[ExI] Written for another list

Keith Henson hkeithhenson at gmail.com
Wed Aug 1 04:52:01 UTC 2012


On Tue, Jul 31, 2012 at 3:26 PM,  Adrian Tymes <atymes at gmail.com> wrote:
>
> On Tue, Jul 31, 2012 at 12:00 PM, Keith Henson <hkeithhenson at gmail.com> wrote:
>> Adrian Tymes <atymes at gmail.com> wrote:
>>> On Mon, Jul 30, 2012 at 7:36 AM, Keith Henson <hkeithhenson at gmail.com> wrote:
>>>> That's acceptable for powering
>>>> lasers, but I can't see much of a market for multiple GW at 20 times
>>>> the cost of electricity from nuclear plants.
>>>
>>> Nope.  If the cost of electricity is higher than what you get
>>> off the grid, then it's not acceptable for powering lasers,
>>> since you could just get it off the grid instead.
>>
>> That would be lovely.  Do you have any suggestions of how to draw 7 GW
>> off the grid from GEO?
>
> Ah, I thought you were talking groundside.
>
> But if it's 20 times the cost of nuclear reactors in GEO,
> then it's cheaper just to put nuclear reactors in GEO,
> and the higher cost is still not acceptable even for
> powering lasers.

Let me try again.

In the bootstrap plan I have outlined, the power from the first
expensive (built with conventional rockets) power sat is used to power
propulsion lasers.  That lets you build more power sats at a much
lower cost than the first one.  The energy the first one generates is
worth around 100 times as much bringing up parts for more power
satellites as it would be to sell it to ground markets.

So it is worth building the first one even though it costs more than
it would be worth if all you could do is sell the power to earth
markets.

>> It is possible to send the laser beams up to GEO and bounce them from
>> there down to the launch vehicle path some 4000 km along the equator.
>> It may even be more economical because tracking mirrors should be
>> lighter, but the risk factor of it not working due to atmospheric
>> distortion is high.  After looking at alternatives, putting a first
>> power plant and laser in GEO with conventional rockets looked to be
>> the lower total cost and risk.  This is subject to further analysis.
>> Want to do it?
>
> No, because I don't buy your premise that it is not
> practical to demonstrate this at a much lower scale.
>
>>>> I can send you the spreadsheet for the financial model if you want to
>>>> try making money on conventional propulsion.
>>>
>>> I might be interested in taking a look *if* your spreadsheet
>>> includes the costs of the intermediate conventional
>>> propulsion stage.  If it does not, then you need to add that
>>> before it is a complete, realistic model.
>>
>> It does, of course.  But only to build the first propulsion laser, not
>> to sell power for a dollar a kWh.  Now if you have a *market* for 5 GW
>> of power at a dollar a kWh, we are in business.
>
> Hmm, let's see.  You know of any customers who
> 1) are operating in a mostly-sunny environment,
> 2) have a high cost of importing fuel to their bases,
> 3) have a need for mobility (like, say, being able to
> repoint their incoming power),
> 4) already have a no-fly zone above the places they
> would like to get power, and
> 5) have a history of "price is no object" mentalities?
>
> Granted, their current need won't last forever.  But
> if they can pay to get the sat up there, and they
> take most of the power - you only need to tap it for
> a few minutes (maybe up to half an hour) to do the
> launch, and these intervals can be scheduled for
> moments of anticipated low demand - then, well...

There has been a lot of looking at selling power from space to the
military.  Never reached the big study phase because there are just
too many problems.  The military wants power in MW or sub MW chunks.
Microwave power sats at 2.45 GHz don't scale below 5 GW, that's 5000
MW, so you have to use lasers, relatively small ones compared to 3.5
GW for propulsion.  Laser about doubles the losses from 50% to 75%.
It also means the power is cut with every passing cloud.

But if we were to build laser power sats for the military, they would
probable use them as weapons rather than power.

>>>> Perhaps.  I freely admit the model may have errors in assumptions or
>>>> formula.  Though saying that without looking at the model seems a bit
>>>> over the top.
>>>
>>> No, it's basic business theory.  If you can make a profit with
>>> practice X after it's developed, and a lesser profit with
>>> practice Y that does not take significant development, then
>>> you can compare the efficiency of developing X and then
>>> making more money over time versus just using Y.
>>
>> I f can figure out a way to get this started on a smaller scale,
>> please do.  If you can, I will support your efforts.  I am not welded
>> to this approach.
>
> What is the gating factor on cost per kg, that makes a
> 10 ton solar satellite that much more efficient than a 1
> ton, or a 0.1 ton?
>
> (Side note: "a 0.1"?  "an 0.1"?  Spike, maybe you could
> clue me in as to which is correct?)

For a given frequency at one of the atmospheric windows such as 2.45
GHz, the minimum size to focus the microwave power beam is 1 km on the
power sat and 10 km on the ground, or the other way around.  This is
what leads to the huge size of these things, the microwave optics.
This has been understood for over 200 years, look up Airy disk.
http://en.wikipedia.org/wiki/Airy_disk

>> As an analogy, it is not possible to mine a low grade ore body and
>> make money if you try to do it on a small scale.  Sunlight is a low
>> grade (dilute) energy source.  Collecting it on a scale of tens of TW
>> is a big task no matter how you slice it.
>
> If tracking mirrors are that much cheaper and lighter,
> then how about replacing most of the PV cells with
> those?

Some of the proposals do concentrate light.  The limit for
concentrating light on PV depends on how hot the PV cells can run and
how well they are cooled.  I favor thermal designs (Carnot engines)
largely because they can be substantially smaller (they are more
efficient) for a given power output.

>>>> I am mainly interested in making a case that there *is* a way out of
>>>> the energy/carbon problems without an 80% die off.
>>>
>>> The theoretical case has long been made.  The challenge now
>>> is the litmus test: actually doing it.
>>
>> That's news to me.  Where?
>
> The theoretical case having been made?  Here, for one.
> This very list.

I don't remember such discussion.  Could you point me to an actual
post?  In a case where we are talking about avoiding an 80% die off,
it's not much use to say they sun puts out plenty of energy so we
don't have a problem.  We do have a problem because sunlight is a
dilute energy source.  It's not easy make economical use of it, and
uneconomical modes cause the 80% die off.

> Most of the general public does not seriously believe
> that humanity will have an 80% or worse die off in the
> next century, but they do believe that the energy/carbon
> problems will get, if not solved, at least not very much
> worse than they already are - which generally requires
> at least a partial solution.
>
>>>> The cost of power at current $10,000/kg is dominated by the lift cost
>>>> of ~50,000/kW.  Cost of power at that transport rate is ~$2/kWh.
>>>>
>>>> For zero lift cost, the cost would be around 1.4 cents per kWh.  The
>>>> derivation of this is in the paper.
>>>
>>> It sounds like you have bigger concerns than the lift cost, if
>>> that is not a majority of the cost.
>>
>> It's around a third of the total cost.  I really don't understand your
>> objection to the other costs.  Do you know of a less expensive way to
>> make and transmit power to the Earth?
>
> Strawman.  My objections regarding the other costs have
> to do with the other costs themselves, not the benefits.
> To wit: what *are* the other costs, and how big of a factor
> are they?

You mean you don't understand the cost of generation equipment?  It's
a highly developed field with razor thin margins and a long history of
slow improvements.  You can get a feel for the cost by looking up what
an emergency generator costs.   PV cells are around a $1000/kW (peak).

snip

>> If you can follow the physics, why not is in here:
>> http://www.sspi.gatech.edu/aiaa-2009-0462_ssp_alternatives_potter.pdf
>>
>> There are scaling problems that are due to the transmission of
>> microwaves through the atmosphere and the distance to GEO.
>>
>> It is about the same expense to built one demonstration power
>> satellite with conventional rockets as it is to set up the
>> transportation pipeline and build them by the hundreds.
>
> The paper was written by Boeing.  Boeing institutionally
> does not believe in cheap solutions: they threaten
> Boeing's business model.  They have a history of making
> assumptions to justify keeping things expensive.

I know these people personally.  They are not influenced by Boeing
business model in this assessment.

> A demonstration - pure demonstrator - satellite can be built
> and launched for tens of thousands - not millions - of dollars.
> It's called a CubeSat.

You could pack a few mW of microwave transmitter into a CubSat.

Why bother when there are communication satellites pouring down as
much as ten kW of microwaves?

> Yes, it would provide tiny - infinitesimal - power to the ground.
> Yes, it would not be cost-efficient to do just 1U CubeSats.
>
> But as a pure demonstrator?  There is a world of difference
> between "trivial" and "nothing".
>
> Besides, if lasers really are that much better, and provide
> you energy for every launch thereafter, could you bootstrap:
> launch something small, then use it to power launching
> something a bit bigger, then gang them up to power
> launching something yet bigger, and so on exponentially?
>
> How much mass could you launch at once with, say, a
> single kW?  How many such launches would you need to
> make your next kW?

You have the right idea, but the scaling laws won't let you do it that
small.  There is a size, 5-10 tons depending on the technology, below
which you can't get *any* payload to orbit.  It has to do with the
amount of atmosphere you have to punch through, and the square/cube
scaling of the vehicles.  I assure you that if I could start with 1 kW
I would.

Keith



More information about the extropy-chat mailing list