Index Home About Blog
From: (Dani Eder)
Subject: Re: Jet-Assisted Launch Vehicles 
Date: Jul 18 1995
Newsgroups: (Marcus Lindroos INF) writes:

>What, exactly, is a "jet boosted launcher?" 

The Sanger assumes a Mach 5 carrier aircraft for a single rocket stage.
The carrier aircraft is much larger than the rocket.

A jet-boosted launcher keeps only the minimum necessary jet propulsion -
the engines - and dispenses with wings and landing gear in favor of
parachute recovery.

The concept we did a study of used ten Pratt & Whitney F100-229 jet
engines (used on the F-15 fighter).  They have a sea-level static
thrust of 29,100 lb on afterburner each, or a combined liftoff 
thrust of 291,000 lb.  Each bare engine weighs 3700 lb.  When a
parachute system, landing legs, fuel tank, and fuel are added, you
get a 6000 lb unit at takeoff.  The rocket core weighed 85,000 lb
and had 3 pieces.  Each piece had the same diameter tanks (about
2 meters) and 2 RL10A-3 rocket engines (except for nozzle size,
identical to those used on the DC-X, and identical to those on the
Centaur upper stage).  The middle piece had a longer fuel tank,
and goes all the way to orbit.  The other two had shorter tanks
and were staged off when they got empty.

The entire launcher thus had a weight of 145,000 lb, and under
jet power only took off at 2 g's.  As opposed to rockets which
avoid high dynamic pressure because it makes more drag, if you
are flying on jet power you get more thrust, so you want to
take off fast and accelerate fast to get the maximum acceleration
before you run out of air.

We used a sophisicated trajectory analysis program to find out what
the best ascent trajectory was, considering maximum allowable
dynamic pressure, variation of jet thrust with altitude and speed,
and drag and lift as a function of vehicle speed and attitude.

We found that the optimal way to fly this vehicle was to take off
under pure jet power up to 50,000 ft and Mach 1.7, at which point
you drop off the jets and light up the rocket.  Assuming a 15%
harware weight fraction for the rocket stages, we get a payload
to orbit of 6,000 lb.

The jet engines run for 60 seconds or so at full afterburner, so 
they are good for many many launches.  The parachute is mounted
so the engine descends in a horizontal position, and legs with
shock absorbers pop out and keep the landing shock to 6 g's, i.e.
less than what the engines can stand.

The jet engines burn about 15 lb of fuel per second, so they consume
about 960 lb (160 gal) of jet fuel each per launch, for a total
of $1600 in fuel cost per launch.  Depreciation and maintenance on
the jet engines runs about $2000 per hour, and you have used 1/6
of an hour of run time, so that is about $330.  The big costs are
interest on the engine cost and repacking the parachutes.

The goal we had for such a vehicle concept was to have no engine
development cost (off-the-shelf jet and rocket engines), a modular
design (designing one engine module and building ten of them costs
less than designing one module with ten engines)(the rocket stages
shared many features except tank length and re-entry insulation).

The rocket stages are small enough to be truck or air shipped back
to the launch site, and have better weight margins than an SSTO
(15% vs 10% inert weight).  All of this was intended to keep the
development cost to a couple of hundred million $, and the 6000 lb
payload size was optimized to fit the satellite delivery market
and incidentally big enough to carry a crew of 2 to orbit.

At a price of $15 million per launch and a cost of perhaps $3 million,
the gross profit at 6 launches per year would pay off the development
in 3 years.  Since all the vehicle parts are recovered, the operating
cost is more how many hours it takes to re-assemble the vehicle,
which is minimzed by making each part smart, so there are only 
structural connections.

A precursor vehicle, which we did not do the analysis of, might
use two jet engines (perhaps from Eastern Europe or some other
cheap source), strapped to a single RL-10 core, which carries
a solid rocket or perhaps two on top.  This would be a Pegasus-class
(on the order of 1000 lb) launcher.  This vehicle should cost
well under $100 million to build.

Dani Eder

From: (Dani Eder)
Subject: Re: "End of manned spaceflight"
Date: Jul 21 1995
Newsgroups: (T. Hancock) writes:

>In article, (Dani Eder) writes:
>ED>This is so far off base I have to reply.  Given the studies I have worked
>>on on low-capital launch systems, I am pretty sure a manned launcher on
>>the Gemini capsule scale can be built for $200 million at most.  Bill
>>Gates could afford this out of pocket change, and there are on the order
>>of hundreds of people worldwide who could afford to do this if they
>>were serious, all by themselves.  

>OK you write the paper on how it is done and I will be happy to read it and
>help get it published.
>Love to see your design, launch costs, ground support, insurance etc..,

Non-Capital Intensive Launchers

by Dani Eder
[draft for comments 21 July 1995]


We describe two launch systems capable of carrying humans to Low Earth
Orbit with low capital acquisition cost.  One is a conservative system
using off-the shelf technology, the other uses more agressive technology
and has higher performance.

System 1:

This launcher is based on Jet fighter type booster and cryogenic 
rocket upper stages.  The overall vehicle weighs 145,000 lb at
takeoff, consisting of 10 jet engine modules @ 6,000 lb each, and
a 2-stage rocket core weighing 85,000 lb.  A jet engine module
contains one Pratt&Whitney F100-229 jet engine (29,100 lb sea-level
thrust).  The bare engine weighs 3,700 lb.  With 1000 lb of jet
fuel, fuel tank, engine housing, separation hardware, parachute,
and landing legs, the complete module weighs 6000 lb.  

On takeoff, the 10 jet engines are running at full afterburner,
providing a liftoff thrust/weight of about 2.0.  The vehicle 
ascends to 50,000 ft under jet power, at which time the velocity
is 500 m/s (Mach 1.7) and the flight path angle is 35 degrees
above horizontal.

The jet engines separate at this point using bomb-release type
hardware and steer away from the core rocket using residual thrust.
The jet engines will run out of fuel or flame out shortly thereafter.
They will coast and begin to descend.  A set of pilot and main
parachutes are deployed as the engine modules near the ground.
The lines attaching the prachutes to the engine module hold the
module in a horizontal position, which is more stable for landing.
At the same time a set of 3 or 4 landing legs are deployed which
have shock absorbers to limit the landing load to 6 g's, which is
within the structural capability of the engines.

The engines will land about 10 miles from the launch site, and
are retrieved by truck with loading crane and returned for the
next launch.

The rocket core consists of 3 pieces.  Two make up the 2nd stage
and the third is the 3rd stage.  The three pieces are similar
except the 3rd stage has a longer fuel tank and more re-entry
insulation, and provisions for a payload on the front.  Each
rocket module has 2 Pratt & Whitney RL-10A-3 Rocket Engines, of 
16,500 lb thrust each, and all three modules burn in parallel
initially.  When the 2nd stage modules run out of fuel, they
stage off and re-enter, while the third stage module continues
to orbit.  The combined vacuum thrust is 93,000 lb vs an initial
weight of 85,000 lb, giving a T/W > 1 at 2nd stage ignition.

The rocket modules share a common tank diameter and plumbing
arrangement at the back end to reduce design cost.

Since the rocket core is starting from a high altitiude, drag
losses, g-losses, and Isp reduction due to back-pressure are
minimized.  Trajectory simulations using the Boeing POST program
indicate a net payload of 6,600 lb assuming the rocket stages
have inert weight fractions of 15%.  We assume 6,000 lb net

The Gemini capsule weighed about 7,000 lb using 1960's technology.
We feel confidant that a 2-person capsule can be built today
for 6,000 lb simply by substituting higher strength materials
and lighter weight diplays & controls.

All components of the vehicle are recovered by parachute and
reused.  The rocket modules are 2m in diameter and weigh no
more than 6000 lb empty, so all components of the vehicle will
fit in standard shipping containers for re-use.

We recommend South Texas as a launch site, since large ranches
are available there with low human population density.  The jet
engine modules will land close enough to be on the same property,
while the 2nd stage modules will land in Florida.  The launch site
can be leased and used in parallel with ranching in the near vicinity.

As much as possible, off-the shelf hardware from other aerospace
programs will be used.  The expected new/modified percentages
are 20% for the jet engine modules and 50% for the rocket and
capsule portions.  Multiplying by the module weights, we have an
estimated 1000 lb + 3000 lb + 2500 lb = 6500 lb of equivalent new
design hardware.  The 777 aircraft cost about $5 billion to develop,
and contains 240,000 lb of hardware.  Since aerospace hardware
development cost scales (based on past program experience) as
the 0.75 power of hardware weight, we would expect the development
cost to be 6 2/3% as much, or $300 million.  Since we will not need
to build as extensive a production line or need to certify the
vehicle to airline passenger standards, we arbitrarily assume that
we can reduce the development cost by 1/3 to $200 million.

The DC-X program demonstrated a ground crew of 10 or so for a
4 RL-10 launch vehicle, and we add another crew of 10 for the
jet engine module handling and installation (based on fighter
bomb crews, since we are using that type of mechanism.  With 
another 10 people for overhead functions, we have a total crew
of 30 at a rate of $80K per year each, of $2.4 million per year.

If flights are once per week, the ground crew cost is $50,000
per launch.  Jet and rocket fuel is around $40,000 per launch.
The jet engines are assumed to be acquired used/surplus from
military downsizing at $2 million each, and the RL-10s cost
$3.5 million each.  The balance of the vehicle should cost
about $20 million per unit in small quantities, scaled from
aircraft production cost modified for the quantity produced.

If we amortize the vehicle at 200 launches, the RL-10s at 40
launches, and the jet engines at interest cost (the run time
is negligible, 1 minute vs. 2000 hr remaining service life), 
we get costs of $100,000, $525,000, and $32,000 per launch.

Thus we have a total operating cost of $747,000 per launch.
If we amortize the development over 5 years, this adds $40
million per year, or $800,000 per flight, for a per-launch
price of $1.5 million.  This comes to $250/lb for cargo or
$750,000 per person.

Dani Eder

From: (Dani Eder)
Subject: Re: Jet engine first stages
Date: Sep 14 1995

Simon Rowland <> writes:

>Why don't people use a jet engine for the primary stage in rockets? Not 
>only can jet fuel be stored more densly then H2, but it can take the O2 
>out of the air, then when it's too high to operate, the jets are 
>jettisioned and rockets take over. Or even the fuel could still be LOX 
>and LH2, but with a tiny LOX tank and a normal size LH2 one. The rockets 
>just feed off the air for the oxygen, then when the air gets too 
>tenuous, it starts using the stored LOX. Makes sense to me (since 81% of 
>the fuel weight is LOX)...

Because 'people' don't build launch vehicles.  Most of them have been
built by government agencies as derivatives of ballistic missiles
that were first designed in the 1950s.  In the 1950s jet engines were
considered for launching things, but they only had a thrust/weight
of around 4 or so, which was too low to be useful.  They figured
out back then that you needed a T/W of around 10 for a good jet
booster.  Well, guess what, such jet engines are now available.

Last year we studied a jet-boosted launcher using P&W F100-229
jet engines (the engine in the F-15 fighter).  It works just fine,
taking you up to 50,000 ft and Mach 1.7.  Since this engine on
full afterburner has about 4 times the fuel efficiency of a rocket
engine, your best performance is achieved by running pure jet
propulsion until you run out of air, then dropping the jets and
lighting the rocket engines.

Unless you want to spend a lot of money, you wnat to stick with
existing engines.  Engine development, be it car, jet, or rocket,
is an expensive proposition.

Dani Eder

From: (Dani Eder)
Subject: Re: Jet engine/rotor first stages
Date: Sep 20 1995

William B Patterson <> writes:

>Why not put your engines on the tips of large rotor systems and gently
>lift your upper stages to 25000 ft or so. We lifted a 250 lb human powered
>helicopter with two 3 lb thrust propellors on the tips of 50 foot radius

Because while a jet engine stages at 50,000 ft, the speed you have
(Mach 1.7) at that point is equivalent to coasting up to 90,000 ft.
So, if you are comparing lift capability, the jets get you a lot higher.

Dani Eder

From: (Dani Eder)
Subject: Re: Jet engine first stages
Date: Sep 27 1995

Daniel Woodard <> writes:

>For vertical takeoff a solid fuel booster has better thrust to weight 
>ratio and lower cost over the brief period a jet can function.  For true 

How do you figure that a solid has lower cost?  The shuttle SRM
provides 3 million lb thrust for $25 million dollars per SRM per
flight, for a cost of $8.33 per lb thrust.  An F100-229 jet engine
provides 29,100 lb sea-level thrust for under $5000 per launch,
for a cost of $0.17 per lb thrust.  The Shuttle SRM has a liftoff
weight around 1.5 million lb, for a liftoff T/W of 2, and a half-
empty T/W of about 4.  The jet engine module (jet+fuel+tank+landing
gear+parachute of) weighs 6,000 lb at takeoff, for a T/W of 4.85.
Halfway through it's boost, the thrust is actually a bit higher
and the weight is down to 5,500 lb, for a T/W of 5.4.

>It is difficult to see a cost advantage unless the jet system is 
>reusable, and there has been little work with winged flyback boosters.  

Of course the jet system is reuseable.  Who in their right mind
throws away a jet engine after 60 sec of runtime?  Even our 
expendable cruise missiles get 3 hrs out of the jet engine before
they blow up.

Dani Eder

From: (Dani Eder)
Subject: Re: Jet engine first stages
Date: Oct 02 1995
Newsgroups:, (John Stoffel) writes:

>Where do you get your numbers for the jet engine?  I don't think you
>can buy an F100-229 for $5000, or does this include the cost of the
>engine amortized over X flights?  I'm also wondering where the 6,000
>lb comes from too.  And since we'd need 105 (rounding to nearest 5)
>F100-229s to equal the thrust of _one_ SRM, I'm not sure I see where
>the savings would happen.  I for one would hate to try and integrate
>all those engines (think of the air intakes you'd need!) into a
>package that would push the capsule up, drop off, parachute down, and
>be be easily and cheaply recovered.

>I'm not saying it couldn't be done, but I'm saying it isn't as cheap
>as you make it out to be.  

I work for Boeing.  When I did the study I got my weight & performance
data from a Boeing propulsion engineer, who in turn got his data from
Pratt & Whitney, who makes the engine.  A new F100-229 engine costs
about $4.5 million to buy.  It has a rated operating life of 4000 hours.
Since you use it for about one minute per launch, engine life is not
an issue.  You will burn about 160 gallons of jet fuel per engine 
per launch, which comes to about $160.  If you parachute recover,
you have the cost of repacking the parachute, say $1000.  If you fly
twice a week, the interest on the money you borrowed to buy the
engine comes to $3500 (at 8% per year).  The time it takes for
two guys with a truck to drive 10 miles, load the engine module, take
it back to the pad, hoist it up with the truck mounted crane, and
re-attach it to another rocket is about 3 hours, so that's another

The 6000 lb comes from:

	3700 lb for bare jet engine
	1300 lb for other equipment (landing legs, parachute, fuel
            tank, mounting structure)
	1000 lb jet fuel

I used the SRB as a cost per pound of thrust comparison.  Trying to
match the SRB in total thrust is unreasonable with existing jet
engines.  We didn't try.  We sized a launcher using 10 jet engines
producing 291,000 lb sea-level thrust, with 154,000 lb takeoff weight
(60,000 in jet engine modules, 94,000 in rocket core), that delivered
an estimated 6,600 lb to LEO.  That size payload can deliver a couple
of people (It's about the weight of a Gemini capsule that carrier
2 people), or carry a fairly large communications satellite.

Dani Eder

Index Home About Blog