Veröffentlicht am

Transporting Heavy Duty to the Moon

I have published a proposal for building powerplants on the Moon. This plants would solve mankinds energy problem, and could provide the cheapest energy ever, cheaper than waterpower. They base on a radical approach to make the well proven technology of thermonuclear bombs available for engergy production. In stone caves, deep beneath the lunar surface they produce huge amounts of hot steam that powers turbines and generators on the surface of the moon which send their electric energy by means of microwave beams to relais sattelites in earth orbit which distribute the energy to receiving stations near cities on earth. At this time I’m 80 percent sure that the plants will work. For most technical problems I could find a practical solution, sometimes even alternatives. But there are still three technical problems to solve, that could endager a realization. From a physical point of view the plants will surely work. If You are interested in this subject You can read about the Nomad Fusion Reactor here [1].

One of the three technical problems that are still open is to find a way to bring the materials for each plant to the Moon for about less than $10 billion dollars. That seems to be a lot for raw transportation cost, at the first view. But it is not so much money, if we have to carry turbine parts of up to 200 tons to the lunar surface. And not only one. Many of them! I will show a way to solve this huge technical problem in this article.

For example the Apollo lunar excursion modules (LEM) carried approximately 5 tons to the Moon. Their payload was simply the upper stage of the „moon spider“. But there is a big difference between 5 tons and 200 tons. If we simply extrapolate the weight of the Saturn V rocket – it was 2800 tons on the launch pad – we get for 200 tons to the moon a 112,000 tons rocket. But it must be much less. When we think about the moon flights, we remember, there was a return possibility for the astronauts. A waiting space ship in the moon orbit: the Apollo command and service module (CM,SM). How much tons could we land with a Saturn V on the moon, if we would do the flight without a return vehicle, and just delivered as much as possible on the lunar surface? Don’t worry, I have calculated it for You. It’s possible to land about 10.5 tons on the moon, if we put a raw landing stage with a typical liquid hydrogen engine with a specific impulse of 450s on top of the Saturn V third stage S-IVB. If we extrapolate it lineary, it would be still a 53.000 tons rocket on a launch pad to carry 200t to the moon.

The Expendable Approach

But is this the right way, a simple linear extrapolation? When I remember my rocket science studies: no, completely wrong. Anything in rocketry is highly non-linear. We have to calculate more. To be able of it, I have built my own theoretical Saturn V. This is a theoretical model, which can be simply changed and modified. It consists of four sub-models: a lunar lander model, a third stage S-IVB model, a second stage S-II model and the final S-IC first stage model. It has the following properties:

  1. the rocket model is a simple cylinder model
  2. any stage has a certain percentage by volume of empty space without fuel tanks and of used space
  3. any stage has the sam mass fraction like the Saturn V stages had: 3.85, 3.85, 2.18
  4. any stage has a certain percentage by surface area of aluminium and of steel
  5. the stages are build of 1/4 inch, 1/5 inch and 1/8 inch metal sheet plates
  6. stringers and frames to stiffen the metal sheets use additional a certain percentage of the sheet metal areas
  7. the surface areas of the rockets interstages (the empty space) must have double thickness metal plates
  8. all surface areas of LOX/LH2 tanks need double walls and foam between the walls
  9. the model varies a start mass dependent quantity of F-1 kerosene engines and J-2 hydrogen engines

With all this assumptions I get the values in the following table as a result:

S-IC S-II S-IVB Lunar lander
Stage diameter [m] 10 10 6.6 3.3
Stage length [m] 42.1 24.8 18.8 6 spherical tanks
R 3.85 3.85 2.18 2.2
Start mass [t] 2810 600 120 45
Fuel mass [t] 2080 444 65 25
Burnout mass [t] 730 156 55 20
Payload mass [t] 600 120 45 11.5
Structures and devices mass [t] 84 27 8.2 8
Engines mass [t] 46 9 1.8 0.7
Dry mass stage [t] 130 36 10 8.7

which fit very well to the true-to-life values of the Saturn for such a simple cylinder model.

The lunar lander model is a little bit different. It is not a cylinder it is a spacial frame structure with 6 spherical tanks. The tanks are made of aluminium with double walles of 1/8 inch each, that have a thick layer of super-insultation as heavy as the aluminium surface. The framework is made of steel box sections of a side length of 1/100 tank radius, a total length of 1000 times the tank radius and a material strength of 1/8 inch. I have calibrated the model with the dry mass of the real lunar lander of Apollo to get realistic values of an extreme lightweight structure that meets moon environment conditions.

For the Saturn V Encyclopedia Astronautica says: „Development Cost $: 7439.600 million. Launch Price $: 431.000 million in 1967 dollars in 1966 dollars“ [2] for the Saturn V, that is development and building cost for the three stages: S-IC, S-II, S-IVB. That is $431 million for a sum of 176t dry wight or $2.45 million per ton for building one rocket in 1966 dollars and $7439.6 million for 176t or $42.27 million per ton for developing one rocket in 1966 dollars.

The consumer prize index (CPI) for 1966 was 32.92, and the current CPI (2012) is 202.5. So we can calculate the building and development cost by multiplying it with 202.5/32.92=6.15 and get for today:

  • $15.07 million per ton for building rockets comparable to Saturn V
  • $259.96 million per ton for developing rockets comparable to Saturn V

With this values it is now easy to estimate the building and buying cost for todays rockets.

After I had calibrated my rocket calculation model by comparing it with the real values of Saturn V, I used it to extrapolate the Saturn V technology to a launcher, that is capable to land 200t payload on the surface of the moon. Here is the table with the resulting values:

S-IC S-II S-IVB Lunar lander
Stage diameter [m] 31 31 22 7.7
Stage length [m] 38.7 24.8 18.8 6 spherical tanks
R 3.85 3.85 2.18 2.2
Start mass [t] 24880 5806 1331 557
Fuel mass [t] 18400 4298 720 306
Burnout mass [t] 6480 1508 611 251
Payload mass [t] 5830 1336 564 200
Structures and devices mass [t] 250 85 27 43
Engines mass [t] 400 87 20 8
Dry mass stage [t] 650 172 47 51
Building cost $ million 2012 9795.5 2592.04 708.29 768.57
Development cost $ billion 2012 168.974 44.71312 12.21812 13.25796

You see, the expendable launcher on basis of Saturn V technology had a start mass of 24,880 tons. That is less than the half of 53,000 tons as the linear estimation has resulted. But 24,880 tons is still a huge lift off mass. Most of the dry mass would be 44 F-1 engines that would give it the same acceleration as a Saturn V had. It seems unbelievable that the first stage can lift 25,000 tons with only 555 t structures, but it is the result of the lightweight-technology used in the Saturn V. Because I didn’t change the length of the stages in my model, the static pressure of the fuels would be the same and so they would need the same wall thickness for the tanks. The volume of the tanks grows faster as the surface areas when increasing the radius. So the result seems to be correct.

The estimated development cost for the launcher would be $237 billion. The cost for building only one of these huge expendable launchers would be $13.9 billion. That is too much. I need several flights of 200t payload to the moon with $10 billion altogether. So I have to go a complete other way. I need a giant reusable launcher that I have to build once and then only have to pay for the fuel and operations cost for each flight.

The Reusable Approach

One of my rocket science teachers, Heinz Hermann Koelle, allways said „Think big!“ I didn’t agree at this time, I was thinking: small is beautiful. What did he mean? When you build big structures you benefit of a simple geometrical phenomenon: areas grow small, volumes grow fast by increasing the diameter. This means you can put much more fuel into your rocket with relatively sparse surface area effort around it. Structures of shell constructions like rockets are just a kind of surface area. You only have to be sure, they are stiff enough for stability reasons and they can stand the inner pressure of the vessels. Therefor it is much easier to build lightweight with huge structures than with medium structures or even small structures. This means it is more simple to build big lightweight things than small lightweight things. It seems to be an antagonism at the virst view. But when we realize that we are talking about density or relative weight of the structures, because that is the important value, then it is completely understandable, I think.

When we want to build a spacecraft that is reusable and plan to build reentry technology we get into the problem with the whole density of the vessel and the resulting heat density. This means our relatively dense spacecrafts fall down to earth fery fast, penetrate the denser atmosphere deeply, heat up fast and suffer high heat densities. This is why we need special heat shields that are thick-walled and can stand more than 2000° celsius. A huge spacecraft that returns without fuel, has a big empty volume that gives it a high drag coeficient and so a low ballistic coefficient. That means it can’t penetrate the athmosphere so deeply with high velocity because it is breaked long before. So the heat density is by far not as high as with smaller and denser vehicles. This means on the other hand, we don’t have to use such thick-walled heat shields and not such high temperature materials. This is a big advantage.

And there is a another advantage of size: the sensitivity on design changings is much lower than with medium or small structures and therefor the margins are much higher. This is also because of the geometric phenomenon I mentioned. To summarize the three advantages of building big spacecrafts:

  1. lightweight constructions are natural and can be achieved without effort
  2. low ballistic coefficients and resulting low heat density at reentry
  3. low sensitivity on design changings, higher margins

Designing big has of course also disadvantage: high absolute masses and resulting high investment cost. So it makes absolutely no sense, if you don’t have an application for it. But that is exactly what we have now: big, cheap power plants on the moon.

It seems to be good to start from the first design with a two stage rocket with both stages reusable. A one stage rocket will become very huge that the much more fuel it consumes will overhaul fast the additional recovery cost for a first stage. Because of the size of the stages they can only fly back to Earth in a ballistic way. Wings or parachutes are not possible at this size. Both stages land on the sea and are tugged back to the launch pad. Heinz Hermann Koelles [3] Neptun design in a bigger version and with modern engines may be appropriate. The first stage will have an Isp of 300s with high pressure RP-1 engines, the second stage 470s Isp with long nozzle vacuum LH2 engines.

To complete the reusable approach there must be a reusable moon lander which lands with 200t payload on the moon and afterwards flies back into a moon parking orbit where it waits for the next payload and refueling. The transporting system is closed with a reusable moon ferry which commutes between earth orbit and moon orbit. The ferry has its parking orbit in earth orbit. Here it waits for payload and refueling. Then it flies to the moon orbit, where it refuels the moon lander and assigns the payload. Then it flies back to the earth and waits in its parking orbit for the next operation. Here’s a table of the complete mission:

Neptun 1st stage Neptun 2nd stage Moon ferry earth accelleration Moon ferry moon decelleration Lunar lander moon decent Lunar lander moon ascent Moon ferry moon accelleration Moon ferry earth decelleration
DeltaV [m/s] 2079 7564 3400 1050 2500 1900 1050 3400
Isp[s] 300 470 470 470 470 470 470 470
R 2 5 2.1 1.25 1.7 1.5 1.25 2.1

Here comes a table with all components of the transporting system, the mass fractions, mass balances and payloads. A first estimation of building and development cost by means of the factors from the last chapter is included:

Neptun 1st stage Neptun 2nd stage Moon ferry earth accelleration Moon ferry moon decelleration Lunar lander moon decent Lunar lander moon ascent Moon ferry moon accelleration Moon ferry earth decelleration
Stage diameter [m] 50 to 70 30 to 50 12 12 6.5 6.5 12 12
Stage length [m] 21 75 6 spherical tanks 6 spherical tanks 6 spherical tanks 6 spherical tanks 6 spherical tanks 6 spherical tanks
R 2 5 2.1 1.25 1.7 1.5 1.25 2.1
Start mass [t] 20892 9621 1860 884 425 50 315 252
Fuel mass [t] 10446 7697 974 177 175 17 63 132
Burnout mass [t] 10446 1924 886 707 250 33 252 120
Payload mass [t] 9700 900 764 587 217 0 132 0
Structures and devices mass [t] 441 932 113 113 25 25 113 113
Engines mass [t] 304 92 7 7 8 8 7 7
Dry mass stage [t] 745 1024 120 120 33 33 120 120
Building cost $ million 2012 11227.15 15431.68 1808.4 497.31
Development cost $ billion 2012 193.6702 266.19904 31.1952 8.57868

For the moon ferry and for the lunar lander I have used the same model as for the Saturn V technology lunar lander. They are both extrem lightweight structures built of spherical tanks with super-insulation that their fuel does not evaporate too fast.

For the Neptun, I had to build a new slightly more complicated model, because the Saturn V simple cylinder model did not fit for such a reusable system. The Neptun calculation model looks like this:

The still simple but sufficient calculation model for the Neptun HLLV first and second stage. A simple cylinder model did not work, Fig: Author

The first stage of the Neptun model consists of 19 kerosene and oxygen tanks of 10m diameter and 7 meter hight. Below them is a cone with 7 meter hight and 70m diameter, around them is a cylinder with 14m hight and 70m diameter. 60% of the tanks are carbon reinforced plastic tanks for the kerosene with stainless steal liner and 40% of the tanks are of aluminium and contain liquid oxygen. The tanks have low pressure and are made of 2.56mm sheet plates. 25% additional sheet metal is used for tank reinforcement. The skin of the vessel is of 6.4mm sheet plates, and has a double wall: the inner wall of carbon reinforced material, the outer wall of stainless steel. Engines of F-1 size, mass and thrust are used to provide the appropriate thrust to lift the giant vessel. They are also considered in the calculation.

The second stage of the Neptun model consists of 7 hydrogen and oxygen tanks. They are 50m tall and 10m in diameter. The six outer tanks are raw hydrogen tanks, the tank in the middle is split into an upper part for hydrogen and a lower for oxygen. They all have a double wall of 2.56mm aluminium and foam is between the walls. The outer skin consists of a cylinder of carbon material that has the same diameter than the first stage and is as tall as the tanks, plus 35% more hight that is made of steel. I use this 35% steel cylinder as margin. The upper stage has 3 base plates, one is made of carbon reinforced material, one is made of high temperature steel and one is made of a thin layer of several centimeters of a ceramic high temperature material. All outer structures are made of 6.4mm material thickness and additional 80% reinforcement like stringers and frames. The second stage uses Space Shuttle main engine (SSME) like engines in size, mass and thrust.

I calculated my model in a Matlab simulation and got hundreds of those charts:

The visualization of the simple Neptun model, Fig: Author

We see the mass in tons over the growing radius for the dry mass (black), the payload (red) and the burnout mass (blue). The green line is the fraction of structure mass to payload mass which is drawn with a factor of 100 in this diagram, this means e.g. 500 is actually 5. The diagram is actually the visualization of the second stage model. Very interesting is the vertical line, where the fraction jumps from negative to positive values. At the same radius the red payload line becomes positive and the burnout mass curve is cutting the dry mass curve. This means, if we would try to build a second stage that is nearly describable with my model and would make the outer diameter approximately 13m it wouldn’t carry any payload. If we would build it smaller than 13m it would get negative payload, this means it wouldn’t even reach its altitude. A good structure to payload value is well below 1, above 1 is allways bad. My Neptun 2nd stage is rawly 1 for the structure to payload ratio. That is moderate and has it’s origin in the effort for making it reusable without high refurbishment cost. In my model increasing the radius means of course increasing the volume and start mass. This triggers the mentioned geometrical phenomenon of size and so the system is getting better with radius.

With that model I calculated the values in the above table. I drawed a design, to visualize, how it could look like:

A two stage heavy lift launch vehicle (HLLV) based on ideas of the rocket engineer H.H. Koelle. It has a lift-off weight of 21,000t, 34 Saturn-V-F1-like-engines power the first stage. It has a payload of 900t into LEO, for example two moon Caterpillar heavy duty trucks. Both stages are reusable and land on the sea, from where they are tugged to their launch site, Fig: Author

In the lower stage we see the tanks for RP-1 and oxygen. 34 F-1 like engines with 750t thrust are mounted at a lightweight beam construction. They are of course no 1966 type F-1, they use modern technology and have a specific impulse of 300s. They burn only 2/3 of the time the F-1 burned. Below the engine we see a hatch that is open and will immediately close by springs and latest by aerodynamic forces, when the first stage returns and falls back to earth. Because of its shape it will fall at very low speed, like an actively stabilized parachute. 4 engines can open their hatches again and fire a short impulse just before the first stage touches down on the ocean. A waiting tug pulls it back to the space port. We see also the double wall of the first stage. The inner wall is of 6.4mm lightweight carbon material, just like all beam constructions. The outer wall is made of 6.4mm stainless steel to withstand the medium aerodynamic heating during falling back to earth and to be immune against sea water. After engine start, Neptun is held down until all vehicle systems are verified to be functioning normally before release for liftoff, this means there must be a heavy concrete structure for the launch pad of at least 25,000 tons.

In the upper stage we see the tanks for liqud hydrogen and the 26 SSME like engines. The engines use a higher chamber pressure, for the sake of high chamber density and low mass but they are not as extreme as the SSME was, because they are vacuum engines with long nozzles and the nozzles give them a high specific impulse of 470s. So they are rugged and can be used many, many times. As if with all engines it is not a problem, if some of them fail during flight. There’s enough reserve. So an intense maintenance of the engines will allways be done after a failure occured but not after any single flight. The upper stage also has a double wall on the basement. It’s made of carbon material, high temperature steel and a ceramic layer on the high temperature steel plates. All outer walls are of 6.4mm material. The big aerodynamic side walls of the upper stages are made of carbon reinforced material, they are only used to reduce the drag of the vessel. The upper payload compartment has a payload shroud which opens and closes automatically and a lightweight beam structure to hold the payloads up to 900 tons. On the picture we see a 200t heavy duty truck on it’s way to the moon, which is cable-tied to the payload platform. The tanks of the upper stage have enough additional volume, that they can also carry the fuel for the ferry and moon lander.

A commuting space ferry will fly between Earth and Moon and will become so huge, that it also makes no sense to design it expendable. It will be fueled in low earth orbit (LEO) by the two stage rocket and will fly to the Moon orbit. Here it will fuel the waiting lunar lander and assign the payload to the lander. Here is a design of the moon ferry:

The moon ferry just on the way from the earth to the moon. It carries a heavy duty truck and a small space station. Inside the space station are three astronauts, that are traveling „piggy back“ to the moon, Fig: Author

We see three of six giant but very light spherical aluminium tanks mounted on an inner super light weight grid structure. Two engines with a specific impulse of 470s fly the ferry to the moon. It’s a redundant solution for the case that one engine fails. Astronauts and plant staff flies to the moon using a kind of miniature space stations that was installed within the Neptun on its platform and now on the payload platform of the ferry. So they actually fly „piggy-back“ on the payload. It’s possible that there will never again be a dedicated mission only for bringing humans to the moon. Here, it’s just a side effect of the moon plants project that we need people there to observe and repair all the many automatic and robotic devices.

After the ferry reached the moon orbit the lander will rendezvous with the ferry in the moon orbit and gets its payload and the fuel that is needed for landing to and for ascension from the moon. The lander now will descent, bring the material to the surface, starts again and wait in the lunar orbit for the next ferry. The ferry will fly back to low earth orbit (LEO). There it waits empty for its next usage in its parking orbit. Now comes a design of the lunar lander:

A Moon lander, suitable to carry heavy payloads to the Lunar surface. Afterwards it flies back to Moon orbit, where it waits in standby until it is refueled by a ferry, Fig: Author

The lunar lander just descending in the drawing. It consists also of 6 spherical tanks. But they are smaller than those of the moon ferry and are installed in two lines on a super-lightweight carbon fibre beam structure. The payload hangs beneath the light framework, but don’t worry: on the moon the cargo of 200t will only weigh 33t. That plus an acceleration value plus a secure margin is the load the framework was designed for. Four long nozzle engines propell the lander. The long nozzles give them the maximum possible specific impulse of 470s, which is very important to get the transportation cost as low as possible.

The lander has 425t start weight, the ferry has 1857t start weight. They are fueled by 2 flights of the Neptun rocket. Therefor the Neptun has a maximum payload of 900t. It delivers 1345t fuel for the ferry, 192t fuel for the lunar lander and the 200t payload for the plant on the lunar surface, together 1737t distributed into two flights from earth to LEO.

Transportation Cost

Now let’s see the cost estimation chart. For the fuel we assume 2$/kg which is a typical prize for liquid hydrogen [4], the most expensive fuel of our three fuels. The other two, kerosene and liquid oxygen, are cheaper.

Neptun 1st stage Neptun 2nd stage Sum for 2 launches of Neptun [$] Moon ferry earth acceleration Moon ferry moon deceleration Lunar lander moon decent Lunar lander moon ascent Moon ferry moon acceleration Moon ferry earth deceleration Sum [$Mio]
Fuel mass [t] 10446 7697 36286 974 177 175 17 63 132
Payload mass [t] 9700 900 764 587 217 0 132 0
Dry mass stage [t] 745 1024 120 120 33 33 120 120
Fuel cost $ million 2012 20.892 15.394 72.572 1.948 0.354 0.35 0.034 0.126 0.264 75.648
Operation cost $ million 2012 5 5 20 5 5 5 5 5 5 50
Start campaign cost $ million 2012 5 5 20 20
Recovery cost $ million 2012 10 20 60 60
Refurbishment cost $ million 2012 50 50 200 200
Total sum [$Mio]: 405.648
Specific cost [$/kg]: 2028.24

The specific cost for landing payload on the moon would become $2028/kg when we use this kind of reusable approach. The cost is dominated by refurbishment, recovery, start campaign and operation cost that are together 4.4 times the fuel cost. So there is still enough margin. For example $200 million would include the total loss of one engine after two flights and additional $64 million for repairing several things. We use 34 main engines that burn 123 seconds, so we can calculate that any main engine must be designed for 68 cycles or 2.3 hours burning time with 123 second phases and cooling down between. I think that is very strict and I don’t dare to reduce the refurbishment cost further.

If we could provide this specific cost, this would mean the following for the power plants on the moon: If our power plants would cost $20 billion and $10 billion is the minimum prize for buying the materials and installing them, then $10 billion for the transportation cost would be enough budget for landing 4930 tons material on the moon. This should be enough. I estimated up to 2000t for one lunar plant in my proposal. The plants are extremely lightweight (for power plants) because they don’t need any material for their huge pressure vessels and don’t need radiators. It seems now, the raw transportation cost will be a maximum of $4 billion for one plant with the reusable approach.

The transportation budget for operation for 10 years was $2Billion. For this sum it is possible to transport 986t to the moon, so 4 to 5 flights with heavy duty and „piggy-back“ astronauts. A typical value for high power cables is 930kg/70m, so it should be possible to close 10km distances, when the Nomad Reactor [1] is marching again, with two strands and their power masts with 400t.

The reusable approach will work, the expendable approach will not. But the reusable approach will only work if we keep a strict eye on the refurbishment cost of the heat shield and most of all the operating life time of the engines. Both problems result from heat density problems and both problems made the Space Shuttle much more expensive than it was planned. Our chokepoint will be the first stage main engines. That is totally clear from the beginning. But we have a chance, because they are as huge as the F-1 were and the margins are big, and as well we have much better materials than 1965 and much more experience with turbomachinery today. But it is a challenge, of course.

Another interesting question is, what it would cost now to send material to LEO. I have calculated it in the following chart:

Neptun 1st stage Neptun 2nd stage Sum [$Mio]
Start mass [t] 20892 9621
Fuel mass [t] 10446 7697
Burnout mass [t] 10446 1924
Payload mass [t] 9700 900
Dry mass stage [t] 745 1024
Fuel cost $ million 2012 20.892 15.394 36.286
Operation cost $ million 2012 5 5 10
Start campaign cost $ million 2012 5 5 10
Recovery cost $ million 2012 10 20 30
Refurbishment cost $ million 2012 50 50 100
Total sum [$Mio]: 186.286
Specific cost [$/kg]: 206.984

So we get a specific cost of $207/kg to LEO at 900t payload, and $1553/kg at only 120t payload. This is also very good. The Saturn V had a start cost of $431Mio at 120t payload to LEO, so it was $3592/kg in 1966 dollars and would be $22.089/kg in 2012 dollars. We would be cheaper with a factor of 14, the result of the reusable approach. You see anyone in the world would try to get a piggy-back opportunity with a Neptun for his payload and would try to reach his intended orbit with an additional upper stage. These upper stages have of course to be certified that they do not endanger the payload for the moon plants. Maybe they could be relatively save re-ignitible hybrid rocket engines with storable oxydators.

But Neptun rockets would never become a threat for the sattelite transportation market, because: the Neptuns would only lift off, when they have a total sum of several hundred tons payload and most sattelite companies can’t wait for such a long time. And the payloads for the moon plants would have 100% priority.

And what would a trip to LEO cost, now? I have mentioned before, that the staff for the plants would use a kind of small piggy-back space stations to travel on top of the huge 200t payloads to the moon. We could also use this piggy-back space stations to send tourists for a certain time to LEO. After the Neptun would have delivered its payload and fuel to the moon ferry it would leave its payload bays wide open and so the people in the small space station on top of the payload platform would have a nice view. After some days – it depend on for how long they have oxygen and energy in their modules – the Neptun would close its payload bays again and return to earth. Let’s assume we would need 1000kg for each tourist as part of space station structure and supply material, then the cost would be $207,000 for a space tourist company per passenger, but not the prize for the tourists, of course. This could be in the range of $400.000 dollar per person and would be therefor 100 times cheaper than today. And I think we are touching the lower limits of chemical rockets with such a prize.

Development and Building Cost

I have summarized the building and development costs for all vehicles to develop and a certain number of them to build. I used again

  • $15.07 million per ton for building one rocket
  • $259.96 million per ton for developing one rocket

for the estimations. The following table is the result:

Neptun 1st stage Neptun 2nd stage Sum for 5 Neptun Moon ferry Sum for 7 ferries Lunar lander Sum for 7 landers Sum
Building cost $ billion 2012 11.22715 15.43168 133.29415 1.8084 12.6588 0.49731 3.48117 149.43412
Development cost $ billion 2012 193.6702 266.19904 459.86924 31.1952 31.1952 8.57868 8.57868 499.64312
Total sum: 649.07724

It is a simple cost estimation by a linear factor, of course. We see the most expensive part for the development and for building will become the Neptun 2nd stage. But I believe it will not get such expensive, because of two arguments:

  1. we calculate with typical cost assumptions of extreme sensitive high tech products. But I have chosen many parts and materials of the Neptun design that are actually not such complicated. They are rugged and tolerate much, and they have high manufacturing tolerances, too.
  2. we calculate with typical hand manufactured unique items. But actually we have to develop only 2 engines and build dozens of them. The same is with the tanks and casing pieces and internal structure pieces.

This is why I think a better cost estimation would be the following for the Neptun:

Neptun 1st stage
mass for one building cost development cost [$Mio] number of items mass for all parts building cost for all parts [$Mio]
Engines 9 135.63 2339.64 34 306 4611.42
Kerosene tanks 1.4 21.098 363.944 19 26.6 400.862
Inner carbon structure elements 19 286.33 4939.24 6 114 1717.98
Outer steel structure elements 50 226.05 3899.4 6 300 1356.3
Control sum: 746.6
Sum: 11542.224 8086.562
+ 1 prototype 8086.562
Total sum development: 19628.786
Neptun 2nd stage
mass for one building cost development cost [$Mio] number of items mass for all parts building cost for all parts [$Mio]
Engines 3.5 52.745 909.86 26 91 1371.37
Hydrogen tanks 32.6 491.282 8474.696 7 228.2 3438.974
Inner carbon structure elements 58.6 883.102 15233.656 6 351.6 5298.612
Outer steel structure elements 58.6 883.102 15233.656 6 351.6 5298.612
Control sum: 1022.4
Sum: 39851.868 15407.568
+ 1 prototype 15407.568
Total sum development: 55259.436

In this table I have used only one time point 1. of my criticism of the simple cost estimation. The outer steel structure elements of the 1st stage are made of stainless steel and it is a well proven industry technology, not much to develop. So I have assumed for these elements only a third of the $15.07 million per ton for building one rocket and of $259.96 million per ton for developing one rocket. But for all the rest in the table I used this values.

The highest reduction of the development cost results from the high modularity of Neptun, point 2. of my criticism of the simple cost estimation. H.H. Koelle was wise when he designed Neptun. He knew exactly that the huge masses of HLLVs would let explode their cost, so he designed the Neptun as a modular concept. Here we have in each stage a certain numer of identical engines, that we have to develop only once, also the tanks. But the most interesting characteristic of the Neptun design is, that it is build not axially symmetric but angled. It has at least 6 identical sections that you have to develop only once and assemble them together. This makes the development much cheaper. It’s becoming so cheap that I have added an additional prototype to the development results because I couldn’t trust it otherwise. After all I think: Neptun is the perfect HLLV. No wonder, Koelle was the primary designer of the Saturn.

Please notice the very important control sums within the accuracy that we don’t forget to build and develop parts of the rocket. Our total cost estimation will now look much more realistic:

Neptun 1st stage Neptun 2nd stage Sum for 5 Neptun Moon ferry Sum for 7 ferries Lunar lander Sum for 7 landers Sum
Building cost $ billion 2012 8.865 15.431 121.48 1.8084 12.6588 0.49731 3.48117 137.61997
Development cost $ billion 2012 19.628 55.259 74.887 31.1952 31.1952 8.57868 8.57868 114.66088
Total sum: 252.28085

But the development costs are still very high. So it should be clear, that the Neptun HLLV’s, the ferries and the lunar landers are very expensive and only NASA could master such a project. The public authorities must pay for the development of the infrastructure. But it is worth it! After a few years the extensive tax payments of the new U.S. power plant companies, which would sell their cheap electric power to all corners of the world, would bring back the multiple of the public investments.

It will not be easy to build such big reusable, reliable vessels, for constructing fusion power plants on the Moon, but it is possible. Without any doubt.

Launching the Relais Sattelites to GEO

The geostationary orbit (GEO) velocity increments are 2.43km/s for the first manouvre of the Hohman transfer trajectory from 300km circular orbit to the transfer elliptic orbit, and 1.47km/s for the second manouvre to inject in 36.000km circular orbit. It’s a sum of 3.9km/s and with an Isp of 470s it is a mass fraction R of 2.3, more than escaping from the earth but less than escaping from earth plus injection into moon orbit. So the moon ferry should have at least a transport capability of 392t (the mass of the lunar lander fuel and the payload to the moon) to the geostationary orbit and can then fly back again to its parking orbit. This is needed for building the relais satellites, that are of a size in the several hundred meters range and of a mass of several hundred tons.

I published my text for the proposal of power plants on the moon in a public forum. A reader, obviously a laymen, mentioned: “Lunar transmitters would not work because you forgot the moon has its own orbit around the earth. It does not sit over a fixed point of the planet. So how many relay stations would you need? [..]”

This is an interesting question and I was thinking of it before I even started thinking about the rest. The moon orbit and the ecliptic have an inclination, that means they are not aligned. Lunar eclipses occur if there is a full moon within 11° 38′ of a ascending or descending node, and solar eclipses occur if there is a new moon within 17° 25′ of a ascending or descending node. The moon has a revolution time of 27.3 days. This means by first view, that there are 4*18°=72° of 360° or  0.2 * 27.3 days = 5.5 days each month when Earth can eclipse the relais sattelites. The sattelites themselve do a full revolution in 24 hours. It’s again a 72°/360°=0.2 fraction of the orbiting time. So 0.2 * 24 hours = 4.8 hours 5.5 days a month the relais satellites might be eclipsed. Statistically is this 50% at night.

But it remains 4.8 hours at daytime in more than two days a month. That is too much and too long, I think.

So there should be a second relais satellite exist from the beginning. Later, if we have more relais satellites there is only one reserve system for a certain number of relais neccessary. Because of the large reusable Moon ferry, we need anyway, I don’t think it will become a crucial argument.

Summary

When thinking about nuclear power plants on the moon to solve the energy problem, we quckly get to the problem of transportation cost. Simple mass and cost calculations for expendable launchers show, that this approach has no chance. At 200t payloads to the moon, the launchers become so huge that the cost for building them is extremely high. So we are forced to another way: the reusable approach. The Space Shuttle experiences showed us, that we have to face very high refurbishment cost. They result of the big technical problems with reentry technology. But as we get bigger with our designs the problems vanish: It can be shown, that the reentry is relatively harmless from a heat density point of view. The lightweight structures become lighter with size. The margins and the tolerances grow with size. If we design the rocket rugged and conservative, the refurbishment cost will be moderate, and the complete reusable launcher becomes reality. But it is much bigger than most people had thought. Additionally a complete reusable moon ferry and as well a lunar lander must be provided to complete the cheap lunar tranportation system. With all that we get tranportation cost of $2028/kg to the moon surface with 200t payloads and $207/kg to LEO with 900t payloads. The development and building cost of this ultimate chemical rocket tranportation system to the moon is $252.3 billion dollars. In simple terms: Two Neptun HLLV would launch just after the other, would refuel the waiting moon ferry in orbit and assign the payload of 200t. Then the moon ferry would fly it to the moon orbit. Here the ferry would refuel the waiting lunar lander and assign the payload. The lander would then descend and deliver the 200t payload at the plant building site.

Resources and Explanatory Notes

[1] The Nomad Fusion Reactor https://monstermaschine.wordpress.com/2012/07/23/a-revised-version-of-the-fusion-steam-machine/#more-2228

[2] Encyclopedia Astronautica on the cost of the Saturn V rocket http://www.astronautix.com/lvs/saturnv.htm

[3] Heinz Hermann Koelle, inventor of the Neptun design: http://de.wikipedia.org/wiki/Heinz_Hermann_Koelle

[4] Prize of liquid hydrogen http://www1.eere.energy.gov/vehiclesandfuels/facts/favorites/fcvt_fotw205.html

Über monstermaschine

Blogger, Diplom-Ingenieur, TU, Raumfahrttechnik, Embedded Systems, Mitglied VDI, DGLR

Eine Antwort zu “Transporting Heavy Duty to the Moon

  1. Patricia

    How nice that my father H.H. Koelle is not forgotten and people continue to work on the subject.

Schreibe einen Kommentar

Trage deine Daten unten ein oder klicke ein Icon um dich einzuloggen:

WordPress.com-Logo

Du kommentierst mit Deinem WordPress.com-Konto. Abmelden / Ändern )

Twitter-Bild

Du kommentierst mit Deinem Twitter-Konto. Abmelden / Ändern )

Facebook-Foto

Du kommentierst mit Deinem Facebook-Konto. Abmelden / Ändern )

Google+ Foto

Du kommentierst mit Deinem Google+-Konto. Abmelden / Ändern )

Verbinde mit %s