Boomerang Air-Launched TSTO RLV Concept (Part I)

Back when I left Masten to start Altius, I originally thought I was starting a launch vehicle company focused on reusable nanosat launchers. While we pivoted away from that to focus on space robotics, I’ve continued to dabble with a smallsat RLV concept that I started developing from ideas from this blog back while I was still at Masten. Since we’re definitely not in the launcher business at this point, I wanted to share the idea in the hopes that at least some of the crazy ideas get picked up by someone at some point. Without further ado, I’d like to introduce you to the Boomerang Air-Launched TSTO RLV.

What is Boomerang?
Boomerang is a concept for a low-cost, eventually fully reusable, two-stage to orbit launch system. Boomerang had the following three main system elements, which will be described in more detail in subsequent blog posts:

  • A subsonic airbreathing carrier aircraft
  • A glide-forward recovery liquid rocket powered first stage
  • A hypersonic plasmadynamic decelerator-recovered liquid rocket powered upper stage

All three components are designed to return for a precision landing at the originating air-field, with design decisions that should optimize the probability of gas and go reuse.

Boomerang CONOPS

BoomerangCONOPSWhile I will go into the CONOPS in more details in the posts about the individual system elements, the following key steps are involved in a typical mission:

  1. The two rocket stages are prepped horizontally and integrated with the planned payload to the carrier aircraft at the originating airfield.
  2. The carrier aircraft then takes off and heads up-range (ie in a direction opposite the desired launch azimuth). Once the carrier aircraft is far enough up-range for a glide-forward return, it turns around and preps for launch.
  3. The aircraft then begins to pull-up. As it nears the attitude where it can no-longer retain airspeed, the first stage begins igniting engines (what I call a “Gamma Maneuver”, although like many of my good crazy ideas, John Hare and Kirk Sorensen both beat me to the punch), enabling the aircraft to continue increasing its flight path angle to the optimal separation angle. Crossfeed from tanks on the carrier aircraft keep the first stage topped.
  4. Immediately prior to separation, the crossfeed disconnects are separated, and the rocket vehicle “drops” the aircraft off its back, and immediately begins throttling its engines up to full flight power.
  5. The first stage then performs a burn to near the optimal staging altitude and velocity. After the upper stage is released and on its way, the first stage can perform a braking burn if necessary or could potentially use a hypersonic plasmadynamic decelerator (more on that later) for some of the braking depending on what reentry speed limit minimizes the reuse complexity.
  6. The upper stage continues to orbit and delivers its payload as with a typical launch vehicle.
  7. The first stage glides forward toward the landing site. Once in the general vicinity it performs a course correction maneuver to approach the originating airfield. It then lands using either vertical or horizontal powered landing, or via an autogyrating helicopter system (still an open trade).
  8. After delivering the payload to orbit, the upper stage waits for the originating airfield to pass under its ground track again, then performs a retro-burn, and uses a hypersonic plasmadynamic decelerator to slow itself down sufficiently for safe recovery. Recovery at this point could be via mid-air recovery, horizontal powered landing, or autogyrating helicopter landing.
  9. Lather, rinse, repeat.

High Level Benefits of This Approach
While this isn’t the most orthodox approach to air launch, and does include some key, non-trivial challenge areas, it has a lot of benefits I like:

  1. The Gamma Maneuver enables launching at high subsonic speeds and flight path angles without requiring a military aircraft for launch. This can knock over 1000m/s of gravity and drag losses off of the system, which makes a huge difference for a small rocket. The gamma maneuver may also eliminate the need for large aerosurfaces or structures to accommodate large bending loads as is a traditional challenge for drop-and-light air-launched rockets.
  2. VTVL with glideforward recovery means that staging can take place at a closer to performance optimum stage velocity than is possible with boost-back or glide-back systems, since you don’t have to completely kill and then reverse first stage velocity. At worst you have to decelerate a little to make recovery easier. These two items greatly decrease the performance hit of recovering the first stage.
  3. The Gamma Maneuver with crossfeed also enables launching the rocket nearly full after having checked out all of the engines sequentially for several seconds prior to separation. Why on earth would you drop a reusable rocket before making sure its engines were all operating nominally if you could?
  4. While they’re still a low TRL technology, hypersonic plasmadynamic decelerators (HPDs, another name for MSNW’s Magnetoshell Aerocapture technology as applied to aeroentry) may greatly simplify upper stage recovery if they pan out. While they don’t give a lot of cross-range maneuverability, they do allow pretty precise up/down-range targeting, and might eliminate or at least greatly reduce the need for TPS on the upper stage.
  5. Boomerang can be initially flown with an expendable upper stage, with the HPDs added as a post-delivery experiment kit, allowing upper stage reuse to be tested-out gradually.
  6. If the stages are recovered via horizontal landing (rocket powered, mid-air recovered, or autogyro landed), that would make reintegration with the carrier airplane even easier, potentially eliminating the need for any complex restacking launch infrastructure.

Key Challenge Areas
Obviously there are some key challenges that would need to be addressed both in the design and risk reduction phases:

  1. The Gamma Maneuver scares the crap out of most pilots. I have a few suggestions for how to deal both via system design and how one could do subscale demonstrations with this that I’ll discuss in follow-on blog posts related to the carrier plane and first stage.
  2. A related issue is separation dynamics with a lit rocket vehicle. I’ll discuss this a bit in the post on the first stage.
  3. Another related issue is lighting rocket engine while attached to an aircraft, for fear of hard starts. I have some ideas for how to deal with this as well.
  4. Storing cryogenic propellants for air-launch can be a pain in the neck. I have some great ideas for this, that I’ve discussed a bit previously, and which I’ll update in one of these follow-on posts related to the first stage.

There are a lot of nuances I probably won’t be able to get into in this series, but I wanted to lay out the concept so that if someone wants to borrow/steal ideas, they can.

Next Up in Part II: Carrier Plane Considerations and Options

The following two tabs change content below.
Jonathan Goff

Jonathan Goff

President/CEO at Altius Space Machines
Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and the founder and CEO of Altius Space Machines, a space robotics startup that he sold to Voyager Space in 2019. Jonathan is currently the Product Strategy Lead for the space station startup Gravitics. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew. Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
Jonathan Goff

About Jonathan Goff

Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and the founder and CEO of Altius Space Machines, a space robotics startup that he sold to Voyager Space in 2019. Jonathan is currently the Product Strategy Lead for the space station startup Gravitics. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew. Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
This entry was posted in Boomerang TSTO RLV, Commercial Space, Launch Vehicles, Orbital Access Methodologies, Rocket Design Theory. Bookmark the permalink.

20 Responses to Boomerang Air-Launched TSTO RLV Concept (Part I)

  1. Nathan says:

    Jon, nice post. Feels like old times around here 🙂

    One difference between now and than is that then all you air launch ideas were rather hampered by the choice of carrier aircraft. You either had to squeeze your rocket stages on to a WK2 – whose stats you weren’t too sure about – or build a new larger aircraft. I was actually a little surprised that you weren’t more excited by WK3. I guess you had other things going on at that time!

    If you use the WK3, what kind of payload are you projecting? Is a hypersonic plasmadynamic decelerator more mass efficient that a conventional TPS as well as operationally better? Are you still considering air-launch SSTO?

  2. Nathan,
    Honestly, I haven’t run the numbers, but I wouldn’t be surprised if you couldn’t get most of the way to an Atlas V 401’s capabilities with a WK3 Boomerang Grande. But I’m more interested in the small RLV that can get at least 1-2 people to orbit.

    As for HPDs, a lot depends on the existing launcher and what capabilities it already has native on it, what propellant choice you use, etc. I think they can be competitive with a TPS plus aerostructures if done right, but I haven’t been able to dig that deep yet. And a lot depends on how far the physics holds together on the slow end.

    As for Air-launched SSTO, I still think it’s an interesting idea that could work pretty well with the general Boomerang approach. A lot would depend though on if I thought I could close the design, with useful payload, for one on a WK2 scale first.


  3. Paul451 says:

    I would worry about releasing the main launcher while it’s under thrust. Essentially the aircraft is falling back through the rocket exhaust. This is not like firing off a missile from a fighter, the scale of that rocket engine is much much greater compared to the scale of the aircraft.

    Can you put the first stage engines on the aircraft? Do a more conventional ballistic staging above the atmosphere?

    Which reminds me…

    Many of the issues with the size of carrier aircraft, especially if you want a decent payload (hence large rocket), come from the mass of fuel you are trying to take off with. Aircraft require much more thrust to take off in a runway length and climb to altitude than they need for flight. Fully fuelled on the ground, they need huge wings, lots of engines. Much cost. Wow. (Ahem. Sorry.) But what if you could shift some of the mass burden to a later part of the flight… So I wondered if aerial refuelling would help. Take off empty – just the weight of the carrier, second stage dry-mass, and payload mass – and add the fuel once you reach speed and altitude. Operating costs for the tankers should be in the single-digit millions even for a large launch system.

    Combined with the idea of putting main engines on the carrier-“air”craft, you’d have a HTOL semi-airbreathing first stage. But without the wing-size, or hybrid-engine development cost of most HTOL concepts. Huge cross range flexibility. Vastly simplified range safety. More flexible launch windows. RTLS from any downrange distance. Reusability of the carrier by default, particularly in testing. (You could test the thing several times a day if you want.)

    You should also be able to scale the concept up. Single rocket-engine small scale launcher (to cut your teeth), perhaps on a repurposed off-the-shelf airframe, as a suborbital test vehicle. Then add a small expendable second stage. Then develop a larger multi-engine carrier, pushing a larger expendable second stage. Then upgrade to a reusable second stage once you’ve proven the basic concept.

  4. Paul,
    In the case of a gamma maneuver separation like this, when you drop the rocket (or when the rocket drops the aircraft), you actually lose more than half of the mass of the system without losing the lift, the aircraft is if anything likely going to want to peal away pretty quickly. Definitely something that needs some thought (and which I’ll touch on in the next post), but I’m skeptical it’s a show-stopper.

    As for putting rockets on the aircraft–sure there are tons of other ways to increase the thrust of the aircraft without lighting the rocket, but I’m trying to make the argument that you *want* to light the rocket. Why on earth add an extra in-air ignition step (and the associated gravity losses of drop-and-light) if you don’t need to?

    As for in-air-fueling, ala Pioneer Rocketplane, I’m a fan of using that for an upgraded system, but I’d want a baseline system that didn’t require it.


  5. Dave Salt says:

    Nice post, Jon… good to see you’re still thinking about launchers.

    Subsonic air-launch is an extremely interesting concept that’s been employed since the late 1950’s (Cf. NOTSNIK) but has never played a significant role in space launch to date (e.g. Pegasus). I think this is mainly because it has so far only been applied to ELVs, which cannot leverage its many advantages –especially the delta-v reduction – as effectively as an RLV, where improved performance can be more effectively exchanged for improved design margins.

    Subsonic air-launch is now more relevant than ever because it’s the basis for XCOR’s orbital RLV concept that, with the exception of the HPD, incorporates all the features you discuss. However, as you say, this could eventually be incorporated as a upgrade.

    The other advantage of this concept is that it’s amenable to future upgrades such as LOx transfer or in-flight harvesting, which could also be incorporated as ‘evolutionary’ upgrades to boost payload performance.

    Having said all that, it’s clear that the vast majority of the engineering effort will relate to the RLV that launches off the back of the aircraft, which is best thought of a zero-stage or a mobile air bourn launch pad.

  6. Ian says:


    This is one of those ideas that looks good on paper but falls apart as soon as you look at the implementation. You have swept under the rug the issues that kills the idea on arrival.

    1) The gamma maneuver doesn’t just “scare the hell out of pilots”. It’s a non-starter. Nobody will let you do that. You have to make it a UAV, buy your own, and make sure it’s cheap, because you’ll lose quite a few. For example, White Knight 2 cannot do a gamma maneuver, period.
    2) Airlaunched cryogenic propellants are a bit more of a “pain in the neck”. It makes the operations such a “pain in the whole body” or, in hard numbers, so expensive that it eats up any other possible advantages immediately. So yeah, it may seem feasible unless you actually look at the numbers. At that point just add a booster stage and forget the plane.
    3) Also, cross feeding from the plane? See item #1.

    Where does that leave you? No gamma maneuver, no cryogenics. Where is the advantage?

  7. Jonathan Goff Jonathan Goff says:

    I do have some ideas for how to address these issues. But I’d love to dig more into your assertions.

    1- Why do you say WK2 can’t do the gamma maneuver, period? Is it due to structural issues, or just because it is piloted, or something else?
    2- Which particular issue with cryogenic propellants for air-launch do you think makes them a non-starter? Is it just the boiloff or something else?

    I do agree with you that if there aren’t good ways of doing the gamma maneuver, and if you can’t use cryogenic propellants (ie if you only do air-launch as a drop-and-light solid rocket), then I agree that air-launch is nearly worthless. I’m just not as convinced as you are that it’s impossible to make work.


  8. john hare says:

    I’m also curious about the problems with the gamma maneuver and airborne cryogenics. Several experienced pilots didn’t seem to have a problem with one, and several experienced rocket men didn’t seem to have a problem with the other.

  9. NA says:

    If WK2 can’t be used for a gamma fling, what other available aircraft are left, for a nanosat/smallsat class TSTO? WK1, Proteus, Orbital’s L-1011 they use for Pegasus, maybe Stratolaunch’s Roc? Else you are stuck building something new and purpose built for gamma, like a Crossbow style UAV carrier aircraft.

  10. Dave Salt says:

    Here’s a reasonably thorough study of subsonic air-launch design issues…
    …which doesn’t flag any major show-stoppers for either the pull-up manoeuvre or on-board cryogenic storage.

    Concerning the pull-up, I think this may be less of a penalty for RLVs because they will likely need wings for re-entry and landing and so can utilize them during ascent without any direct mass penalty. Since ELVs don’t need wings for re-entry/landing, the addition of a wing to support the pull-up is a clear mass penalty and so getting the aircraft to perform the pull-up before separation will obviously be very beneficial to any ELV concept.

  11. Pingback: Engineer Spitballing

  12. Jonathan Goff Jonathan Goff says:

    You want me to put wings on my RLV? Heresy!


  13. Paul451 says:

    “As for in-air-fueling, ala Pioneer Rocketplane”

    Manned SSTO? God no. Sending wings into orbit is cra-cra. SSTO is el loco. No I very specifically limited my wing’d first stage to Mach 3 (and only subsonic under jets) and even then I worry about that reentry speed for a wing’d vehicle of that size.

    And the mass penalty of the jets (and wings) is only there to allow the aerial refuelling. The air-carrier advantages (simpler ground facilities, simpler range safety, cross-range and launch window extension, etc) are a bonus.

    (As an aside, I intended my proposed refuellable rocket-carrier first-stage to be unmanned. It means developing an automated aerial refuelling system, but that’s hardly a product without secondary markets… Even current generation drones refuel while under direct human control.

    Speaking of… Have you considered aerial refuelling (both boom and probe’n’drogue systems) as a potential non-space market for Altius? A robotic “tip” that provides the fine movement to augment the cruder motion of the aircraft (and the simple vane-steering of flying booms). You can see what I mean by this clip: Developing such a system for flying booms gives you access to the USAF market (especially as they are trying (very expensively) to eliminate the boom-operator’s station at the back of next-gen refuellers.) But developing it for drogue systems gives you US Navy and the bulk of international markets.)

    Re: Boomerang.

    During the gamma manoeuvre, the carrier won’t lift away from the release of the weight of the rocket-stage as it would in a conventional air-launch release. Because the rocket is under thrust, the rocket-stage’s “weight” (net force) is entirely forward. It isn’t hanging, it’s pushing the whole carrier. By definition, its thrust exceeds its gravity-weight (indeed, exceeds the combined weight of the rocket-stage and carrier.) So the force vector is not downwards – it’s forwards, at 30-50 degrees above the horizon (or whatever your optimum release angle is.)

    At the moment of release, the aircraft will initially fall directly back (it won’t bounce up.) With the lost thrust, it will be in a soft stall, so it will automatically pitch nose-down. The net vector will be into the rocket-stage, or if not, certainly into the exhaust plume.

  14. Paul,
    The Rocketplane comment was just mentioning that they too have advocated in-air LOX transfer. Nothing more, nothing less.

    As for your gamma maneuver discussion, there are several buried assumptions that you seem to think “by definition” have to be so, that I disagree are hard and fast. Specifically, you seem to be assuming non-throttleable rockets, and aircraft whose thrust is insufficient to sustain their own forward velocity at the release flight path angle…


  15. Paul451 says:

    “Specifically, you seem to be assuming non-throttleable rockets, and aircraft whose thrust is insufficient to sustain their own forward velocity at the release flight path angle”

    I’m just going by your description (that’s all I meant by “by definition”): “As it nears the attitude where it can no-longer retain airspeed”

    Beyond that point, the aircraft is being dragged by the rocket-stage. The behaviour on release is therefore not the same as a conventional carrier dropping an unlit rocket-stage, it’s more like an aircraft losing all-engines hard while mid-climb. (Specifically, it would be like a super-manoeuvrability fighter flaming out during a tail climb.)

    (I can’t see where I’m making any assumptions about rocket throttling. If the rocket-stage is allowing the carrier to increase altitude beyond its conventional limits, and climb at 30-50 degrees, then the thrust must be greater than the weight of the combined vehicles.)

  16. Paul,
    I’ll try to be clearer, as I was apparently more ambiguous than I intended. If you take an aircraft without payload, it will have a certain angle it can reach before it can’t maintain speed (at a given altitude). If you add a heavy payload, that angle is going to decrease significantly. If that payload can provide enough thrust to exactly offload it’s gravity component in the direction parallel to the engine, you can get back to that original angle that the aircraft could have achieved without the payload. In that case the payload is neither dragging or pushing, it is just offloading its weight. If the aircraft was capable by itself of reaching the altitude and attitude of separation (or hopefully a little steeper attitude), then when you let go, the forces will induce separation not reconvergence.

    If I have the time, a free-body diagram might help.


  17. James C Cooper says:

    I wonder where you’d be able to operate such a system. At CC or Brownsville, unless the uprange launch site is only a few miles away, it’s going to put the flight track over some people, and the FAA doesn’t much care for that. Vandenburg might be better, but even that may be a problem. And while the uprange track may be okay in NM, downrange is a problem; I don’t think inland orbital launch is going to work in the US anytime soon. (The Russians seem to be fine launching over Siberia, so I’m sure you could find a track near Vostochny or Baikonur but that’s even more problematic from a legal perspective.) You could do Kwajalein or Kodiak, but those are logistically annoying.

  18. Paul451 says:

    “If that payload can provide enough thrust to exactly offload it’s gravity component in the direction parallel to the engine”

    Ah. That’s the part I missed. Sorry.

    (The image I has was: You use the carrier aircraft to carry the rocket-stage as high as you can. Once you reach the aircraft ceiling, you light the rocket-stage but continue to use the carrier aircraft as an external fuel-tank, which you jettison as soon as the benefits of the external tank are outweighed by the mass of the carrier or where the altitude will damage the carrier, whichever comes first.)

  19. Peterh says:

    If the upper stage overflying the continent is permitted, the southern California coast may be a good place (geographically) to operate launches of this profile. The second stage should already be at substantial altitude with engines operating before it crosses the coastline.

  20. Jonathan Goff Jonathan Goff says:

    Peter, James,
    For land overflight the key things are:
    1- keeping your instantaneous impact point away from populated areas, especially during high risk operations such as staging, engine ignition, max-Q, etc.
    2- having your IIP moving quickly when you pass over relatively populated areas
    3- not intentionally dropping stuff like fairings or stages, unless it is over a really, really uninhabited area.

    I haven’t run the analysis, but I would actually be surprised if you couldn’t find multiple land-overflight corridors in the US that you couldn’t use for something like this, especially if you follow rules #1 and #3.


Leave a Reply

Your email address will not be published. Required fields are marked *