Barely affordable SPS using ISRU in LEO

[a presentation by Paul Rosenman at the ISDC 2007]

I’d like to talk to you this afternoon about space solar power, and to show you how we can collect solar power in outer space and use it on earth. This is going to be primarily about the revenues and costs involved.

There are many benefits to being able to collect energy from outer space. It is clean, renewable energy that, unlike most renewable energies would be available day and night, so could provide baseline power. This energy has no carbon footprint, nor does it create any hazardous radioactive materials. If we could collect energy from outer space it would be wonderful, but of course, there is a catch. It is a very expensive proposition. This idea was created by Dr. Peter Glaser, and looked at in the 1970s but deemed way too costly

Let me start with an explanation of this talk; Barely affordable SPS using ISRU in LEO. The SPS refers to a solar power satellite project. ISRU refers to “in situ resource utilization”, and LEO is low earth orbit. Barely affordable speaks for itself. So, a barely affordable space solar power project by using in-situ resources in low earth orbit. Let’s examine a solar power project, and then ways to reduce the costs.

FIRST SLIDE of SSI cover Solar Power Project picture

This first slide shows an artist’s rendition of the Dr. Peter Glaser plan for a large space solar power project. It comes from the cover of a Space Studies Institute conference publication of 1993.

It shows the basic pieces of a space solar power project. This picture takes place in geosynchronous orbit, about 22,000 miles into space. It shows solar collector cells being mounted onto a frame by astronauts. Those cells are always facing the sun and collecting solar power. They transfer that power to the transmitter, which transforms that power into microwave energy, and beams it to earth. The inset shows the microwave beam arriving on earth. There it is collected, and turned back into electricity and delivered to the electric grid. Because of the geosynchronous location, the satellite always stays over the same location at the equator on Earth. This solar power project consists of 2 parts. The first part is the large, solar collector satellite in geosynchronous orbit that collects the solar power and beams it to earth as microwave energy. The second part is the receiving facility on earth.

Let’s look at the revenues from a 5 Gigawatt version of this project. The highest cost that you could charge for this clean power, and have customers willing to pay, is about ten cents a kilowatt hour. That is high for most Americans, but not unreasonable. I live in New York City, and sometimes pay more than that for electricity, and sometimes less. If you had a 5 gigawatt power plant and ran it 24 hours a day for a year, your revenue would be about $4 billion dollars. It would actually be a bit more than that, but this makes the numbers easy. So, in ten years of operation, you would sell $40 billion dollars of energy, in 15 years, $60 billion. If we can get costs into this range, the project might be possible.

That sure seems like a lot of money! Then the costs must really be large. Let’s take a look at those problems.

The launch cost from Earth to low earth orbit is the greatest impediment to this project. It is currently about $5,000 per pound to low earth orbit, and it has been about that cost for a long time. One important reason for this is that the rocket booster that launches the payload is destroyed after it’s task is finished. I get that $5,000 per pound to low earth orbit figure by dividing the delivered payload of a launch by the cost of that launch. As an example, published on the net, the European Space Administration’s Ariane V costs $180 million dollars per launch, and delivers 16,000 kg to low earth orbit. That works out to about $5,100 per pound.

Going back in time, the Space Shuttle was sold to the American public as a way to lower the costs to low earth orbit by being able to be reused. As it turned out, it was widely said to cost about double the standard cost to low earth orbit. The payload is 40,000 pounds and it cost $400 million dollars, making the launch cost $10,000 dollars a pound to low earth orbit.

Since then, the work done on Single Stage to Orbit (SSTO) vehicles hasn’t changed that cost to low earth orbit yet, and NASA is no longer paying much attention to that field, though private industry is. But for the foreseeable future, launch costs to low earth orbit are going to be in the area of $5,000 per pound. Now what does that really mean for the costs of this project?

One part of a solar power satellite is solar cells. One way to rate these cells is in kilowatts of power collected per kilogram of weight of the cell (kW/Kg ). Current cells are 2 kW/Kg. To launch 5 gigawatts of solar cells to low earth orbit would cost $22.5 billion at $5,000 per pound launch costs, and that is just for the solar cells. If you launch them to geosynchronous orbit, where they need to be, the cost doubles to $45 billon. That is why it is so expensive to do this project. To compare, the solar cells cost about $1 apiece or about $5 billion for 5 gigawatts of collecting capacity.

The hardware that has to be delivered to geosynchronous orbit and assembled to do this project consists of the solar cells, the wiring and power management hardware, the structural parts, and the transmitter. The total weight that goes to geosynchronous orbit comes to about 3 times that of the solar cells, making the cost of delivering just the parts to geosynchronous orbit about $135 billion. And they still have to be bought, and assembled. How can we make those costs less?

Ion engine technology can reduce the costs of going from low earth orbit to geosynchronous orbit by a factor of 7, if we ignore the cost of energy to the ion engine. Since we are using solar cells, if they can be made usable in low earth orbit, then our cost of energy is free. That means we can deliver our materials to geosynchronous orbit for about $75 billion, plus their costs, Add what the solar power collector assembly in low earth orbit costs, and the ion engine vehicles. Still way too costly to be practical, but on the way.

There is an untapped materials stream available to us in low earth orbit if we can re-manufacture it into useful parts. It consists of the empty booster rockets that are discarded and destroyed in re-entry. Many boosters are launched with “strap-on” boosters. If these “strap-ons” are strong enough, they can boost the rocket boosters into low earth orbit. Also, depending on the size of the “strap-ons”, there may be no penalty in payload delivered.

The Space Shuttle has very powerful “strap-ons”. In a Shuttle launch, the Shuttle takes it’s External Tank (ET) 95% of the way to space, then turns downward toward earth and releases the External Tank, so it gets destroyed over the Indian Ocean. Then the Shuttle climbs into orbit. With little or no payload loss, the External Tank could be brought to low earth orbit for remanufacturing. As a cost example, the weight of the External Tank is over 60,000 pounds, so the Shuttle’s total payload would be over 100,000 pounds. That would be $4,000 per pound to low earth orbit; contrasted to the standard $10,000 per pound.

For the Ariane V, with it’s standard “strap-ons”, the payload penalty is from 16,000 kilograms payload to 6,000 kilograms delivered to low earth orbit. But, to the 6,000 kilograms, add the weight of the empty booster, plus the weight of the upper stage, The total payload comes to over 21,000 kilograms, or about $3300 per pound.

This untapped materials stream is made of refined, lightweight materials and we know the composition of all the pieces. We can cheaply and handily turn this materials stream into the structural parts we need, and save on the launch costs from earth. That saves at least $20 billion dollars, if we get the re-manufacturing facility to be cheap to build and run.

We do this by using the materials stream to create a re-manufacturing structure in low earth orbit that has artificial gravity by rotation. We populate it with earth-normal machinery and earth-normal closed ecology equipment to save on costs of development and operations. We take advantage of the work already done on earth with closed ecology, like the NASA LMLSTP structure that held 4 people with full air and water recycling for three months. We don’t have to develop any manufacturing techniques in microgravity to do this project. We take empty booster rockets in low earth orbit and connect them together, stuff them with closed ecology, habitat and manufacturing modules from earth. Then we spin them to create an artificial gravity.

So we’ve added to our 2 location project, geosynchronous orbit and earth. We’ve added a manned, low earth orbit facility. Here we combine solar cells from earth with the remanufactured structural materials from booster rockets. We re-manufacture them into solar power panels. These panels are then moved to geosynchronous orbit by unmanned ion engine craft that use the solar power collected by the solar power panels.

Let’s examine an example of this technique. I chose as the empty booster the Shuttle External Tank. It has the most square footage available, and only oxygen and hydrogen propellants are used, so people won’t have any local environmental issues. For the artificial gravity, people can best accept rotation when it is not higher than 2 times per minute. I chose a 100 meter structure diameter, so that at 2 rotations per minute, an artificial gravity of 4 tenths of earth normal gravity would be created, quite like that of mars. An earth normal gravity would require an 800 foot diameter at 2 rotations per minute, which would be much more expensive.

A simulated martian gravity in low earth orbit would be an excellent test bed for mars missions, for closed ecology and re-manufacturing testing, things needed anyway for this project. That should help reduce hardware costs significantly.

SECOND SLIDE : LEO Re-manufacturing Facility

Here are some brief slides as an example. This first slide shows 4 External Tanks linked together. Only the 2 External Tanks on the outside are significant for this example, the others being used only for structure. But, in those boosters used for the structure’s trunk, anywhere a floor is added will have a lower artificial gravity than a martian gravity.

THIRD SLIDE : External Tank Cutaway diagram

This slide shows a cutaway of the External Tank. It shows the empty hydrogen tank, about 100 feet long and on 2, levels about 5000 square feet of space. I show modules for habitat, foundry, machine shop, and closed ecology.

FOURTH SLIDE : Structural pieces created

This slide shows the pieces needed to be created, pipes and pipe holders for frames and trusses. Also created are thin sheets of aluminum to be used to attach the solar cells.

FIFTH SLIDE : LEO Re-manufacturing Facility with solar power panel

This slide shows a 100 meter square solar collector panel, created inside the re-manufacturing structure, and assembled on the outside of the structure.

SIXTH SLIDE : Group of solar power panels linked with an ion engine tug.

This slide shows a group of solar collector panels being ferried to geosynchronous orbit by an ion engine tug. They are then joined to the group of solar power panels there.

SEVENTH SLIDE : Cost Estimate

This is the important slide. It shows launch costs for the materials, costs to get them to geosynchronous orbit, the cost of the parts, and a note about the cost of operations.

The total is $55 billion dollars, which could be paid off in under 15 years of production of this 5 gigawatt solar power station at $0.10 per kilowatt hour. Plus, you get a manned, closed ecology, artificial gravity facility in low earth orbit. That is worth quite a lot.

The numbers for the solar cells are more solid than the others. I estimate the weight of the structure is equal to that of the solar cells. I show a delivery cost to low earth orbit as $700 per pound. I also estimate the weight of the wiring, power modules and transmitter elements is equal to the weight of the solar cells. but I show less cost for launch because much of the wiring will come from of the recycled boosters.

The total cost of the project could range between 40 and 65 billion dollars. That is not enough to tempt private industry, they will want at least 20% on their investment. But, this project could be possible with a federal loan guarantee, for example. The government will have to be involved due to the requirement to put the collector satellite in geosynchronous equatorial orbit, where ITU rules apply.

This analysis is preliminary. It shows that this project is possible. What is needed is a feasibility study to resolve many of these issues. I look for your help in making the feasibility study happen.

I want to begin the question and answer phase with a question I’ve often been asked. Some have said “The Shuttle doesn’t make a good example, it’s going out of business, and it is booked thru then.”

First, The External Tank is being incorporated into the new heavy lift booster, so will still be available.

And second, and more important: When the Shuttle program ends in 2010, what will happen to the 2 remaining Shuttles? I expect one will go to the Air and Space Museum, and the other will grace the entrance of some NASA center.

If, when they are done, they should not mothballed, and not refurbished, but launched again into space to permanently reside. The 2 External Tanks launched with them could be the ones in my diagram that are stuffed with the modules for re-manufacturing, and the two Shuttles could be the original habits for the construction crew.

I’d also like to add that a lot of attention has been focused on the problem of the debris in low earth orbit. In 1997 I gave a talk at a Space Studies Institute conference. I showed that debris could be collected profitably. On a 10 year payout, at 20% interest, the break even price for 3000 pounds per month would be $200 per pound. For 1500 pounds of debris collected per month, it would be $400 per pound. I used an unmanned, space robot already in testing and added ion engines to it. I used the robot to capture debris in low earth orbit, and return it to a re-manufacturing facility, also in low earth orbit.

So, this project can help with many of our current problems. It can begin to help with the current energy problems It can be a test bed for systems going to mars and the moon. It can help mitigate the debris problem in low earth orbit. It can help to open up another pathway to space. I look forward to your help in making the feasibility study happen, to fully quantify the costs.

Paul Roseman May, 2007

2 Replies to “Barely affordable SPS using ISRU in LEO”

Leave a Reply

Your email address will not be published. Required fields are marked *