08/97 - preliminary

Cost of Space Access - FARs Opinion

Even the advocates of space travel admit it - and its opponents have said it all the time: space exploration is too costly. In that discussion it helps very little to compare space budgets with our societies' expenditures on opera and musical tickets, the consumption of potatoe chips or pulling Teflon pans out of the hat. All our dreams of conquering the cosmos will forever remain just that: dreams - if we don't succeed in bringing the cost of space travel down substantially.

Cost of space exploration is closely related to the cost of space access, that is access to Low Earth Orbit = LEO. This cost is usually expressed in $ per kg payload. It is somewhat difficult to obtain exact data, whereby one reason may be that officials and protagonists like to obscure the cost to make it look less expensive. Along that line, up-front development cost, cost of infrastructure etc. is commonly omitted from many cost calculations and is often written off as "general expenditure to advance the sciences". It is thus difficult to pin down cost. Nevertheless, a figure of appx. $ 18,000.- / kg payload is commonly accepted as the Space Shuttle's price ticket (NASA estimates the cost of a single launch totay at $ 550 million. Prices in 1993 US-$'s). The Shuttle, by the way, which was originally constructed to bring the cost down, is today the most expensive vehicle in the US arsenal. Thus, in a sense, the Shuttle is somewhat like a high school assignment - brilliant, but the subject was missed.

To put these numbers into context: $ 18,000.- / kg is more than we pay for a kg of gold. In other words - if space was full of gold and we had to transport it with the Shuttle we might as well leave it where it is. An average 85 kg astronaut with 15 kg of "hand luggage" would have to pay $ 1,800,000.- for a ticket to space. Obviously, at these prices, space tourism is for Bill Gates and friends only. Clearly, we need to cut the cost at least 90 to 95 %.

But how low could transport cost actually be ? What price can be expected on the basis of existing technology ? And what price would we accept as being "reasonable".

Papers on this topic are often somewhat nebulous and general at best (whereby we admit that it is difficult to say anything about cost if you don't have a tangible vehicle design on which to base your estimate). We would like to take an approach whereby we compare a space launch with aviation: what if we could operate rockets like air planes ? Let's elaborate on this approach.

Cost of air travel

We all are air travellers. Almost all of us have been flying as commercial passengers. Insofar we would all agree that the cost of air travel is generally acceptable. Obviously none of us will expect a trip to space to become as normal as a flight from LA to Hawaii is today any time soon. Rather, trips to space will remain special for some time to come. Thus, if we want to compare air and space travel, we have to look at air trips which are also special in their own respect. Such a trip would be a flight on the Concorde. Not exactly cheap, but still there are enough people who can afford it. A trip on the Concorde is special even though it is an every day event.

The following data is available on the Concorde: Max take-off weight 186.000 kg fuel weight 96.000 kg passengers 100 For a maximum distance trip (Paris - New York plus apprx. 10 % fuel reserve as safety margin) we thus have to calculate 1000 kg of kerosine per passanger (when the airplane is fully booked). Under an energy-concious point of view this is not exactly an efficient method of flying, but it is commonly accepted. (it might be noted that Paris - NY is not the longest distance commonly flown - rather, trips to Australia are typically twice the distance. They are just not flown by the Concorde because many countries are not granting that aircraft landing rights due to environmental considerations. If the Concorde did fly to these destinations, fuel consumption would of course be double the above figure).

But wait:

If you fly to New York, you would usually come back the same way, right? Thus, we are talking about 2000 kg of fuel per passanger per round trip. This is by no means "splitting hairs", because in a fair comparison with a trip into space we must note that the return trip from LEO is always free (from an energy standpoint) - an important difference, as we will see.

2000 kg of kerosine per 100 kg of passanger are thus accepted. But in order to burn that kerosine we need another 7000 kg of oxygene (in a 3.5 to 1 mixture ratio). The airplane takes its oxygen simply "out of the air" as it goes. Air oxygen is (today) free for any country's aircraft to take.

Ultimately, the rocket also takes its oxygen out of the air, whereby the difference is obviously that the rocket has to take its oxygen along in a tank in liquified form. As LOX it is still "free" except for the cost of liquification. In a fair comparison of planes and rockets we cannot simply forget about the oxygen. Rather, it is part of the energy and fuel balance.

To finish this off, the Concorde burns 2000 kg of kerosine plus 7000 kg of oxygene = 9000 kg of fuel in total per 100 kg of passengers. Assuming that passengers and their luggage are the Concorde's only payload (it is not a freight plane), we can say that the fuel to payload ratio of the Concorde is 90 to 1 (kg of fuel used per kg of payload).

Are Rockets Inefficient ?

Let's now look at the rocket. Almost certainly, for the next 30 to 50 years to come, rockets will be our only way of getting people and large amounts of payload into space. At least it doesn't make sense to discuss other means as long as there are no prooven or at least theoretically viable concepts on the table. For our comparison, we will look at a rocket which is based on existing and well-established technology and which will stand a comparison with air planes. Such a rocket should burn kerosine and liquid oxygen (just as most Russian launchers do). No exotic fuels, technological break-through's, no liquid hydrogen, no particular infrastructure (kerosine and LOX are available in any major city worldwide).

A mathematical and physical analysis of rocket flight (link under preparation) shows that such a rocket cannot be a single stage vehicle. Two stages are however realistic. We will further assume that both stages can be re-used, ignoring for the moment how. The efficiency of a rocket is usually expressed via the mass ratio, i.e. the weight of the fully loaded and fueled rocket divided by its dry weight. It is possible today (with a reasonable amount of optimizing) to build such rockets to a mass ratio of 30:1. (Sample calculation page under construction). This ratio will go down with the number of stages (i.e. more stages mean better fuel economy) and it will go up with less technical sophistication, for ex. 50:1 or worse for larger pressure-fed vehicles.

This means that we have to spend 30 kg of fuel (kerosine AND oxygen) for the transport of 1 kg of payload to LEO. This is surprisingly much better (3 times better !) than the Concorde's consumption which we have calculated above to 90 kg. How can that be ?

Obviously, besides its payload, the Concorde has to carry its own weight, too. I call this the payload carrying structure, as it does not count as payload. Similarly, if we were to ride a rocket, we also cannot simply sit at the top, more or less "topless". Rather, we will need some kind of surrounding structure, seats etc. there, too. The Concorde's structural mass is 100.000 kg (i.e. the plane's dry weight) for 100 passangers á 100 kg. I.e. the payload to structure ratio is 1:10 Now, an airplane has to carry a lot of weight which the rocket must not, for ex. wings (which are of no use at all during ascent and while in space !) or a landing gear. Rather, an Apollo-type capsule will suffice for the purpose. Also, many features which today's air traveller expects such as comfortable stretch-out seats, on-board moovie, a pantry, toilets etc., even stewardesses serving food and drinks, can be omitted from our rocket design - trips into space typically take an hour or so until you dock at a space station. Also, please note that the vehicle's second stage, which is also space-going, is not part of our payload calculation.

Thus, we can assume a payload to structure ratio of significantly less than 10. Let's assume a ratio of 1 to 4 (which may sound optimistic, but is actually not. It is much worse than a typical passanger car's (not weight-optimized) ! It is also worse than the Space Shuttle's (1 to 3) or the Space Shuttle's if it were a wing-less craft (1 to 1)). So I think this figure is "save". It means that we have to launch a total of 5 kg per 1 kg of payload with an overall mass ratio of 30:1

Sounds complicated ? Well, it isn't. 1 kg of "true" payload takes 1 x 5 x 30 kg of fuel = 150 kg of fuel. This is now not as good as the Concorde's fuel balance, however it is still much better than most would have thought when looking at those gigantic rockets ! Considering what flight in a rocket really is, such a fuel expenditure seems very acceptable !

Conclusion: rockets are not that fuel-inefficient. For what they are, they are not too big.

We now have a realistic measure for fuel expenditure. At today's cost for kerosine and LOX (apprx. 15 c per kg for large quantities, without taxes) this comes out to apprx. $ 25,- per kg. A 100 kg passenger would thus have to pay $ 2500,- for the fuel he spends to go into LEO. Obviously, fuel is not the end of a cost analysis. Rather, we must add to it cost for infrastructure, cost of capital, equipment cost, wages, promotion etc. as well as calculatory earnings. To stick with our aviation comparison where fuel cost is between 10 % and 25 % of total cost, a ticket would be between $ 10,000.- and $ 25,000.- Interestingly, this is still near the price of a concorde ticket. Less than what many spend on a boat cruise.

Conclusion

The above is a screening calculation only. Perhaps the reader would like to "play" a bit with the numbers himself. One may calculate with a 50:1 mass ratio, even 100:1 (now this would be really a clumsy rocket!) or change some of the other parameters. Most likely you will find for yourself that the size of a rocket (i.e. tank space and fuel) does not make it's cost! An unusual finding for those who have ever stood in front of a rocket and concluded that big must be expensive. But a big fuel tank is not necessarily expensive. Look at an oil tanker. Much bigger than the Saturn V, but not more expensive. Big is not expensive as long as you KISS - keep it simple stupid!

Now, nobody will seriously expect tickets to LEO for US-$ 10,000.- any time soon. Also, the above calcuations are based on a number of assumptions such as an overall mass ratio etc, which must remain vague as long as no vehicle design has been specified. The same is true for other assumptions such as the payload to structure ratio. Yet, I feel that we have some sound estimates here. Nevertheless, there is one element in the above calculations which is certainly far-off. It is in the final step where we conclude from fuel prices to the final ticket cost. Though a quote of 10 to 25 % (depending on oil prices, exchange rates, global production etc.) is pretty correct for aviation, there is no way we can simply apply it to space travel. Flying is a mass market and even the most courageous among us don't see a similar development in space travel any time soon, that is, certainly not for the next 20 to 30 years. On the other hand, we set out to examine "what could be" if rockets could be operated like air craft. I do not mean that space craft or rockets should look like air craft. This is a different discussion as you will see further down.

At least the above calculation shows the following:

1) from an energy standpoint, rockets ase not as inefficient as most people would think

Rather, mere fuel consumption is within dimensions which we all accept for aviation. Clearly, the popular assumption that rockets are so expensive because they are so big (big = a lot of fuel) is wrong! One reason they appear bigger than aircraft is because they have to carry their oxygene. If aircraft had to, they would truely become Jumbos !

2) For the economic access of LEO we don't need any new, yet-to-develope or exotic technologies. Technology does not even have to be taken to its limits.

3) The above reduction in ticket prices (2 M$ to 20K$) is 99 % !

Even if the end result was one order of magnitude "worse", we would still get a reduction by 90 to 95 % as compared to current prices, making a trip to space affordable (though still not cheap) for many.

A 90 % launch cost reduction (for the US, spending apprx. 5 Billion $ on Shuttle launches) would translate into a 4.5 Billion $ savings, now free to house the poor, save the wales or whatever is dear to your heart. There would not be less space travel.Or the US might choose to put those 4.5 Billion $ into building payloads such as moon bases, Mars hardware etc. And remember - that money is saved every year. Thus, even if the construction of a cheap booster costed 4,5 Billion $ up-front, it would ammortize within just one year. What a perspective !

4) We can immagine transport scenarios where the structure to payload ratio is actually much better than assumed above. This is the case for all payloads where you don't need a surrounding structure. Interestingly, today, this is the bulk of payloads such as satellites, space probes, space labs etc where the payload fairing is discarted early in the flight. In this case, the overall mass ratio and thus cost results are actually even better than those calculated.

Payload is not Payload

Orwell knew it: not all animals are equal. The USA today transports practically all payload with the Space Shuttle. Occasionally, some planetary probes, satellites and military hardware go on the Delta rocket and a few others. To put all loads on the Shuttle was the declared goal of US space policy, in an assumption that only then could the Shuttle "fly" it's development cost back in. As a sequel of that policy, the US has shortsightedly abandoned other already existing launch alternatives such as the Saturn V (150 tons into LEO) - and how well could we use a heavy launcher today! Of that great rocket, not even a full set of blueprints exists any more.

Apart from the fact that the Shuttle, due to its poorer-than-expected turn-around times and other reasons couldn't even carry all payloads, it doesn't make sense to have just one method of transport for all types of payloads. The Shuttle is "man rated", which contributes heavily to its complexity and cost. In reality, most payloads don't need this sophistication. Back on earth, we have a number of transport choices and we usually don't haul "a sack of coal" the same way we transport humans. Rather, we use a mix of gear. Coal goes on the slow boat, so do soy beans. Computers and pig halfs are shipped by truck, people fly or drive.

Suppose we want to conquer space (whatever that means) we can expect largely 3 types of payloads:

1) humans

2) gear and equipment such as computers, space suits, lunar rovers, communication equipment, life support, solar cells, experiments etc.

3) big structural components such as aluminum cylinders serving as space station modules or habitats. Most of all, we will have to launch fuel (hydrogen/oxygen/or whatever = the "coal" of the space age). Most of it (by weight) would be LOX. We would also need large quantities of water (fuel and Oxygen when electrolyzed), food stuff, beer (to raise crew spirits) etc.

Note that in that sequence 1) to 3) weight increases while the cost of goods to be transported decreases significantly.

Without doubt, humans are our most precious "goods" to be carried. Only maximum safety will do.

Have you ever thought about the direction of transport in space flight? Silly question ? Well, think about it - it's only the humans (and perhaps a few diskettes, film or soil samples) we want to return to earth. The "rest", in the above list 2) and 3) can and should even stay up there so we don't have to launch it again and again. Interestingly, this "rest" makes up for at least 99 % of the weight we need to moove. Expedition gear (group 2) will certainly be expensive, but only because most of it are originals, developed and built just once. One moon rover is expensive, but 20 moon rovers would not be 20 times one ...

Group 3 holds only items downright worthless when compared with their launch cost. An interesting finding, isn't it? Take water, for ex. Water costs no more than a few bucks per ton here on earth. When launched with the Shuttle, all of a sudden, this ton of water costs $ 18.000.000 That figure would change drastically with a cheap launcher, but even at $ 100,- per kg, space-born water would be incredibly expensive compared to its "true" market value.

The same is true for LOX (apprx. $ 250,- per ton), for fuel, for structural elements or for all foods, unless it's caviar, truffles or a 57 Dom Perignon Champagne. A strong case for only the most exquisite wines and finest foods as rations, by the way.

Summary: most payloads are totally worthless when compared with their launch cost.

Group 3) makes for at least 80 % of any major Moon or Mars class mission (including Zubrin mission types). Fuel and LOX alone is typically 60 to 70 % for any mission to escape earth's gravity. Thus, for the major part, we have to moove worthless mass. Obviously, it doesn't make much sense then to moove that with the same level of caution and sophistication as humans. Hence the Shuttle seems not right for 80% of all payload. Neither is the Ariane V.

Rather, we need a real cheap transporter, a rugged space truck, if you like, capable of throwing stuff up there like there is no tomorrow. The cheaper, the better. If an occasional transport is lost - so what !

We can carry this line of thinking further: since it is only the humans we want back, who fly rarely, every space launch concept preoccupied with bringing things back obviously makes only limited sense. This includs all concepts with wings and a landing gear, whether its called Space Shuttle, Sänger, Hermes or Hotol. NASA's new X-33 also falls into that category. Wings and landing gear simply make no sense in space. They are just ballast at the cost of true payload. Period. Someone has called the directive that space craft must be able to land like air craft "as sensible as expecting air planes to be able to land at train stations". Wings, flaps, hydraulics and the like only contribute unnecessarily to the complexity of a space craft. It seems to me that NASA is heading directly into the next billion dollar flop. For the most part of the space program humans have returned to earth in capsules (once again: return from space is a transport exception). This method of return is well prooven. It uses existing technology (parachutes), needs no pilot intervention and resorts to existing infrastructure (the ocean).

Most of the above seems very logical. Yet, most institutional concepts look different.

What do you think? Send us an e-mail - we'll publish it.

Cost of Space Access - second round: flaws in thinking

You may wonder why institutions and corporations engagegd in space activities haven't actually come to the same conclusions. The answer may surprise you - they have !

There have been numerous studies, done by the US space industry and the DoD (Department of Defence) as early as the 60ies. They have all come to the same conclusion: space access will become less expensive when we apply less high tech in the design of launch vehicles.

Any discussion must start with an analysis of the cost-driving factors. As we have seen above, cost does not correlate with the size of the rocket. Rather, it is

• the number of parts
• a strong drive to optimize and use the highest tech
• an aversion against standard off-the-shelve components, everything is "done new", there is little component evolution
• a linear application of aviation design practices to space launcher design
• an almost ideological compulsion to build reusable craft
• requirements for documentation and project management practices
• political interference
• the use of hydrogen as a fuel

Complexity and number of parts

Cost driver number one is the complexity with which today's rockets are being built. The Shuttle's main engine, for ex. consists of 70.000 parts. Roughly half of those are in the turbo pumps. The Shuttle has 3 such engines, ammounting to 210.000 parts. If it had just one (more powerful scaled-up) engine, obviously, part count would be reduced by 66 %. Now, one engine means that in the case of failur the mission would have to be aborted. On the other hand, with a 66 % reduced part count, an engine faliur is 3 times less likely. In a manned mission, benefits and risks must be weight carefully. When launching worthless payloads (see above), they must not.

It seems logical that "many parts" cost "more money" and less parts less money. This relation holds for most technical items. Even for car engines, where the manufacturing cost of each item makes it's price, not it's metal price. Car engines are also complex and difficult to design, but this overhead cost is regained through mass production - a benefit unavailable in rocket design unless we were to mass-produce a standardized expendable launcher. The cost / part count relation, however, is not linear. Rather, it seems to be exponential. This is because many parts mean many sources for errors, necessitate more testing and controlling, more documentation, and create more interfaces between components in a technical as well as a human sense. Many interfaces translate into many contacts between designers, engineers, testers, operators, the man in the shop etc. This means trips, conferences, meetings, discussions, documents, signatures and the like. According to a DoD study this type of activity which, after all, produces nothing tangible, makes for up to 2/3rd of system cost.

High Tech at any cost

Most space engineers have either an aviation or a military background. This is true, for ex. for all engineers of "the first hour" (i.e. von Braun etc.). Nobody had to keep an eye on money. With that background in mind it is not surprising that many space design practices actually come out of aviation. In a sense, these people see space travel as "the continuation of aviation by other means" (pardon me, Clausewitz).

This attitude is also reflected in the language: we talk about the aerospace industry, the space plane, the Aeronautics and Space Administration etc. It is best expressed in the Space Shuttle which is - visible for everybody - half air craft. It deemd the engineers unacceptable for a second generation space vehicle to just let astronauts in a capsule drop into the ocean. It had to be more elegant. Landing on a runway. This design decision alone consequented wings, landing gear, complicated avionics, pilot training etc. If the Shuttle were unwinged, it could way 50 % (= 50 tons !) less or carry 3 times (75 instead of 25 tons) more payload. All this potential was wasted for the trade-off of an elegant landing. And not only that - during launch the wings are just in the way and in space they are obviously totally useless. They make for the most area of the troublesome heat protection tiles ... Once again, the Shuttle is an incredible flying machine and an engineering marvel, but it is not a good space craft.Once again, European engineers are hardly better - they still batteled for Hermes after the Shuttle had already failed.

Actually air and space travel have very little in common.

A rocket launch is an acceleration process up to a predetermined final velocity. From thereon it is all coast with minor corrections. The underlying idea is to pass the atmosphere as quickly as possible for an unhindered accelleration. Aircraft, on the other hand can only fly because of the atmosphere. The atmosphere is also needed for direction changes. Hence, aircraft must have wings, they must take rapidly changing loads, they must posess light yet strong structures etc., and they must work over many years of almost constant operation. Aircraft must be able to land anywhere - this is the very reason for their existence. Aircraft must be light. Weight savings translate directly into operational profits. Each ton trimmed off the air plane's mass is another ton of payload. For a rocket, each ton of first stage weight trimmed off will give us just a fractional gain in payload.

For air planes, optimizing makes further sense due to their high flight frequencies. A little example: we all know those small wing tiplets, which look like little wings themselfs, designed to smooth out turbulences. Those have entered aviation about 10 years ago when fuel saving pressure was great. They save up to 2% fuel. Such an innovation makes sense. Let's look at the 747 as one type of aircraft. It flies since 30 years. There is an operational fleet of some 1000 units of this type alone, each burning some 200 tons of kerosine daily. Optimizing 747 design to save 2% fuel thus means 2 % x 1000 aircraft x 365 days x 30 years of operation. Not even the most optimistic space buffs would expect similar flight frequencies any time soon. A comparison: the Saturn V flew only some 15 times, the Space Shuttle 100 times in 15 years, the (economically very successful) Ariane IV 100 times in 10 years. What sense does weight optimizing make to save fuel at such flight frequencies ? Quite to the opposite. Saving weight means exotic materials (Al-Lithium coming now), complicated processing, high testing efforts, low fault tolerance etc. Nobody would seriously consider building a Jumbo Jet from sheet metal to save money, but on the same token, nobody would consider building an oil tanker from carbon fiber composites just so it is able to carry a few tons more oil.

Air craft and rocket engines also have nothing in common. Aircraft engines are highly complicated turbines which have to withstand heavy loads (extreme rotation, heat, temperature changes, ratteling and bangs etc.) over extended time periods (years). I don't see how one could reduce the part count here drastically. Rocket engines, by contrast, are principally very simple. Remember the Chineese built rocket engines a millenia ago. Could they have built a Diesel engine, provided they had known its principle of operation ? Rocket engines have been characterized as "essentially single cylinder combustion engines without a piston" ... and a high speed gas coming out the hole. Thus, rockets seem to have more in common with flying boilers. The whole setup must hold for a few minutes, for a first stage typical burn times are 2 to 3 minutes.

What we are saying in essence, admittedly quite provocatively, is that aircraft engineers, due to their line of thinking, probably cannot make good rocket engineers, much as they would also not be good ship builders. Building large rockets, however, seems more like a job for our ship yards.

Optimized launch vehicles also affect payload cost.

If launch cost were cheaper then we wouldn't need to pack so much into every single payload. This cost aspect has also been subject to analysis. Experts have calculated that a 50 % launch cost reduction would effectively also lower payload cost by 50%. In other words: we could do the same "amount" of space exploration on half the budget - or do twice as much with today's money. Or we could build more of the same payload. If NASA looses 1 out of 1 Sojourner, it looses face. If it looses 1 out ot 20, then people would say: "gee, that's pretty good, eh ?".

All new ....

Space engineers like to do things new rather than give evolution a chance. Again, the Space Shuttle makes a good example. Here, in order to save money, someone had the insipration to develop those big strap-on boosters. NASA would have had an alternative in using the F1 engines of the Saturn's first stage. Those had functioned flawlessly in about 15 launches, the development cost was paid for, there was an existing production machinery, trained maintainance teams etc. The engine used cheap kerosine and LOX (an estimated 40:1 cost savings over solid fuel, not counting production and transport hazard overhead cost). Two F1 engines deliver the same thrust as one Shuttle solid booster. In addition, the F1 uses turbo pumps and could have really saved weight in the range of a few hundred !!! tons over the solid boosters, if that was desired. Not speaking of the extra safety that an engine gives which can be turned off. Yes, the F1 was optimized and not cheap and designed for one time use. But it could certainly have been adapted for reuse. It could have been very easily turned into a pressure-feed engine, if desired, saving a lot of money while maintaining a cut-off safety option. Or it could have been further optimized, mass-produced etc. Many choices based on something that actually existed, that was paid for and that had always worked well. Not speaking of safety aspects which ultimately costed peoples lives or of environmental hazzards with view to some 100 tons of hydrochloric acid HCl which the Space Shuttle sets free in each launch.

A Saturn V F1 eingine

Enginers like to do things new rather than look around what's available. This also applies to the Ariane V, having a payload capacity of some 20 tons to LEO, a range which has been "cracked" by Americans and Russians decades ago. Development cost so far some 6.6 Billion US-$. Of course Europe needed a state-of-the-art hydrogen/LOX engine (1.1 Billion US-$ so far), of course it had to have strap-on boosters. My personal estimate is that with the consequent application of MCD (Minimal Cost Design) criteria the whole system could have been done for no more than 2 Billion US-$. This means that we would have had another 5 Billion (almost) for the construction and launch of hardware, a little space station perhaps, a moon rover, some planetary probes. Even if the actual operative launch cost were not lower than anticipated now (20 tons for around 50.000.000 US-$), then we could have launched 2000 tons of payload). Or we could have devoted some of that money to fusion studies. Or to cancer research. Now we won't have any of this. And what's even worse - we must fear that the Russians with their kerosine/LOX technology will snach our market share (and they will once they get their act together).

Reusability

Reusability is seen by many as the key to cheap space access. This seems logical, however, it only makes sense under certain conditions, namely high flight frequencies. Again, we will look at the Space Shuttle. Here, the two strap-on boosters are reused. NASA has two vessles specially built for the purpose of recovering the spent boosters and towing them to shore. The crews must be paid at all times, wether the Shuttle flies or not. There is a little harbor and special buildings for disassembly and cleaning, a special building for washing the parachutes as well as special transporters for the shipment to Utah (were the boosters are refilled). Each step requires personell standing by. With flights averaging once a month this is not very efficient.

In fact there are many examples in "the real" economic world, where only bucks count and where reusability would be feasable but is not practiced. A good example we all know is McDonalds. They rather throw styrofome boxes away than put their burgers on washable plates. I am certain McDonalds has calculated their cost down to a fraction of a cent and you bet they would do dishes like hell if it made them money. But they have obviously come to the conclusion that for their type of business "ex and hopp" is cheaper.

There are also examples where goods are transported in containers much more expensive than their "payload". Again, an example we all know from everyday life is a glass of pickels. Here, the filling costs apprx. 5 cents (so I was told), whereas the glass and lid cost 10. Hence, the "transport system" is 100 % more expensive than the "payload". And we accept it and have never even thought about it. It all makes sense, because reusing the "transport system" would be more expensive than "building" another one. There are many more examples. Many could be taken from the military where even expensive flying machines (missiles) are thrown away rather than reused. I am not advocating a throw-away mentality. Rather, what I want to say that in economic terms what counts is what's under the bottom line: profit. We have to do a lot of calculating ....

Nevertheless, we would look for reusability whenever possible. A launch stage usually could be recovered and refurbished. It is interesting to note that this seens ti be the easier the heavier and more rugged this stage is. After all, it has to survive reentry, a fall from great height and a splash into the water. It seems to me that a steel structure is more apt to survive than an Al-Lithium structure so thin that you can indent it with your finger.

The tip of the rocket can also be reused. This would be the payload bay, a transport capsule etc. It should hold the "brain" of the system, that is, computers, guidance and control, communications, power supply, life support when manned etc., everything that is expensive. This system part should be a capsule landing in water via parachute. A conventional ablative heat shield should be replaced in one piece after each or every other flight.

This leaves us with stage two (we will need at least two stages unless we use a hydrogen SSTO). If that stage is to be reused, it must enter the atmosphere, it will need a heat shield, some landing system (parachute) etc. The velocities are much higher than for a first stage. This upper stage could also be integrated into the capsule to form a combined landing vehicle. It should water rather than land on legs (what is the benefit of that, anyways ?) to keep things simple.

Or we try a new approach in thinking and substitute the term reusability by full usability.

Now what is that supposed to mean? Well, the upper stage is also weight which is transported into orbit. Even given a substantially lower-than-today cost of, say, $ 500,- per kg, it would still make no sense at all to bring anything back that has already made it up there and could be used for something else. Spent stages, if they are of a certain size and standardized, would make good habitats, building blocks for space stations, mars transporters etc. Spent stages would be mostly empty LOX tanks. Those could already have a 19" frame work in them for accepting the standard 19" scientific racks. In this manner, the upper stage becomes payload, much to the improvement of the overall mass ratio of the rocket.

Political and lobby influences - a touchy subject

Those are very bad cost drivers. It is obvious that a company, having won a space contract, does not have the slightes interest in building a piece of equipment for half the negotiated cost. To the contrary: the more expensive - the better. That is, once you have the job. Now there are not all that many companies, in my country certainly, that are at all capable of building large rocket components, at least not when you are talking about aviation high tech precision and not about ship yard tolerances. Insofar there is probably less competition than most people think. An ideal setup for price drivers and cost overruns.

Now you would think that politicians as the taxpayer's advocats would having an interest in keeping cost down. Actually, this one has already been answered by Kennedy in 1961: "we choose to go to the moon ... not because it is easy but because it is hard ...". He might as well have said: we go because we want it to be expensive rather than cheap (perhaps to put national industry into the advantage - or simply buy the Russians out).

Political control mechanisms fail for a number of reasons (in no way restricted to aerospace projects). First, most politicians don't have the slightest mathematical and technical background. I remember having seen a report on SDI, where a scientist was discussing laser power advances with some Reagan people. The scientist said that they already had the energy output up to a level of 10 to the power of 10, but in order to do what they wanted they would have to get up to 10 to a power of 20. And the Reagan people rejoyced "fine, so you already got half ...". That says it all.

Further along, politicians like expensive programs, too. Only big expensive projects mean "fame and honor", they bring money and work to your state, they consequent campaign contributions etc.

High cost also means increased bureaucracy. You need steering commities, supervisors, personal referents, conferences, trips and meetings etc. Bureaucracy grows like cancer, if you let it. It never limits itself. And it tends to choke the really productive. All this is adversly affecting getting the job done.

But federal structures as such also cause cost. Again we will look at the Space Shuttle, but we might as well look at the Ariane or Airbus consortium (for worse examples). Federal structure means that with such a gigantic sum of subsidies to be cashed in as space is (more than 10 Billion US-$ in the US), every state more or less must receive a fair share. Otherwise, you are in for trouble. It thus came that a Utah based company, Morton Thiokol, got the contract to build the strap-on boosters. Morton Thiokol is more than 1000 km from any coast, more than 2500 km from Cape Canaveral. Not ideal conditions for hauling a piece of equipment weighing almost 1000 tons, loaded with a high explosive. You might think that whoever gets the refill job would have to be situated on (or at least be ready to moove to) the East Cost. Clearly, not a logical decision.

Hydrogen as fuel

Hydrogen as fuel actually deservs a discussion of its own. The "pure mathematics" of rocket flight tells us that we should use hydrogen, being a very powerful fuel. This is why engineers wanting to build elegant systems use it. The rocket equasion, however, does not take price aspects into account. Hydrogen is a real price driver. First of all, the fuel itself is not exactly cheap (on a per-kg basis). The price is said to be not very quantity-sensitive, because the prime cost in producing hydrogen is energy expense. Hydrogen is also difficult and dangerous to handle. At hydrogen's near-absolute zero temperatures, many metals become brittle, lines and tubes have to be pre-cooled (usually done with Helium, very expensive), hydrogen molecules are very small and can diffuse through some metals, welds etc., leakage gas must be blown off so it doesn't accumulate and cause explosions. Since hydrogen has a very low specific weight, tank volumina must be very big, thus offsetting part of the gains.

We have - like many others - come to the conclusion that kerosine is an ideal rocket fuel. Or, alternatively, hybrid fuels. Kerosine is available worldwide, it is cheap, has many uses, is safe, environmentally clean etc. If hydrogen was a commonplace fuel, for ex. in aviation, this attitude might be reevaluated. It seems, however, that the aircraft industry is currently backing off from the use of hydrogen for aircraft, which is not encouraging.

Conclusion:

We must totally re-think our approach to rocket building. Best if we let people do that who are relative novices without an extended work background in aviation or the aerospace industry. We should scan for well-established technologies and existing hardware, as well as established manufacturing processes. Construction material should be steel or aluminum or fiberglass (also relatively cheap). As much as possible should stay in as few hands as possible. Construction and tests should be conducted by small teams. They should be working goal-oriented only, otherwise fully independent in their decision making process. Targets are rugged systems with a minimum of parts. Economy is a design factor of equal rights. Each design decision must stand economic considerations. Documentation would be thin and mostly paperless. The rule of thumb should be: complex is expensive - fuel is cheap.

In the 60ies and 70ies there were a number of US firms, for ex. TRW, which had build the lunar descend engine on the basis of a true low tech approach (pressure feeding system, no turbo pumps, near 100 % reliablility, high tolerance), experimented with low cost designs. TRW build an 80 ton thruster using kerosine and LOX. This engine was supposedly build by a local boiler company for an estimated 30.000 US-$, welded together as any ordinary piece of hardware would be. The engine worked great. Other companies including names like Boeing have also experimented. But when it became evident that the Shuttle would get all payloads there was all of a sudden no market for a cheap carrier and all work was abandoned. Truely a missed chance. All pledges to reduce launch cost have since then been nothing but lip service. For us space enthusiasts, a real dream killer.

What we need to open up space is a robust truck, the DC-3 of space. We don't have to start with a Lear Jet.

Incidently, all parties would profit from a cost reduction, including the aerospace "establishment", by an overall growth of the market. This is how it was in EDP, which took a breathtaking rise after the prices for computers had gone down by a few powers of 10. IBM is no longer leading the market, but they make more money than ever.

On the other hand: If we don't manage to turn direction, traditional aerospace firms must be careful not to loose all their busines to some newcomer. Rocket building has undergone an enormous proliferation in the past decade - and it is not just the bad boys like Iran and Irak. Rather, a number of striving, pragmatic and forward-looking nations like Brazil, Australia and of course those industrious Asians are on it. Japan, Taiwan, Malaysia, Indonesia, even India. If those were to implement low-tech rocketry they would beat us right out of the field, being too expensive and not producing what the customer needs.

What the customer needs is clearly defined in the space launch business: get that payload up there cheap and safely. Everything else doens't matter in the end.

Within the next decades a lot of money will be earned in space. New communication systems are currently being designed. Where money is earned, there usually is also some surplus for science, for exploring Moon and Mars. But even here, the signs are that the times of the visionaries promoting 500 billion $ Mars landing schemes is coming to an end. In the end, we owe it to those people that we (mankind) are still sitting here. What we need are economically viable concepts, such as Zubrin's and strategies for acting - not visions. But remember - everything starts with a cheap launcher.

We a FAR believe that the time of cheap rockets is finally coming.

At least we will do our bit and try to promote the idea in Europe. Even our countries, being relatively small, have a chance of developing launchers, if we get it right. The thesis is that it can be done on a limited budget. Thus, risks should also be limited while chances are high. So what are we waiting for ?

Write us what you think:

We will publish your oppinion and answer your questions. There are no dumb questions, so ask. This discussion belongs to all of us because it is our money and our future!

PS:

A good book on Low-Cost Design is LEO on the Cheap von Lt. Col. John R. London III, published at Air University Press, 1994

(c) FAR 1997 - back to our Home page