Japanese Rocket Engine Explodes: Continuously And On Purpose

Image of detonation engine firing

Liquid-fuelled rocket engine design has largely followed a simple template since the development of the German V-2 rocket in the middle of World War 2. Propellant and oxidizer are mixed in a combustion chamber, creating a mixture of hot gases at high pressure that very much wish to leave out the back of the rocket, generating thrust.

However, the Japan Aerospace Exploration Agency (JAXA) has recently completed a successful test of a different type of rocket, known as a rotating detonation engine. The engine relies on an entirely different method of combustion, with the aim to produce more thrust from less fuel. We’ll dive into how it works, and how the Japanese test bodes for the future of this technology.

Deflagration vs. Detonation

Humans love combusting fuels in order to do useful work. Thus far in our history, whether we look at steam engines, gasoline engines, or even rocket engines, all these technologies have had one thing in common: they all rely on fuel that burns in a deflagration. It’s the easily controlled manner of slow combustion that we’re all familiar with since we started sitting around campfires.

A diagram of the JAXA rotating detonation engine, showing the intended operation in which the shock wave from detonating fuel travels around the engine in the annular channel to continue the combustion cycle. Source: JAXA

However, there are potential efficiency gains to be had by combusting fuel in a detonation instead. This is where the combustion creates a shock wave that travels faster than the speed of sound that rapidly propagates the detonation reaction further, and comes with a huge pressure increase to boot. The key advantage of burning fuel in this manner is that there is more energy to be gained from that huge pressure increase. Thus, by releasing more energy from the same amount of fuel, engines operating on a detonation-based process could theoretically be more energy efficient.

There are several issues with operating an engine on a detonation-based cycle, however. It can be difficult to sustain a continuous detonation reaction. Additionally, large spikes in temperature and pressure from the detonation process and the associated shockwaves can easily damage or destroy parts made of even very tough materials. Thus far, engineers in many fields have struggled to tame and control detonation processes to the point where they can be used successfully.

The rotating detonation engine consists of a combustion chamber that has a annular, ring-type construction. In this ring, fuel and oxidizer is injected, and ignited in such a way to detonate the mixture. The aim is for the shockwave of this detonation to travel around the ring-shaped combustion chamber causing further detonations as it goes in a continuous cycle.

Getting the concept to work has proved difficult; despite the concept being first developed in the 1950s at the University of Michigan, it was only in recent years that engineers had successfully demonstrated a rotating detonation engine in continuous operation. A team at the University of Central Florida demonstrated a hydrogen-oxygen fueled engine in 2020, producing up to 200 lbf (890 N) of thrust in testing. The feat was achieved through careful tuning of the size of the jets that inject the propellants to get the mixture just right for controlled detonation to go on. Get the mixture wrong, and the fuel will burn in a slower deflagration, with no benefits to thrust or efficiency.

Japan’s Live Test

Unlike the experiment by the University of Central Florida, the Japanese effort involved launching an actual rocket. The test used a standard sounding rocket with a conventional engine to launch the test payload hundreds of kilometers above the Earth, with the second stage of the rocket mounted the rotating detonation engine. The mission took place using sounding rocket S-520-31, launched from the JAXA Uchinoura Space Center on July 27, 2021.

The JAXA rotating detonation engine in operation over Earth. Source: Nagoya University, JAXA

The second stage fired successfully, running for six seconds and producing 112 lbf (500 N) of thrust during that period, 56% of the Florida team’s ground demonstrator. Data collected from the experiment confirmed that the engine operated as expected, combusting its fuel in the detonation regime.

JAXA hopes to put the technology into practical applications within five years, given the successful demonstration of the flight hardware. Built in collaboration with a team from Nagoya University, the hope is to develop the technology further to create more efficient spacecraft in future. It could find application in a variety of areas, from first stage and second stage rocket motors, to being applied to deep space missions to make the most of limited fuel resources.

The technology has come a long way in the past few years. With multiple independent groups now demonstrating working engines, it’s shaken the “impossible” title that had become attached to the rotating detonation concept for half a century. Obviously, much engineering will be required to build practical engines that outperform existing designs. However, with the recent strides made in the field, there’s now a spark of hope that tells us it could be done.


62 thoughts on “Japanese Rocket Engine Explodes: Continuously And On Purpose

  1. Several years ago there was a detonation powered airplane displayed at the experimental aircraft flyin in Oshkosh Wisconsin. The plane used part of a car motor and a series of tuned pipes to create reoccurring explosions in a series of pipes. A modified car engine was used to create the proper air/fuel mixture which was charged into a tuned pipe and then ignited by a spark plug. There were no cylinders. The pipes were substituted for the cylinders.
    This engine was mounted on a Rutan style plane modified for this purpose. It looked like a Rutan Vari-easy. This was a government program and they said the plane did fly and very effectively. If you search around you can find a report on this detonation engine. Later I heard that a major disadvantage of this technology was its operating noise level. The supersonic explosive gasses are exhausted directly into the air which can not be silenced since this high speed gas is what makes this technology work. Yet it’s a very cost effective way to generate supersonic propulsion.

          1. The V1 was a pulse conflagration ramjet. People have been trying to make pulse detonation engines for a long time, starting just before WWII, but prevoiusly there had been no successful (known) tests of one. https://en.wikipedia.org/wiki/Pulse_detonation_engine for some details.
            A pluse detonation engine would be even louder than the Fiesler pulsejet, which as some wag put it, was a way of converting fuel to noise with a slight side effect of thrust.

            There were speculations that the rumored Aurora spyplane may have used a PDE, but there’s not even any good evidence for the Aurora itself, much less the PDE. With that said, I and a number of other people saw some weird airplane flying into Warren AFB in the late 1990’s that was leaving a lumpy contrail, like beads on a string, and was making a sound unlike any jet we’d ever heard, which corresponds with other descriptions of claimed pulse detonation engines. Too bad this was before I started carrying a video camera with me at all times…

          1. Interesting that they’d named the pulse detonation plane Borealis – seems like that might have been a deliberate reference to rumors about the Aurora. The timeframe looks like the rumors would have predated any design work on the plane – perhaps somebody was thinking “It would be cool if we could have what everyone thinks we have – let’s try it for real!”

    1. Can’t speak to the PDE aspect of this, but the aircraft is at the National Museum of the US Air Force in Dayton, OH. Likely on display in Hangar 4, but I last knew of it’s whereabouts prior to them opening Hangar 4.

        1. Only air is heated. The diesel is injected at the end of compression and the tiny droplets burn in the superheated air. If the fuel was pre-mixed, it would detonate and the engine would be destroyed in a short order.

  2. @Dude (putting it explicitly, since I don’t know what threading is going to do with me)

    So I was intrigued and searched for “supersonic pulse jet”, because my mental image was that of the WW2 V1 [1], and that didn’t rhyme with “supersonic” at all… and it turns out that @David Beck is onto something [2]. It’s not a pulsejet (deflagration), but a pulse detonation engine. So potentially even louder :-o

    But faster.

    Reference [2] also probably shows the Rutan mentioned by David.

    [1] https://en.wikipedia.org/wiki/V-1_flying_bomb
    [2] https://en.wikipedia.org/wiki/Pulse_detonation_engine

  3. >” by releasing more energy from the same amount of fuel, engines operating on a detonation-based process could theoretically be more energy efficient”

    My thermodynamics is rusty, but surely combusting the fuel always releases the same amount of energy?

    1. 1: you have to manage to combust all the fuel fully. (This is a design problem with scramjets because the length of the flying body may not be long enough to fully mix the fuel with the incoming oxygen so you dump unburnt fuel or have combustion outside the body of the vehicle.)
      2: there are differences in how efficiently you couple the combustion energy into producing thrust. For instance, most rocket engine designs spend some of their fuel energy spinning turbopumps to pump more fuel into the engines, which is lost energy, where a full flow staged combustion design ends up dumping all its fuel and oxidizer out the main engines, increasing the energy efficiency.
      3: similar to carnot efficiency, which is a delta T thing, and the larger the delta T the better the efficiency, you can also design rockets that have poor heat conversion into thrust through combustion chamber design. The expanding nozzle is optimized for one set of internal and external pressure conditions, and at any other condition it’s operating at lower efficiency,
      I bet there are others I don’t know about, too…

    2. You get the same amount of enthalpy/heat, but more work available – if you think of ideal heat engine efficiency, where adding heat at a higher temperature increases the efficiency, detonation means the chemistry happens at higher pressure and temperature (it’s fast enough that there’s not time for the heated gas to expand as it burns, so you get a pressure rise like you get in constant-volume combustion), so more of the energy gets converted into work / kinetic energy

  4. “hether we look at steam engines, gasoline engines, or even rocket engines, all these technologies have had one thing in common: they all rely on fuel that burns in a deflagration.”
    No. Take steam engines out of the list, please.

          1. Solar/Nuclear – Sun is burning, Nuke burning but burning
            Geothermal – Heat left over from start of Earth, it was burning, possible Nuke burning still.
            Nuke Burning – stuff is burnt, produces heat and other things, stuff is gone. Refuel put more stuff in. Burning.

  5. Did I read about a different experimental rocket engine the Japanese were developing, using pulse explosions in a manner similar to the experiments (with conventional explosives rather than nuclear) conducted under the cancelled Project Orion (not the spacecraft) in the early 1960’s)?

    1. Not exactly. The plug (spike geometry) in the combustion chamber is necessary to create the annulus that the detonation waves travel around. If you removed the conventional bell nozzle from around and added a throat geometry, the spike it would be an aersopike engine.

      My understanding is that an aerospike engine would have to use the ambient air pressure to control the expansion of the products of combustion.

      If you look at the test article (Figure 1 in the paper) there is no spike and no bell nozzle.

  6. would a lightning bolt, compressing air magnetically and releasing it, count as a detonation. think it could be made purely electric. Guess you need a quick way to extinguish the spark, so that you can have tousends of detonations a second, guess that was what tesla also was working on with the tesla valve.

    1. Short answer – no.

      Lightning goes bang because it heats the air, compressing and releasing it is ver much secondary. Magnetic forces of the average 32kA strike are not all that impressive. But heating several km of a thin channel up makes for an impressive sound source.

      It can have higher currents which would lead to more interesting magnetic effects. But not a detonation. Squeezing and releasing something isn’t a detonation. Detonation is a form of combustion which results in a shockwave.

      You appear to be suggesting a tesla coil powered pulse jet?

    1. Deflagration is subsonic combustion. That happens in car engines when performing propperly. They can detonate if running too hot, too much boost or timing too advanced. Thats bad and kills the engine. Petrol engines have to have good or high octane fuel to make sure the combustion stays within deflagration and does not transition to detonation.

  7. Actually rocket engines, that is liquid fueled ones, the idea goes back to 1926, when Goddard flew one from a corner of his Aunt’s farm. They then considered all of their research from a ranch in Roswell NM, because the current perception of his work was, well, not what he wanted.

  8. One thinks that to cope with the issues related to the impulse forces of detonation we could use a cushion of say, conflagrating mixture behind the detonating mixture. Or even using the exhaust gas of the previous detonation to cushion the next. So instead of injecting the explosive mixture into the solid area of the engine, inject it further forward in the chamber so that the explosive pressure pushes back into a sort of “accumulator” flattening the Peak pressure of the pulse into something less damaging.

    1. Even in deflagrating engines, a lot of the time combustion is separated from the solid surfaces (like film cooling in jet engine combustors/turbines) – there’s not a lot that can keep decent structural strength at those temperatures

      I’d imagine the falloff of a shockwave takes enough distance that it’s hard to get enough separation between the shock and the surfaces to make much of a difference – unlike the heat which you can shield from, a shockwave will happily propogate through whatever fluid you put in its way without losing much energy, unless you find a way to make it oblique

      In the black-and-white diagram early on in the article, it does look like the blobs where whatever it’s plotting (presumably pressure/density/temperature?) are round and fuzzy enough that they’re not directly touching a wall, so there might be some sort of film cooling going on

  9. The (ultimate) art of detonation rocket/jet/ram detonation propulsion vests in perfecting tens/millions bubble explosions a second and not kaboom/boom/boom organpipes because bubbles in nature performs/aligns/reacts in perfect harmony in accordance with the 1st and 2nd law of thermodynamics vs. organpipes driven by gears and and plugs. MMM MAYBE SOMEONE IS ALREADY DOING IT SECRETLY??

  10. Fascinating read though I’m sending most of remaining grey matter in for service. Happy to suggest F117 flew 20 years before public awareness & it’s likely development of 6gen ahead of public awareness is still secure

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.