Len Cormier
Third Millennium® Aerospace, Inc.
Reno, Nevada
Note: This paper is basically similar to a professional paper --
AIAA-96-2773 -- given to the 32nd AIAA/ASME/SAE/ASME Joint Propulsion Conference,
July 1-3, 1996, Walt Disney World Dolphin, Lake Buena Vista, Florida
Abstract
Pre-operational investment is the dominant economic consideration for fully
reusable launch systems. Assuming a business-investment environment, Third
Millennium Aerospace has found that traffic levels must build to more than
a few hundred flights per year before recurring operational costs become
as important as return on investment. Our studies suggest that this finding
is largely independent of vehicle size.
Because of the importance of minimizing pre-operational investment, purely
single-stage-to-orbit concepts are generally unpromising at this time from
the business point of view. Single-stage-to-orbit is technically feasible
with current technology if the system is large enough, e.g. 800 metric tons
gross liftoff mass. However, this size vehicle is likely to cost $6,000,000,000
or more to bring to operational status. If this investment is made at the
rate of $1,500,000,000 per year for four years, and is followed by a ten-year
operational recovery period with ten percent interest on unrecovered investment
for the 14-year period, then the return on investment (ROI) should be about
$1.25 billion per year. By comparison -- even at 125 flights per year --
recurring costs for a properly designed, 800-tonne SSTO should not cost more
than $375 million for one year's operations.
At a more realistic venture-capital interest rate of 26 percent per year,
ROI should be about $3.2 billion per year. This is more than eight
times more than total recurring costs for 125 flights per year -- which is
comparable to the current world-wide total number of launches per year for
all launch vehicles.
Third Millennium has long felt that the most promising way of minimizing
pre-operational investment is to minimize the size of the orbiter stage.
The assist-stage concept promises to reduce the size of the orbiter by a
factor of ten or more -- without significantly increasing the recurring cost
per kilogram of payload. Third Millennium defines an assist stage as one
that greatly reduces the size and cost of the expensive orbiter without greatly
increasing the cost and complexity of the total system. We believe that viable
assist-stage concepts include: drop-off engines used only for takeoff; sled
launchers; flyable sled launchers; subsonic carrier aircraft; and relatively
simple, pressure-fed, reusable boosters that stage at low dynamic pressures
and low supersonic speeds.
Introduction
Aerospace engineers who have only worked in a government-sponsored environment
are likely to have misconceptions of what constitutes proper return on investment
in a business environment.
In the foregoing abstract, we have alluded to a 10 percent return on investment
(ROI) and a 26 percent ROI. A 10 percent ROI is more typical of relatively
low-risk equity investments, while a 26 percent ROI is more typical of a
venture-capital investment in a low inflation period. Potential venture-capital
ROI may be much higher -- for example, 40 percent compounded per year --
for new, risky projects in order to make up for a significant chance of complete
failure by at least some of the projects being funded.
The sections that follow attempt to show what impact these rates of returns
on investment should have on the requirements for, and the design of, reusable
launch vehicles.
Economies of scale generally favor larger
expendable launch vehicles. Users have come
to expect to pay a premium for launch of smaller payloads. The economics
of reusable launch vehicles, however, is
quite different. At least initially, small appears
to be much better.
A Simplified Financial Model
Spreadsheet software such as Quattro Pro or Excel permit rather sophisticated
analysis of the assumed use of funds during any quarter, as well as varying
interest rates, varying development schedules, varying market buildup rates,
etc. However, a simplified model -- with a few simplifying assumptions --
is useful for illustrating the importance of investment with respect to reusable
launch vehicle economics. The following simplified model will be used to
illustrate the importance of investment to reusable launch vehicle
economics:
With these simplifying assumptions, one can calculate the equivalent
investment and the payment schedule corresponding to the entire fourteen
year period for a given ROI. The payment schedule -- together with recurring
costs -- determines the price per flight and price per kilogram of payload.
We then apply this method to two different reusable launch vehicles.
where I = total investment; and i = assumed interest rate (ROI)
Note: The numbers after each set of parenthesis are supposed to
be exponents; not all browsers handle exponents and superscripts
properly.
Applying Equation (1), Table 1 indicates the equivalent investment that must
be paid off during the revenue operations period.
Table 1. Equivalent Investment (EI).
| ROI |
EI Factor |
EI for $6B |
EI for $500M |
| 10 percent |
1.276 |
$7.7B |
$638M |
| 26 percent |
1.842 |
$11.1B |
$921M |
| 40 percent |
2.486 |
$14.9B |
$1.24B |
The "present value (PV) of an annuity" equation can now be used to calculate
the annual payments (PMT) required for various expected interest rates or
ROIs:
PV = PMT[1 - (1 + i)-n] / i (2)
Note: -n is supposed to be an exponent; not all browsers handle
exponents and superscripts properly.
Table 2. Portion of Annual Revenue
Required for ROI.
| ROI | SSTO | Assisted SSTO |
| 10 percent | $1235M | $104M |
| 26 percent | $3204M | $266M |
| 40 percent | $6173M | $514M |
Table 2 shows the results of applying Equation (2) to the results of Table
1.
Tables 3a, 3b, and 3c show the price per flight and the price per kilogram
that must be charged for ROIs of 10 percent, 26 percent and 40 percent,
respectively. These tables are constructed by applying the results of Table
2 and adding in the recurring costs for various assumed traffic levels.
Tables 3a, 3b, and 3c assume that recurring costs follow a 90 percent learning
curve. These tables assume further that -- at traffic levels of 125 flights
per year -- recurring costs are $3,000,000 per flight for the SSTO and $1,000,000
per flight for the Assisted SSTO.
Table 3a. Price per Flight for 10 Percent ROI and Assumed Traffic Levels.
|
ROI |
Tonnes to LEO/yr |
SSTO |
Assisted SSTO |
||||||
|
No. Of Flts/ Year |
Recurring Cost/Flt |
Total Price per Flt |
Price /Kg of Payload |
No. Of Flts/ Year |
Recurring Cost/Flt |
Total Price per Flt |
Price /Kg of Payload |
||
|
10 percent |
100 |
5 |
$4.9M |
$252M |
$12,600 |
25 |
$1.3M |
$5.5M |
$1,375 |
|
500 |
25 |
$3.8M |
$53.9 |
$2,695 |
125 |
$1.0M |
$1.8M |
$450 |
|
|
1000 |
50 |
$3.4M |
$28.5 |
$1,425 |
250 |
$0.9M |
$1.3M |
$325 |
|
|
2500 |
125 |
$3.0M |
$13.0M |
$650 |
625 |
$0.8M |
$1.0M |
$250 |
|
|
5000 |
250 |
$2.7M |
$7.7M |
$385 |
1250 |
$0.7M |
$$0.8M |
$200 |
|
Table 3b. Price per Flight for 26 Percent ROI and Assumed Traffic Levels.
|
ROI |
Tonnes to LEO/yr |
SSTO |
Assisted SSTO |
||||||
|
No. Of Flts/ Year |
Recurring Cost/Flt |
Total Price per Flt |
Price /Kg of Payload |
No. Of Flts/ Year |
Recurring Cost/Flt |
Total Price per Flt |
Price /Kg of Payload |
||
|
26 |
100 | 5 | $4.9M |
$646M |
$32,300 |
25 | $1.3M |
$11.9M |
$2,985 |
| 500 | 25 | $3.8M |
$131M |
$6,550 |
125 | $1.0M |
$3.1M |
$775 |
|
| 1000 | 50 | $3.4M |
$67.5M |
$3,375 |
250 | $0.9M |
$2.0M |
$500 |
|
| 2500 | 125 | $3.0M |
$28.6M |
$1,430 |
625 | $0.8M |
$1.2M |
$300 |
|
| 5000 | 250 | $2.7M |
$15.5M |
$775 |
1250 | $0.7M |
$0.9M |
$225 |
|
Table 3c. Price per Flight for 40 Percent ROI and Assumed Traffic Levels.
|
ROI |
Tonnes to LEO/yr |
SSTO |
Assisted SSTO |
||||||
|
No. Of Flts/ Year |
Recurring Cost/Flt |
Total Price per Flt |
Price /Kg of Payload |
No. Of Flts/ Year |
Recurring Cost/Flt |
Total Price per Flt |
Price /Kg of Payload |
||
|
40 |
100 | 5 | $4.9M |
$1240M |
$68,000 |
25 | $1.3M |
$22M |
$5,465 |
| 500 | 25 | $3.8M |
$251M |
$12,550 |
125 | $1.0M |
$5.1M |
$1,278 |
|
| 1000 | 50 | $3.4M |
$127M |
$6,350 |
250 | $0.9M |
$3.0M |
$739 |
|
| 2500 | 125 | $3.0M |
$52.4M |
$2,620 |
625 | $0.8M |
$1.6M |
$406 |
|
| 5000 | 250 | $2.7M |
$27.4M |
$1,370 |
1250 | $0.7M |
$1.1M |
$278 |
|
Tables 3a, 3b, and 3c suggest that the assist-stage concept may be a powerful
mechanism for alleviating the "investment barrier problem" common to reusable
launch vehicles.
The operational desirability of an SSTO is undeniable. However, the large
investment required for an SSTO -- or for a reusable two-stage vehicle with
a large payload, for that matter -- demands that the market build very rapidly
in order to justify the investment.
In gross terms, the current launch vehicle market might be characterized
as follows:
With reference to table 3, the SSTO only
starts to become attractive when it captures the total market of 500 metric
tons per year -- even at a very modest 10 percent return on investment. At
this point, ROI still represents about 93 percent of the price that must
be charged per flight. At a more realistic 26 percent ROI, capture of the
current total traffic level of 500 tonnes per year would only reduce cost
per kilogram by a factor of two, compared to current prices. At this point,
ROI represents 97 percent of the price that must be charged per flight.
By comparison, for a ROI of 10 percent, the Assisted
SSTO reduces cost per kilogram of payload to about one-tenth current
prices with capture of only one-fifth the current market in terms of tonnage:
i.e. at 100 tonnes per year. For 26 percent ROI, prices are reduced to about
one-fifth at 100 tonnes per year, and to nearly one-twentieth at 500 tonnes
per year.
Ironically, the assist-stage concept may be one of the best ways to build
the market for an eventual SSTO. As tables 3a, 3b, and 3c suggest, the Assisted
SSTO shows the benefits of low recurring costs even at a fraction of current
tonnage levels. This allows the market to grow rapidly by enabling new
applications that require much lower cost
access to space. Assist-stage concepts are also less sensitive to requirements
for advanced technology -- another reason for starting small.
Actually, current tonnage levels -- or current number of flights -- are very
poor market criteria for reusable launch vehicles. Total dollar volume level
is probably much more meaningful. A dollar level criteria also presents both
the SSTO and the Assisted SSTO in much more favorable light.
The figure below presents comparisons of how much cargo could be carried
by: a) current launch vehicles; b) an SSTO; and c) an Assisted SSTO -- for
20, 40, 60, 80, and 100 percent of current dollars spent on space transportation.
The prices for both the SSTO and the Assisted SSTO assume a ROI of 26 percent
is required; current launch vehicles assume investment is a sunk cost or
that any additional investment is already factored into the price.
This first figure suggests several important observations. First, reusable
launch vehicles are likely to make current tonnage levels completely irrelevant
as a market indicator. Third Millennium has maintained for nearly three decades
that the availability of low cost, frequent, and reliable space transportation
by means of one or more fully reusable launch vehicles will completely change
the nature of what we do in space. The next section on the nature of payloads
in the reusable era will treat this in more detail.
Second, a reasonable ROI mitigates against the introduction of large reusable
launch vehicles until the market has had a chance to adjust to this vastly
different way of doing business. Unless and until a larger reusable launch
vehicle like an SSTO captures about half of the current dollar market as
soon as it is ready to operate, then a 26 percent ROI does not appear to
be possible. A smaller reusable launch vehicle like an Assisted SSTO, however,
becomes attractive as soon as it captures even a few percent of the current
market as measured in dollars.
And third, a smaller reusable launch vehicle appears to be an appropriate
way to transition the launch vehicle market place from a market measured
in hundreds of tons per year of cargo to one measured in tens of thousands
of tons or more per year of cargo.
Of course, one can make an argument for a lower -- or higher -- ROI than
26 percent per year. The next figure estimates what might occur at a 10 percent
ROI; the final figure estimates what might occur at a 40 percent ROI.
The second
figure shows that the SSTO can be attractive at lower market levels if the
ROI expectations are low. This second figure also indicates that the SSTO
can catch up with and surpass the Assisted SSTO, if ROI expectations are
low enough. Although recurring costs per kilogram of payload are relatively
independent of size, the larger SSTO does seem to have a slight advantage
because of economies of scale. If the market grows sufficiently -- perhaps
with the initial use of an Assisted SSTO -- then investment becomes relatively
less important. Accordingly, the SSTO can then be very cost effective even
at higher ROI levels. Eventually, the market will probably support different
size reusable launch vehicles -- just as the airline market now supports
many different sizes and types of airliners and cargo aircraft.
Unless the government subsidizes the effort, Third Millennium does not
believe that a 10 percent ROI is appropriate. Investors satisfied with a
10 percent ROI can achieve that kind of return with a lower perceived risk.
Government subsidy, on the other hand, distorts the market place, and this
has been one of the main problems with the development of a truly competitive
commercial space transportation market.
The third
figure indicates that at higher ROI levels, the timing for the market buildup
required to support larger reusable launch vehicles like the SSTO is even
more critical. Even if measured in terms of dollars rather than tons, the
current market must grow significantly to support a 40 percent ROI. An Assisted
SSTO, however, appears to be attractive to potential customers even with
a high ROI.
The Nature of Payloads in
the Reusable Era
Low-cost, frequent, reliable access to space should change the nature
of launch-vehicle payloads. One of the main changes should be a payload managers'
attitude toward rendezvous. Rendezvous should become far more practical.
With practical assembly on orbit, 4-tonne payloads should be adequate, initially
at least -- with less dependence on much more costly transport of payloads
in large packages, e.g. 20 tonnes or larger.
With respect to very large payloads, rendezvous and some on-orbit assembly
is necessary even for the largest of launch vehicles. With greatly increased
practicality, assembly on orbit should become far more attractive. The main
concern will likely be with total cost. If an assembled payload is far lower
cost, there should not be any great concern with how it was delivered. The
fact that an Assisted SSTO may have to make many flights for a particular
customer makes the economics of reusability that much better.
With respect to large payloads now carried on a single launch, there may
-- or may not -- be valid arguments for paying perhaps twenty times as much
for launch costs for delivery of the payload in one piece. To the extent
that this argument is valid, such payloads could be carried on current launch
vehicles. Third Millennium expects that much of the real growth in the tonnage
market will be for low-cost transport and assembly of very large payloads
that cannot be carried on single flights of current or projected launch vehicles.
With respect to small payloads, small launch vehicles with dedicated payloads
are preferable -- especially for replacement of satellites in varied orbits.
If such payloads can be launched at much lower costs using a small, reusable
launch vehicle, so much the better.
Payloads should also become much less expensive. Payload designers seem
to believe that they are still launching on Vanguard at $1,000,000 per pound
mass of payload (1958 dollars). Even with current launch vehicles, launch
costs have become two orders of magnitude or more less expensive. In the
reusable launch vehicle era, there will be even less justification for high
payload costs. If reliability is the reason, the same reliability can be
achieved at far lower cost by allowing the payload to be heavier with redundant
equipment. Similarly, there will be little justification for using exotic
materials to reduce the weight of a satellite, when launch costs are only
a few hundred dollars a pound.
Third Millennium has often proclaimed a basic paradox of space transportation
economics: When launch costs are high, the predominant cost is likely to
be for spacecraft. However, when launch costs become very low, the predominant
cost is likely to be for space transportation. Consider deep space missions:
the dominant payload for chemical rockets will be propellants, mostly liquid
oxygen. Propellants are very inexpensive relative to even very low-cost space
transportation.
Assist Stage
Concepts
Third Millennium has long felt that the most promising way of minimizing
pre-operational investment is to minimize the size of the orbiter stage.
The assist-stage concept promises to reduce the size of the orbiter by a
factor of ten or more -- without significantly increasing the recurring cost
per kilogram of payload.
Third Millennium defines an assist stage as one that greatly reduces the
size and cost of the expensive orbiter without greatly increasing the cost
and complexity of the total system. We believe that viable assist-stage concepts
include: drop-off engines used only for takeoff; sled launchers; flyable
sled launchers; subsonic carrier aircraft; and relatively simple, pressure-fed,
reusable boosters that stage at low dynamic pressures and low supersonic
speeds. Over the years, Third Millennium has advocated various forms of assist
stages -- generally progressing from drop-off engines, through sleds and
subsonic air launch, to our current concept for a reusable, pressure-fed
booster that stages at mach 3 at about 30 km altitude.
Our 1967 concept was a winged SSTO that used six drop-off H-1 liquid
oxygen/kerosene engines for vertical liftoff. These engines were recovered
within a few kilometers of the launch site. After pitchover, an aerospike
engine designed around J-2 turbopumps powered the SSTO into orbit. This concept
was similar, in some respects, to the current X-33 concept.
Our 1971 concept was a HTOL winged vehicle that used a sled to "stage"
the "landing gear" that would have been ten times heavier than it would have
to be at landing weight. The U.S. Air Force became interested in this concept;
it became the direct precursor to a 20-year effort on the Boeing Reusable
Aerospace Vehicle (RASV).
In the late 1970s, we considered subsonic air launch from an airbreathing carrier. Our 1980 concept used a modified Boeing 747 to launch a relatively small orbiter powered by eight RL10 engines. In recent years, we also considered launch from an Antonov AN-225.
Third Millennium Aerospace now believes that its reusable pressure-fed booster that boosts an 86 tonne orbiter to mach 3 at low dynamic pressure represents the lowest financial risk and technical risk for the near future. Over the past decades, Third Millennium has considered essentially all of the concepts now being considered by others. Third Millennium also believes that all of these other concepts have significant drawbacks either from the business or technical point of view, or both -- relative to Third Millennium's current Space Van.
Conclusions
Copyright © 1996 by Third Millennium® Aerospace, Inc.