| Home | Traveller | BARD | SARA |

Commercial Airbreathing Shuttle

design by Bruce Alan Macintosh

As an exercise (and to go with some starship designs I'm working on) I tried to design TL 9- (TL 9 without contragrav) ground-to-orbit shuttles. It mostly worked, without breaking the FFS rules more than slightly. The only major difference in fact was that I used the real delta-V formula rather than the simplified FFS one (see, especially the article from, though her transfer orbit formula looks wrong to me.) I did use FFS engines (see below.) Ways in which the design changes with alternative rules, or CG, are also discussed at the end.

TL-8 versions of these designs would look almost identical; if anyone is really interested in the subtle differences I can post them.

The one major difference to published TNE rules is of course the delta-V to orbit. The TNE rule-book number-of-G-hours to orbit, as has been noted, are completely wrong. Delta V from earth's surface to a 150km orbit is 8.00 km/s. (Not 0.64 G-hours, which is 22.5 km/s.) Getting back from orbit requires much less delta-V, too; all you have to do is change your orbit so it scrapes the atmosphere more, and then aerobrake (like the Space Shuttle does.) This raises total delta-V to 8.04 km/s. If you start near the equator delta-V goes down; 7.64 km/s is the figure I used, appropriate for a 30-degree inclination takeoff. For a size 9 world of normal density delta-V is 8.52 km/s, for a size 7, 6.72 km/s.

These are delta-V in a vacuum and assuming very high instantaneous acceleration. Real atmosphere and gravity complicate this somewhat. FFS talks about non-CG spacecraft using 1-g of their thrust to negate gravity until they reach orbit, which is spectacularly nonsensical, even by GDW standards. *Anything* that's moving is in an orbit; a thrown baseball is in an orbit - just a *very* long and skinny elliptical orbit that happens to intersect the Earth's surface. The trick for a rocket is to make sure your orbit never actually reaches the ground. For finite accelerations this means some amount of delta-V is spent going upwards as the spacecraft lifts off, which is less efficient than a pure Hohmann transfer orbit. In addition, you lose delta-V due to atmospheric friction. Together these add about 1.5 km/s to the delta-V requirements for a classical rocket, and require that the takeoff acceleration be significantly higher than 1G.

The shuttle designs below are horizontal-takeoff designs which use their airbreathing AZHRAE engines for takeoff and to climb to high altitudes (and, in the case of the first design, for some extra delta-V.) This is similar to the (now-cancelled) X-30 Aerospace Plane. This approach has been pretty much abandoned in the real world, but FFS makes it easier to design than most other single-stage-to-orbit designs (see below.) It also means that the acceleration required of the engine is only 1G (or even less.) I've allowed 0.6 km/s delta-V for remaining atmosphere drag (8.1 km/s round trip to 150-km orbit from 30 degrees latitude), which is typical for the Pegasus air-launched rocket. There's a huge range of operational parameter space for these vehicles depending on the atmosphere, density and diameter of the world they're operating on; I've designed for an Earthlike world.

Terminology for those who don't have FFS: AZHRAE is the acronym for a combination turbojet/ramjet/rocket engine. HRF is hydrogen rocket fuel. HCD is hydrocarbon distillates (petrochemical fuels.) EAPlaC is the incredibly efficient solid-rocket-like TL-9 engine in FFS, normally used in missiles.

FFS engines have low thrust-to-weight but also lower fuel consumption than the Real World. This is actually an advantage for craft like this one which have quite low accelerations. Using more realistic engines (see the URL above) bites heavily into the payload; designs are available if anyone is really interested.

One can argue that craft that aerobrake should have higher armor values than normal craft, but one can also argue that AV=1 and especially the internal structure is far heavier than real world craft needed, so I've just left it at AV=1.


These designs don't use EAPlaC for a variety of reasons:
  • I started with the exploration shuttle, which needed to be completely field-refuellable.
  • I wanted them to be basically practical at TL-8
  • I wanted the commercial shuttle to emphasize rapid turnaround, thinking that this would decrease operations costs. (I was wrong.)
  • Even though the rules implicitly assume EAPlaC rockets can be turned off and re-ignited (unlike solid rockets in the real world), they're still a type of solid, and no-one in their right mind uses solid rockets on a manned vehicle. (The NASA Space Shuttle is a case in point.)
  • EAPlaC is poorly-explained magic, not extrapolated technology.

    However, EAPlaC has an *incredible* ISP. It's very tempting. If the safety considerations don't apply, TL-9 shuttles would probably at least use expendable strap-on EAPlaC boosters. A shuttle that gets part of its delta-V 4 ton (28 m^3) strap-on EAPlaC boosters can carry 390 tonnes into orbit at a fuel cost of MCr 1; commercial shipping charges would be about Cr 2500 per tonne.

    Additionally, pure EAPlaC unmanned rockets would almost certainly dominate the bulk cargo market, where safety is irrelevant. A 100-ton unmanned disposable EAPlaC cargo carrier can carry 1280 tonnes into orbit, and costs MCr 0.42! It's so cheap - dominated by fuel costs - that it's not even worth reusing. Commercial cost to orbit would be about Cr 500 per tonne.

    The Real World

    In the real world, SSTO's are unlikely to look like this. Partially this is because we don't know how to make an AZHRAE engine, and partially it's because of various details like landing gear that FFS doesn't model. Without AZHRAE engines horizontal takeoff has no advantages and in fact is a net disadvantage, because you have to design the wings and landing gear to lift the whole loaded weight of the spacecraft, not the empty weight.

    Current SSTO paper designs are either VTHL (vertical-takeoff/horizontal landing) rocket takeoff/glide-landers, kind of like the Space Shuttle with no external tank or boosters, or VTVL (vertical-takeoff/vertical landing) craft which look a lot like 1950s science fiction and hover/land using thrust from rocket engines (the DC-X and proposed follow-ons.) See and for perpetual debates as to which approach is better. Neither works well with FFS engines. Also, a VTVL doesn't work very well for a shuttle that starts in orbit - it has to land/hover while carrying all the (heavy) fuel it needs to get back into orbit. Realistic rather than FFS engines are better for these high thrust-to-weight designs, but they're still very hard to do (I suspect because the interior structure mass in FFS is too high for small craft.)

    See also the Exploration Shuttle

    Commercial Shuttle

    General Data

    Displacement: 100 tons Hull Armor: 1 (Internals stressed to 1G)
    Length: 28 meters Volume: 1400 m3
    Price: 32.39 MCr Target Size: S
    Configuration: Cylinder AF Tech Level: 9
    Mass (Loaded/Unloaded): 577.32/88.02

    Engineering Data

    Power Plant: 0.34 MW fuel cell (0.34 MW/hit), 8 hour duration at full power
    Jump Performance: 0
    Fuel: 43.44 m3 of HCD, 1086.18 m3 of HRF
    G-Rating: 1 (AZHRAE rocket mode), 0.66 (AZHRAE ramjet), 0.4 (AZHRAE turbojet) (gives maximum speed 1400 km/h with turbojet, 2324 km/h ramjet)
    G-Turns: 7.36 km/s delta-V from rockets. 8 minutes turbojet flight 2.2 minutes ramjet flight (0.64 km/s delta V, limited by maximum ramjet speed.) Total=8.0 km/s delta-V loaded, 12.5 km/s with no cargo.
    Maint: 141


    Computer: 2xTL-9 Mod Flt Computer (0.03 MW each)
    Commo: 3000km radio (0.1 MW)
    Avionics: TL-8 avionics (0.1 MW)
    Sensors: Radar (3 km; 0.02 MW)
    ECM/ECCM: None
    Controls: 2 open crewstations


    Offensive None
    Defensive: None
    Master Fire Directors: None


    Life Support: Basic, does not cover engines or fuel.
    Grav Compensation: None
    Crew: 2 (1x Maneuver, 1xElectronics)
    Crew Accommodations: None other than crewstations
    Passenger Accomadation: 14 adequate seats
    Other Facilities: None
    Cargo: 140 m3 Large cargo Hatches
    Small Craft and Launch Facilities:
    Air Locks: 1


    The shuttle takes off and lands like an aircraft. It uses 4 minutes of turbojet fuel to take off and climb, then fires the ramjets for 2.2 minutes, accelerating to 0.64 km/s and punching through most of the atmosphere, then firing the rockets for the rest of the trip to orbit. It de-orbits with a brief rocket burn (0.05 km/s), aerobrakes like the shuttle, glides most of the way to its landing sight, then activates the turbojets for the last part of the landing. Enough turbojet fuel is available for four minutes of full thrust at landing, but since it weighs much less while landing and doesn't need to run the engines at full thrust it can run for more than half an hour at 350km/h during landing, a comfortable safety margin.

    Operating costs are dominated by fuel (HRF costs Cr 1000 per m3.) Cost to orbit for commercial service would therefore be around Cr 10,000 per tonne of cargo and 2,000 per passenger (possibly somewhat less if there is enough traffic that shuttles always fly full both to and from orbit.) Compare to the EAPlaC disposable discussed below - shuttles would probably only be used for people and fragile/urgent cargoes.

    ®1996. Traveller is a registered trademark of FarFuture Enterprises. All rights reserved.
    BARD Logo Copyright ©1996 by Lawrence C. Cox.