sky could play in
combat. I'll give you that. That doesn't make my power distribution
laser a weapon system nor does it make it militarily useful. If you
think it is, why don't you just say so and quit with all these
atmospherics?

> And this assumes that you
> don't overlap lasers...

Again, your lack of knowledge of optics is apparent.

> wouldn't take too long to pile up 200 solar
> lasers to make something capable of damaging tanks and sinking ships.

And how would you do that with my beam steering device pray tell? I'll
give you the answer to that one. You can't!

> > > Now, if the original statment had been: "If you ever see someone
> > > RATIONAL building, or even proposing..."
> >
> > My system is rational. Sadly I can't say the same of your
criticisms.
> > Sorry.
>
> Well, it's obviously because "rational" in your universe only applies
to
> nuclear war.

Zing! I don't know what you're talking about there. Fact is you have
yet to give a single cogent military use my powersat could support. Of
course to begin that sort of analysis you'd have to first understand
how my powersat works, and sadly, you don't have a clue about that.

> Well, sorry, that's a minor concern. Wars will continue to
> be fought with things other than ICBM's and nukes.

Pardon me, I got a little motion sick from the quick turn in
conversation.

What the hell are you talking about? Or uh... tell you what, strike
that question. This has nothing to do with what we're talking about,
so lets not even continue down this path.

wm...@my-deja.com

unread,
Aug 1, 2000, 3:00:00 AM8/1/00
to
In article <20000801013451...@ng-cv1.aol.com>,
brian...@aol.com (BrianF5070) wrote:

[snip!]

> They take 3 - 6 months to develop a design, cost
> estimate, and contract (for Mr. Mook's benefit - about 3/4ths of all
chemical
> plant contracts are reimbursable. Yet cost effective). Then you run
it
> through your in house design teams and business teams, make a go/no-go
> decision, arrange financing, select a contractor, and start
construction.
>
> This is a product that has *not* been commoditized.

Well, I guess you're referring to my comment that the network of SPS
peak power users would trade their rights to a portion of the energy
available from the first SPS to meet their peak power needs.

Its clear that a power trading system as I envision could easily be set
up. The volume, value, and similarity to existing power trading
systems all tend to favor this conclusion.

[snip!]

> Sigh. Well, we won't agree here. I know, with complete certainty,
that you
> can infer what space solar cells will cost ten years from now, based
solely on
> the cost of today's cells and recent trends. Why don't we get back
together in
> 2010? If space solar cells cost less than $10/kW-hr on satellites
launched at
> that time, I will publicly abase myself to your superior intellect,
experience,
> and all around good-fellowness.

Wow! I get all tingly just thinking about that! :)

But seriously, this requires that the person taking this bet must be
able to predict with certainty that the demand for space power will
increase dramatically. And no one knows that for certain.

Can't we both do something more reasonable and dispense with the self
flagellation?

Consider, one can draw a cost/volume curve for a variety of space power
technologies. Agreed?

Now, all we really have to disagree about is where to place these
curves for each technology, how steeply the measure you mentions
declines with total capacity on orbit, and so forth.

But, I think you'll find we're reading off pretty much the same page
here.

A PV array fed by a lightweight concentrator in orbit has associated
with it a specific cost at a specific total demand. I think you'll
find that if we put up about 1 Terawatt of big powersats (ignore our
ongoing argument about launch costs for a minute, lets settle one thing
at a time!) using the curve we already agreed upon - we can both
calculate what the cost per KW-hr is likely to be.

Surprise - its well below $0.01 per kw-hr!!! This was found in the
early studies done by Arthur D. Little, and also by ONeil and others.

Amazing ain't it.


> But they won't.
>

<shrug> Depends on total amount of energy produced in space doesn't it?

Scott Lowther

unread,
Aug 1, 2000, 3:00:00 AM8/1/00
to

Wrong. It was one of your typical Mook Points...overly verbose non
sequiturs about shooting down ICBM's, when the fact remains that wars
have yet to be fought with any sort of ICBM.

> >An
> > M-1 Abrams CANNOT SHOOT DOWN AN ICBM.
>
> Yes it can. If its sitting right at the launch center and fires
> somewhere very close to ignition.

Please. Then that means that the optimal SDI system is to park a tank
next to every enemy ICBM site. Good luck.

> By your reasoning anything that's dangerous is a weapon. That's
> extraordinarly foolish. By this standard a pack of matches, a cutlery
> set, and a baseball bat constitute meaningful weapons.

Hey, guess what? they've all been used as meanigful and effective
weapons. And just try walking onto an airplane with a bunch of cutlery,
and see how security takes that.

> Fact is, in the sense of modern military uses, they don't.

I'll be sure to tell the next Marine I meet that he doesn't need to
carry a knife anymore.

> > An aerosol can of anthrax CANNOT SHOOT DOWN AN ICBM.
> > This also classifies it as a "William Mook certified non-weapon."
>
> Bull. A can of anthrax doesn't cost $60 billion, and besides, if
> deployed in the launch center well before launch, it can inactivate a
> large number of ICBMs.

So, again,. the William Mook Definition Of A Weapon includes high cost
and anti-ICBM capability.


> If you believe my powersat system is a useful and effective military
> device giving its owners supreme control over the Earth (so much so
> you've said previously that the moment it was deployed it would become
> a universally abhorred target!) - then why don't you quit all this
> offensive and meaningless falderah and tell us all just precisely in
> your mind makes it such a gawdawful weapon?

Already done.

> You've yet to provide a clear and meaningful and cogent reply to this
> question, and until you do, we can all safely write your postings off
> as that of a meaningless crank!

Hey, that's my line!

> > Look, a 200 solar laser spot 30-50 centimeters wide is a dandy weapon.
>
> By this standard a 230 V three phase and a set of bare wires make a
> dandy weapon. I have yet to see them as standard issue in any military
> force.

You can't aim an electric fence, dipshit. And certainly not over several
hundred kilometers. Just as a small lump of lead doesn't seem to be much
of a weapon until it's coming at you at 2,000 feet per second.

> > It'd do a great job setting forest fires, melting asphalt roads,
> > blinding people, burning people, burning down wooden structures
>
> So does 230 V three phase, if things get out of control. So what?
> Electrical distribution can be dangerous if improperly handled. That
> doesn't make an electrical distribution system a weapon.

Again, you can't aim and shoot electricity. The effort required to start
fires with electric wires at ground level does not compare with the
relative ease with which this could be accomplished from orbit.

I weep for the future if knuckleheads like you get any sort of power...
such a basic lack of understanding of weapons potential is frightening.


> You have consistently shown a total ignorance and lack of understanding
> of four wave mixing, nonlinear optics, and all the other critical
> factors that go into making my powersat work.

Actually, I have a vague understanding... not enough to build such a
system, but enough to know how to fuck with it. Have you built a
demonstrator yet?

Until you come to
> understand these factors why don't you bite your tongue, reign in your
> stupid and foolish comments, and try to recover some modicum of respect
> here? You really don't know what you're talking about on this one
> Scott. I'm sorry to see you play the rabid fool so often.

Rabid? I'm hardly rabid. I'm simply laughing at the funny little
crackpot who can't think two steps ahead.

> > (wouldn't LeMay have LOVED to have such a thing), damaging delicate
> > structures (radar setups and the like), confusing IR sensors, setting
> > petroleum plants on fire, causing stampedes, altering weather,
> > evaporating ponds, destroying crops, dazzling pilots on final
> approach,
> > lighting up enemy forces at night, psychological warfare ("Booga
> booga,
> > we control the Sun"), etc. Such a system would be useful in any of a
> > number of foreseeable military situations.
>
> <shrug> There may be a role a big laser in the sky could play in
> combat. I'll give you that. That doesn't make my power distribution
> laser a weapon system nor does it make it militarily useful. If you
> think it is, why don't you just say so and quit with all these
> atmospherics?

Atmospherics? The examples listed above would be very useful military
lasetsat applications.


> > wouldn't take too long to pile up 200 solar
> > lasers to make something capable of damaging tanks and sinking ships.
>
> And how would you do that with my beam steering device pray tell? I'll
> give you the answer to that one. You can't!

Yes, you can, and quite easily so. You claim your system will use a
small laser mounted on a target vehicle to direct the power laser. Fine,
maybe that might even work. But if so... _two_ (better still, three)
lasers operated by ground troops, separated by some distance, tuned to
the same frequency and putting out appropriate information, will cause
the power laser to be directed to a source OTHER than either of the
directing lasers... say, split the difference and shoot directly in
between, or offset by some angle (basic trig, folks). Several setups
such as this would cause multiple beams to fall on a single target. This
would essentially "confuse" the powersat... instead of the targetting
laser being a single point, it would be multiple points, and the result
would be the laser would point in the "wrong" direction.

This can also be accomplished by having small "targetting lasers" either
mounted to the back of the powersat, or as a free-flyer nearby... then
ground forces would not need to do anything but duck.

There are I'm sure other means of accomplishing this same goal. Your
insistance that this is impossible, mated to your insistacne that this
highly complex system can be cheaply built to withstand thirty years
without a moments maintenance, simply points you out as someone with no
real understanding of the real world.

> > > My system is rational. Sadly I can't say the same of your
> criticisms.
> > > Sorry.
> >
> > Well, it's obviously because "rational" in your universe only applies
> to
> > nuclear war.
>
> Zing! I don't know what you're talking about there.

Hardly surpising. But remember those ICBM/SDI references YOU kept
making?

Fact is you have
> yet to give a single cogent military use my powersat could support.

HA!!!

Michael Kagalenko

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
Bill Bonde (std...@mail.com) wrote
]
]
]BrianF5070 wrote:
]>
]>
]> >No, what I'm saying is, in the world where space solar cells are being

]> >mass produced (as they would have to be for SPS production), instead
]> >of built in a boutique and in limited quantities (as they are now),
]> >they will be orders of magnitude cheaper than they are now, and there
]> >is no way to reliably infer what they will cost from the cost of
]> >today's solar cells.
]>
]> Sigh. Well, we won't agree here. I know, with complete certainty, that you
]> can infer what space solar cells will cost ten years from now, based solely on
]> the cost of today's cells and recent trends. Why don't we get back together in
]> 2010? If space solar cells cost less than $10/kW-hr on satellites launched at
]> that time, I will publicly abase myself to your superior intellect, experience,
]> and all around good-fellowness.
]>
]> But they won't.
]>
]Do a price vs. performance analysis on x86 compatible CPUs using only

]information and trends from before AMD got its act together.


Solar cells do not get cheaper at the same rate as integrated circuits.
In fact, the price is leveling out. There is no reason to assume
that the solar cells will ever get cheaper than the conventional
or nuclear power generator (if you don't list "strnuous wishful
thinking" as a plausible reason the way the space amateurs such as Bonde do)

George William Herbert

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
Michael Kagalenko <mkag...@lynx.dac.neu.edu> wrote:
> Solar cells do not get cheaper at the same rate as integrated circuits.
> In fact, the price is leveling out. There is no reason to assume
> that the solar cells will ever get cheaper than the conventional
> or nuclear power generator (if you don't list "strnuous wishful
> thinking" as a plausible reason the way the space amateurs such as Bonde do)

How on earth are you passing classes with research skills this poor, Michael?

No, they don't get cheaper at the same rate as ICs. This part is correct.

The price is not leveling out. It continues to drop, and basic manufacturing
and technology improvements already in the R&D and production pipelines will
see it through at least 5 years more price drop, down to nearly the price
parity with conventional power.

There is plenty of reason to assume they will get cheaper than conventional
power systems. If we assume *any* new R&D successes then the price will
continue to drop beyond conventional source parity as described above,
into clear price win territory. At this point, given the decades long
history of success after success in developing new technologies in this
area, it would be folly to predict that it's going to come to a screaming
halt in the near future.

There are literally hundreds of relevant documents linked off of:
http://pix.nrel.gov:8020/BASIS/nich/www/public/SDW?M=1&W=subcode='SEP'+order+by+pubyear/desc
http://www.epri.com/targetLinks.asp?program=207933&value=00T053.1
http://www.ases.org/solarguide/factbase.html

I could go on but it would be a waste of electrons, which are still being
produced using fossil fuel, so I want to save a few.


-george william herbert
gher...@retro.com


Michael R. Irwin

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to

Michael Kagalenko wrote:

> Bill Bonde (std...@mail.com) wrote
> ]
> ]
> ]BrianF5070 wrote:
> ]>
> ]>

> ]> >No, what I'm saying is, in the world where space solar cells are being


> ]> >mass produced (as they would have to be for SPS production), instead
> ]> >of built in a boutique and in limited quantities (as they are now),
> ]> >they will be orders of magnitude cheaper than they are now, and there
> ]> >is no way to reliably infer what they will cost from the cost of
> ]> >today's solar cells.
> ]>
> ]> Sigh. Well, we won't agree here. I know, with complete certainty, that you
> ]> can infer what space solar cells will cost ten years from now, based solely on
> ]> the cost of today's cells and recent trends. Why don't we get back together in
> ]> 2010? If space solar cells cost less than $10/kW-hr on satellites launched at
> ]> that time, I will publicly abase myself to your superior intellect, experience,
> ]> and all around good-fellowness.
> ]>
> ]> But they won't.

> ]>
> ]Do a price vs. performance analysis on x86 compatible CPUs using only
> ]information and trends from before AMD got its act together.
>

> Solar cells do not get cheaper at the same rate as integrated circuits.
> In fact, the price is leveling out. There is no reason to assume
> that the solar cells will ever get cheaper than the conventional
> or nuclear power generator (if you don't list "strnuous wishful
> thinking" as a plausible reason the way the space amateurs such as Bonde do)

Bill maybe you could make your point explicitly regarding
the IC chips I am not certain I caught your meaning either.

I interpreted your suggestion to imply that the cost reduction
was very slow when Intel completely dominated the market
and then "miraculously" or suspiciously accelerated
dramatically when they had some effective competition with
compatible chips.

In other words, although the Intel chips were mass
produced in large volumes and in a "mature" market
they were not commodities. Once AMD was
reliably reverse engineering and bringing binary
compatible chips to market, the chips were
"commoditized" and prices came down much faster.

Is this a proper interpretation?

Thanks,
Mike Irwin


Michael R. Irwin

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to

wm...@my-deja.com wrote:

>
> Its clear that a power trading system as I envision could easily be set
> up. The volume, value, and similarity to existing power trading
> systems all tend to favor this conclusion.
>

Say Willian, most of your material I have read here references
your concept of using SPS for both base loading and peak
loading by diverting the power to where it is required dynamically.

Clearly this requires a massive system of SPS, rectennas, etc.

In your studies of this concept have you ever looked at the
benefits of using an SPS or 2 to support a heavy industry
in an infrastructure challenged area.

Its seems to me you might get some benefits from oversizing
the SPS so that it can meet all peak demands and diverting
the surplus power from off peak periods to generating
chemical feedstocks such as HCL, CL, NA, etc.

This might have to be combined somehow with a high
value compact export product such as chips or solar cells
because if high volume cheap transportation is available
it is probably much cheaper inital investment to ship in
oil.

Anyway I was curious as to whether you have looked at
alternate scenarios or know of references to some that
might make a transition period or startup prototype
possible.

Maybe we (U.S.) could give the Chinese an SPS for central
China if they agree our military could man or control
it and install a NMD radar/laser sensor suite on it ... :-)

Sure make me feel safer at night knowing we
can meet treaty obligations to Taiwan and target
missiles in Boost Phase .... might be worth
20 or 30 Billion ... LOL

We could do that same with North Korea
and India/Pakistan and hold out for integrated
manufacturing or tourism (R&D prototype) facilities.

Sure be fun to pitch, if a President made interested
noises we would probably lose some reactionary
Generals/Admirals to apoplexy!

What do you think?
Mike Irwin


Michael R. Irwin

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to

Scott Lowther wrote:

>
> > If you ever see someone building, or even proposing, an actual solar
> > power satellite with an antenna big enough to serve as a space-based
> > death ray weapon, fell free to cry foul, scream bloody murder, and
> > whatnot. If you've done your math right, odds are a lot of us here
> > will join you."
>
> Foul. Bloody murder. Whatnot.
>
> Take a look at William Mook's proposals for 200+ solar intensity laser
> SPS systems. Sure, not quite instant death, but a dandy weapon
> nonetheless.
>

> Now, if the original statment had been: "If you ever see someone
> RATIONAL building, or even proposing..."

So Scott, according to your calculations regarding Mr. Mook's
design how long would it take his operational system to target
and set my house on fire?

Would this time be significantly affected if I used a silver
metal roof instead of a dark asphalt shingle?

How many houses could he set on fire in an hour?

Do you think he could effectively harden his transceivers
or optics against a retalitory U.S. ASAT strike?

Do you think it would be possible for the U.S. military
to monitor his facilities to perform their function of
providing for the "common defense" or should we
organize some Boy Scouts with telescopes?

Are you concerned that he has secret over-the-horizon
capabilities or shall we assume line of sight
is necessary?

Can you give me an estimate of the sun block
necessary to protect against his death ray?

What is your estimate of the available warning
time available from notification from the Boy
Scouts through the Emergency Broadcast
System?

Do you think we can get by with
just a few tests and practice sessions in putting
on our aluminum caps and sunscreen or would
there be a large potential profit in setting up
seminars for the technologically challenged
among us?

Thanks in advance for your assistance in clarifying these
matters; we must after all remain eternally vigiliant
because we do have a few inimical adversaries
out there.

Sincerely
Michael R. Irwin

Paul F. Dietz

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
George William Herbert wrote:

> How on earth are you passing classes with research skills
> this poor, Michael?

Don't confuse poor Michael. He couldn't even tell the
difference between a demand curve and an experience curve.

Paul

wm...@my-deja.com

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
In article <3987FCE8...@harborside.com>,

"Michael R. Irwin" <mir...@harborside.com> wrote:
>
>
> wm...@my-deja.com wrote:
>
> >
> > Its clear that a power trading system as I envision could easily be
set
> > up. The volume, value, and similarity to existing power trading
> > systems all tend to favor this conclusion.
> >
>
> Say Willian, most of your material I have read here references
> your concept of using SPS for both base loading and peak
> loading by diverting the power to where it is required dynamically.

Yes. Peak load power matching at first through dynamic reallocation
from the first few sats. Then as capacity grows, migrate to
intermediate and then base load.

> Clearly this requires a massive system of SPS, rectennas, etc.

I don't know about the word 'massive', but yes. More receivers than
satellites. That's why you must make your beam energy density high,
to reduce the size of the power receivers and consequently their costs.

> In your studies of this concept have you ever looked at the
> benefits of using an SPS or 2 to support a heavy industry
> in an infrastructure challenged area.

Yes. As a matter of fact there would emerge a two tier market. The
first tier would be those who would buy power on demand. The second
teir would be those who would buy power when available. And within
each tier, there would be flexibility in pricing.

For example, a movie house sells tickets at matinee pricing, senior
pricing, student pricing, and so forth. This is to accomodate the
demand curve for tickets within each group. To the extent that unique
groups exist, and the supply is cheaply available, we can expect
similar pricing to emerge.

The ultimate outcome of this process - as in all market driven
processes - is the flow of capital to undercapitalized regions.

> Its seems to me you might get some benefits from oversizing
> the SPS so that it can meet all peak demands and diverting
> the surplus power from off peak periods to generating
> chemical feedstocks such as HCL, CL, NA, etc.

You won't be significantly oversized until you start sopping up large
segments of baseline load. Then, you will find that if SPS is your
only utility about 20% of your power over the year has no ready
customer.

As you approach this condition you will do the pricing strategies I
mentioned.

As you beat down the demand for oil, oil derived products like
fertilizer through the haber process, should also decrease in price.

When the entire world demand for energy is saturated. That is, when
everyone shares largely the same living standard, this 20% or so will
reappear.

At this point - or well before it - you could use surpluses to drive
desalination plants, or even provide synthetic fertilizer and fuels in
competition with oil. But, these would be a loss leader for SPS
operators.

Of course new technologies, like laser powered jets and rockets, would
change this picture. Dramatically raising the demand side of the
equation and moving SPS practice away from that needed to manage this
surplus.

On an interplanetary scale, SPS power, now imagined driving
interplanetary commerce and settlement, would eventually have surpluses
emerge again for the reasons you cite. Although we can't tell the size
or timing of these today, we will likely have them.

In this case, these surpluses would likely be used to support
interstellar probe launches, bulk conversion of asteroid feedstocks
into more highly valuable feedstocks, and so forth.


> This might have to be combined somehow with a high
> value compact export product such as chips or solar cells
> because if high volume cheap transportation is available
> it is probably much cheaper inital investment to ship in
> oil.

With a large and growing SPS industry, oil prices will fall.

> Anyway I was curious as to whether you have looked at
> alternate scenarios or know of references to some that
> might make a transition period or startup prototype
> possible.

A smaller system of about 100 MW might be usefully deployed.

> Maybe we (U.S.) could give the Chinese an SPS for central
> China if they agree our military could man or control
> it and install a NMD radar/laser sensor suite on it ... :-)

I doubt if the Chinese would agree to that. But they don't agree to
our overflights now! So, its not a biggie. But, there's a larger
issue here.

I would prefer to keep the military side of things separate from the
commercial side of things. That pays more dividends in the long run wrt
commercial SPS.

> Sure make me feel safer at night knowing we
> can meet treaty obligations to Taiwan and target
> missiles in Boost Phase .... might be worth
> 20 or 30 Billion ... LOL

So, put up a separate military sensor satellite if it makes sense to do
that. Its not clear from what you're saying here, but understand that
my SPS cannot be used as a weapon to *do* anything about boost phase
stuff. And, I don't know of any way to make a solar pumped laser of
the type I've designed for my SPS into a weapon.

As far as Asia is concerned, I guess I would favor working with the
Japanese to supply energy for the indefinite future. They import about
98% of their energy needs, and their need is growing.

They already use about half the available fleet of supertankers to
service their need. They've got to drag the oil through the straits of
Hormuz and the Malucca straihts, both surrounding by unstable powers.

I think a properly presented and backed program would find many many
takers in Japan. Japan already spends a lot for energy, already is
technically innovative, already has strong strategic interests to
reduce their reliance on Arabian crude.

> We could do that same with North Korea
> and India/Pakistan and hold out for integrated
> manufacturing or tourism (R&D prototype) facilities.

I don't know what you're getting at. My SPS cannot be used as a
weapon. I don't know how it could be used as a weapon. I don't think
its a good idea to hang military hardware of any type on a commercial
satellite.

> Sure be fun to pitch, if a President made interested
> noises we would probably lose some reactionary
> Generals/Admirals to apoplexy!

Feeding Japan, Taiwan, Singapore and so forth more sense to me.

> What do you think?
> Mike Irwin

I think a small 100 MWatt satellite could be launched on a Titan class
system. This satellite could feed power to a small island group like
the Cayman islands - who are capable of paying top dollar and absorbing
most of the output most of the time.

From here, places like Japan, for peak power matching first, for
example, make a dandy follow on market.

wm...@my-deja.com

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
In article <398737...@ix.netcom.com>,

Scott, you're confusing. So what if no wars have been fought with
ICBMs? It has nothing to do with the weapons potential of a powersat.
Nada.

> > >An
> > > M-1 Abrams CANNOT SHOOT DOWN AN ICBM.
> >
> > Yes it can. If its sitting right at the launch center and fires
> > somewhere very close to ignition.
>
> Please. Then that means that the optimal SDI system is to park a tank
> next to every enemy ICBM site. Good luck.

This has nothing to do with the weapons potential of a powersat.

>
> > By your reasoning anything that's dangerous is a weapon. That's
> > extraordinarly foolish. By this standard a pack of matches, a
cutlery
> > set, and a baseball bat constitute meaningful weapons.
>
> Hey, guess what? they've all been used as meanigful and effective
> weapons.

Militarily useful Scott? Worth the investment of billions? I don't
think so.

> And just try walking onto an airplane with a bunch of cutlery,
> and see how security takes that.

Security will wrestle you to the ground if you look dangerous in their
opinion. So what?

> > Fact is, in the sense of modern military uses, they don't.
>
> I'll be sure to tell the next Marine I meet that he doesn't need to
> carry a knife anymore.

I didn't say knife friend, I said cutlery. Please feel free he doesn't
need to take a carving knife into battle.

> > > An aerosol can of anthrax CANNOT SHOOT DOWN AN ICBM.
> > > This also classifies it as a "William Mook certified non-weapon."
> >
> > Bull. A can of anthrax doesn't cost $60 billion, and besides, if
> > deployed in the launch center well before launch, it can inactivate
a
> > large number of ICBMs.
>
> So, again,. the William Mook Definition Of A Weapon includes high cost
> and anti-ICBM capability.

No, just pointing out your inconsistencies! Fact is, the weapons
potential of my powersat is nil, despite the fact that if its
improperly used it might pose a danger to someone.

> > If you believe my powersat system is a useful and effective military
> > device giving its owners supreme control over the Earth (so much so
> > you've said previously that the moment it was deployed it would
become
> > a universally abhorred target!) - then why don't you quit all this
> > offensive and meaningless falderah and tell us all just precisely in
> > your mind makes it such a gawdawful weapon?
>
> Already done.

I don't think so.

> > You've yet to provide a clear and meaningful and cogent reply to
this
> > question, and until you do, we can all safely write your postings
off
> > as that of a meaningless crank!
>
> Hey, that's my line!

Well its a good one in this case! :)

> > > Look, a 200 solar laser spot 30-50 centimeters wide is a dandy
weapon.
> >
> > By this standard a 230 V three phase and a set of bare wires make a
> > dandy weapon. I have yet to see them as standard issue in any
military
> > force.
>
> You can't aim an electric fence, dipshit. And certainly not over
several
> hundred kilometers.

Don't go apeshit sir. I can put a set of bare wires precisely where
you tell me. Just give me several hundred kilometers of insulated wire
and I'll give anyone you want a hot foot. Ever hear of a TASER?

> Just as a small lump of lead doesn't seem to be much
> of a weapon until it's coming at you at 2,000 feet per second.

Right. You get my point. A lump of lead moving at 2000 fps is
deadly. Its not a weapon. A lot of details of how that lead got
moving, its cost, ease of repetition, etc., all determine that.

> > > It'd do a great job setting forest fires, melting asphalt roads,
> > > blinding people, burning people, burning down wooden structures
> >
> > So does 230 V three phase, if things get out of control. So what?
> > Electrical distribution can be dangerous if improperly handled.
That
> > doesn't make an electrical distribution system a weapon.
>
> Again, you can't aim and shoot electricity.

What do you think insulated wires are?

> The effort required to start
> fires with electric wires at ground level does not compare with the
> relative ease with which this could be accomplished from orbit.

So you want to hang insulated wires from orbit? Hmm...

> I weep for the future if knuckleheads like you get any sort of
power...
> such a basic lack of understanding of weapons potential is
frightening.

<shrug> Weapons potential is not a weapon. And if you understood the
physics of what I propose, you'd see the potential you credit my
powersat with doesn't even exist!

> > You have consistently shown a total ignorance and lack of
understanding
> > of four wave mixing, nonlinear optics, and all the other critical
> > factors that go into making my powersat work.
>
> Actually, I have a vague understanding... not enough to build such a
> system, but enough to know how to fuck with it.

Might I suggest you date more?

> Have you built a
> demonstrator yet?

Yes.

> Until you come to
> > understand these factors why don't you bite your tongue, reign in
your
> > stupid and foolish comments, and try to recover some modicum of
respect
> > here? You really don't know what you're talking about on this one
> > Scott. I'm sorry to see you play the rabid fool so often.
>
> Rabid? I'm hardly rabid.

Fuck no dipshit. :) You're the acme of sanity. Not!

> I'm simply laughing at the funny little
> crackpot who can't think two steps ahead.

If it makes you happy, if it makes you feel beautiful, who am I to stop
you?

> > > (wouldn't LeMay have LOVED to have such a thing), damaging
delicate
> > > structures (radar setups and the like), confusing IR sensors,
setting
> > > petroleum plants on fire, causing stampedes, altering weather,
> > > evaporating ponds, destroying crops, dazzling pilots on final
> > approach,
> > > lighting up enemy forces at night, psychological warfare ("Booga
> > booga,
> > > we control the Sun"), etc. Such a system would be useful in any
of a
> > > number of foreseeable military situations.
> >
> > <shrug> There may be a role a big laser in the sky could play in
> > combat. I'll give you that. That doesn't make my power
distribution
> > laser a weapon system nor does it make it militarily useful. If you
> > think it is, why don't you just say so and quit with all these
> > atmospherics?
>
> Atmospherics? The examples listed above would be very useful military
> lasetsat applications.

If you solved some very difficult problems that have yet to be solved,
maybe. But, I doubt you even know what these problems are, let alone
appreciate their impact on the weapons potential of lasersats.

> > > wouldn't take too long to pile up 200 solar
> > > lasers to make something capable of damaging tanks and sinking
ships.
> >
> > And how would you do that with my beam steering device pray tell?
I'll
> > give you the answer to that one. You can't!
>
> Yes, you can, and quite easily so.

Do it. Combine the laser beams of a dozen or so satellites and see how
easy it is.

>You claim your system will use a
> small laser mounted on a target vehicle to direct the power laser.
Fine,
> maybe that might even work. But if so... _two_ (better still, three)
> lasers operated by ground troops, separated by some distance, tuned to
> the same frequency and putting out appropriate information, will cause
> the power laser to be directed to a source OTHER than either of the
> directing lasers... say, split the difference and shoot directly in
> between, or offset by some angle (basic trig, folks). Several setups
> such as this would cause multiple beams to fall on a single target.

They'd have to intercept and interpret the pilot beam direct it to a
big damn laser source and have it work in conjunction with my power
plant for this to work. So, even while one might imagine something
along the lines you envision, as a practical matter, *my* network of
powersats wouldn't work that way. And, a ground based interploper
would have to fly overhead to intercept the pilot beam information.
Doing so is instantly detected by the power sat, and power beam will be
shut off as a consequence. Overpower conditions in the receiver shut
off the pilot beam, thus the 'enemy' loses their 'lock' on
their 'target'. Sorry.

Actually a variation of this idea is something I've thought about. In
the limit you have what is known as 'smart smoke'. Colloidal particles
that absorb sunlight and emit laser light toward recievers emitting
pilot beams that allow the entire cloud to operate as a single phased
array serving a wide variety of users.

> This
> would essentially "confuse" the powersat

It would turn off the power sat if you were using the collimated pilot
beam for targeting information.

... instead of the targetting
> laser being a single point, it would be multiple points, and the
result
> would be the laser would point in the "wrong" direction.

You have to intercept the pilot beam first, which would turn off the
power beam from the powersat. You would have to return 'fire' to the
reciever which would cause an overpower condition and cause the
receiver to turn off the pilot beam. The 'target' would then be lost.
The most this exercise would do is shut a power receiver down. Which
might be useful if folks didn't pay their bills.

> This can also be accomplished by having small "targetting lasers"
either
> mounted to the back of the powersat,

You'd have to clear that with the owners bub.

> or as a free-flyer nearby...

This would attract attention of the owners, and if the free flyer made
use of the power/pilot beam coupling, all they'd do is shut down power
transfer for customers - which would get EVERYBODY'S attention.

Power companies already claim ownership of the EM fields around their
power lines. A farmer many years ago erected a magnetic coupling to a
power line to tap power from it. He argued that since the coupling was
on his land, and it didn't touch the power line, he was in his rights.
He lost his case when the power company convinced the judge and jury
that they, not the farmer, owned the EM fields around the line.

I think this case and others like it would easily show that any
interloper, attachment, or interception of any powercoupling is theft
of energy and information, and the owners of the powersat would not
only operate their couplings safely, but they would take legal action
against all perpetrators. BTW, the phase information that is
introduced by the interlopers pinpoints their position as well, so
finding them is really rather simple.

> then
> ground forces would not need to do anything but duck.

<shrug> I don't think so.

> There are I'm sure other means of accomplishing this same goal. Your
> insistance that this is impossible, mated to your insistacne that this
> highly complex system can be cheaply built to withstand thirty years
> without a moments maintenance, simply points you out as someone with
no
> real understanding of the real world.

Lessee, I bought a watch 5 years ago at a department store on Father's
day. Its pretty impervious to anything I do, some might say impossible
to destroy, its highly complex with mechanical parts and everything, it
cost less than $20, I've never maintained it. I dunno. It looks
pretty real to me!

> > > > My system is rational. Sadly I can't say the same of your
> > criticisms.
> > > > Sorry.
> > >
> > > Well, it's obviously because "rational" in your universe only
applies
> > to
> > > nuclear war.
> >
> > Zing! I don't know what you're talking about there.
>
> Hardly surpising. But remember those ICBM/SDI references YOU kept
> making?

Yes. Do you remember the context? A laser in orbit might be dangerous
if used improperly. It is not a weapon as a result.

> Fact is you have
> > yet to give a single cogent military use my powersat could support.
>
> HA!!!

HA HA!!! That doesn't count either! :)

Scott Lowther

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
Michael R. Irwin wrote:

> So Scott, according to your calculations regarding Mr. Mook's
> design how long would it take his operational system to target
> and set my house on fire?

According to Mr. Mook, it could be targetted in a split second;
according to some simple experiments, it could, depending on what it's
made of, be set alight within a second or two; brick, adobe or metal
structures would not burn, of course, but they could heat up enough that
interior flamables would go.

> Would this time be significantly affected if I used a silver
> metal roof instead of a dark asphalt shingle?

Yes.

> How many houses could he set on fire in an hour?

Well, in Mr. Mook's system, every single car, airplane, boat, motorcycle
and structure on the planet could be simultaneously powered by these
space based lasers... so I suppose the answer to your question would be
"all of them."

> Do you think he could effectively harden his transceivers
> or optics against a retalitory U.S. ASAT strike?

Nope. Then again, there's no such thing as a truly hardened satellite of
any type.

> Do you think it would be possible for the U.S. military
> to monitor his facilities to perform their function of
> providing for the "common defense"

Yes. But considering that Mr. Mook is an American and would likely be
using US funds and US facilities 9in whatever fantasy universe that he
recieves a few trillion dollars to fund this), the US military would not
be too worried about defending itself against this system, but looking
at how to best use it.

> Are you concerned that he has secret over-the-horizon
> capabilities or shall we assume line of sight
> is necessary?

With the 30,000+ satellites in low Earth orbit he's proposing, there'd
be total, line of sight coverage.

Scott

Bill Bonde

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to

Michael Kagalenko wrote:
>
> Bill Bonde (std...@mail.com) wrote
> ]
> ]
> ]BrianF5070 wrote:
> ]>
> ]>

> ]> >No, what I'm saying is, in the world where space solar cells are being


> ]> >mass produced (as they would have to be for SPS production), instead
> ]> >of built in a boutique and in limited quantities (as they are now),
> ]> >they will be orders of magnitude cheaper than they are now, and there
> ]> >is no way to reliably infer what they will cost from the cost of
> ]> >today's solar cells.
> ]>
> ]> Sigh. Well, we won't agree here. I know, with complete certainty, that you
> ]> can infer what space solar cells will cost ten years from now, based solely on
> ]> the cost of today's cells and recent trends. Why don't we get back together in
> ]> 2010? If space solar cells cost less than $10/kW-hr on satellites launched at
> ]> that time, I will publicly abase myself to your superior intellect, experience,
> ]> and all around good-fellowness.
> ]>
> ]> But they won't.

> ]>
> ]Do a price vs. performance analysis on x86 compatible CPUs using only
> ]information and trends from before AMD got its act together.
>
> Solar cells do not get cheaper at the same rate as integrated circuits.
>

No, they don't. But this just shows that there are lots of different
rates at which prices change.

> In fact, the price is leveling out.
>

Look at what pushed the CPUs into such huge gains, massive demand and
real competition. Do we have that for solar cells?


> There is no reason to assume
> that the solar cells will ever get cheaper than the conventional
> or nuclear power generator (if you don't list "strnuous wishful
> thinking" as a plausible reason the way the space amateurs such as Bonde do)
>

I'm not strenuously wishfully thinking about solar cells. I'm pointing
out that claims that all products operate on the same sort of price
change scale are flawed. Solar cells don't drop in price as quickly as
CPUs because CPUs can increase their die size AND their transistor count
per area. Solar cells can only increase their efficiency. That is hard
to do when you are talking about order of magnitude changes.

But what if it suddenly doesn't matter how big the solar cell is because
you've found a way to make them in volume from the material your
manufacturing device is moving over? And you are powering your
manufacturing device with the cells you've already made.

wm...@my-deja.com

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
In article <398871...@ix.netcom.com>,

Scott Lowther <lex...@ix.netcom.com> wrote:
> Michael R. Irwin wrote:
>
> > So Scott, according to your calculations regarding Mr. Mook's
> > design how long would it take his operational system to target
> > and set my house on fire?
>
> According to Mr. Mook, it could be targetted in a split second;
> according to some simple experiments, it could, depending on what it's
> made of, be set alight within a second or two; brick, adobe or metal
> structures would not burn, of course, but they could heat up enough
that
> interior flamables would go.

I'd like to see your numbers on that one Scott.

> > Would this time be significantly affected if I used a silver
> > metal roof instead of a dark asphalt shingle?
>
> Yes.
>
> > How many houses could he set on fire in an hour?
>
> Well, in Mr. Mook's system, every single car, airplane, boat,
motorcycle
> and structure on the planet could be simultaneously powered by these
> space based lasers... so I suppose the answer to your question would
be
> "all of them."

Dumb! Putting aside the issue of targeting and control, consider
energy! Do you know how much energy it would take to set every house
and structure ablaze on the planet? Far more than even a 20 Trillion
watt power network can collect thats for sure! Need I restate the
targeting issues? Every powerbeam is commingled with a pilotbeam and
as a powerlink they can't be used to set fires and such in the first
place.

> > Do you think he could effectively harden his transceivers
> > or optics against a retalitory U.S. ASAT strike?
>
> Nope. Then again, there's no such thing as a truly hardened satellite
of
> any type.

How easy would it be to take a powersat out of commission relative to
say a hardened spy satellite?

> > Do you think it would be possible for the U.S. military
> > to monitor his facilities to perform their function of
> > providing for the "common defense"
>
> Yes.

Excellent response. So, my SPS isn't the monster you paint it.

>But considering that Mr. Mook is an American and would likely be

> using US funds and US facilities (in whatever fantasy universe that he


> recieves a few trillion dollars to fund this),

Overstating there, a hundred billion, that's all I ask! :)

> the US military would not
> be too worried about defending itself against this system,

Right again! Wow!

> but looking
> at how to best use it.

And after studying my powersats in detail, they'd find there is no
militarily useful role for them.

> > Are you concerned that he has secret over-the-horizon
> > capabilities or shall we assume line of sight
> > is necessary?
>
> With the 30,000+ satellites in low Earth orbit he's proposing, there'd
> be total, line of sight coverage.

Only in a fully populated system. By that time, the benign nature of
the power links will have been fully established.

> Scott

Good questions Mike, reasonably sane answers Scott! Excellent!

William Mook
Orbatek, Inc.

Michael Kagalenko

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
George William Herbert (gher...@gw.retro.com) wrote
]Michael Kagalenko <mkag...@lynx.dac.neu.edu> wrote:
]> Solar cells do not get cheaper at the same rate as integrated circuits.
]> In fact, the price is leveling out. There is no reason to assume

]> that the solar cells will ever get cheaper than the conventional
]> or nuclear power generator (if you don't list "strnuous wishful
]> thinking" as a plausible reason the way the space amateurs such as Bonde do)
]
]How on earth are you passing classes with research skills this poor, Michael?

How on earth are you hoping to deceive anyone to give you money for your
projects when you are so obviously dishonest ?

]
]No, they don't get cheaper at the same rate as ICs. This part is correct.


]
]The price is not leveling out. It continues to drop,

"The price is leveling out" and "the price continues to drop" are
not mutually exclusive statements. The former means that the rate of
price declines is decreasing.

] and basic manufacturing


]and technology improvements already in the R&D and production pipelines will
]see it through at least 5 years more price drop, down to nearly the price
]parity with conventional power.

Wishful thinking, unfounded assertions. Baring breakthroughs in basic
research which can not be predictd, or guaranteed by increase in funding,
solar cells will not be competitve with conventional and nuclear power.

]There is plenty of reason to assume they will get cheaper than conventional
]power systems.

Wishfu thinking is not the same as reason , no matter how strenous it is.

] If we assume *any* new R&D successes

If you assume new R&D successes, you are guilty of wishful thinking,
as such are never guranteed. Furthemore, you'd also have to assume that
there won't be any R&D successes in conventional and nuclear power
production.

]then the price will


]continue to drop beyond conventional source parity as described above,
]into clear price win territory. At this point, given the decades long
]history of success after success in developing new technologies in this
]area, it would be folly to predict that it's going to come to a screaming
]halt in the near future.
]
]There are literally hundreds of relevant documents linked off of:
]http://pix.nrel.gov:8020/BASIS/nich/www/public/SDW?M=1&W=subcode='SEP'+order+by+pubyear/desc

]http://www.epri.com/targetLinks.asp?program 7933&value=00T053.1
]http://www.ases.org/solarguide/factbase.html


]
]I could go on but it would be a waste of electrons, which are still being
]produced using fossil fuel, so I want to save a few.

Are you still trying to scam investors into funding your fraudulent
space schemes ?

Scott Lowther

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
wm...@my-deja.com wrote:
>
> In article <398871...@ix.netcom.com>,
> Scott Lowther <lex...@ix.netcom.com> wrote:
> > Michael R. Irwin wrote:
> >
> > > So Scott, according to your calculations regarding Mr. Mook's
> > > design how long would it take his operational system to target
> > > and set my house on fire?
> >
> > According to Mr. Mook, it could be targetted in a split second;
> > according to some simple experiments, it could, depending on what it's
> > made of, be set alight within a second or two; brick, adobe or metal
> > structures would not burn, of course, but they could heat up enough
> that
> > interior flamables would go.
>
> I'd like to see your numbers on that one Scott.

Easy experiment: take a simple lens, and in direct sunlight hold it
over paper, wood, asphalt, whatever and locate it so that the "sunspot"
is 1/14 the diameter of the lens itself. Paper and dried vegetable
products burn in seconds, asphalt melts. Considerably more damage would
be achieved if the spot was half a meter wide, of course..s slight
breezes would not provide the same cooling properties, and melted tar
wouldn't nbe able to run out of the sunlit spot before they overheated.


> > > How many houses could he set on fire in an hour?
> >
> > Well, in Mr. Mook's system, every single car, airplane, boat,
> motorcycle
> > and structure on the planet could be simultaneously powered by these
> > space based lasers... so I suppose the answer to your question would
> be
> > "all of them."
>
> Dumb! Putting aside the issue of targeting and control, consider
> energy! Do you know how much energy it would take to set every house
> and structure ablaze on the planet? Far more than even a 20 Trillion
> watt power network can collect thats for sure!

Nonsense. Colorado just lost something like 30,000 acres in Mesa Verda
because someone flicked a lit cigarette. And as Dresden showed us, the
more fires you start, the less energy is needed to start another... it's
a self-propogating "living" system. Firestorms are wonderfully
energy-efficient weapons of war.


> How easy would it be to take a powersat out of commission relative to
> say a hardened spy satellite?

It'd be a snap, as I'm sure iraq or North Korea would be happy to
demonstrate if we start lobbing 40,000 satellites into the sky.

> > > Do you think it would be possible for the U.S. military
> > > to monitor his facilities to perform their function of
> > > providing for the "common defense"
> >
> > Yes.
>
> Excellent response. So, my SPS isn't the monster you paint it.

But it can be. That' the point. You have, as you always do, completely
missed the point of the post to which I repsonded. The quote: "If you


ever see someone building, or even proposing, an actual solar
power satellite with an antenna big enough to serve as a space-based
death ray weapon, fell free to cry foul, scream bloody murder, and

whatnot. " Nobody said that the SPS WOULD BE a weapon, but that IT CAN
BE a weapon.


> > the US military would not
> > be too worried about defending itself against this system,
>
> Right again! Wow!
>
> > but looking
> > at how to best use it.
>
> And after studying my powersats in detail, they'd find there is no
> militarily useful role for them.

And that's exactly what the DoD's public position will be, at the same
time they are covertly modifiying your system to serve their needs. And
at the same time our opposition is planning on bringing the system down
because they know what it's capable of. Because unlike your childish
idea of dropping electrified copper lines from orbit onto the enemy, a
laser beam from orbit isn't nealry so silly, and is much more militarily
useful. A few trillions watts worth of orbiting lasers would be a dandy
weapons system, as any sane person can see. And no amount of "no way,
uh-uh" is going to convince America's adversaries that it is a benign
system.

>
> > > Are you concerned that he has secret over-the-horizon
> > > capabilities or shall we assume line of sight
> > > is necessary?
> >
> > With the 30,000+ satellites in low Earth orbit he's proposing, there'd
> > be total, line of sight coverage.
>
> Only in a fully populated system. By that time, the benign nature of
> the power links will have been fully established.

Just as nuclear power plants have been shown to be completely benign.
Not a single person has been killed by a nuclear power plant weapons
system in the 50 years or so they've been around. Try launching massive
nuclear power plants into LEO and see what people think.

Scott Lowther

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
wm...@my-deja.com wrote:

> Since when is delivering copious power at very low cost across the globe
> a problem for anyone?

Ask OPEC. Care to take a guess at how well Saudi Arabia would handle no
longer recieving large sums of Western money?

What was the last war the US got involved in in a serious and determined
way? And was or was not oil (i.e. energy) a major part of it?

> > Because unlike your childish
> > idea of dropping electrified copper lines from orbit onto the enemy, a
> > laser beam from orbit isn't nealry so silly,
>

> If you want a nice clean laser beam capable of delivering lots of power
> from orbit in a compact space (but not nearly so much to be a military
> threat!) you need some sort of method to control the blooming and
> atmospheric distortions of doing that. This is achieved in my system
> through the use of pilot beams.

Whicvh you have yet to demonstrate. How do you control
blooming/distortion in a dust storm? Blizzard? Driving under trees?
Under serious cloud cover? Under smoke? While covered in mud? While
these 40,000 LEOsats keep bumping into the kilotonnes of trash left in
orbit from their launch and erection?


Without this technology the idea of
> using lasers as weapons from space to strike at ground based targets is
> nearly as silly as dropping copper wires from orbit! That's why SDI had
> pop up lasers and so forth. Even nuclear powered beams millions of
> times more powerful than the one's I'm proposing weren't militarily
> useful for ground targets!

That's because SDI would have had only a relative few. You're proposing
worldwide coverage with the capability of perhaps billions of these
beams at any one time.


> > And no amount of "no way,
> > uh-uh" is going to convince America's adversaries that it is a benign
> > system.
>

> <shrug> Ever read the stuff folks said about steam engines on boats
> when they first appeared? Especially after the explosion of one during
> a Presidential review on the Mississippi?

Ever read some of the horror stories the environmental movement put out
regarding nuclear power plants? When was the last time the US built one
of these plants? You ignore public and worldwide opinion at your peril.

> Nuclear power plants have had a bad rap, their association with nuclear
> weapons doesn't help.

Lasers satellites are equally badly regarded, and for similar reasons.

> > Not a single person has been killed by a nuclear power plant weapons
> > system in the 50 years or so they've been around.
>

> You speak as if half a million people weren't killed in the US atomic
> bombing raids against Japan. Duh!

A) We didn't drop a nuclear power plant on Japan.
B) See A).
C) Hal;f a million people weren't killed in Japan by nukes in any event.
200,000 conservative, 400,000 ravign liberal.
D) They deserved it in any event (but that's another debate).

As far as power comparisons:

I'll admit to not having exact numbers ready for one of the
Dresden/Hamburg/Tokyo class raids. But let's take a guess: 1,000
bombers, each with 10,000 kg of bombs (Lancaster). Assume for simplicity
that that's 10,000,000 kg of TNT, or 10 kilotonnes. This is 42E12
Joules, spread over several hours.

In comparison, assume laser beams of 200 (space) solar at .5 meters
diameter. That works out to about 55 kilowatts per beam. To equal 42E12
Joules, you'd need 764 million beam-seconds. Sounds like a lot. But
since to power a major city each LEO rectifying satellite (assumeing
the 40K sat constellation) must power, I dunno, 100,000 cars plus say
500,000 other targets (and this is likely quite conservative; and keep
in mind that with effectively only one satellite in the sky at any one
time, each sattelite must be capable of powering Las Vegas or NYC at
rush hour, all by itself), this means that at any one time AT LEAST
600,000 lasers beams are available. These 600K laser beams will have to
operate for about 1273 seconds to dump Dresden class energy into a
target city. In that 21 minute period, several LEO rectifying satellites
would pass overhead, so there'd be considerable hand-off, and none of
these satellites would have to work any harder than their design called
for anyway (remember, this is simply the power require to run a major
city and all vehicles therein).

So your satellite SPS system would be quite capable of dumping as much
destructive energy into a target as an old-style bombing campaign. Of
course efficiency would rapidly drop off, as smoke cover blankets the
target; but by then there might not be much need to continue.

And as to "conspiracy theories:" assume your SPS system goes forward.
This will very rapidly become a matter of national security, just as the
interstate highway system, the current power grid, air traffic control,
the rail systems and the phone lines are. Like it or not, Uncle Sam will
be leaning over your shoulder. Design changes will be mandated, and
don't act all shocked and surprised if every few LEOsats has a few minor
changes in it's targetting system. It's called "emminant domain." "They"
can come in and take your toys away from you if they so like, and have
repeatedly done so in the past.

George William Herbert

unread,
Aug 2, 2000, 3:00:00 AM8/2/00
to
Michael Kagalenko <mkag...@lynx.dac.neu.edu> wrote:
>George William Herbert (gher...@gw.retro.com) wrote
>] [...]

>]No, they don't get cheaper at the same rate as ICs. This part is correct.
>]The price is not leveling out. It continues to drop,
>
> "The price is leveling out" and "the price continues to drop" are
>not mutually exclusive statements. The former means that the rate of
> price declines is decreasing.

The rate of price declines continues to match the 1997 and 1998
predictions which I posted starting around that time, and the volume
of sales continue to match the 1997 and 1998 predictions.

This is some variant on "lying with statistics", Michael.

Is the price "leveling out" in the sense of no longer matching
the predicted decline? No. Is it "leveling out" in that it's
not getting $1/watt cheaper each year? Yes. Is it "leveling out"
in not getting the predicted percentage cost drops? No.

Pick a meaningless measure and you can say anything you want to.
Fact remains, costs are dropping and will continue to drop for
at least the forseeable future.

>] and basic manufacturing
>]and technology improvements already in the R&D and production pipelines will
>]see it through at least 5 years more price drop, down to nearly the price
>]parity with conventional power.
>
> Wishful thinking, unfounded assertions. Baring breakthroughs in basic
> research which can not be predictd, or guaranteed by increase in funding,
> solar cells will not be competitve with conventional and nuclear power.

This is absolutely untrue, and you would know if if you bothered to
read even the executive summaries of the references I posted, Michael.
Why won't you do that? Why won't you do research? Is your ego so
big that it encompasses all of engineering, physics, economics,
and policy analysis?

The cost breakeven point is around 2006, if the predicted cost drops
continue for at least that long. The fundamental breakthroughs for
the next 5-8 years worth of productizing are already demonstrated and
in hand in labs. Ergo, it's unlikely to stop improving before breakeven,
even if you assume no more new basic underlying new innovations or
discoveries are made.

>]There is plenty of reason to assume they will get cheaper than conventional
>]power systems.
>
> Wishfu thinking is not the same as reason , no matter how strenous it is.

Failure to look down and read reference reports and studies basically
handed to you on a silver platter with URLs and all is no excuse
for lousy understanding of the actual state of solar photovoltaic
production, research, and economics.

Look, Michael, it's this simple: I'm not claiming to be Right.
I'm no more perfect than the next guy. But I have done my homework,
and I'm perfectly willing to provide research and references when
appropriate, and have done so several times recently when you made
unsupportable statements. You can't wipe that away by calling me
a wishful thinker or fraud. No matter what *I* may be, the source
materials I am quoting exist and say what I'm saying they do.
They may not in fact be correct. I am willing to accept that I
may from time to time use as a reference materials which do not
turn out to be entirely accurate... I've had it happen before.

But I am not going to take the unreferenced word of an undereducated
snot that professional researchers and industry experts whose work
I reference are wrong. If you're not going to bother to either bring
up counterargument references from equivalently competent sources
or *gasp* use some actual numbers in your own analysies, you are
sitting here looking like a grade A idiot rather than contributing
usefully to any higher understanding on anybodys part, starting with
your own and certainly ending with the hundred-thousand odd readers
of the newsgroup.

If you want to continue posting things which make you look like
a grade A idiot, then I certainly can't stop you. But consider
what it's doing to you, your reputation and credibility.
Being wrong but having had good evidence and reasons for it
isn't shameful, but being wrong and having nothing to justify
it but being a headstrong ass about it is going to make it
hard to find employment among the net savvy... will your next
job interviewer have run a Deja search on your name?
A Google search, Altavista, etc? A lot of them do.
If your next potential employer asks about it,
can you pull something from elsewhere than your ass to
explain and justify what you've posted recently?
If you can, why haven't you posted that here?

You don't have to like it. But I've had people I was
being interviewed by pull out Usenet posts of mine and
on at least one occasion a technical paper I'd written
and start challenging me about its methods, data,
and validity. If you get into that hotseat and don't
have credible research methods, sources, and detailed
arguments to justify your conclusions, you're going to
get fried in it.


-george william herbert
gher...@retro.com


wm...@my-deja.com

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
In article <3988B6...@ix.netcom.com>,

Melting tar won't set the world ablaze, and houses aren't generally made
of dried vegetable products or other tender suitable for burning.

A more reliable measure would be the energetics involved. The US Army
relates the destructiveness of NAPALM and other fire sources to the
quantity used. To set a building ablaze, reliably, requires a certain
amount. This can be related to the energy contained in the NAPALM.
Using these figures one can determine the amount of NAPALM and hence the
amount of energy (assuming an equivalent efficiency of energy delivery,
which is questionable, because a burning gel is much more efficient at
starting fires than radiant heat!) the system would have to deliver to
set the world on fire. (again ignoring practical problems related to
targeting and control).

> > > > How many houses could he set on fire in an hour?
> > >
> > > Well, in Mr. Mook's system, every single car, airplane, boat,
> > motorcycle
> > > and structure on the planet could be simultaneously powered by
these
> > > space based lasers... so I suppose the answer to your question
would
> > be
> > > "all of them."
> >
> > Dumb! Putting aside the issue of targeting and control, consider
> > energy! Do you know how much energy it would take to set every
house
> > and structure ablaze on the planet? Far more than even a 20
Trillion
> > watt power network can collect thats for sure!
>
> Nonsense. Colorado just lost something like 30,000 acres in Mesa Verda

Well "all of them" aren't in Colorado are they?

> because someone flicked a lit cigarette. And as Dresden showed us, the
> more fires you start, the less energy is needed to start another...
it's

Yes a firestorm. And how much energy was expended over what period of
time to start this firestorm? What is the efficiency of the energizing
mechanism relative to radiant energy? How do you target that much
radiant energy through a variable and dispersive atmosphere from a large
number of sources? Clearly achieving a Dresden or a Hiroshima style
firestorm using lasersats is nearly impossible, if its doable at all!

> a self-propogating "living" system. Firestorms are wonderfully
> energy-efficient weapons of war.

<shrug> But how to start them reliably, and control them once started
to achieve the ends desired. Tactical nukes might be useful here, but
certainly not solar powered laser sats limited to 200 solars controlled
by pilot beams.

> > How easy would it be to take a powersat out of commission relative
to
> > say a hardened spy satellite?
>
> It'd be a snap, as I'm sure iraq or North Korea would be happy to
> demonstrate if we start lobbing 40,000 satellites into the sky.

I don't think so.

> > > > Do you think it would be possible for the U.S. military


> > > > to monitor his facilities to perform their function of
> > > > providing for the "common defense"
> > >
> > > Yes.
> >
> > Excellent response. So, my SPS isn't the monster you paint it.
>
> But it can be.

Not!

> That' the point.

If it can't be, I've responded to your point. Its only your pig
headedness that doesn't allow you to see that!

> You have, as you always do, completely
> missed the point of the post to which I repsonded. The quote: "If you
> ever see someone building, or even proposing, an actual solar
> power satellite with an antenna big enough to serve as a space-based
> death ray weapon, fell free to cry foul, scream bloody murder, and
> whatnot. "

Well, your screaming is not warranted in the case of my SPS, and that's
my point!

>Nobody said that the SPS WOULD BE a weapon, but that IT CAN
> BE a weapon.

Well, you're wrong here Scott, what more can I say?

> > > the US military would not
> > > be too worried about defending itself against this system,
> >
> > Right again! Wow!
> >
> > > but looking
> > > at how to best use it.
> >
> > And after studying my powersats in detail, they'd find there is no
> > militarily useful role for them.
>
> And that's exactly what the DoD's public position will be, at the same
> time they are covertly modifiying your system to serve their needs.
And

Yeah right! What other conspiracie theories do you believe in?

> at the same time our opposition is planning on bringing the system
down
> because they know what it's capable of.

Since when is delivering copious power at very low cost across the globe
a problem for anyone?

> Because unlike your childish


> idea of dropping electrified copper lines from orbit onto the enemy, a
> laser beam from orbit isn't nealry so silly,

If you want a nice clean laser beam capable of delivering lots of power
from orbit in a compact space (but not nearly so much to be a military
threat!) you need some sort of method to control the blooming and
atmospheric distortions of doing that. This is achieved in my system

through the use of pilot beams. Without this technology the idea of


using lasers as weapons from space to strike at ground based targets is
nearly as silly as dropping copper wires from orbit! That's why SDI had
pop up lasers and so forth. Even nuclear powered beams millions of
times more powerful than the one's I'm proposing weren't militarily
useful for ground targets!

>and is much more militarily


> useful. A few trillions watts worth of orbiting lasers would be a
dandy
> weapons system,

There are a whole host of difficulties you are totally unaware of.

> as any sane person can see.

As any person ignorant of these difficulties might fear. That ignorance
won't change the reality of the system.

> And no amount of "no way,
> uh-uh" is going to convince America's adversaries that it is a benign
> system.

<shrug> Ever read the stuff folks said about steam engines on boats
when they first appeared? Especially after the explosion of one during
a Presidential review on the Mississippi?

We won't start with 20 trillion watts. We'll start with a few hundred
million. Over time people will come to accept power from the sun
delivered and channeled through a satellite system as being as safe and
reliable as the sun.

> >
> > > > Are you concerned that he has secret over-the-horizon
> > > > capabilities or shall we assume line of sight
> > > > is necessary?
> > >
> > > With the 30,000+ satellites in low Earth orbit he's proposing,
there'd
> > > be total, line of sight coverage.
> >
> > Only in a fully populated system. By that time, the benign nature
of
> > the power links will have been fully established.
>
> Just as nuclear power plants have been shown to be completely benign.

Nuclear power plants have had a bad rap, their association with nuclear
weapons doesn't help.

> Not a single person has been killed by a nuclear power plant weapons


> system in the 50 years or so they've been around.

You speak as if half a million people weren't killed in the US atomic
bombing raids against Japan. Duh!

>Try launching massive


> nuclear power plants into LEO and see what people think.

We're not talking about nuclear power plants in LEO are we? We're
talking about SPS. Don't change the subject.

BrianF5070

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
Bill Bonde wrote:

>Do a price vs. performance analysis on x86 compatible CPUs using only
>information and trends from before AMD got its act together.

Bill,

I'm not certain precisely what you are getting at here. Certainly Moore's Law
has been known for almost 40 years, and not many people dispute that it
predicts the performance and price of microprocessor circuits remarkably well.
To my knowledge, this is true with or without AMD.

AMD may have pressured Intel's margins, but it didn't really change the
technology curve in a meaningful way.


James Hunter

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to

BrianF5070 wrote:

Could be, but some are beginning to suspect Moore's Law is rigged.
Packing more "stuff" into CPU's to support the "information" superhighway
is not what we originally had in mind.


Bill Bonde

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to

James Hunter wrote:
>
> BrianF5070 wrote:
>
> > Bill Bonde wrote:
> >
> > >Do a price vs. performance analysis on x86 compatible CPUs using only
> > >information and trends from before AMD got its act together.
> >
> > Bill,
> >
> > I'm not certain precisely what you are getting at here. Certainly Moore's Law
> > has been known for almost 40 years, and not many people dispute that it
> > predicts the performance and price of microprocessor circuits remarkably well.
> > To my knowledge, this is true with or without AMD.
> >
> > AMD may have pressured Intel's margins, but it didn't really change the
> > technology curve in a meaningful way.
>
> Could be, but some are beginning to suspect Moore's Law is rigged.
>

Rigged? I'm claiming that it is borderline antitrust. (Yes, this is
offtopic)

Thomas F. Radloff

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to

Scott Lowther <lex...@ix.netcom.com> wrote in message
news:3988DA...@ix.netcom.com...
>... While

> these 40,000 LEOsats keep bumping into the kilotonnes of trash left in
> orbit from their launch and erection?

You really think that by the time folks _could_ build in space on that
scale, they'll still have the orbital debris problem?

> That's because SDI would have had only a relative few. You're proposing
> worldwide coverage with the capability of perhaps billions of these
> beams at any one time.

You keep jumping to a fully developed system. But said system could only
grow from early prototypes. There is no public hysteria about the issue now,
and it's hard to see it developing over one, or a handful of prototypes.
Once in place and the benefits become visable, it will be even more
difficult to fan the flames of hysteria.

Your argument is essentially that because of the raw amounts of power
involved, powersats _could in principle_ be used as weapons. Mooks argument
is that A) although there's lots of power involved, it's still not enough to
cause large scale damage, and B) that there are other considerations that
weapons systems have to address, that his design does not.

But suppose you _had_ to design some powersats. What managerial/political
control mechanisms could you design into it to thwart such use? The system
that controls the overall operation could be _easily_ (relative to the scale
of the entire problem) be designed from day 1 to detect and avoid
concentrations of power. Early prototypes will have small numbers of
legitimate receivers. While maintaining the pilot/power beam link, the
system is also aware of _where_ the power is going. The factory that
consumes lots of power is not going to all of a sudden be located underneath
the Whitehouse. How about physical constraints in the design that couldn't
be worked around merely with software?

As for Uncle Sam hijacking it: The Feds are good at twisting arms behind the
scenes, in order to make minor changes to accomodate their interests.
Notice that there is a new Carnivore at large. If the system even _could_
direct lots of beams in concert, simply by making minor changes to the
control software, the Feds could very well become a nuisance for the reasons
you state. But if they had to go to all the trouble of confiscating an
enormous amount of private wealth, and totally redesigning the control
software in order to specifically turn it into a weapon, THEN they'd have a
public uproar. If BMD scares some folks, imagine the outcry over
MadUncleSam's mission to build death beam weapons!


> ... You ignore public and worldwide opinion at your peril.

But with powersats, we can ignore said opinion with IMPUNITY! :)

TFR

wm...@my-deja.com

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
In article <3988DA...@ix.netcom.com>,

Scott Lowther <lex...@ix.netcom.com> wrote:
> wm...@my-deja.com wrote:
>
> > Since when is delivering copious power at very low cost across the
globe
> > a problem for anyone?
>
> Ask OPEC. Care to take a guess at how well Saudi Arabia would handle
no
> longer recieving large sums of Western money?

Fact is, the reason OPEC can't raise prices more is that by doing so
they cut their own throats because their wealth is tied to the economic
productivity of the world.

Fact is, global productivity is directly tied to the cost and amount of
energy available.

Fact is, economic growth worldwide is constrained by limited supplies
and high cost of energy.

Therefore, the addition of another low cost source of energy to our
energy mix right now, will actually INCREASE the amounts of money taken
in by OPEC. Their economic analysts know this. Which is why they
won't complain as you suggest. They may change their product mix, more
plastics and fertilizer, less fuel oil, slightly less gasoline. But
their total revenue will rise as the economic activity of the world
increases. They may get a smaller percentage of the pie, but the pie
will be growing if energy is cheap.

They may kibbitz about the *rate* of growth of the SPS system - and
that will be tied to economic growth rates, and *that* to aggregate
rates of taxation and regulatory limits on international capital
formation. But, that won't be a problem until after the first 100 SPS
units are operational even in a world as mismanaged and corrupt as ours.

Of course in a well managed world without corruption, things would be
even better! :)


> What was the last war the US got involved in in a serious and
determined
> way? And was or was not oil (i.e. energy) a major part of it?

When someone threatened to cut off *our* supply! Duh! We didn't give a
fig about OPEC and all those sheiks.

[snip]

> > Without this technology the idea of
> > using lasers as weapons from space to strike at ground based
targets is
> > nearly as silly as dropping copper wires from orbit! That's why
SDI had
> > pop up lasers and so forth. Even nuclear powered beams millions of
> > times more powerful than the one's I'm proposing weren't militarily
> > useful for ground targets!
>
> That's because SDI would have had only a relative few. You're
proposing
> worldwide coverage with the capability of perhaps billions of these
> beams at any one time.

Eventually, not at first.

[snip]

> > Nuclear power plants have had a bad rap, their association with
nuclear
> > weapons doesn't help.
>
> Lasers satellites are equally badly regarded, and for similar reasons.

No one has been killed as a result of a laser attack during wartime.
The psychodynamics are just not the same as for nuclear weapons. You
are creating problems where none need exist by drawing these parallels.

> > > Not a single person has been killed by a nuclear power plant
weapons
> > > system in the 50 years or so they've been around.
> >
> > You speak as if half a million people weren't killed in the US
atomic
> > bombing raids against Japan. Duh!
>
> A) We didn't drop a nuclear power plant on Japan.

No, just a nuclear weapon fueled by the same material, and it was
highly publicized, easily remembered, and the public's first broad
introduction to the uses of nuclear power. ATOMIC in the 1940s became
equivalent to words like virile, strong, powerful, etc. Just look at
the ads. Mushroom clouds with a Boron atom festooned everything from
candybars to soap to paint.

> B) See A).

Still just as wrong as when you first said it.

> C) Hal;f a million people weren't killed in Japan by nukes in any
event.
> 200,000 conservative, 400,000 ravign liberal.

Well, there are the immediate deaths, and then the long term effects of
radiation through the generations. Its not liberalism, just science.

> D) They deserved it in any event (but that's another debate).

This really marks you off as a raving lunatic! Even the defense
department, and President Truman, and all the powerful national leaders
at the time called the use of the bomb a regretable act that was done
to save lives, both Japanese and Allied, not take them. An invasion of
the islands by ground troops would have cost millions of Allied dead
and perhaps tens of millions of Japanese dead.

Well you've proven my point. 21 minutes isn't instantaneous is it.
And your comments about one satellite being able to do this are all wet.

And, you've forgotten the effects of cooling, radiation, convection,
etc. Over 21 minutes, over how many square meters? you didn't
calculate that, what's the albedo? thermal conductivity? air temp?
humidity? material being heated? combustion temperatures?

Fact is, you haven't shown that laser irradiation - even on this scale -
could create a firestorm. Sorry.

> So your satellite SPS system would be quite capable of dumping as much
> destructive energy into a target as an old-style bombing campaign. Of
> course efficiency would rapidly drop off, as smoke cover blankets the
> target; but by then there might not be much need to continue.

Even assuming it were physically possible to target non-receivers,
which I don't believe is possible to do, uncontrolled irradiation of
say an asphalt shingle, or roadway, would drive off a lot of smoke, but
cause very little collateral damage. So, by your own argument, you
couldn't continue irradiation.

Understand dropping napalm on a target is different than irradiating
it. Napalm will burn with a flame and that flame will spread as the
material fuels it. Irradiating a target will not necessarily cause
combustion, and combustion will not necessarily continue once
irradiation is removed, and smoke and dust and charring will cut the
effectiveness of irradiation.

> And as to "conspiracy theories:" assume your SPS system goes forward.
> This will very rapidly become a matter of national security, just as
the
> interstate highway system, the current power grid, air traffic
control,
> the rail systems and the phone lines are. Like it or not, Uncle Sam
will
> be leaning over your shoulder. Design changes will be mandated, and
> don't act all shocked and surprised if every few LEOsats has a few
minor
> changes in it's targetting system. It's called "emminant
domain." "They"
> can come in and take your toys away from you if they so like, and have
> repeatedly done so in the past.

Only if their a threat or uniquely and supremely useful or in some
other way destabilizing to our way of life. You have yet to
demonstrate any of this.

brian...@my-deja.com

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
In article <398663ae...@nntp.ix.netcom.com>,
simberg.i...@trash.org (Rand Simberg) wrote:
> On 01 Aug 2000 05:34:51 GMT, in a place far, far away,
> brian...@aol.com (BrianF5070) made the phosphor on my monitor glow
> in such a way as to indicate that:
>
> >>>If you ever have the chance to walk the floor at LockMart's
Titan/Atlas
> >>>assembly facility, you'll see the degree to which the vehicles are
> >>>commoditized.
> >>
> >>You obviously don't understand the meaning of commoditized.
> >>
> >>You mean that I can buy Titan futures on the Chicago Exchange? Can
I
> >>buy generic Titans?
> >
> >I don't know if you can by Titan futures, but I don't believe there
is any
> >reason why someone couldn't create that market if desired.
>
> Then let me inform you that you cannot buy Titan futures. The reason
> is that there is insufficient volume in the market. It is not a
> commodity.

I stated I was uncertain about Titans. However, I also do not see any
reason why one would not be able to make a futures market if desired.
In fact, this is an interesting idea - there might be a profit in it.

And let me inform you that you *can* buy Atlas options.

> >And yes, you can buy generic Titans.
>
> No, you can't. They don't exist, in the sense that commodities do.

You certainly *can* buy generic Titans and Atlas's and Protons, etc.
They are manufactured in generic configurations for standard prices.
That is what ILS's little launch book is for.

> >Go to Lock/Mart, say that you want this model Titan, with
> >such and such an upper stage, and the manager will pull out his
little book, go
> >bing, bang, say this is how much it will cost, this is when it will
be ready,
> >then you enter the order and X months later it rolls off the
assembly line.
>
> That's not a commodity. You continue to demonstrate a profound lack
> of understanding of the term "commodity."

<grin> I find this ironic. Since there is no standard definition for
“commodity”, I've been trying to fit myself into the context you are
using. Which is tricky because it seems to be somewhat variable. But,
to get this down, let's run by some of the definitions of commodities
that are commonly used:

1) Dictionary says A) That which affords convenience or profit, esp. in
commerce, including everything movable that is bought or sold or B) an
element of wealth; an economic good.

2) A market definition of a commodity that is often used is any item
that is bought and sold on an open, licensed, and regulated exchange.

3) A functional definition that is often used is any product or service
that customers perceive
to be exactly the same.

4) A contractual definition of a commodity is a standardized contract
for future delivery of a specific product traded through licensed
regulated exchanges.

None of these definitions is perfect, nor is any universally used. The
term commodity is highly dependent on context. I have been trying to
follow your context, but it seems to keep changing. It should be
apparent, though, that rockets are at least partially commoditized
based on the definitions above.

> >Better yet, go to ILS and order a Proton.
> >
> >This is a product that has been commoditized.
>
> No, it's not.

A Proton certainly fits definition 1 or 3, and options are written on
Proton launches, so definition 4 is partially fulfilled. These options
are negotiable instruments that can be traded, although there is
currently no public market set up to do that. Which might be an
interesting business possibility – I can see a potential profit in
setting up such a market.

> >Rockets are commoditized in much the same way as automobiles.
>
> Automobiles are not a commodity.

Depends on your definition. In business school, you'll find that
automobiles (along with many other manufactured items - ICs, solar
panels,e tc.) are, in fact, defined as secondary commodities. And
dealers certainly use typical commodity-type risk hedging instruments.

> >Or airplane
> >tickets. There is a certain amount of haggling and arguing and
customization,
> >but by and large both buyer and seller know the cost and schedule
and come
> >rather quickly to agreement.
>
> Airplane tickets are commoditized, but not as fully as the other items
> mentioned.

You seem stuck on this commodity black and white thing. You have to
realize that commoditization is a process, not a result. I will repeat
that *nothing* is ever fully commoditized. Not even the examples you
gave.

Just as an example, I used to live in Colorado. Colorado Public
Service Co had a program called Windsource, where customers could sign
up to receive blocks of "environmentally friendly" wind generated
electricty for a premium (around 30%, IIRC). The program was
immediately oversubscribed, and CPSCo has been busily installing
turbines ever since. The actual commodity received by the customer was
a kW-hr like any other, but enough customers felt the urge to buy
branded kW-hr into the grid that the program was (and is) a huge
success.

> >These markets are less mature than pork bellies,
> >as you eloquently state, but they are pretty mature.
>
> Yes, and rockets and access to LEO are nowhere near that state.

Near the state of pork bellies? No. Considerably beyond the state
where it can be considered immature? Yes, very much so. That is
rockets, of course, and access to LEO for satellites.

Access to LEO for humans is very much immature and not commoditized.

> >Chemical plants are a different story. If you want an ethylene
plant, you go
> >to two or more engineering and construction firms. You give them
the location
> >and environmental parameters. They take 3 - 6 months to develop a


design, cost
> >estimate, and contract (for Mr. Mook's benefit - about 3/4ths of all
chemical
> >plant contracts are reimbursable. Yet cost effective). Then you
run it
> >through your in house design teams and business teams, make a go/no-
go
> >decision, arrange financing, select a contractor, and start
construction.
>

> Chemical plants are not commoditized, but chemicals (their outputs)
> are.

Oversimplification again. Oil is oil, right? Perceived by customers
to be exactly the same? Wrong. Saudi sour, light sweet Texas, North
Sea Brent, and syncrude are all different and valued differently. The
very same platform will have different prices for different loads for
different customers and feeds. However, in general the spot market
allows transparent transactions with minimal market inefficiencies so,
in that sense, it is a market commodity - it has been commoditized far
more than, say, dimethoxymethane.

Again, commoditization is a process, not a result. For the last 30
years, upstream facilities and refineries have been hashing out common
ground for valuation of materials based on upgrading efforts and
results. A lot of good progress has been made, but it will never be
completed.

> >This is a product that has *not* been commoditized.
>

> Never claimed otherwise. Nor has a rocket plant.

Rocket plants have not been, but rockets (their outputs), in many
cases, *have* been.

> >>>You'll see how economies of scale are employed. You'll realize
> >>>that there is no way this manufacturing facility (or industry) is
immature.
> >>
> >>I've already said that ELV's are mature. This has nothing to do
with
> >>the future price of access. Access to space will be commoditized,
and
> >>the industry mature, when I can buy a ticket on Priceline.
> >
> >I think this is a rather narrow definition. Is it not possible to
commoditize
> >cargo delivery to LEO before it is possible for human delivery there?
>
> Yes, but I cannot buy cargo to LEO on Priceline either, nor is there a
> futures market for it.
>
> >>>I know what you will say. I also don't agree with it. There have
been
> >>dozens
> >>>of programs. They all have received investment.
> >>
> >>Many of your examples, particularly the ones that had some hope of
> >>significant cost reductions, received little or no investment. None
> >>of your examples were relevant to my claim.
> >
> >But just to note, your originally claim was that as soon as someone
made an
> >investment in reducing launch costs by an order of magnitude, they
were 100%
> >guaranteed of results. You didn't include qualifiers such as, "a
sufficient
> >investment."
>
> Well, excuse me. I thought that this would be obvious.
>
> Did you really think that I meant that it would happen as soon as
> someone put up one percent, or ten percent of the money? Do you really
> want people in the newsgroup to think that you are that dim?

Do you really want people in the newsgroup to think you are dim enough
to believe that a weak argument can be strengthened by ad hominem
attacks? Or are you merely dimsagreeable? Hey, Rand, do you enjoy
kicking puppies, or do you just do it from habit?

You made a statement that was clearly false. I responded to it *as
written*, pointing out that it was demonstrably false. Don't blame me.

Your revised statement isn't much better, as it is essentially
meaningless. With a sufficient investment, just about anything that is
physically possible can be accomplished, from SPS, to multi-TW fusion
plants, to strip mining the entire surface of the Moon to a depth of 10
km and launcing the debris into the Sun. A sufficient investment to
achieve the goal of order of magnitude launch cost reduction would
involve not only uncertain R&D expenses to overcome whatever technical
obstacles remain, not only the capital cost of the vehicle(s), not only
orders of magnitude improvement in operational efficiencies, but also
the building of sufficient market to justify it.

This sort of investment cannot achieve a workable return and thus,
using the rational actor model of economics, is impossible. If you
want to postulate this sort of investment, you'll have to also
postulate either a dimwit, a kook, or the government. Any of these
three choices has other difficulties associated with it that would,
IMO, make the investment unlikely in the extreme to achieve an order of
magnitude cost reduction. Furthermore, the mere act of choosing an
irrational actor means that you cannot then turn around and make cost
predictions based on standard economics.

Essentially you have two options that are mutually exclusive. Either:
1) The market is rational, in which case it is impossible, with current
market conditions, to make an investment with a desirable ROI in a
sufficiently advanced vehicle and a sufficiently large supporting
market that costs of space access would be reduced by an order of
magnitude

or:
2) The market is irrational, in which case the investment in the
vehicle and market may be made, but you may no longer use economics to
reliably predict what costs will be.

Of course, if you want to make a “Rand Simberg Science Fiction and
Fantasy” school of economics, then perhaps you could use that to
predict costs in case 2.

> >Part of the reason you are repeating this is that we are talking
past each
> >other. You say that the vehicles weren't invested in with the goal
of reducing
> >launch costs by an order of magnitude. I say, "No kidding!"
Investors DON'T
> >invest for goals like reducing launch costs. They invest to make a
profit. To
> >make a profit, they only have to be cheaper than their competitors.
They
> >normally tend to choose the least expensive, lowest risk means of
accomplishing
> >this.
>
> I have never claimed otherwise. All I claimed was that if someone
> made such an investment, that the costs would be reduced. Whether or
> not that is an economically sensible thing to do is an entirely
> different matter, as I explained to James Davis.

Whether or not it is an economically sensible thing is the crux of the
matter! You can create wondrous technical gizmos out the wazoo. But,
if it is not economically sensible, then, to achieve standard ROI’s,
you will have to jack the launch price of the vehicles so high that,
guess what!, you will be competing with vehicles currently on the
market. The “economic sensibilityness” of the vehicle is *part* of the
cost. I don’t see how you can possibly separate this.

> >>When someone makes an investment (that is, a full investment) in a
> >>vehicle with the intent of reducing costs an order of magnitude (or
> >>more), it will happen. You still haven't provided any examples in
the
> >>last two decades where anyone has even made the attempt, other than
> >>Beal, for whom the jury remains out (though I notice that you left
out
> >>Kistler, another vehicle on whom the jury is out).
> >
> >At the bottom of the list I noted that I have left out dozens of
other
> >projects. Beal is attempting to drop launch costs by around 30% or
so.
>
> No, they're attempting to drop prices by that amount. If they want to
> make a good profit to amortize their investment, they'll drop costs
> quite a bit more.

Your comment is a bit dim. If you are claiming that you can drop
launch costs by an order of magnitude by not amortizing your R&D and
capital expenditures, then you've made it quite a bit easier. But
management will take dim view of such a practice. Launch costs are
more than just operating costs. Amortization costs are and must be
included in the overall cost structure of a vehicle. Which then, after
a sufficient ROI, determines price.

> >As is
> >Kistler (though I was under the impression their project was close to
> >moribund). These are real projects following tried and true rules
of economics
> >and engineering development. Personally, I'd love to see either
succeed. But
> >you won't get order of magnitude here.
>
> You continue to confuse cost and price.

I assure you that I have absolutely no confusion on the matter. You
certainly must be aware that in a perfectly competitive economic world,
margins are driven to zero and price and cost track precisely. Or
perhaps this isn’t the case in the “Rand Simberg Science Fiction and
Fantasy” school of economics? At any rate, perhaps my usage was a bit
sloppy, but your nitpicking cannot disguise the fact that Beal will not
reduce either price or cost of the new vehicle by an order of magnitude
from current levels.

In fact, if the R&D program goes on any longer, I would not be
surprised if the amortization costs completely wipe out any cost
advantage of the Beal vehicle. The longer the program goes on, the
more chances seem dim that there will be any cost (or price) reduction
at all.

> >>>My point is only that new
> >>>investments DON'T get orders of magnitude reduction in costs.
There are
> >>market
> >>>reasons for this, technical reasons, etc. Doesn't matter. You
WON'T get
> >>the
> >>>cheap launch costs you want without going through the whole process
> >>predicted
> >>>by your typical cost curves.
> >>
> >>You do if you take a new approach. It's called jumping the curve.
> >
> >Well, jumping the curve is talked about a lot. But it is far harder
to do than
> >you seem to think. And, in reality, it usually ends up being more
of a
> >curve-acceleration rather than a true discontinuity.
>
> If it's harder to do, it's because of the uncertainty of financial
> return--not because it's intrinsically difficult to do.

Again, how do you get off separating the financial risk from intrinsic
difficulty? Cost of risk must be included in vehicle cost and,
ultimately, vehicle price. There is no getting around this. It is as
intrinsic to the problem as can be. If you cannot reduce the financial
risk to an acceptable cost, then you cannot make the investment.

> >Energiya-M had a real chance to be a discontinuity in the launch
cost curve.
> >It went broke, as is normally the case.
>
> Why? That is, why did Energiya-M have a real chance?

I overstated the case, really. From a technical perspective, it is
possible to see how Energiya-M could have a recurring flight cost on
order of $50 MMUSD. At a not monstrously unreasonable flight rate, and
including the fact that many of the development costs were borne by the
Soviets and would not have to be amortized, I could see prices on order
of $2000/kg, including a nice ROI, by the Energiya-M. The reason
nobody did so, of course, is simply because of the reasons I’ve been
stating – the cost of financial risk was too high, and thus the
investment was impossible by any rational actor. In short, the vehicle
was probably 20 years ahead of its time.

> >>>Still doesn't matter. Shuttle and current ELV's are perfectly
legitimate
> >>>points to use on a cost curve for future vehicles. And current
space solar
> >>>panels are perfectly legitimate points to use to cost future
systems.
> >>
> >>Not when you simply assume, as you did, that the current costs *are*
> >>the future costs.
> >
> >I never assumed this. I assumed that future costs can be predicted
from
> >current ones. Rather accurately for the most part.
>
> Not at all. You didn't even postulate what you thought that future
> costs would be. You simply stated present costs of space solar cells,
> and compared them to present terrestrial solar cell costs, as though
> there was no other analysis necessary. It was, in the blunt (but as
> this discussion goes on, apparently appropriate) words of another
> poster, a "stupid analysis."

As the discussion goes on, it does become more and more appropriate.
But I don’t hold it against you – I think it is more ignorance than
stupidity. Buck up! As you gain experience, you’ll see how these
things work.

I stated the current cost of solar cells. Any business decision on the
concepts that were being proposed would factor this into the business
plan. If the proponent stated that the solar cells that currently cost
$50/kW-hr would only cost $0.01/kW-hr on his design, then the business
managers would quietly call security and have him/her escorted from the
building.

Good conservative utility and energy businesses do not launch wild
projects that depend on orders of magnitude cost reductions in the key
technology(s) to make it work. They do not increase manufacturing
rates between technological generations by multiple orders of magnitude
to achieve the necessary economies of scale to make the project
financially attractive. You are kidding yourself if you think that the
current cost of space solar cells, factored downward by some
conservative percentage (5 years from now? Maybe $25/kW-hr) would not
be germane to this project.

> >>No, what I'm saying is, in the world where space solar cells are
being
> >>mass produced (as they would have to be for SPS production), instead
> >>of built in a boutique and in limited quantities (as they are now),
> >>they will be orders of magnitude cheaper than they are now, and
there
> >>is no way to reliably infer what they will cost from the cost of
> >>today's solar cells.
> >
> >Sigh. Well, we won't agree here. I know, with complete certainty,
that you
> >can infer what space solar cells will cost ten years from now, based
solely on
> >the cost of today's cells and recent trends.
>

> Well, aren't you smart. Why ain't you rich?

Actually, I’m quite well off. Thank you for your concern.

> >Why don't we get back together in
> >2010? If space solar cells cost less than $10/kW-hr on satellites
launched at
> >that time, I will publicly abase myself to your superior intellect,
experience,
> >and all around good-fellowness.
>

> I have no expectation that space solar cell costs will be
> significantly lower in 2010, because I don't expect anyone to make the
> investment necessary to cause that to happen. My only point is that
> it could happen if it were required or desired, and that *this* is the
> most important criterion, not what they cost today.

The investment can *not* happen because the factors that would make
this investment profitable are not present.

Scott Lowther

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
Thomas F. Radloff wrote:

> But if they had to go to all the trouble of confiscating an
> enormous amount of private wealth, and totally redesigning the control
> software in order to specifically turn it into a weapon, THEN they'd have a
> public uproar. If BMD scares some folks, imagine the outcry over
> MadUncleSam's mission to build death beam weapons!

The basic point I'm trying to make is this: using lasers as the power
transfer medium means that the beam, unlike a microwave system, is a
_potentially_ dangerous device. Coupled with the fact that this
dangerous device has the high ground and coverage of the entire planet,
this sounds like a potential weapons system. And if there is any way,
any way at all for this system to be modified to serve as a weapon,
those not in control of it cannot be considered insane for seeing the
potential.

Ask yourself this: instead of saying "It's impossible to use this system
as a weapon," ask yourself "How would I modifiy this sytem so I could
use it as a weapon?" The history of engineering is that of closed-minded
"Impossiblists" waking up to the hard realization that others have
figured out how to modify an unmodifyable system.

brian...@my-deja.com

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
In article <8m77j7$9qa$1...@nnrp1.deja.com>,
wm...@my-deja.com wrote:

> In article <20000801013451...@ng-cv1.aol.com>,
> brian...@aol.com (BrianF5070) wrote:
>
> [snip!]
>

> > They take 3 - 6 months to develop a design, cost
> > estimate, and contract (for Mr. Mook's benefit - about 3/4ths of all
> chemical
> > plant contracts are reimbursable. Yet cost effective). Then you
run
> it
> > through your in house design teams and business teams, make a go/no-
go
> > decision, arrange financing, select a contractor, and start
> construction.
> >

> > This is a product that has *not* been commoditized.
>

> Well, I guess you're referring to my comment that the network of SPS
> peak power users would trade their rights to a portion of the energy
> available from the first SPS to meet their peak power needs.

Actually, I was referring to your message expounding on the
disadvantages of the cost plus contract and why they have destroyed
NASA. Just pointing out that the majority of contracts in the oil,
gas, and chemical business are cost plus.

> Its clear that a power trading system as I envision could easily be
set
> up. The volume, value, and similarity to existing power trading
> systems all tend to favor this conclusion.
>

> [snip!]


>
> > Sigh. Well, we won't agree here. I know, with complete certainty,
> that you
> > can infer what space solar cells will cost ten years from now, based
> solely on

> > the cost of today's cells and recent trends. Why don't we get back


> together in
> > 2010? If space solar cells cost less than $10/kW-hr on satellites
> launched at
> > that time, I will publicly abase myself to your superior intellect,
> experience,
> > and all around good-fellowness.
>

> Wow! I get all tingly just thinking about that! :)
>
> But seriously, this requires that the person taking this bet must be
> able to predict with certainty that the demand for space power will
> increase dramatically. And no one knows that for certain.

For certain? No. With broad outlines that are nevertheless narrow
enough to get a general idea of what costs will be? Yes.

> Can't we both do something more reasonable and dispense with the self
> flagellation?

Sure.

> Consider, one can draw a cost/volume curve for a variety of space
power
> technologies. Agreed?

Agreed. Typically, the curves are drawn in threes - base case, high
and low. Agreed?

> Now, all we really have to disagree about is where to place these
> curves for each technology, how steeply the measure you mentions
> declines with total capacity on orbit, and so forth.
>
> But, I think you'll find we're reading off pretty much the same page
> here.

I expect you are correct, although I think we probably have
disagreements over the precise steepness of the curve.

> A PV array fed by a lightweight concentrator in orbit has associated
> with it a specific cost at a specific total demand. I think you'll
> find that if we put up about 1 Terawatt of big powersats (ignore our
> ongoing argument about launch costs for a minute, lets settle one
thing
> at a time!) using the curve we already agreed upon - we can both
> calculate what the cost per KW-hr is likely to be.
>
> Surprise - its well below $0.01 per kw-hr!!! This was found in the
> early studies done by Arthur D. Little, and also by ONeil and others.

Ah, but the curves they use to obtain these estimates are filled with
assumptions about future technologies - they are not simply
extrapolations of current curves.

O'Neill has postulated that solar cells are made from lunar resources,
which he states makes a quantum leap in reducing costs. This is, at
best, an unproven assertion.

> Amazing ain't it.
>
> > But they won't.
> >
>
> <shrug> Depends on total amount of energy produced in space doesn't
it?

To a large degree, yes. And this amount of energy is part of what I
have been talking about. The usage of solar power in space has
followed a predictable upward path which matches the downward pressure
on prices very well. Although the curve shapes can be altered to some
degree - made steeper or flatter - they don't suddenly leap by orders
of magnitude.

Solar cells in space may well cost $0.01 per kW-hr at some point in the
future. But that point is pretty far off, and certainly well beyond
the 10 year time horizon that companies plan on for their most blue-sky
research. If you want to get from here to there, you need an
incremental plan - next generation get the cost of solar cells to
$30/kW-hr. Make them profitable. The generation after that to $20/kW-
hr. Make them profitable. And so on.

Michael Kagalenko

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
George William Herbert (gher...@gw.retro.com) wrote
]Michael Kagalenko <mkag...@lynx.dac.neu.edu> wrote:
]>George William Herbert (gher...@gw.retro.com) wrote
]>] [...]
]>]No, they don't get cheaper at the same rate as ICs. This part is correct.
]>]The price is not leveling out. It continues to drop,
]>
]> "The price is leveling out" and "the price continues to drop" are
]>not mutually exclusive statements. The former means that the rate of
]> price declines is decreasing.
]
]The rate of price declines continues to match the 1997 and 1998
]predictions which I posted starting around that time, and the volume
]of sales continue to match the 1997 and 1998 predictions.

As I mentioned earlier, the price of solar panles is leveling off.
Time to start backpedaling furiously from your earlier attempts
to challenge this fact.

]
]This is some variant on "lying with statistics", Michael.

Well, nope.

]Is the price "leveling out" in the sense of no longer matching


]the predicted decline? No. Is it "leveling out" in that it's
]not getting $1/watt cheaper each year? Yes. Is it "leveling out"
]in not getting the predicted percentage cost drops? No.
]
]Pick a meaningless measure and you can say anything you want to.
]Fact remains, costs are dropping and will continue to drop for
]at least the forseeable future.

As I said earlier, the price is leveling off. The only plausible reading
of this is that the rate of price decrease is decreasing.

]>] and basic manufacturing


]>]and technology improvements already in the R&D and production pipelines will
]>]see it through at least 5 years more price drop, down to nearly the price
]>]parity with conventional power.
]>
]> Wishful thinking, unfounded assertions. Baring breakthroughs in basic
]> research which can not be predictd, or guaranteed by increase in funding,
]> solar cells will not be competitve with conventional and nuclear power.
]
]This is absolutely untrue, and you would know if if you bothered to
]read even the executive summaries of the references I posted, Michael.
]Why won't you do that? Why won't you do research? Is your ego so
]big that it encompasses all of engineering, physics, economics,
]and policy analysis?

Refertences that you posted do not establish anything
like your assertions.

]The cost breakeven point is around 2006, if the predicted cost drops


]continue for at least that long. The fundamental breakthroughs for
]the next 5-8 years worth of productizing are already demonstrated and
]in hand in labs. Ergo, it's unlikely to stop improving before breakeven,
]even if you assume no more new basic underlying new innovations or
]discoveries are made.

Nonsense. You are assuming continued reductions in the price of solar
cells, which you can not do. you are frther assuming no reduction
in the price of competing energy source, which you can not do either.

]>]There is plenty of reason to assume they will get cheaper than conventional

Thank you for your very touching concern for my professional
well-being. Are you still trying to scam potential investors
into funding your unsound space projecteuring ?

Rand Simberg

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
On Thu, 03 Aug 2000 15:15:53 GMT, in a place far, far away,
brian...@my-deja.com made the phosphor on my monitor glow in such a

way as to indicate that:

>> Then let me inform you that you cannot buy Titan futures. The reason


>> is that there is insufficient volume in the market. It is not a
>> commodity.
>
>I stated I was uncertain about Titans.

Yes, and I accordingly provided you with the information.

>However, I also do not see any
>reason why one would not be able to make a futures market if desired.
>In fact, this is an interesting idea - there might be a profit in it.

As I said, it would be too thinly traded.

>And let me inform you that you *can* buy Atlas options.

Not on the Chicago exchange.

>> >And yes, you can buy generic Titans.
>>
>> No, you can't. They don't exist, in the sense that commodities do.
>
>You certainly *can* buy generic Titans and Atlas's and Protons, etc.
>They are manufactured in generic configurations for standard prices.
>That is what ILS's little launch book is for.

Those are not generic. They are still Lockmart and Energia products.
Generic means that they are unbranded, and I can buy them in bulk at a
considerably lower price than the branded variety, though I might have
to bring my own bag to take them home with...

><grin> I find this ironic. Since there is no standard definition for
>“commodity”, I've been trying to fit myself into the context you are
>using. Which is tricky because it seems to be somewhat variable. But,
>to get this down, let's run by some of the definitions of commodities
>that are commonly used:
>
>1) Dictionary says A) That which affords convenience or profit, esp. in
>commerce, including everything movable that is bought or sold or B) an
>element of wealth; an economic good.

This is not what I mean.

>2) A market definition of a commodity that is often used is any item
>that is bought and sold on an open, licensed, and regulated exchange.
>
>3) A functional definition that is often used is any product or service
>that customers perceive
>to be exactly the same.
>
>4) A contractual definition of a commodity is a standardized contract
>for future delivery of a specific product traded through licensed
>regulated exchanges.

These last three are what I'm describing. All three.

>None of these definitions is perfect, nor is any universally used. The
>term commodity is highly dependent on context. I have been trying to
>follow your context, but it seems to keep changing. It should be
>apparent, though, that rockets are at least partially commoditized
>based on the definitions above.

Rockets do not come close to satisfying definitions 2, 3, or 4, which
the the ones that I've been using. I hope that helps.

>> >Better yet, go to ILS and order a Proton.
>> >
>> >This is a product that has been commoditized.
>>
>> No, it's not.
>
>A Proton certainly fits definition 1 or 3,

Not 3. Everyone knows that it's a Proton.

>and options are written on
>Proton launches, so definition 4 is partially fulfilled.

Not on a licensed, regulated exchange.

>These options
>are negotiable instruments that can be traded, although there is
>currently no public market set up to do that. Which might be an
>interesting business possibility – I can see a potential profit in
>setting up such a market.

That's not the kind of option being referred to in the definition.

In any event, for past and future reference, I am using the term
commodity to mean a product that satisfies *all three* definitions 2,
3 and 4, not one that might partially satisfy one or two of them.
Rockets don't cut it, nor does cargo delivery to orbit.

>> Automobiles are not a commodity.
>
>Depends on your definition.

Yes, and now that we've established the definition, my statement
stands.

>> >Or airplane
>> >tickets. There is a certain amount of haggling and arguing and
>customization,
>> >but by and large both buyer and seller know the cost and schedule
>and come
>> >rather quickly to agreement.
>>
>> Airplane tickets are commoditized, but not as fully as the other items
>> mentioned.
>
>You seem stuck on this commodity black and white thing. You have to
>realize that commoditization is a process, not a result. I will repeat
>that *nothing* is ever fully commoditized. Not even the examples you
>gave.

In what way is winter wheat not fully commoditized?

>> >These markets are less mature than pork bellies,
>> >as you eloquently state, but they are pretty mature.
>>
>> Yes, and rockets and access to LEO are nowhere near that state.
>
>Near the state of pork bellies? No. Considerably beyond the state
>where it can be considered immature? Yes, very much so. That is
>rockets, of course, and access to LEO for satellites.

The access-to-space industry is not mature, for cargo or passengers.

If it were, it would not be possible to make significant cost
reductions with each generation, but EELV demonstrates that it is.

>> Chemical plants are not commoditized, but chemicals (their outputs)
>> are.
>
>Oversimplification again. Oil is oil, right?

No, but regular gas is regular gas. Some people will be willing to
spend more for a branded version, but many buy generic. It's a
commodity, as are the individual types of crude.

>> >This is a product that has *not* been commoditized.
>>
>> Never claimed otherwise. Nor has a rocket plant.
>
>Rocket plants have not been, but rockets (their outputs), in many
>cases, *have* been.

Nope. See definitions above.

>> >But just to note, your originally claim was that as soon as someone
>made an
>> >investment in reducing launch costs by an order of magnitude, they
>were 100%
>> >guaranteed of results. You didn't include qualifiers such as, "a
>sufficient
>> >investment."
>>
>> Well, excuse me. I thought that this would be obvious.
>>
>> Did you really think that I meant that it would happen as soon as
>> someone put up one percent, or ten percent of the money? Do you really
>> want people in the newsgroup to think that you are that dim?
>
>Do you really want people in the newsgroup to think you are dim enough
>to believe that a weak argument can be strengthened by ad hominem
>attacks?

It's not an ad hominem attack. It was a question. I didn't say that
you were dim, and in fact do not believe that you are. I asked if you
really wanted people in the newsgroup to think that by pursuing such
an absurd tack.

>You made a statement that was clearly false.

>I responded to it *as
>written*, pointing out that it was demonstrably false. Don't blame me.

It was not clearly, or even demonstrably false. You just chose to
misinterpret it. Most reasonable people would interpret it the way I
intended it.

>Your revised statement isn't much better, as it is essentially
>meaningless. With a sufficient investment, just about anything that is
>physically possible can be accomplished, from SPS, to multi-TW fusion
>plants, to strip mining the entire surface of the Moon to a depth of 10
>km and launcing the debris into the Sun.

These are all different goals, qualitatively, than reducing the cost
of something.

>A sufficient investment to
>achieve the goal of order of magnitude launch cost reduction would
>involve not only uncertain R&D expenses to overcome whatever technical
>obstacles remain, not only the capital cost of the vehicle(s), not only
>orders of magnitude improvement in operational efficiencies, but also
>the building of sufficient market to justify it.

No, it doesn't require a market. This is only required to make a
profit. The cost will be reduced whether there is market or not.

>This sort of investment cannot achieve a workable return and thus,
>using the rational actor model of economics, is impossible. If you
>want to postulate this sort of investment, you'll have to also
>postulate either a dimwit, a kook, or the government. Any of these
>three choices has other difficulties associated with it that would,
>IMO, make the investment unlikely in the extreme to achieve an order of
>magnitude cost reduction.

I didn't claim that it was likely--just that it was possible.

>Furthermore, the mere act of choosing an
>irrational actor means that you cannot then turn around and make cost
>predictions based on standard economics.

One can make cost predictions by analyzing the system that is being
designed and built, and estimating its costs. The revenues are
entirely independent of this analysis.

>Essentially you have two options that are mutually exclusive. Either:
>1) The market is rational, in which case it is impossible, with current
>market conditions, to make an investment with a desirable ROI in a
>sufficiently advanced vehicle and a sufficiently large supporting
>market that costs of space access would be reduced by an order of
>magnitude
>
>or:
>2) The market is irrational, in which case the investment in the
>vehicle and market may be made, but you may no longer use economics to
>reliably predict what costs will be.

Nonsense. I can predict costs totally independently of the price.

>> I have never claimed otherwise. All I claimed was that if someone
>> made such an investment, that the costs would be reduced. Whether or
>> not that is an economically sensible thing to do is an entirely
>> different matter, as I explained to James Davis.
>
>Whether or not it is an economically sensible thing is the crux of the
>matter! You can create wondrous technical gizmos out the wazoo. But,
>if it is not economically sensible, then, to achieve standard ROI’s,

I have never made any claims about ROI's. I don't understand why you
have so much trouble with this concept.

>you will have to jack the launch price of the vehicles so high that,
>guess what!, you will be competing with vehicles currently on the
>market. The “economic sensibilityness” of the vehicle is *part* of the
>cost. I don’t see how you can possibly separate this.

No, it's not. The economic sensibility of the vehicle is part of the
business case, as is the cost, but cost and price are independent of
each other.


>> >At the bottom of the list I noted that I have left out dozens of
>other
>> >projects. Beal is attempting to drop launch costs by around 30% or
>so.
>>
>> No, they're attempting to drop prices by that amount. If they want to
>> make a good profit to amortize their investment, they'll drop costs
>> quite a bit more.
>
>Your comment is a bit dim. If you are claiming that you can drop
>launch costs by an order of magnitude by not amortizing your R&D and
>capital expenditures, then you've made it quite a bit easier. But
>management will take dim view of such a practice. Launch costs are
>more than just operating costs. Amortization costs are and must be
>included in the overall cost structure of a vehicle. Which then, after
>a sufficient ROI, determines price.

No, price is determined by market demand.

>> You continue to confuse cost and price.
>
>I assure you that I have absolutely no confusion on the matter. You
>certainly must be aware that in a perfectly competitive economic world,
>margins are driven to zero and price and cost track precisely. Or
>perhaps this isn’t the case in the “Rand Simberg Science Fiction and
>Fantasy” school of economics? At any rate, perhaps my usage was a bit
>sloppy, but your nitpicking cannot disguise the fact that Beal will not
>reduce either price or cost of the new vehicle by an order of magnitude
>from current levels.

It's not nitpicking--it is fundamental to my point. They won't reduce
price that much, because they need to make a return on their
investment--they'll reduce it only to the level they need to to get
the business from their competitors. But you have no idea how much
they'll reduce cost, unless you're privy to their design team.

>In fact, if the R&D program goes on any longer, I would not be
>surprised if the amortization costs completely wipe out any cost
>advantage of the Beal vehicle. The longer the program goes on, the
>more chances seem dim that there will be any cost (or price) reduction
>at all.

That is a function of market and flight rate.

>> >Well, jumping the curve is talked about a lot. But it is far harder
>to do than
>> >you seem to think. And, in reality, it usually ends up being more
>of a
>> >curve-acceleration rather than a true discontinuity.
>>
>> If it's harder to do, it's because of the uncertainty of financial
>> return--not because it's intrinsically difficult to do.
>
>Again, how do you get off separating the financial risk from intrinsic
>difficulty? Cost of risk must be included in vehicle cost and,
>ultimately, vehicle price.

Technical risk, not financial risk. There is very little technical
risk in designing a launch system for an OOM cost reduction. There
mostly market risk.

>There is no getting around this. It is as
>intrinsic to the problem as can be. If you cannot reduce the financial
>risk to an acceptable cost, then you cannot make the investment.

Yes, I can. If I have the money, and choose to do it, I can make the
investment. Whether or not I get my money back is a different issue,
but the investment will reduce the launch cost nonetheless.

>> >>Not when you simply assume, as you did, that the current costs *are*
>> >>the future costs.
>> >
>> >I never assumed this. I assumed that future costs can be predicted
>from
>> >current ones. Rather accurately for the most part.
>>
>> Not at all. You didn't even postulate what you thought that future
>> costs would be. You simply stated present costs of space solar cells,
>> and compared them to present terrestrial solar cell costs, as though
>> there was no other analysis necessary. It was, in the blunt (but as
>> this discussion goes on, apparently appropriate) words of another
>> poster, a "stupid analysis."
>
>As the discussion goes on, it does become more and more appropriate.
>But I don’t hold it against you – I think it is more ignorance than
>stupidity. Buck up! As you gain experience, you’ll see how these
>things work.
>
>I stated the current cost of solar cells. Any business decision on the
>concepts that were being proposed would factor this into the business
>plan. If the proponent stated that the solar cells that currently cost
>$50/kW-hr would only cost $0.01/kW-hr on his design, then the business
>managers would quietly call security and have him/her escorted from the
>building.

Not if he can put together a good story on how those costs are
achieved.

>Good conservative utility and energy businesses do not launch wild
>projects that depend on orders of magnitude cost reductions in the key
>technology(s) to make it work. They do not increase manufacturing
>rates between technological generations by multiple orders of magnitude
>to achieve the necessary economies of scale to make the project
>financially attractive. You are kidding yourself if you think that the
>current cost of space solar cells, factored downward by some
>conservative percentage (5 years from now? Maybe $25/kW-hr) would not
>be germane to this project.

No space solar cell costs based on today's limited quantities are
relevant to the costs of solar cells that are produced in quantities
many orders of magnitude higher. There is absolutely no relationship.
That's where the discussion started, and that's where it will end.

>> I have no expectation that space solar cell costs will be
>> significantly lower in 2010, because I don't expect anyone to make the
>> investment necessary to cause that to happen. My only point is that
>> it could happen if it were required or desired, and that *this* is the
>> most important criterion, not what they cost today.
>
>The investment can *not* happen because the factors that would make
>this investment profitable are not present.

It can happen, but it probably won't.

************************************************************************
simberg.interglobal.org * 310 372-7963 (CA) 307 739-1296 (Jackson Hole)
interglobal space lines * 307 733-1715 (Fax) http://www.interglobal.org

"Extraordinary launch vehicles require extraordinary markets..."
Replace first . with @ and throw out the "@trash." to email me.
Here's my email address for autospammers: postm...@fbi.gov

Scott Lowther

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
wm...@my-deja.com wrote:
>
> In article <3988DA...@ix.netcom.com>,
> Scott Lowther <lex...@ix.netcom.com> wrote:
> > wm...@my-deja.com wrote:
> >
> > > Since when is delivering copious power at very low cost across the
> globe
> > > a problem for anyone?
> >
> > Ask OPEC. Care to take a guess at how well Saudi Arabia would handle
> no
> > longer recieving large sums of Western money?
>
> Fact is, the reason OPEC can't raise prices more is that by doing so
> they cut their own throats because their wealth is tied to the economic
> productivity of the world.

Fact is, OPEC's wealth is tied to the oil economy. Ending the oil
economy
means taking away OPEC's livelyhood, with many OPEC nations not having
any really viable alternatives. How would you respond if you were faced
with suddenly losigin your job, home, possessions and never again being
employed, but only getting the occasional handout?

We'll have to face up to this sooner or later, no matter what. But a
vast low altitude, easily targetted SPS system such as yours would make
a supremely obvious target for them.

> They may change their product mix, more
> plastics and fertilizer, less fuel oil, slightly less gasoline.

"Slightly?" You've proposed replacing every single IC motor in the world
with an electric one. If that occurs, there will be no more fuel oil or
gasoline; and without the vast economy of scale available from that,
palstics and petro fertilizers will be simply too expensive to buy. So
the oil producers will ahve nothing to produce, no money to make,
nothing to lose.

> Of course in a well managed world without corruption, things would be
> even better! :)

In a well managed world without corruption there'd be no humans.

> > What was the last war the US got involved in in a serious and
> determined
> > way? And was or was not oil (i.e. energy) a major part of it?
>
> When someone threatened to cut off *our* supply! Duh! We didn't give a
> fig about OPEC and all those sheiks.

You oversimplify. WHY was that supply threatened?

> > > Without this technology the idea of
> > > using lasers as weapons from space to strike at ground based
> targets is
> > > nearly as silly as dropping copper wires from orbit! That's why
> SDI had
> > > pop up lasers and so forth. Even nuclear powered beams millions of
> > > times more powerful than the one's I'm proposing weren't militarily
> > > useful for ground targets!
> >
> > That's because SDI would have had only a relative few. You're
> proposing
> > worldwide coverage with the capability of perhaps billions of these
> > beams at any one time.
>
> Eventually, not at first.

Well, initially the US had only a few nukes. Why should the Soviet Union
have cared? "It's not like the nuclear power of the United States will
grow in time..."

> > > Nuclear power plants have had a bad rap, their association with
> nuclear
> > > weapons doesn't help.
> >
> > Lasers satellites are equally badly regarded, and for similar reasons.
>
> No one has been killed as a result of a laser attack during wartime.
> The psychodynamics are just not the same as for nuclear weapons. You
> are creating problems where none need exist by drawing these parallels.

"Death Star."

> > > > Not a single person has been killed by a nuclear power plant
> weapons
> > > > system in the 50 years or so they've been around.
> > >
> > > You speak as if half a million people weren't killed in the US
> atomic
> > > bombing raids against Japan. Duh!
> >
> > A) We didn't drop a nuclear power plant on Japan.
>
> No, just a nuclear weapon fueled by the same material, and it was
> highly publicized, easily remembered, and the public's first broad
> introduction to the uses of nuclear power.

Well, I guess it's a good thing your system won't be beaming down
concentrated SUNSHINE, which has caused how many cancer deaths? DUH!!!


> > C) Hal;f a million people weren't killed in Japan by nukes in any
> event.
> > 200,000 conservative, 400,000 ravign liberal.
>
> Well, there are the immediate deaths, and then the long term effects of
> radiation through the generations. Its not liberalism, just science.

If you want to count deaths "through the generations," how many hundreds
of millions died because of the Japanese? How many billions because of
Napoleon? How many trillions yet to come because of Julius Ceasar?

You count those killed, not those who might have been.

> > D) They deserved it in any event (but that's another debate).
>
> This really marks you off as a raving lunatic!

Japan declared war on the United States. Fortunately we had leaders at
the time who wouldn't listen to advisors such as yourself, who counsel
surrender in the face of aggression. They listened to advisors who said
that striking back was appropriate.

And yes, the Japanese people deserved everything they got, just as the
German people did. If the German or Japanese people did not want the
wars they started, they wouldn't have fought them. Tojo and Hilter would
have been hung from street lights as Mussolini was. But the German and
Japanese people backed the wrong horse.

Well, you've proven MY point that you don't understand the world... the
firebombing raids of WWII LASTED FOR MANY HOURS. So 21 mintues just
isn't that long, is it?

> And your comments about one satellite being able to do this are all wet.

One satellite can do all this. One followed by another followed by
another... exactly the setup YOU PROPOSE!!!

> And, you've forgotten the effects of cooling, radiation, convection,
> etc.

Nope, wrong.

With a laser intensity of 280kwatts/square meter, temperatures rise
rapidly in stuff like asphalt (dark with reasonably poor thermal
transfer), and have impressively high blackbody temperatures in the
range of 2700K and beyond. Convection cooling is hard pressed to counter
this.

> Fact is, you haven't shown that laser irradiation - even on this scale -
> could create a firestorm. Sorry.

I'm sure the firefighters at Mesa Verde will be glad to hear that a
single cigarette can't create a firestorm. Look, you don't need to cause
massive blazes to create a firestorm. All you need to do is create
enough small ones to overwhelm the fire fighters, if any. Say, every
building in town catches fire at once. Not a big fire, just a fire.
Pretty soon the whole twon will be up in smoke.

> Even assuming it were physically possible to target non-receivers,
> which I don't believe is possible to do,

Okay, here's the challenge: since we're talking about YOUR
highly-steeped-in-the-hypothetical SPS system, presumably you know more
about it than I do. Well, if you want to claim to be a halfway decent
scientist or engineer, you have to be able to effectively play Devil's
Advocate. It is left to the crackpots of the world to ignore negative
aspects of their inventions, so prove you are not a crackpot and examine
the negatives: how would YOU go about modifying your system to make it
capable of firing at ground targets? What would _you_ do with regard to
multiple pilot beams, or rectifier sat modifications, whatever? Are you
clever enough to turn this benign system into something more
interesting?

uncontrolled irradiation of
> say an asphalt shingle, or roadway, would drive off a lot of smoke, but
> cause very little collateral damage.

You think melting asphalt roadways is just a minor inconvenience?

> Understand dropping napalm on a target is different than irradiating
> it. Napalm will burn with a flame and that flame will spread as the
> material fuels it. Irradiating a target will not necessarily cause
> combustion, and combustion will not necessarily continue once
> irradiation is removed,

Well, I'm sure the sainted citizens of Hiroshima thank you for pointing
out that irradiating a target is a poor way to start a fire.

Rand Simberg

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
On Thu, 03 Aug 2000 10:40:55 -0700, in a place far, far away, Scott
Lowther <lex...@ix.netcom.com> made the phosphor on my monitor glow

in such a way as to indicate that:

>Fact is, OPEC's wealth is tied to the oil economy. Ending the oil


>economy
>means taking away OPEC's livelyhood, with many OPEC nations not having
>any really viable alternatives. How would you respond if you were faced
>with suddenly losigin your job, home, possessions and never again being
>employed, but only getting the occasional handout?
>
>We'll have to face up to this sooner or later, no matter what. But a
>vast low altitude, easily targetted SPS system such as yours would make
>a supremely obvious target for them.

More likely it would make a supremely obvious investment for them...

Scott Lowther

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
Rand Simberg wrote:
>
> On Thu, 03 Aug 2000 10:40:55 -0700, in a place far, far away, Scott
> Lowther <lex...@ix.netcom.com> made the phosphor on my monitor glow
> in such a way as to indicate that:
>
> >Fact is, OPEC's wealth is tied to the oil economy. Ending the oil
> >economy
> >means taking away OPEC's livelyhood, with many OPEC nations not having
> >any really viable alternatives. How would you respond if you were faced
> >with suddenly losigin your job, home, possessions and never again being
> >employed, but only getting the occasional handout?
> >
> >We'll have to face up to this sooner or later, no matter what. But a
> >vast low altitude, easily targetted SPS system such as yours would make
> >a supremely obvious target for them.
>
> More likely it would make a supremely obvious investment for them...

Agreed. Myself and others have repeated put forward the notion of
getting US oil companies involved with SPS systems; they are about the
only corporate entities with the funds to attempt something like this,
they need to look towards the future, and the SPS system can be hooked
inot syntehtic fuel plants to produce hydrocarbon fuels for existing
vehicles and infrastructure so that we don't need to repave the whole
world to use a completely new (and, in my opinion, completely insane)
system of laser-powered cars.

And consider this: assume an SPS powered future. Who do you want running
it? The western democracies, or middle eastern potentates?

Ian Stirling

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
wm...@my-deja.com wrote:
>In article <3988B6...@ix.netcom.com>,
> Scott Lowther <lex...@ix.netcom.com> wrote:
>> Easy experiment: take a simple lens, and in direct sunlight hold it
<snip>

>> products burn in seconds, asphalt melts. Considerably more damage
>would
>> be achieved if the spot was half a meter wide, of course..s slight
>> breezes would not provide the same cooling properties, and melted tar
>> wouldn't nbe able to run out of the sunlit spot before they
>overheated.
<snip>

>A more reliable measure would be the energetics involved. The US Army
>relates the destructiveness of NAPALM and other fire sources to the
>quantity used. To set a building ablaze, reliably, requires a certain
>amount. This can be related to the energy contained in the NAPALM.


A more reliable measure might be the flash ignition radius of a large
fusion weapon.
The energy is spread over a second or two, so it'd give an idea of how
much power you need to put over a whole house in order to ignite it
in a second or two.
Differences would be due to the spectrum, pulse shape, illumination
direction, but it should be in the ballpark.

--
http://inquisitor.i.am/ | mailto:inqui...@i.am | Ian Stirling.
---------------------------+-------------------------+--------------------------
The fight between good and evil, an epic battle. Darth vader and Luke,
suddenly in the middle of the fight, Darth pulls Luke to him, and whispers
"I know what you'r getting for christmas!" Luke exclaims "But how ??!?"
"It's true Luke, I know what you'r getting for christmas" Luke tries to ignore
this, but wrenches himself free, yelling "How could you know this?",
Vader replies "I felt your presents" -- The Chris Evans breakfast show ca. 94

Scott Lowther

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
Ian Stirling wrote:

> A more reliable measure might be the flash ignition radius of a large
> fusion weapon.
> The energy is spread over a second or two, so it'd give an idea of how
> much power you need to put over a whole house in order to ignite it
> in a second or two.
> Differences would be due to the spectrum, pulse shape, illumination
> direction, but it should be in the ballpark.

Good point. Pulled mt copy of "the Effects Of Nuclear Weapons" off the
shelf and looked this up. There is an extensive table for "approximate
radiant exposures for ignition of:" one for "fabrics" and one for
"house-hold materials and dry forest goods." Sadly, the exposures are
listed in cal/sq. cm units, so I hope I did the conversion correctly
(I'm tired and my brain is fuzzy) to J/sq. meter. A few examples:

Wood from rotten logs to pine needles ranges from 4 to 21 (167440 to
879060).
newspaper ranges from 4 to 11 (460460)
white paper from 10 (418600) to 50 (2093000)
Cotton ranges from7 (293020) to 34 (1423240), with similar ranges for
rayon.

The needed exposure levels were about 2-3 times higher for a 10 megaton
bomb as for a 40 kiloton device; I assume that this has something to do
with the spectrum (higher energy radiant less likely to cause fire...
maybe penetrates further?). The SPS laser will be a fairly low-energy
wavelength in comparison, so I'd suspect the lower ends of the estimates
might be more accurate.

Perhaps the most interesting are still photos of the house you always
see in test footage films: Bright flash, the front of the house bursts
into smoke (but not flame) and a second later it gets mashed by the
shock wave. The ignition exposure was 25 cal/sq. cm (104650)

Compare this to a 200 (space) solar laser... 280,000 watts/sq. meter. A
house fire would start in ~0.5-1 seconds, trees would burn in 0.5-3
seconds exposure, cotton in 1-5 seconds. Remember that the SPS system is
supposed to hold it's lasers on moving targets for hours, if need be.
Also, assuming 1 km diameter GEOcollectors, to get that 200solar
concentration means a beam about 71 meters in diameter. Assuming that
this will spread, but then recondensed at the LEOsat, a single
block-wide beam could in principle be slowly wandered across a target
region at ~10 meters per second or so, laying waste to anything but
metal and mineral structures.

Paul F. Dietz

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to
Scott Lowther wrote:

> The needed exposure levels were about 2-3 times higher for a 10 megaton
> bomb as for a 40 kiloton device; I assume that this has something to do
> with the spectrum (higher energy radiant less likely to cause fire...
> maybe penetrates further?).

More likely, it has to do with the time over which the thermal
energy is delivered. The large fireball from a 10 megaton
bomb takes longer to cool than the small fireball from a
40 kiloton bomb, so the heat is delivered over a longer
period (and out to a much larger distance).

Paul

Scott Lowther

unread,
Aug 3, 2000, 3:00:00 AM8/3/00
to

Yeah, that makes sense. So for Bomb Y with 100 times the yield of Bomb
X, 10 times the distance will get you the same energy per unit area...
but at a lower power per unit area. Reading the book in a little more
depth tells much the same thing... more burning with smaller, faster
nukes or bigger nukes at higher altitude (less air to absorb the higher
energy and re-radiate as thermal).

Jonathan Thornburg

unread,
Aug 4, 2000, 3:00:00 AM8/4/00
to
In article <39857de1...@nntp.ix.netcom.com>,
Rand Simberg <simberg.i...@trash.org> wrote:
>Pork bellies are fully
>commoditized.

Part (a very _large_ part) of being a "commodity" is that there are
no non-price discriminants between different suppliers. But the last
time I looked, the US and the EU were in the middle of a trade war
over hormones in beef, pork, et al, with tarrifs, labelling laws,
counterveiling duties, et al, all being bandied about. Doesn't look
quite like a "commodity" to me...

--
-- Jonathan Thornburg <jth...@thp.univie.ac.at>
http://www.thp.univie.ac.at/~jthorn/home.html
Universitaet Wien (Vienna, Austria) / Institut fuer Theoretische Physik
Seen on usenet (dueling .signature quotes):
#1: "If we're not supposed to eat animals, why are they made of meat?"
#2: "If we're not supposed to eat people, why are they made of meat?"

BrianF5070

unread,
Aug 4, 2000, 3:00:00 AM8/4/00
to
Scott Lowther lex...@ix.netcom.com wrote:

>Ian Stirling wrote:
>
>> A more reliable measure might be the flash ignition radius of a large
>> fusion weapon.
>> The energy is spread over a second or two, so it'd give an idea of how
>> much power you need to put over a whole house in order to ignite it
>> in a second or two.
>> Differences would be due to the spectrum, pulse shape, illumination
>> direction, but it should be in the ballpark.
>
>Good point. Pulled mt copy of "the Effects Of Nuclear Weapons" off the
>shelf and looked this up. There is an extensive table for "approximate
>radiant exposures for ignition of:" one for "fabrics" and one for
>"house-hold materials and dry forest goods." Sadly, the exposures are
>listed in cal/sq. cm units, so I hope I did the conversion correctly
>(I'm tired and my brain is fuzzy) to J/sq. meter. A few examples:
>
>Wood from rotten logs to pine needles ranges from 4 to 21 (167440 to
>879060).
>newspaper ranges from 4 to 11 (460460)
>white paper from 10 (418600) to 50 (2093000)
>Cotton ranges from7 (293020) to 34 (1423240), with similar ranges for
>rayon.

Just to throw more fuel onto the fire (so to speak), the American Petroleum
Institute (API RP 521) lists radiation standards for flare sizing and spacing.
For personnel exposure:

4.7 kW/m^2 is the maximum allowable level for continuous exposure of a properly
attired person.

6.3 kW/m^2 is the stated level of exposure to a properly attired person that
will incapacitate the person in 1 minute unless shelter is available.

38 kW/m^2 is the maximum allowable continuous exposure for equipment.

Not putting any judgement on any proposed system whatsoever, but if it was ever
possible in whatever manner for 200 solars of laser light to hit a person, they
would be incapacitated nearly instantaneously.


BrianF5070

unread,
Aug 4, 2000, 3:00:00 AM8/4/00
to
>Agreed. Myself and others have repeated put forward the notion of
>getting US oil companies involved with SPS systems; they are about the
>only corporate entities with the funds to attempt something like this,
>they need to look towards the future, and the SPS system can be hooked
>inot syntehtic fuel plants to produce hydrocarbon fuels for existing
>vehicles and infrastructure so that we don't need to repave the whole

(Snip)

OK. Since I work for a company whose competitors have names like ExxonMobil
and Texaco, let me comment briefly.

First of all, yes, the industry does need to (and is) looking toward the
future. Yes, we would be interested in technologies that could supply cheap
and plentiful electrical energy. Yes, we are looking at alternative energy
sources. Yes, to the degree that it can fit into the existing infrastructure,
that is a benefit, but we are not really averse to changes in the
infrastructure, as long as the overall project economics are favorable. And
the existing distribution infrastructure is a smaller part of most project
economics than you'd suspect.

But no, the industry does not have the funds to attempt anything that has been
proposed yet.

To be blunt, a megaproject in the oil industry typically runs about $2 billion
CAPEX and on average is about 65% debt/35% equity financed. Projects of this
size are about the largest that the three very largest oil companies (BP Amoco,
ExMob, and RD Shell) can attempt alone, but normally would be done as part of a
large JV with three or more partners. The very largest project that has *ever*
been proposed in the industry was $38 billion total CAPEX, was a hugely complex
(at least 12 companies, I think) JV, and collapsed under its own weight because
finanincing couldn't be arranged. In addition, oil companies have found that
the megaprojects typically don't get the desired discounted cash flow ROR, and
have been placing increasing emphasis on driving CAPEX down for large projects,
and spreading cash through many more small, higher DCFRR projects.

If you want to interest an oil company in a SPS project, make sure it meets
these requirements:
1) Reduce CAPEX to absolute maximum of $2 billion.
2) Demonstrate technological maturity such that P10 to P90 difference in CAPEX
is max $500 million.
3) Demonstrate a DCFRR of min 20%. More like 25% if you really want to get
interest.
4) Demonstrate a NPV (based on _discounted_ cash flows, Mr. Mook - say 10%
discount rate) of at least $ 3 billion.
5) Demonstrate P50 time from initial capital expense to operability <40 months.

There would be other requirements, but if the project can meet these first
hurdles, it would go a long way toward solving some problems.


Scott Lowther

unread,
Aug 4, 2000, 3:00:00 AM8/4/00
to
BrianF5070 wrote:
>
> >Agreed. Myself and others have repeated put forward the notion of
> >getting US oil companies involved with SPS systems; they are about the
> >only corporate entities with the funds to attempt something like this,

> But no, the industry does not have the funds to attempt anything that has been
> proposed yet.

Well... maybe the tobacco industry?

BrianF5070

unread,
Aug 4, 2000, 3:00:00 AM8/4/00
to
simberg.i...@trash.org (Rand Simberg) wrote:

>>And let me inform you that you *can* buy Atlas options.
>
>Not on the Chicago exchange.

Shocking! And yet, you can't buy electricity options and futures there,
either. I still consider electricity a commodity.

>>2) A market definition of a commodity that is often used is any item
>>that is bought and sold on an open, licensed, and regulated exchange.
>>
>>3) A functional definition that is often used is any product or service
>>that customers perceive
>>to be exactly the same.
>>
>>4) A contractual definition of a commodity is a standardized contract
>>for future delivery of a specific product traded through licensed
>>regulated exchanges.
>
>These last three are what I'm describing. All three.
>

Alright. But then airline seats (which you mentioned)) don't meet any of the
criteria.

>>None of these definitions is perfect, nor is any universally used. The
>>term commodity is highly dependent on context. I have been trying to
>>follow your context, but it seems to keep changing. It should be
>>apparent, though, that rockets are at least partially commoditized
>>based on the definitions above.
>
>Rockets do not come close to satisfying definitions 2, 3, or 4, which
>the the ones that I've been using. I hope that helps.

It helps. Although, a satellite maker mostly cares about getting the payload
to orbit. They couldn't care less about *how* it gets there as long as it is
safe and cheap. They can choose Ariane, or Long March, or Atlas, or Delta, or
whatever, and the decision is often made largely on price. This indicates that
the buyer sees access to orbit as more or less equivalent among sellers, which
means that it at least partially satisfies definition 3.

>>> >Better yet, go to ILS and order a Proton.
>>> >
>>> >This is a product that has been commoditized.
>>>
>>> No, it's not.
>>
>>A Proton certainly fits definition 1 or 3,
>
>Not 3. Everyone knows that it's a Proton.
>
>>and options are written on
>>Proton launches, so definition 4 is partially fulfilled.
>
>Not on a licensed, regulated exchange.

Well, we have some disagreement here. I see that ExxonMobil wants an oil
finding satellite. So they go to Loral, who designs and builds it, then it
gets launched and ExMob gets data. The ExMob executives may not know and
certainly won't care that it is or isn't a Proton, as long as the price is the
lowest they can get.

>In any event, for past and future reference, I am using the term
>commodity to mean a product that satisfies *all three* definitions 2,
>3 and 4, not one that might partially satisfy one or two of them.
>Rockets don't cut it, nor does cargo delivery to orbit.

Ah. OK. Then rockets don't cut it. Neither do many other things that I would
normally think of as commodities.

>>You seem stuck on this commodity black and white thing. You have to
>>realize that commoditization is a process, not a result. I will repeat
>>that *nothing* is ever fully commoditized. Not even the examples you
>>gave.
>
>In what way is winter wheat not fully commoditized?

I'm not familiar with winter wheat so I couldn't tell you.

However, I did provide an example of branded electricity. Since electricity no
longer satisfies definition three, then is it safe to assume it is not a
commodity?

>The access-to-space industry is not mature, for cargo or passengers.
>
>If it were, it would not be possible to make significant cost
>reductions with each generation, but EELV demonstrates that it is.

Now you are defining maturity of an industry based on where it is on the
learning curve? OK. That doesn't seem unreasonsable - I guess I can go along
with it. But the question becomes, where do you define maturity? What is a
"significant" cost reduction? What does the learning curve characteristic have
to be? 85%? 95%? Certainly not above 95%, because almost no industry, not
even the infamous pork bellies, have characteristics above this.

Currently the oil and gas industries, by and large, have characteristics about
90%. I would consider them mature.

The rocket industry (ELV at least), to my knowledge (and EELV bears this out),
is around 85%. Not as mature as the oil and gas business, but mature enough to
make it in my book.

Do you have a specific characteristic on the learning curve that would define a
mature industry?

>>Do you really want people in the newsgroup to think you are dim enough
>>to believe that a weak argument can be strengthened by ad hominem
>>attacks?
>
>It's not an ad hominem attack. It was a question. I didn't say that
>you were dim, and in fact do not believe that you are. I asked if you
>really wanted people in the newsgroup to think that by pursuing such
>an absurd tack.

Fine. I'll tone down the stridency of my responses. Aren't we just as polite
as can be? ;-)

>>A sufficient investment to
>>achieve the goal of order of magnitude launch cost reduction would
>>involve not only uncertain R&D expenses to overcome whatever technical
>>obstacles remain, not only the capital cost of the vehicle(s), not only
>>orders of magnitude improvement in operational efficiencies, but also
>>the building of sufficient market to justify it.
>
>No, it doesn't require a market. This is only required to make a
>profit. The cost will be reduced whether there is market or not.

I guess that is true. In fact, it doesn't even require any R&D or a new
vehicle. If you want to make a "sufficient investment" you can:

A) Endow a trust fund with several $10's of billions of dollars.
B) Use the trust fund to pay 90% of the price of any launch.

And voila! - the cost of access to orbit is reduced by an order of magnitude
for any user of launch services.

To me this simply demonstrates that if you consider only the investment without
any context, it reduces the problem to triviality.

>>Furthermore, the mere act of choosing an
>>irrational actor means that you cannot then turn around and make cost
>>predictions based on standard economics.
>
>One can make cost predictions by analyzing the system that is being
>designed and built, and estimating its costs. The revenues are
>entirely independent of this analysis.

Entirely independent. Hmm...

So costs of a flight vehicle are not dependent on flight rate? And flight rate
is not dependent on having something to fly? And this doesn't depend on
customers, which doesn't depend on what you are charging, which doesn't
determine your revenues?

To me, it seems like these factors are so intertwined that they can never be
considered independently.

>>Whether or not it is an economically sensible thing is the crux of the
>>matter! You can create wondrous technical gizmos out the wazoo. But,
>>if it is not economically sensible, then, to achieve standard ROI’s,
>
>I have never made any claims about ROI's. I don't understand why you
>have so much trouble with this concept.

I have trouble with this concept because in typical business analyses the
return on capital with minimum profit is *included* in the costs of the
project. If you just go ahead and lop these costs off the vehicle, then I'd
say hell! we've probably *already* reduced the cost of access to space by an
order of magnitude.

You keep differentiating between cost and price. OK, let's look at Sea Launch.
I believe they are currently charging on order of $10,000 per kg or a little
less to LEO. They capitalized $1.6 billion to make the system, took 3 years to
build it and the current flight rate looks to be near 6 per year for the next
couple of years. If you assume this price includes amortization of the capital
and 15% ROI, then your *cost* is already down below $2000/kg to LEO. This
doesn't sound right to me, so I assume that at this flight rate they are not
actually making a profit and their costs are actually a little higher (probably
enough to drive them out of business, IMO). But, without detailed knowledge of
their cost structure, we are probably talking about $3000/kg to LEO cost, as
you define it. This is a trivial definition, without much utility.

Not only that, but it lowers the bar on your claim. Your hypothetical
irrational investor making a sufficient investment would now have to get costs
(again, as you define it, without capital or profit) to below $300/kg. At that
level, I think you'd start to run into significant technical problems.

>>you will have to jack the launch price of the vehicles so high that,
>>guess what!, you will be competing with vehicles currently on the
>>market. The “economic sensibilityness” of the vehicle is *part* of the
>>cost. I don’t see how you can possibly separate this.
>
>No, it's not. The economic sensibility of the vehicle is part of the
>business case, as is the cost, but cost and price are independent of
>each other.

Alright, fine. Let's then get this defined straight, as well.
A) We live in the real world, where capitalist economics demand that an
investment make a suitable profit.
B) Amortization of both fixed and working capital is a requirement of a real
project.
C) Investment will not be made until economics are favorable; which statement
means that the investing agency has to be able to include all costs - including
capital costs and profit - into the price that they will charge, and they will
have to be able to get sufficient customers at that given price.

Given those (IMO, highly rational) criteria, then:
A) The price of cargo to orbit is currently in the $8,000 to $10,000 per kg
range.
B) That price cannot be dropped by an order of magnitude in the next
generation vehicle.
C) That price *will* drop, in a reasonably predictable manner, based on a
learning curve with a given characteristic.

Does this cover all the bases?

>It's not nitpicking--it is fundamental to my point. They won't reduce
>price that much, because they need to make a return on their
>investment--they'll reduce it only to the level they need to to get
>the business from their competitors. But you have no idea how much
>they'll reduce cost, unless you're privy to their design team.

Again, I just have to ask: you do realize that a minimum profit and the capital
costs *are included* as an expense along with other standard expenses when you
use the return based on capital recovery with minimum profit method?

So, unless you are assuming that they are making a ROI tremendously huger than
the minimum ROI to get investment, they are not reducing costs by an order of
magnitude.

>>There is no getting around this. It is as
>>intrinsic to the problem as can be. If you cannot reduce the financial
>>risk to an acceptable cost, then you cannot make the investment.
>
>Yes, I can. If I have the money, and choose to do it, I can make the
>investment. Whether or not I get my money back is a different issue,
>but the investment will reduce the launch cost nonetheless.

It is not fundamentally impossible for an individual to act irrationally. But,
according to economic theory (which postulates rational actors) it *is*
impossible. And you are trying to mix and match - you say that an irrational
actor messes up economic predictions to fund something without a return, and
then you turn around and say that you can predict costs based on rational
market assumptions. It seems inconsistent to me.

>No space solar cell costs based on today's limited quantities are
>relevant to the costs of solar cells that are produced in quantities
>many orders of magnitude higher. There is absolutely no relationship.
>That's where the discussion started, and that's where it will end.

The relationship is based on the learning curve, which just a little while ago
you postulated as determining market maturity.

If you postulate a characteristic of 87% for solar cells, which doesn't seem
completely out of the question, then for evey doubling of production, price of
the cells (based on *cost* in a rational market) will decrease 20%.

Since the current rate of manufacture of space solar cells is about 300 kW
installed/year, and cost (optimistically) is $50/kW-hr, then you can find that
cell cost at a production rate of 1 TW installed/year is $0.90/kW-hr. Of
course, this is far in the future, and predictions out that far are extermely
uncertain. But it is certain to say that, unless you have a characteristic
far, far lower than any yet seen in any industry, your cell costs will not be
$0.01/kW-hr ten years from now, even if you have ramped production to these
rates.


BrianF5070

unread,
Aug 4, 2000, 3:00:00 AM8/4/00
to
>Part (a very _large_ part) of being a "commodity" is that there are
>no non-price discriminants between different suppliers. But the last
>time I looked, the US and the EU were in the middle of a trade war
>over hormones in beef, pork, et al, with tarrifs, labelling laws,
>counterveiling duties, et al, all being bandied about. Doesn't look
>quite like a "commodity" to me...

Thank you!

The truth is that by and large these are commodities. But commoditization,
being a process instead of some sort of final state, is *never* complete.


Michael R. Irwin

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to

BrianF5070 wrote:

> >Agreed. Myself and others have repeated put forward the notion of
> >getting US oil companies involved with SPS systems; they are about the
> >only corporate entities with the funds to attempt something like this,

> >they need to look towards the future, and the SPS system can be hooked
> >inot syntehtic fuel plants to produce hydrocarbon fuels for existing
> >vehicles and infrastructure so that we don't need to repave the whole

Brian, thanks for the excellent outline of the required economic
model for likely investment in SPS from existing oil companies.
Could you define the acronyms I have extracted at the bottom,
while I get most of the drift it would be extremely useful to me
to have your definitions. Thanks!

>
>
> (Snip)
>
> OK. Since I work for a company whose competitors have names like ExxonMobil
> and Texaco, let me comment briefly.
>
> First of all, yes, the industry does need to (and is) looking toward the
> future. Yes, we would be interested in technologies that could supply cheap
> and plentiful electrical energy. Yes, we are looking at alternative energy
> sources. Yes, to the degree that it can fit into the existing infrastructure,
> that is a benefit, but we are not really averse to changes in the
> infrastructure, as long as the overall project economics are favorable. And
> the existing distribution infrastructure is a smaller part of most project
> economics than you'd suspect.
>

> But no, the industry does not have the funds to attempt anything that has been
> proposed yet.
>

> To be blunt, a megaproject in the oil industry typically runs about $2 billion
> CAPEX and on average is about 65% debt/35% equity financed. Projects of this
> size are about the largest that the three very largest oil companies (BP Amoco,
> ExMob, and RD Shell) can attempt alone, but normally would be done as part of a
> large JV with three or more partners. The very largest project that has *ever*
> been proposed in the industry was $38 billion total CAPEX, was a hugely complex
> (at least 12 companies, I think) JV, and collapsed under its own weight because
> finanincing couldn't be arranged. In addition, oil companies have found that
> the megaprojects typically don't get the desired discounted cash flow ROR, and
> have been placing increasing emphasis on driving CAPEX down for large projects,
> and spreading cash through many more small, higher DCFRR projects.
>
> If you want to interest an oil company in a SPS project, make sure it meets
> these requirements:
> 1) Reduce CAPEX to absolute maximum of $2 billion.
> 2) Demonstrate technological maturity such that P10 to P90 difference in CAPEX
> is max $500 million.
> 3) Demonstrate a DCFRR of min 20%. More like 25% if you really want to get
> interest.
> 4) Demonstrate a NPV (based on _discounted_ cash flows, Mr. Mook - say 10%
> discount rate) of at least $ 3 billion.
> 5) Demonstrate P50 time from initial capital expense to operability <40 months.
>
> There would be other requirements, but if the project can meet these first
> hurdles, it would go a long way toward solving some problems.

CAPEX

JV - Joint Venture

ROR - Rate of Return

DCFRR -

P10, P90, in context of "difference in CAPEX"

P50 time

NPV

Thanks again, Mike Irwin


Michael R. Irwin

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to

Scott Lowther wrote:

> BrianF5070 wrote:
> >
> > >Agreed. Myself and others have repeated put forward the notion of
> > >getting US oil companies involved with SPS systems; they are about the
> > >only corporate entities with the funds to attempt something like this,
>

> > But no, the industry does not have the funds to attempt anything that has been
> > proposed yet.
>

> Well... maybe the tobacco industry?

After recent court decisions, you are joking right?

The tobacco corporations are probably running on
negative credit lines not cash. Reduce their risk
and maximize their profits by actually moving cash
out of the corporate accounts by posting dividends.

Mike Irwin

Scott Lowther

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to
Michael R. Irwin wrote:
>
> Scott Lowther wrote:
>
> > BrianF5070 wrote:
> > >
> > > >Agreed. Myself and others have repeated put forward the notion of
> > > >getting US oil companies involved with SPS systems; they are about the
> > > >only corporate entities with the funds to attempt something like this,
> >
> > > But no, the industry does not have the funds to attempt anything that has been
> > > proposed yet.
> >
> > Well... maybe the tobacco industry?
>
> After recent court decisions, you are joking right?

No, no. They've paid out, or are in the process of paying out, something
over a hundred billion ("pretty soon you're talking real money"), and
aren't quite dead yet. Get Al Gore to continue his support for the
industry and pry open further Asian and South American markets; perhaps
flood the Russian market as well, and bajillions of dollars will be made
available. Get the State Department to strongarm a few key foreign
governments to allow the US tobacco companies to flood their markets,
and pretty soon there'll be enough money to settle the moons of the
planet around Epsilon Eridani.

Many more billions can be had by making cocaine and marijuana legal, but
a governemnt monopoly, and heavily taxed (with steep fines for black
marketeering). It'll also clear out the jails and eliminate the need for
the "drug war," freeign up many more billions.

We can use this money to paper the heavens with SPS systems, like Oliver
Wendel Jones' "Star Trek" ballistic missile defense system. All we need
is creative governing.

Filip De Vos

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to
Scott Lowther (lex...@ix.netcom.com) wrote:
: wm...@my-deja.com wrote:
: >
: > In article <398680...@ix.netcom.com>,
: > Scott Lowther <lex...@ix.netcom.com> wrote:

Could you guys be a little more diligent in snipping?

<example of removable stuff left in>

: > > > > >
: > > > > > Here's what
: > > > > > John Schilling said the most recent time that you did it, a
: > couple
: > > > of
: > > > > > months ago.
[snip]

: > >An
: > > M-1 Abrams CANNOT SHOOT DOWN AN ICBM.
: >
: > Yes it can. If its sitting right at the launch center and fires
: > somewhere very close to ignition.

: Please. Then that means that the optimal SDI system is to park a tank
: next to every enemy ICBM site. Good luck.

Will you guys knock it off? Please.

You could ask whether a WWII battleship is a weapon? Even though it cannot
shoot down an ICBM? (Unless used as a platform for current AEGIS
radar/missile combo, or further engineered kinetic warheads, better
tracking etc)

To try to get this back on track.

Scott: do you recognise that an SPS can be made to deliver power to ground
stations in a safe manner?

and William: do you recognise that the technology for SPS (as opposed to
the system you propose) can be used to build lasers for military
applications?

[snip]

: > > It'd do a great job setting forest fires, melting asphalt roads,
: > > blinding people, burning people, burning down wooden structures
: >
: > So does 230 V three phase, if things get out of control. So what?
: > Electrical distribution can be dangerous if improperly handled. That
: > doesn't make an electrical distribution system a weapon.

: Again, you can't aim and shoot electricity. The effort required to start
: fires with electric wires at ground level does not compare with the
: relative ease with which this could be accomplished from orbit.

Scott the entire discussion revolves around the ability to deliver amounts
of energy. The comparison with a high-tension line that passes over a
forest is an apt one. If all that energy were suddenly delivered to the
forest during summertime, then it could easily be set on fire. This never
happens.

So generation, delivery and use of electrical energy using present
technology is safe. The occasional short-circuit induced fire in old
homes with decrepit electrical installation do not count and reinforce the
argument that electricity use is inherently safe, when proper and
reasonable precautions are taken. It is exactly these type of precautions
that William is designing in his system.

: I weep for the future if knuckleheads like you get any sort of power...
: such a basic lack of understanding of weapons potential is frightening.

Yeah right we cannot allow Saddam/Kim/Whoever access to 'supercomputers'
(Apple Macs and Sony gamesmachines) because they could be used for guiding
missiles and calculating nuclear bombs, right? And trucks currently used
for carrying milk could be converted to carry munition, or even be turned
into rocketlaunchers, right? We cannot allow Iraq to import pencils,
because the graphite could be used for nukulear reactors, right?

[snip rabid atmospherics]

: > And how would you do that with my beam steering device pray tell? I'll
: > give you the answer to that one. You can't!

: Yes, you can, and quite easily so. You claim your system will use a
: small laser mounted on a target vehicle to direct the power laser. Fine,
: maybe that might even work. But if so... _two_ (better still, three)
: lasers operated by ground troops, separated by some distance, tuned to
: the same frequency and putting out appropriate information, will cause
: the power laser to be directed to a source OTHER than either of the

I doubt that the system as designed could deliver power to a location not
where the pilot beam is generated. Although it will have to steer in
advance of a fast moving vehicle, some fancy trickery will be needed to
point it to the putaticve target. This will probably mean that the
would-be terrorists must track the target with their own laser, to
establish distance of themselves (remember they are calling in the SPS
beam) to the target. This wil expose them to detection and
countermeasures.

: directing lasers... say, split the difference and shoot directly in
: between, or offset by some angle (basic trig, folks). Several setups

yeah right

: such as this would cause multiple beams to fall on a single target. This
: would essentially "confuse" the powersat... instead of the targetting
: laser being a single point, it would be multiple points, and the result
: would be the laser would point in the "wrong" direction.

I think this could only occur on setting up a new beam. Fixed ground
arrays will not suddenly jump ten miles. And mobile users will not
suddenly become stationary or suddenly demand twenty times the amount the
usually ask for.

And unique encrypted client IDs can be generated. Your hypothetical
terrs with their portable SPS confusing laser must also have NSA grade
encryption-busting supercomputers to generate bogus client accounts. In
real time. Flags can be set up in the SPS computer to signal unusual or
bogus power demands.

'Uh boss, this new client, an aluminium smelter, asks for 6 GW at location
Nxx.xx.xxxx Wyy.yy.yyyy. on a dish just fifteen meters across. Should I
deliver? Their credit checks out good.' 'Hold on, is there an aluminium
smelter at that location?' 'No, but there is a restaurant and a
disarmament conference is going on with the President an the Chinese party
secretary in attendance. Joe over on SPS #327 had the same request' 'Oh
cancel that, then... And call the Feds'

: This can also be accomplished by having small "targetting lasers" either
: mounted to the back of the powersat, or as a free-flyer nearby... then
: ground forces would not need to do anything but duck.

Scott. Military airforces can make use of civilian supplies of kerosine at
international airports and refineries. Clearly we must ban the use of
fossil fuels. Terrible things can happen. Fleets of gasoline powered four
engined bombers can carpet bomb central European cities. I say outlaw the
the whole concept.

: There are I'm sure other means of accomplishing this same goal. Your
: insistance that this is impossible, mated to your insistacne that this
: highly complex system can be cheaply built to withstand thirty years
: without a moments maintenance, simply points you out as someone with no

Yup, William is rather a bit enthousiastic about that. But admit if you
have a fleet of SPS in orbit worth seveal hundreds of billions of dollars,
then you have the means to deliver maintenance to them, not?

: real understanding of the real world.

--
"Who needs credibility around | Filip De Vos
here?" -- T. L. Elifritz | FilipP...@rug.ac.be

Filip De Vos

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to
Michael Kagalenko (mkag...@lynx01.dac.neu.edu) wrote:
: Bill Bonde (std...@mail.com) wrote

: ]Do a price vs. performance analysis on x86 compatible CPUs using only
: ]information and trends from before AMD got its act together.


: Solar cells do not get cheaper at the same rate as integrated circuits.
: In fact, the price is leveling out. There is no reason to assume
: that the solar cells will ever get cheaper than the conventional

Did you take account of dual-gap cells? multiple layers? use of
semiconductors other than silicium? concentrators?

: or nuclear power generator (if you don't list "strnuous wishful
: thinking" as a plausible reason the way the space amateurs such as Bonde do)

And you ignore that there are other means of converting solar energy to
beamed energy. Solar pumped lasers/masers and Rankine cycle solar
generators, to give two examples, are likely to have quite different
cost-curves.

Filip De Vos

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to
Scott Lowther (lex...@ix.netcom.com) wrote:
: wm...@my-deja.com wrote:

: > Since when is delivering copious power at very low cost across the globe
: > a problem for anyone?

: Ask OPEC. Care to take a guess at how well Saudi Arabia would handle no
: longer recieving large sums of Western money?

The real decisions are made in Washington, New York and London.

: What was the last war the US got involved in in a serious and determined


: way? And was or was not oil (i.e. energy) a major part of it?

[snip]

: > > And no amount of "no way,
: > > uh-uh" is going to convince America's adversaries that it is a benign
: > > system.
: >
: > <shrug> Ever read the stuff folks said about steam engines on boats
: > when they first appeared? Especially after the explosion of one during
: > a Presidential review on the Mississippi?

: Ever read some of the horror stories the environmental movement put out
: regarding nuclear power plants? When was the last time the US built one
: of these plants? You ignore public and worldwide opinion at your peril.

The same PR that demonised civil nukes can be turned against current
fossil fuel use. Just look at the Ford Explorer 'Ford Valdez'. One way the
environmental movement were allowed such play with their horror stories
was that the economic crisis induced by the oil shock of 1973 (there were
other factors such as the ill-advised grain-farming set-aside program
under Nixon, and the general Vietnam-war-spending induced inflation)
reduced demand for (electrical) power. No utility wanted to buy nuke
powerplants anyway, without the demand for the electrity that it would
generate. They had seriously overbuilt (or over-ordered) during the good
times.

The engineering firms could bilk the power companies for additional safety
measures studies and other add-ons.

And the fossil companies liked it too.

: > Nuclear power plants have had a bad rap, their association with nuclear
: > weapons doesn't help.

: Lasers satellites are equally badly regarded, and for similar reasons.

Do you have a CD-player? Ever go to a disco, or rock-concert?
[snip]

: As far as power comparisons:

: I'll admit to not having exact numbers ready for one of the
: Dresden/Hamburg/Tokyo class raids. But let's take a guess: 1,000
: bombers, each with 10,000 kg of bombs (Lancaster). Assume for simplicity
: that that's 10,000,000 kg of TNT, or 10 kilotonnes. This is 42E12
: Joules, spread over several hours.

How much power goes through a major high-tension line? If all that power
were delivered centralised say in a supermarket or an apartment building,
would it catch fire? Given this dangerous use of electricty, should we ban
its use?

[snip fire in the eminent domain]

Actually the deliberate modification on demand of the military might be a
valid concern. Solutions are open standards/engineering, to achieve
complete built in tamperproof safety, and perhaps a complete separate
military system (with real targeting capabilities, real concentrated
energy delibery, and real targetting along the lines you describe). The
tax take on the SPS system will need to be recycled.

wm...@my-deja.com

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to
In article <8mc3af$td1$1...@nnrp1.deja.com>,


Well, this is interesting! You seem to quote this as if I'm all wet
wrt cost plus contracts. Actually, the use of cost plus contracts
actually strengthens what I'm saying, and expands it in more detail.
Here's how!

There is a role for cost plus contracts within a larger context of
competitive profit type contracts. That's when the uncertainties are
large, and the risk is to be transferred to the person paying the
freight.

Some can reasonably argue this was appropriate during the early days of
space vehicle development, when the government wanted space
capabilities and didn't want to burden aerospace firms with the risk of
development. This made sense at the time.

However, where the problem comes for government mandated programs
versus programs of exploration supported by the oil and gas business,
is that government runs by taxes which are involuntary and taken
without competition in a market sense (though totally voluntary in a
political and legal sense), and oil and gas sales are items sold in
competition between providers and hence are voluntary. So, there's
still market discipline among oil and gas companies to run their
exploration vendors efficiently.

What I'm saying is that oil and gas exploration companies are generally
captives of larger energy concerns who ultimately take the risk in a
competitive environment. Large numbers of exploration providers who
are poorly capitalized compared to their customers provide market
discipline, and competition among the energy concerns provide a
motivation for them to contain costs among their exploration
companies.

In fact the whole setup in these cost plus contracts among oil and gas
companies is to make sure the folks who know how to find and develop
the oil and gas don't enter into direct competition with the energy
concerns.

In the aerospace realm the government doesn't have the same impetus to
contain costs, they don't have a market discipline in the same sense as
the energy concerns.

Some also have pointed out that the government has used the cost plus
strategy to maintain abject control of the spacefaring technologies of
the aerospace firms, much as energy concerns have used it to control
those who know how to find and develop oil and gas.

You ignore the potential for paradigm shifts, which have been known to
occur across the technological spectrum. Are solar cells and other
solar conversion technologies immune? I don't think so. In fact, if
there is a paradigm shift in the way we *must* generate energy, due to
resource limitations, I think solar power technologies will be the
focus of much attention in the hopes of creating a paradigm shift from
current practice.

> > Can't we both do something more reasonable and dispense with the
self
> > flagellation?
>
> Sure.
>
> > Consider, one can draw a cost/volume curve for a variety of space
> power
> > technologies. Agreed?
>
> Agreed. Typically, the curves are drawn in threes - base case, high
> and low. Agreed?

Yep.

> > Now, all we really have to disagree about is where to place these
> > curves for each technology, how steeply the measure you mentions
> > declines with total capacity on orbit, and so forth.
> >
> > But, I think you'll find we're reading off pretty much the same page
> > here.
>
> I expect you are correct, although I think we probably have
> disagreements over the precise steepness of the curve.

Yes, especially when one factors into it the total number of dollars
and the precise capitalization structure wrt technologies.

> > A PV array fed by a lightweight concentrator in orbit has associated
> > with it a specific cost at a specific total demand. I think you'll
> > find that if we put up about 1 Terawatt of big powersats (ignore our
> > ongoing argument about launch costs for a minute, lets settle one
> thing
> > at a time!) using the curve we already agreed upon - we can both
> > calculate what the cost per KW-hr is likely to be.
> >
> > Surprise - its well below $0.01 per kw-hr!!! This was found in the
> > early studies done by Arthur D. Little, and also by ONeil and
others.
>
> Ah, but the curves they use to obtain these estimates are filled with
> assumptions about future technologies - they are not simply
> extrapolations of current curves.

But they *are* based upon solid scientific results. Would you agree
with that? I mean, they're not assuming something that's physically
impossible!

> O'Neill has postulated that solar cells are made from lunar resources,
> which he states makes a quantum leap in reducing costs. This is, at
> best, an unproven assertion.

True. But, if you examine the impact of lunar sourcing and apply a
multiplier to that (divisor actually) you'll find you can achieve the
same reduction in mass from earth per unit area - hence cost per unit
area - by a very lightweight concentrator combined with very high
conversion efficiencies - both of which he ignored.

> > Amazing ain't it.
> >
> > > But they won't.
> > >
> >
> > <shrug> Depends on total amount of energy produced in space doesn't
> it?
>
> To a large degree, yes.

And when you suppose the entire Earth's population by 2050 time frame
is consuming power at US per capita rates and extend the curves out
that far, you'll find that they'll all be paying far less than even the
baseline rates the cheapest utilities pay today!

Assuming this as a target, one can then figure out an optimal
investment strategy to reach that target, given the mix of technologies
readily available!@

> And this amount of energy is part of what I
> have been talking about. The usage of solar power in space has
> followed a predictable upward path which matches the downward pressure
> on prices very well. Although the curve shapes can be altered to some
> degree - made steeper or flatter - they don't suddenly leap by orders
> of magnitude.

Underyling all of this is rate of capital formation - or rate of
investment. If this takes a leap, say because oil prices take a leap,
or discovery rates flatten, then you'll get the orders of magnitude
shifts *I'm* talking about!

> Solar cells in space may well cost $0.01 per kW-hr at some point in
the
> future.

Yes.

> But that point is pretty far off,

I suspect we're using different measures. If you say this in terms of
total capital invested, I'd probably agree with your number. If you
say this in terms of years, I'd say you're assuming a certain rate of
capital formation and this is a derivative figure.

Then what we're really disagreeing about is the rate of capital
formation for SPS technologies and the future cost of oil.

> and certainly well beyond
> the 10 year time horizon that companies plan on for their most blue-
sky
> research.

Yes, 2050 is more than 10 years away. But assuming the end point
above, we must invest more today. If we don't, we face a die off of
humans on a massive scale. But this is yet another point of contention
between us I suspect.

> If you want to get from here to there, you need an
> incremental plan - next generation get the cost of solar cells to
> $30/kW-hr. Make them profitable. The generation after that to
$20/kW-
> hr.

Already during peak summer months folks in LA paid $100 per kw-hr,
there was an article about that in the Wall Street Journal a few months
ago.

So, I think we're there.

> Make them profitable. And so on.

Yep. I've detailed such a plan based on my lasersat technology.

William Mook
Orbatek, Inc.

Scott Lowther

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to
Filip De Vos wrote:

<much angry spleen diligently snipped>

> Scott: do you recognise that an SPS can be made to deliver power to ground
> stations in a safe manner?

Of course.

> Scott the entire discussion revolves around the ability to deliver amounts
> of energy. The comparison with a high-tension line that passes over a
> forest is an apt one.

Horse hockey. You can't shoot electricity. If you want to use a high
tension line to create havoc, you have to have it on site. If you want
to use an SPS-generated laser to create havoc, you don't have to bring
the emitter into nearly physical contact with the target. With
electricity you do, unless you have the power of a thunderstorm
available to do lightning shots (somehow).

If all that energy were suddenly delivered to the
> forest during summertime, then it could easily be set on fire. This never
> happens.

I'm sure the forest fighters battling lightning-induced blazes will be
glad to hear they're fighting phantoms.

> So generation, delivery and use of electrical energy using present
> technology is safe. The occasional short-circuit induced fire in old
> homes with decrepit electrical installation do not count and reinforce the
> argument that electricity use is inherently safe, when proper and
> reasonable precautions are taken. It is exactly these type of precautions
> that William is designing in his system.

Not. Look at the other recent posts: a 200 solar laser is quite capable
of causeing common materials to burst into flames. And since the laser
beams would pass through an unontrolled medium, such as THE AIR, rather
than through power lines, any of a number of things can go wrong,
intentionally or otherwise.


> I doubt that the system as designed could deliver power to a location not
> where the pilot beam is generated.

Oh, no? How about this:

An aircraft cruises subsonically at 80,000 feet (think U-2). Towed
behind the aircraft is a specially built flying wing, built from
lightweight materials, with a chord of 4 meters and a span of 35 meters.
Built through the center of the wing is another rectifier, 2.5 meters in
diameter. The pilot beam on the wing brings down a 200-solar laser 2
meters in diameter into the rectifier. This rectifier is different in
that it was specifically built to have it's refractive optics variable
on-board and requires no pilot beam 9and don't even try to suggest such
a thing can't be built). The laser passes through and continues to a
selected ground target, where ~850 kilowatts is available for as long as
the aircraft can stay airborne. With proper precision manufacturing, the
rectifier can re-focus the laser, bringing it from a 2-meter diameter
bean down to a 1 meter diameter beam, producing a metal-melting 800
solar beam.

So, the safest SPS system in the world can be made into a weapons
system, and quite easiliy so.


> And unique encrypted client IDs can be generated. Your hypothetical
> terrs with their portable SPS confusing laser must also have NSA grade
> encryption-busting supercomputers to generate bogus client accounts. In
> real time. Flags can be set up in the SPS computer to signal unusual or
> bogus power demands.

I assume "terrs" are "terrorists." I never said we'd need to worry about
them hijackignt he system. I suggested THEY'd need to worry abotu the US
military doing it to them. And yes, I'm pretty sure the US military has
a measure of access to NSA-grade computers.

>
> 'Uh boss, this new client, an aluminium smelter, asks for 6 GW at location
> Nxx.xx.xxxx Wyy.yy.yyyy. on a dish just fifteen meters across. Should I
> deliver? Their credit checks out good.'

According to Mook, there would be no human interaction. It would all
take place automatically and virtually instantly. Since he's proposing
to use this system to power everything from motorcycles to fighter
planes to ballistic hoppers, it'd have to be damned fast, with no time
to "think" abotu whether to send the requested power... it'd just do it.


> : There are I'm sure other means of accomplishing this same goal. Your
> : insistance that this is impossible, mated to your insistacne that this
> : highly complex system can be cheaply built to withstand thirty years
> : without a moments maintenance, simply points you out as someone with no
>
> Yup, William is rather a bit enthousiastic about that. But admit if you
> have a fleet of SPS in orbit worth seveal hundreds of billions of dollars,
> then you have the means to deliver maintenance to them, not?

I would, yes. But again, we're discussing Mook's system, and he has
repeatedly claimed that not only will there be no need for maintenance,
there will also be no mechanism for doing so.

Thomas F. Radloff

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to

Filip De Vos <fid...@eduserv1.rug.ac.be> wrote in message news:8mgr06

> ... We cannot allow Iraq to import pencils,


> because the graphite could be used for nukulear reactors, right?

Pencils!? Pencils were a vital part of the computational effort to design
the first bomb. If the WackyIraqies get 'em, WERE DOOMED!!!!!!! :)

TFR

Thomas F. Radloff

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to

Scott Lowther wrote:

> The basic point I'm trying to make is this: using lasers as the power
> transfer medium means that the beam, unlike a microwave system, is a
> _potentially_ dangerous device. Coupled with the fact that this
> dangerous device has the high ground and coverage of the entire planet,
> this sounds like a potential weapons system. And if there is any way,
> any way at all for this system to be modified to serve as a weapon,
> those not in control of it cannot be considered insane for seeing the
> potential.

So are you yourself personally _really_ concerned about this _potential_
threat? Or are you just playing devil's advocate? Are you equally (or more)
concerned about the myriad of other future technologies that are also
potentially dangerous?

Also, one thing you've overlooked is how fragile such systems would be. How
would you defend gigantic powersats from retaliation?

TFR

Scott Lowther

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to
Thomas F. Radloff wrote:
>
> Scott Lowther wrote:
>
> > The basic point I'm trying to make is this: using lasers as the power
> > transfer medium means that the beam, unlike a microwave system, is a
> > _potentially_ dangerous device. Coupled with the fact that this
> > dangerous device has the high ground and coverage of the entire planet,
> > this sounds like a potential weapons system. And if there is any way,
> > any way at all for this system to be modified to serve as a weapon,
> > those not in control of it cannot be considered insane for seeing the
> > potential.
>
> So are you yourself personally _really_ concerned about this _potential_
> threat? Or are you just playing devil's advocate? Are you equally (or more)
> concerned about the myriad of other future technologies that are also
> potentially dangerous?

Ah, a rational response. I (as an American) am not worried about
(American) SPS lasers lancing out of the sky and zapping me or my family
or my stuff. However, the simple fact is that if I can come up with a
possible means to use this system as a weapon of mass destruction, so
can those who might be in power with regards to such as system... and
those who would be our enemies would ALSO be able to conclude that such
a system was a very real hazard... not to the US, but to THEM.

As far as serious threats that actually concern me, SPS and nukes are at
the low end of the list. Gengineered virii are much spookier, and more
liekly to become available to unstable elements. While a terrorist
organization might be able to pick up a nuke or two and perhaps take out
a city, that doesn't threaten humanity as a whole. An asshole with a
briefcase full of aerosol cans filled with the Andromeda Stain and a
series of airline tickets is much worse.

> Also, one thing you've overlooked is how fragile such systems would be. How
> would you defend gigantic powersats from retaliation?

It would be virtually impossible. Which is part of my problem with the
William Mook Solar Death Bubble type of SPS that uses 40,000 very large
(on the order of 70+ meters in diameter) rectifying satelites in LEO. It
would be apparent to everyone on the planet that the US has one hell of
a weapons system hanging over their heads, and boy howdy it'd be easy to
bring it down.

Now, if the SPS system was restricted to GEO and only targetted large
stationary recivers, it would not only be more difficult to attack, but
also less likely to serve as a weapons system (a long distance,
virtually fixed system as opposed to a closer range, highly targettable
system).

But nevertheless, a 1 kilometer inflatable structure, even in GEO, would
be a profoundly easy target to destroy.I t'd only take a few pounds of
high explosive to shred (pretty completely) an infaltable collector a
kilometer across. Make the warhead a shaped charge, with the business
end loaded with 1/4" carbon fiber filaments , and aim it at the focus of
thge collector; when it gets there it goes off and sends a hypersonic
shock wave and a clouds of tumbling needles at the mirror. The recoild
blast from the high explosive woudl send chunks of the bus into the
lasing apparatus.

Scott Lowther

unread,
Aug 5, 2000, 3:00:00 AM8/5/00
to
Anthony Roberts wrote:

> Well... Microwaves penetrate clouds. Most lasers don't.

This fact is well established to everyone but William Mook. How cars,
boats, low-flying airplanes, etc. are to be powered by SPS-based lasers
on a cloudy day has never been made all that clear.


> As for destroying the thing, you wouldn't exactly need high explosives and carbon
> fiber filaments. Just collide a spent rocket stage into the thing. You could do
> it with a big sounding rocket. It would probably increase the budget by a factor
> of 10 just to make it able to survive that much.
>
> Look at Mir, a much sturdier structure, when a collision occurs at a couple m/s.

Well, assume that the Mir cluster of partrs was a cluster of thousands
of panels and units, not just a dozen. Dhe damage would have been more,
but nothing in comparison to the size of the whole thing.

It is loading more messages.
0 new messages