Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

High-hypersonic external scramjet aerospike engine

  1. Aug 14, 2013 #1
    Take a look at the PDF I attached to this thread.
    Note that this sketch represents the entire aircraft.
    Also the aircraft may be much longer than in the sketch
    Do you think it would work above Mach ten ?
    What about lower speeds and lower altitude ?
    Has someone designed the same engine before me ?

    Attached Files:

    Last edited: Aug 14, 2013
  2. jcsd
  3. Aug 14, 2013 #2


    User Avatar

    Staff: Mentor

    It doesn't seem very efficient to have the fuel burn external to the engine. Are you counting on the shock wave to provide some pressure to force the expanding burn against the contour of the back half of the engine pod?
  4. Aug 14, 2013 #3
    yes I hoped the incoming air would prevent the burn from expanding too much.
    I think it would work better at low altitude.
    Last edited: Aug 14, 2013
  5. Aug 18, 2013 #4
    Where exactly would you hope this engine to be mounted?
    Don't tell me you expect any living creature to be on board...
  6. Aug 18, 2013 #5


    User Avatar
    Science Advisor
    Gold Member

    This sure looks like you are going to lose an awful lot of exhaust momentum in directions that don't actually contribute to thrust. I can't imagine this would be very efficient. Allowing sufficient time for this to combust would be a real challenge as well, as the air will be moving so fast that by the time the combustion takes place it may very well no longer be near the body. This is a real problem with current scramjets as well. You would also need an extraordinary amount of pressure on the fuel in order to inject it out the nose tip against the stagnation pressure at that point. Generating that kind of pressure won't be trivial.
  7. Aug 18, 2013 #6
    This is not just the engine, it's the entire aircraft. You can see the payload in the sketch.
  8. Aug 18, 2013 #7
    and what exactly would it carry?
  9. Aug 18, 2013 #8
    Take a look at aerospike engines, they're still efficient compared to classical rockets.
    Maybe my engine is still efficient compared to classical scramjets.

    I guess that depends on a lot of variables like the exact shape of the aircraft, speed, altitude, fuel type, fuel temperature, amount of injected fuel.
    To have more control over when combustion happens, one could place small bumps on the surface at the right places, which produce small "turbulences" to mix the fuel and air faster.

    The advantage here is that the fuel is spread over the entire circumference of the aircraft.
    This means the area of contact between the fuel and air is much bigger, hence it would combust much faster, right ?

    Wouldn't the evaporated fuel inside the tank (gas) would provide enough pressure?
    Here's a similar cycle for rockets: http://en.wikipedia.org/wiki/Pressure-fed_cycle_(rocket [Broken])
    The disadvantage of this cycle over other cycles, is that it requires thick walled fuel tanks.
    But we could use the pressure of the flames outside to counteract the pressure inside the tank, so we don't have to use a thick walled fuel tank.
    Last edited by a moderator: May 6, 2017
  10. Aug 18, 2013 #9


    User Avatar
    Science Advisor
    Gold Member

    Yeah, but they are combusting their materials inside a combustion chamber to keep that reaction contained and then exhausting the resulting gas over a concave surface where the pressure gradient will keep the flow attached to the surface and direct the flow in the direction of thrust. Yours combusts externally where much of the energy of combustion may be lost in other directions and then those gases traverse a convex surface where flow separation is quite possible, letting more momentum escape vertically before it can be redirected horizontally.

    You aren't entirely off-base here, but small bumps won't do the trick for you. It is extraordinarily difficult to trip a hypersonic boundary layer. This isn't a golf ball or a 747. For example, the scramjet used on the X-43 required giant ramp-shaped boundary layer trips to try and induce turbulence prior to the engine intake and used a series of reflected shocks to further compress and slow the flow entering the combustor to allow it time to combust and even that was only partially successful. Your design cannot utilize reflected shocks of any sort, and to slow the flow down substantially to help you out using simply the conical shock coming from the tip, you would need to have a larger half-angle on the tip, which would quickly become prohibitive in terms of wave drag at the Mach numbers you are proposing.

    Not really. The keys are getting the fuel and air well-mixed and then having the flow moving slow enough that the combustion reaction will occur in a region where it is still useful (i.e. before the flow is re-expanded). Having it well-mixed will help speed the reaction along, but it only buys you so much since the reaction still occurs at a finite rate. Creating that mixture is very difficult to begin with and is one of the chief challenges in current scramjet technology. Your plan essentially puts fuel into the lower levels of a laminar boundary layer, meaning it isn't going to mix almost at all as long as that boundary layer is laminar, and on a smooth surface at Mach 10, it will likely be laminar for a very large portion of the body.

    You can add boundary layer trips as are used on scramjet vehicles now (not an easy challenge in and of itself), but that raises other problems. These would need to be located very near the tip of the vehicle (or even be part of the fuel injector) so as to maximize the distance along the body where you have turbulent flow. Unfortunately, this is also the hottest part of the body where the flow is compressed across the shock and then adiabatically brought to rest. At 100,000 feet, it would be hot to the tune of some 4600 K, hot enough to absolutely destroy these trips you would hypothetically need (and most other materials used in typical aircraft).

    It could perhaps provide some of that pressure, but remember that the gases and fuel in a rocket engine are combusting and exiting out the rear of the rocket and not having to fight against the pressure at the tip of the rocket, only the static pressure outside the nozzle which is substantially lower. For grins, let's assume that you are flying this thing at 100,000 ft altitude. The pressure at the tip of your vehicle would be some 20.6 psi (142 kPa) absolute. That's not so bad.

    I assume that your vehicle would need to make it up to that altitude though, so let's look at say 50,000 feet, the maximum service ceiling at which a B-52 could fly, which usually would launch a vehicle of this sort where it would be launched. The pressure at the tip of the vehicle would then reach roughly 225 psi (1550 kPa) absolute. That is getting pretty substantial. Maybe your fuel being heated would be able to overcome this, maybe not. If you want to launch this at sea level, the pressure is now 1,900 psi (13 MPa). This would be nearly impossible to generate I would think.

    So yes, you may be able to get it to work at one set of altitudes, but what happens then if you try and change altitudes? If you are relying solely on the fuel pressure, you might imagine flying at 100,000 feet, and having a fuel pressure from heating that is just enough to eject the fuel, but then when it comes down to 50,000 ft it is no longer sufficient. Or go the other direction and now your vehicle is spewing out so much fuel from the tip that you lose a lot of it. You need some way to control the pressure. A pressure regulator could help if one exists for such high temperatures, but then you have no means of pressure relief in your tank since it is going to continually get hotter until either the tank melts, the fuel runs out of the tank bursts from the built-up pressure.

    Plain and simple, it might work but you would need to do a pretty substantial amount of engineering work to make it work. One thing that would certainly help is to just not expel the fuel from the tip and do it somewhere slightly downstream in a fashion that it is injected into the flow at an angle that gives it some pre-existing momentum in the direction of the flow (sort of a wall-parallel injector). That would help lower the required pressure and keep you from losing a lot of fuel and give you a more even fuel mix around the azimuth of the vehicle.

    This doesn't really make sense for two reasons. For one, even if the pressure of the combustion region was enough to counteract the internal pressure of the tank, using a thin-wall tank would lead to thermal failure very quickly. Either the tank would melt or weaken to the point of rupture due to internal pressure since the internal pressure would continue to go up while the flame pressure would remain constant.

    The second reason is more fundamental. Your tank is in a region where the flow is going to be rapidly expanding, meaning that the pressure in that part of your flow is going to be extraordinarily low to begin with. Depending on your altitude and the nature of the combustion, it may even be a lower pressure than the surrounding atmosphere.

    You could maybe take a cue from rocket designers and cool the surface with the fuel, but you can't do it in the fashion you propose where you use the entire fuel reservoir as the heat sink. The pressure would just get way too high most likely. When they cool rocket nozzles with fuel, they do it by pumping the fuel in coils around the nozzle to cool it down and then the resultant heating of the fuel, by the time it reaches the combustion chamber, has caused it to evaporate and increase in pressure in a constructive way. That heat never finds its way back into the fuel reservoir, and if it did it would be potentially quite catastrophic. You will grapple with that same challenge here, meaning your tank needs to be smaller relative to the body and well-insulated from the heat.

    Finally, what about controlling this vehicle? If it is axisymmetric and shrouded in flame, it would be quite difficult to use any sort of control surfaces that would survive very long.
    Last edited by a moderator: May 6, 2017
  11. Aug 18, 2013 #10
    I hoped the pressure inside the tank would also remain constant because of the constantly "sinking" liquid fuel level. But from what you're saying there would be way too much heat.

    Ok, this means I should inject hotter fuel.

    I was thinking maybe use a small mechanism to direct the fuel flow at the nose tip into the desired direction. This would have maximal effect with minimal effort.
    Last edited: Aug 18, 2013
  12. Aug 18, 2013 #11
    I was thinking of having a circular "wing" around the location of combustion to redirect the momentum into the desired direction.
    As soon as the temperatures become unbearable for the wing because of high speed, it gets jettisoned (also reduces some drag). At this high speed the shock wave should be close enough to the aircraft surface, to do the job of containing the expanding gas.

    If we use hydrogen fuel at high-hypersonic speed, having a pure hydrogen layer has two advantages:
    1. after ignition there will be a burning water layer on the outside and a hyrdogen layer inside, the hydrogen layer is heated and expands very fast (fastest gas) towards the surface of the ship, which would make this propulsion a bit more efficient at very high speeds, at the cost of some wasted fuel.
    2. a bit less drag on the nose?

    I don't know what such boundary layer trips look like.
    The 4600 K would let the surface of the fuel layer burn immediately. I think this already hot upper layer would accelerate combustion when mixing later.
    What if we use a bigger, longer aircraft? Would this reduce those problems, thanks to the longer time it takes for the gasses to pass by?
  13. Aug 19, 2013 #12
    If behind this aircraft there's a hydrogen column which is still heated by the radiation of the surrounding flame,
    then this hydrogen column would expand mostly on our flight axis because the surrounding water vapor is a heavier gas which hinders the radial expansion of the hydrogen column. Because of the high speed of sound of hydrogen the axial expansion of this hydrogen column would still provide some thrust.
  14. Aug 30, 2013 #13


    User Avatar

    For the most part, boneh3ad is exactly correct. I do want to correct one thing though (which is irrelevant to his overall point) - pressure fed rockets use the pressure in the fuel tanks to inject the fuel into the combustion chamber, where the pressure is anywhere from a few hundred to a few thousand psi, so it actually requires quite a high internal pressure to make it work. Sure, the exit pressure of the nozzle is lower, but that's irrelevant to the fuel injection pressure (since the nozzle throat is choked). If anything, at least at altitude, the injection on this vehicle would probably be easier (at least strictly from the standpoint of the pressure required) than the fuel feed of a pressure fed rocket.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook