So Close To Landing A Model Rocket On Its Tail

We’ve become so used to seeing SpaceX boosters land themselves back on the pad with clockwork reliability, that it’s easy to forget it took them a good number of attempts to get right. Inspired by SpaceX’s work, [Joe Barnard] of [BPS.Space] started working to replicate it at the model scale five years ago, with no engineering education or experience. On the latest attempt with a brand-new thrust vectoring Scout E rocket, he has gotten tantalizingly close to doing a controlled propulsive landing with a solid-fuel rocket motor.

We’ve all been thrilled to see the SpaceX rockets return to earth, landing elegantly on a floating pad. But those are liquid-fueled. The trick with a solid-fuel rocket motor is it can’t be throttled directly, which is a challenge when you need precision control to land. Thanks to [Joe]’s custom AVA flight computer and the remarkably consistent thrust curve of the Estes F15 black powder motors he used, it becomes a matter of igniting the descent motor at the right moment to make the vertical velocity zero at touchdown. However, [Joe] found that the time between sending the ignition signal and when peak thrust is reached was inconsistent, so he had to work around that. He did this by controlling how much of the thrust is spent in the vertical direction, by vectoring the motor side to side to spend some trust horizontally.

View from rocket of the ascent motor falling away immediately after being ejected

In this attempt, the rocket tipped over on landing due to excessive horizontal movement at touchdown. Joe tracked the cause down to a weak GPS signal caused by antenna position and a possible bug in the Kalman filter that fuses all the sensor data for position and velocity estimation. Thanks to incredibly detailed telemetry and logging done by the flight computer, data from every launch are used for future improvements. We are looking forward to the next flight in a few weeks, during which [Joe] plans to tune and test the control software, among other minor improvements.

Almost every single part of this rocket is a display of engineering ingenuity. The landing struts are designed to absorb as much impact as possible without bouncing while being light and quick to deploy. The ascent motor is ejected simply by moving the thrust vectoring mount to one of its extremes, allowing the descent motor to drop into place. The rocket also features a complete emergency abort system with a parachute, which can be activated manually, or by the flight computer if it calculates that landing isn’t feasible. We already covered [Joe]’s latest launch pad, which is a very interesting project all by itself.

44 thoughts on “So Close To Landing A Model Rocket On Its Tail

          1. Having worked in a company run by a rich engineer, and seen it taken over by a rich non-engineer, there’s no way Elon isn’t good technically. You can’t hire awesome people consistently without having skills yourself.

          2. Good as those he hires is rather unlikely at this point. Once you have that level of funding you can get the absolute best in every field – Where he is I have no doubt smart and educated around the subjects he fronts, if he was really that good at any one of them he most likely would end up an employee not the boss (as if you are really that good at any facet of engineering/science type stuff it tends to be what absorbs your focus – you don’t want to be the boss as its full of other crap you are not that interested in)..

            Personally I find him a somewhat annoying, but no doubting his companies produce good stuff that evolves rather fast, and that can’t happen if the boss only cares about his personal power and paycheck (and thus the share price above all else)… So he gets serious respect for making it happen here…

          3. He’s literally not an engineer. Elon has a dual BA in economics and physics and his only professional work before hitting it big with PayPal and becoming an angel investor was in web design. He pretends to be an engineer, but he’s basically just a “tech enthusiast.”

          1. I look at it from a different angle.

            Let’s forget Elon Musk for a moment. (ignoring Twitter helps with this!)

            I think it’s wonderful that a company like SpaceX exists, to employ the best and brightest engineers of this generation. If it wasn’t for SpaceX, more of them would go and work for the military, or for Facebook. Then what would we have? More weapons and more internet spying. I would much prefer to see a future with more interplanetary spacecraft than either of those things!

        1. Miss.

          I’m Russian. You know, from that crazy Russians who could build a turbojet drone running on vodka with an axe, saw and file in a cold garage somwhere in taiga with bears. And we forgot about space and rockets more than many will ever learn about. :)

          In any case, eventually, space technology will reach DIY level. It is inevitable. And we have to make it cheap, safe, simple, reliable and repairable in field conditions, because Zefram Cochrane will need something to test his warp drive on. :)

        1. Oh, a basic solution is probably published as an example in some Mathworks app note or similar. As long as you’re landing a spherical cow with perfect actuators and sensors, the problem is probably solved in published form. (probably something available from DTIC, no less) What isn’t going to be open source is the details of how you make it work with real actuators, real sensors, and real motors.

        2. You don’t need to solve that problem if you have powerful processor on board. Just correct the thrust and vector on the fly to get the path you want. It’s on the level of quadrocopter firmware.

          1. That’s the thing with the rockets being used they don’t have a throttle – it burns till its done – down a reasonably constant thrust curve. So hes having to use the vector to control the lifting thrust as well meaning its all about knowing in advance, solving the problem for this rockets motor and body under current environmental conditions, as you can’t just correct one with his design – to change one changes the other.

            That said I do think thrust ‘reversing’ plates could be implemented into the tail of his rockets to make it possible – then it becomes a question of just how much diversion of the thrust is wanted and how rapidly you need to be able to control it (I would think even as little as 10% of the exhaust diverted would be enough really as he doesn’t need major throttle control its close to working as it is – this would just give the computer that extra little bit of correction for imperfect burn (Though diverting all of it could be done – when the rocket is at very low velocity near the ground but still has a little burn time to go it can just burn it out safely). I would think the reversing plates could also be used as thrust vectoring for the final stage – have 3-5 flaps to divert the thrust and you can deliberately gain some lateral force by selectively opening just one side to whatever degree you need.

  1. How sophisticated can hobby rockets get before sharing detailed technical information on the Internet runs afoul of ITAR regulations and such? I would hate to see this brilliant man go to prison for something as simple as posting his code, and having foreigners download it, therefore running afoul of export laws. Exporting plans and code for sophisticated rockets is horribly illegal.

    1. Joe’s been very careful to not share the special sauce of the code that controls this thing anywhere, and does not post turnkey designs or STLs either. At most he’s got the restricted to US only sales Signal kit but that’s preburned with code and you’d need to be pretty high tech level to extract the binary, then reverse.

  2. It – of course – depends on state and legislation. I rememeber the story, when PGP was exported in printed book form and then typed in again to circumvent export restriction laws. This was covered by free speech rights. AFAIK information other than program code should be mostly free and unrestricted, except if gained the info under an NDA.

    1. From an export controls standpoint, probably all the detailed design and implementation information is controlled. This is one step from a guided missile. The “printed in a book” doesn’t make it not controlled, unless it is of a general nature and not specific to the application (textbooks are the usual example).

      Whether it’s source code or not makes no difference. As noted above, no turnkey designs or files are distributed either. If he wrote the control laws on a white board and took a picture, that would likely be export controlled information.

      Also note that there’s TWO sets of export control rules to worry about: ITAR and EAR. Lots of things aren’t on the US Munitions List, so they aren’t ITAR, but are still controlled under the EAR.

      Whether it *should* be export controlled is a whole ‘nother can of worms.

  3. Just like to point out that liquid fuel rockets from Armadillo Aerospace ( and the DC-X ( both did this.

    The real trick is solid fuel motors do not give you variable thrust or burn time. Timing to make that work must be tough.

    Also if you included rocket powered props, ROTON ( and Peroxide Propulsion ( also made this work.

    1. And that’s what he’s figured out. The uncertainty is the time between “start” and “thrust begins” – the actual burn is quite consistent. So to “throttle” it, he wiggles the trajectory so that the thrust vector is directed off to the side, and the “vertical component” is reduced by cos(theta). It’s very clever.

  4. Oh great googlies, ITAR! Making something LAND is THE OPPOSITE of making a weapon. We already have things that land. They’re called drones. Predators, Reapers, and the ones you buy at Walmart.

    Of course we’re all screwed if Wherevistan gets our model rocket technology. Newsflash: People have real missiles and nuclear weapons.

    1. The issue isn’t so much whether this is useful, it’s whether they come after you for not following the regulations. There have been plenty of cases where the regulations prohibit something that is not exceptional because technology moves faster than regulation. Motorola 68000 processors in video games for instance. GPS receivers that work at high altitudes.

      There’s a long history of high power rocketry enthusiasts staying away from guidance systems for just this reason, except for systems that basically guide the rocket “up”. It attracts a lot of attention from regulators, and if you are interested in building rockets that go high, you’ve got enough problems without raising the issue of export control problems.

      The few folks who do it are very careful, and I’m sure they’ve had conversations with people who care about such things.

        1. 1990 – Commerce department decontrols IBM AT (i.e. 80286) and Motorola 68000 processors.
          “Exchange Opportunities : Hearings Before the Subcommittee on Economic Stabilization of the Committee on Banking, Finance and Urban Affairs, House of Representatives, One Hundred First Congress, Second Session, February 14 and 15, 1990 · Volume 4”

          Computerworld, 31 July 1989. page 66, “Commerce to remove bars to PC AT trade”


          1. So you couldn’t get PC clones and Apple computers outside the U.S.? I remember it differently. A few bureaucrats stamping their tiny elfin feet doesn’t constitute…. anything. North Korea has missiles, Pakistan and Israel have nukes. Calm the heck down. Johnny Model Rocket is not helping them, and you think North Korea and Pakistan and Israel can’t order their own model rockets?

            Know who helped the Nazis with their rocket program?

            Robert Goddard. Perhaps you’ve heard of him.

            Know who helped the U.S. with their rocket program?

            Werner Von Braun. Perhaps you’ve heard of him.

          2. The concern I would have is NOT for the proliferation of the technology. It’s that the rules get update slowly (more slowly than technology advances) – and you could, without realizing it, wind up breaking the rules, and getting yourself into a whole lotta hassle. Just because technology X or Y or Z is available somewhere does not mean it’s legal to export it from the US. And, “publishing on the web” is an export of technical data. If it happens to be technical data for something on “the list”, life can be really miserable, whether or not any actual damage to national security has occurred.

            AMSAT, a largely volunteer, and definitely non-profit, organization, who build satellites for amateur radio, got into just this sort of trouble.
            Stanford University doesn’t do cube-sats (or anything space related more advanced than TRL 3) for just this reason.

            It’s easy to stumble over the line, and unless you’ve got some serious resources, getting out of that tar pit is difficult.

          3. OKay, when you put it in terms of “hassle” I can see your side. Still didn’t interfere with the Clinton Loral deal, because some animals are more equal than others. He built on the Nixon principle: “When the President does it that means it’s legal.” And the Reagan era stuff and every other President, don’t mean to make it partisan because they’re all guilty.

            But again, Macs available all over the world, so these laws (like secrecy laws in general) are just to oppress U.S. citizens, not to keep stuff from the Soviets or whoever.

            Still, I don’t think Johnny Model Rocket has anything to worry about, but in a country where the BATFE once ruled that a shoestring is a machine gun who knows?

            Years later they changed the ruling to “A shoestring is SOMETIMES a machine gun.” That ruling is still in place.

  5. 2 suggestions, from the video it seems that when the rocket gets very close to the ground the thrust “increases” (I know it doesn’t, maybe it some form of wing in ground effect) so he should take that into account and secondly since he’s landing on grass maybe just put some spikes on the end of the landing gear so when the rocket makes contact it will stick in with more force and (hopefully) avoid bouncing or toppling. This thing is amazing though and I bet he did it for quite a bit less money than Musk did.

  6. The reliance on a solid fuel rocket makes true both challenging and impressive.

    NASA pulled off a reverse rocket landing 51 years ago with 2k of RAM and a steady hand on a joystick. No point even mentioning Elon here

  7. Interesting video…good science in this video however there is a piece of the puzzle that is missing…once you have the piece you can make it land pretty close to perfect….I won’t tell you what it is…. but i will give you a hint…what do you do when you are balancing your weight on one foot? Shift your weight right…what if you don’t have a gimbel -now you should be able to solve the problem with all your equations and such…and knowing the burn rate of the fuel is critical to solving the problem..its just a question on how you apply that knowledge….If you want more than a hint..message me…

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.