Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
A 'self-eating' rocket engine could put satellites in orbit more easily (phys.org)
145 points by curtis on June 1, 2018 | hide | past | favorite | 44 comments


Here's the paper, complete with a cutaway diagram of the engine: https://eprints.gla.ac.uk/155610/7/155610.pdf


Much more sane than the phys.org write-up. The advantage of this isn't "would consume its own structure during ascent" (what does that even mean??), it's that this would have the simplicity, cost, and storability of a solid rocket with the throttleability of a liquid rocket.

(Side note: Great caption on figure 5. "Second firing. The engine was extinguished by an explosion at 142s.")


Phys.org is a low value website that republishes press releases, in this case this one: https://www.gla.ac.uk/news/headline_586616_en.html


Aaaah. Didn't know the source.

Also, I find it funny that it's the UK partner that published the press release and the paper, even though the actual research was done in Ukraine (where there's a lot of Soviet space industry legacy) and the paper seems to indicate that the Glasgow contribution was just to give them modern instrumentation.


This is pretty neat. It would make staging and reusability a moot point if the entire launch vehicle was an engine attached to an expendable fuel rod with a payload adapter on top. You could recover the engine fairly easily since it's so compact. Not sure about the Isp of polyethylene fuels though.


Aluminium is decent fuel as used in NASAs solid boosters. Now if they could just find a way to grind up the structure and stick it in the engine.


How strong would the fuel rod and motor have to be? Since they are the only structural components it seems like they would have to bear the full thrust of the rocket:

The fuel rod is not bolted to engine, so the pressure in the combustion chamber is transmits the force upward. The motor must be attached to combustion chamber so the interface between the motor and fuel rod must counter all that force to keep the rod in the chamber.


The text repeatedly talks about small satellites/cubesats.

I think that’s essential.


Where is the satellite? Is there a framework that holds it above the fuel rod, or what?


This seems to be a solid rocket booster with moving parts. Isn't that the worst of both worlds? Lower ISP than liquid fuelled rockets due to the propellant choices and higher manufacturing complexity than a solid rocket booster...


The number one thing you worry about with a launch vehicle is not ISP, it's the tyranny of the rocket equation. We have tech with VERY high ISP (ion drives, for example), but we can't use them until our vehicles hit space because of the extremely low maximum thrust.

Meanwhile the high thrust options we have are all very heavy, which means we have to carry more fuel, which means we need a bigger rocket, which means have to carry more fuel, and so on. The sum of this infinite series is finite, but it is still large.

If this tech lowers the weight of the first stage, it might actually RAISE the ISP of the rocket overall, even if it lowers the ISP of the engine itself.


>The number one thing you worry about with a launch vehicle is not ISP, it's the tyranny of the rocket equation.

The rocket equation doesn't account for thrust. It's terms are mass and ISP (or exhaust velocity).

>If this tech lowers the weight of the first stage, it might actually RAISE the ISP of the rocket overall, even if it lowers the ISP of the engine itself.

ISP is depends only on exhaust velocity. Changing the mass of the rocket cannot effect it.


> The rocket equation doesn't account for thrust. It's terms are mass and ISP (or exhaust velocity).

Only sort of true. If you naively apply the rocket equation like you're imagining, you would predict a nonzero final velocity for a hypothetical rocket that has enormous ISP but thrust less than its weight. This is true if the rocket is starting out in orbit, but it's totally wrong when you're on the ground. Starting from the ground, you also have to fight gravity, so you care more about (thrust - weight) / mass.


It is absolutely true that 'the rocket equation' doesn't account for thrust. It also doesn't account for gravity gradients, aerodynamic drag, relativity, solar wind, the cosmological constant...

It's very accurate when those things contribute little, and totally inadequate for modeling spaceflight if any are significant.

There are more complex versions that account for those things, but 'the rocket equation' is always understood to mean Tsiolkovsky's equation.


Sure, but "the rocket equation" is 100% talking about the classic ideal rocket equation, which I just discovered is also called the Tsiolkovsky rocket equation.

https://en.wikipedia.org/wiki/Tsiolkovsky_rocket_equation

If you want to be more specific, e.g. include gravity then you can't say "the rocket equation".


    > If you want to be more specific, e.g. include gravity 
    > then you can't say "the rocket equation".
You may not be able to say "the rocket equation" in that context, but NASA does. [1]

[1] https://www.nasa.gov/mission_pages/station/expeditions/exped...


Despite Don mentioning gravity along with deltaV he is obviously attempting to describe the classic rocket equation without scaring people away with the math.


The delta-v also depends on the mass fraction you can achieve. Removing the tanks might mean a much higher mass fraction, so you have the same delta-v with much lower Isp.


I don't think you would even necessarily need stages if you don't have any empty fuel tanks to discard. The engines might not be as optimized for operating both in and out of atmosphere, but that seems like a fair tradeoff.


Part of the reason for staging is the tanks but part of it is the weight of the rocket engines themselves. The original Atlas had a single tank that started off with two large and one small engines attached to it. When it had been lofted enough the two large engines dropped off and the small engine finished pushing it into orbit.


Yeah, tanks are not THAT heavy. But you are forced to use multiple engine types due to the bell nozzle.

If we used aerospikes, then it would be a different matter.


> The sum of this infinite series is finite, but it is still large.

Out of curiosity, if g was much greater, could it be infinite? What order of magnitude would g need to be for that to happen?


Yes; if the acceleration from gravity exceeds the rocket's thrust-to-mass ratio, the rocket cannot make any upward progress against gravity. The best engine listed on the Wikipedia article was the Merlin 1D with 180.1 gravities, so g ≥ 1767 meters per square second would suffice to keep it on the ground, less if you account for the fuel tanks and such.


You can always make a mass driver system (which is not limited by the rocket equation), but given how insanely hard it is at our own gravity, it would be more insanely difficult at a few times great g.

So there's definitely a quite low maximum gravity allowing practical space access.


The ability to launch from the surface is proof of it's finiteness.

As a corralary it should be finite as long as you're not inside a blackhole.

It should be asymptotic up to that point. Mathematically infinite at the horizon and then completely unbounded inside it (in the sense that it converges to infinity versus not converging at all ( see a "flat" universe versus hyperbolic)).


It's a niche application for sure. But niche applications count. I know one place it could be handy: getting ISRU products off the moon and into lunar orbit or over to a Lagrangian point.

The moon is practically made of alumina-oxide solid fuel propellant. Along with its lack of atmosphere and low gravity, it would be preferable to use the literal dirt that is everywhere on the ground to loft precious refined or manufactured products into orbit instead of more efficient yet much more valuable cryofuels.

It also would have a very good shelf life, and a brutally simple machine can have the kind of reliability people would kill for in space. For most missions you'd need something with better ISP, but for some applications it could be just the ticket.

Granted it's not currently as simple as conventional solid rocket motors, but that will probably improve as they work out the kinks. The fact that it eliminates structural mass continuously is almost like a CVT for spacecraft; the staging doesn't happen in big, clumsy chunks, it gradually sheds mass all the way down to the payload.


See the paper's description: "Autophage Engines: Towards a Throttleable Solid Motor".

i.e. a small amount of extra complexity for a big new feature.


How are they going to get a good seal around the rod at it was being consumed? That sounds like it would limit the thrust the engine can achieve.


The engine is exerting tremendous pressure against the rod, wouldn't that help?


This article doesn't answer the question every Kerbal Space Program player wants to know: what's the I_sp?


At the engine design phase, they care about a different number: chamber pressure.


Only slightly related, but are there news on the concept of not taking the energy necessary along at all?

https://en.wikipedia.org/wiki/Beam-powered_propulsion

This was hyped for a time, but then silence returned. Anyone got insider infos?


It's a giant round of caseless ammunition? Neat!!


how would the shrinking "body of the rocket" ie fuel be able to withstand/transfer the thrust/load/accelleration force safely to the rest of the rocket stack?

i mean to say its a 3 to 5-ish G load or something like that, and vibration alone would be...interesting...


One thing i saw was that the rocket can throttle, presumably by changing the rate at which the fuel rod is moved inwards.


How hazardous would igniting the fuel-shell be? How much larger do the engines have to be, if they have to be at all, to consume two different types of fuel?


It's not two different types of fuel and all rockets do this. It's an oxidizer and a fuel. For instance a common rocket fuel is liquid hydrogen and liquid oxygen. React the two and you get energy in the process... and water. So for instance most of the space shuttle's main engine exhaust was water vapor! Quite counter intuitive.


Yeah, water is hydrogen dust :)


Hydrogen rust


Hydrogen ash?


Polyethylene is more or less inert under atmospheric conditions.


It's all about self-refueling self-replicating water-burning probes. Map the Oort Cloud like that and you'll make a penny!!


I was hoping this would be about an aerospike engine.


What about the environmental impact? What are the wAste products of the plastic solid fuel and oxidizer combustion?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: