Rocket Diffused Ram Jet

[saabstory88]

Hello all,

Seeing as there are an abundance of intelligent people who are involved with KSP, and in turn, have a fascination or career involving rocket science and aerospace, I was hoping to have some assistance with a theoretical design. I have been researching the topic of Ramrockets, and I had an idea for something resembling a cross between a NASA style RBCC as described here: http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19970017381.pdf and a more mundane ducted rocket. From what I have been able to research about ram jet physics, the greatest concern at supersonic velocities is the variable geometries of the diffuser assembly required to achieve a subsonic fuel air mixture.

This got me thinking about whether the boundary layer generated by traditional rocket exhaust inside of a ducted nozzle could be varied, by changing thrust, to create ideal conditions for additional combustion. If additional fuel injectors could be added to the duct assembly, could additional fuel be burnt at higher efficiency, being mixed with the incoming air? In the NASA case study, and related designs I have seen, the rocket motor has always been positioned in the aft section of the combustion chamber, where the flame trap is usually located, and variable intake geometry, via active motorized means, has been used. By relieving mechanical actuators and flaps from the responsibility of changing the geometry, could significant weight and complexity be saved?

I have been working up to putting some small scale hybrid rocket motors into practice, and understand most of the principles involved, but the fluid dynamics required to actually design a ramjet shroud, are a little beyond me at the moment. I would love to eventually upgrade a motor to incorporate such a system, but then again, I do not have a supersonic wind tunnel to test it. I have not yet delved into the realm of trying to configure any of the CAD software I have available at work for this task, but even if anyone has any direction on the software portion, that would be greatly appreciated.

Here is a rough diagram of what I would envision this system to look like.

This image is only meant to demonstrate the idea, and is obviously not based on any calculated geometry.

Thanks!

[shynung]

That's an air-augmented rocket. It's not a new idea, but interesting nevertheless. One of the great things about them is that you can install the air ducts on any type of rocket (even solids), and it would still work, boosting thrust and I_Sp whenever in an atmosphere.

All that using a substantially simpler device than the SABRE's precooler.

[saabstory88]

I'm curious about how you calculate the total effect on ∆V when using this system, especially when using additional fuel. I know I can find literature about how to calculate the various cross sectional geometries of ram jets, both sub- and super-sonic. Do i treat the rocket and the additional jet thrust as two separate systems due to the boundary layer? Assuming the main engine burn time is less than the time the rocket will take to reach peak operating altitude and air speed, it seems pretty easy to calculate the amount of extra fuel required. I'm really talking about something beyond using passing air as extra reaction mass for the system, and actually burning additional fuel, stored and injected separately. If I could determine the Isp and thrust of each system, could I simply factor those proportionately to get a total system Isp assuming uniform burn time between the two?

The term, Air-Augmented rocket, appears to only refer to the concept of using thermal energy from the exhaust plume to heat captured air, not actually burning additional fuel. The non mechanical designs I have seem thus far also only increase efficiency of the system at subsonic velocities. I am trying to ascertain whether the exhaust boundary could be controlled to the point where it could allow additional fuel to be burned at supersonic velocities as well, without the use of any actuated mechanical means.

[shynung]

I'm curious about how you calculate the total effect on ∆V when using this system, especially when using additional fuel. I know I can find literature about how to calculate the various cross sectional geometries of ram jets, both sub- and super-sonic. Do i treat the rocket and the additional jet thrust as two separate systems due to the boundary layer? Assuming the main engine burn time is less than the time the rocket will take to reach peak operating altitude and air speed, it seems pretty easy to calculate the amount of extra fuel required. I'm really talking about something beyond using passing air as extra reaction mass for the system, and actually burning additional fuel, stored and injected separately. If I could determine the Isp and thrust of each system, could I simply factor those proportionately to get a total system Isp assuming uniform burn time between the two?

I don't know enough about it to have any opinion. Hope other users will help you there.

The term, Air-Augmented rocket, appears to only refer to the concept of using thermal energy from the exhaust plume to heat captured air, not actually burning additional fuel. The non mechanical designs I have seem thus far also only increase efficiency of the system at subsonic velocities. I am trying to ascertain whether the exhaust boundary could be controlled to the point where it could allow additional fuel to be burned at supersonic velocities as well, without the use of any actuated mechanical means.

On oxygenated atmospheres, one could theoretically run the engine rich, and let the atmosphere burn the rest of the fuel in the ducts. At supersonic speeds, where one wanted the engine to run as a ramjet, it's possible to simply burn an extremely fuel-rich mixture in the rocket, and let the residual heat ignite the excess fuel in the ducts.

Of course, in a non-oxygenated atmospheres, all that would be moot. However, that's where air-augmented rockets excel; they don't need external oxygen.

[GoSlash27]

saabstory,

This looks very much like the sort of thing I used to come up with when I was a teenager back in Pittsburgh. Small world!

IMO the answer is yes and no

What you basically have here is an afterburner. good for thrust, but bad for efficiency.

Now... if you were to use the diffuser not to slow the air, but rather compress it, I'd imagine you could inject fuel into that and have very high efficiency. The trick would be keeping the shock wave stable over a wide range of mach numbers. I'm not sure how badly that would screw up the Isp of the core rocket, but it can't be good.

Suppose you were to start with an aerospike design. That works by creating a small pocket of vacuum in the center of the exhaust stream, thus making the exhaust flow laminar. At low speed, the aerospike is doing all the work. At supersonic speed, your baffle introduces a shock wave behind this, creating a high pressure combustion chamber. This reduces (or more properly wrecks) the efficiency of the aerospike, but introducing fuel to your high pressure pocket of air would probably make up for that.

I hope I'm describing it properly. No diagrams

Best of luck!

-Slashy

Edited July 6, 2014 by GoSlash27

[saabstory88]

saabstory,

This looks very much like the sort of thing I used to come up with when I was a teenager back in Pittsburgh. Small world!

IMO the answer is yes and no

What you basically have here is an afterburner. good for thrust, but bad for efficiency.

Now... if you were to use the diffuser not to slow the air, but rather compress it, I'd imagine you could inject fuel into that and have very high efficiency. The trick would be keeping the shock wave stable over a wide range of mach numbers. I'm not sure how badly that would screw up the Isp of the core rocket, but it can't be good.

Suppose you were to start with an aerospike design. That works by creating a small pocket of vacuum in the center of the exhaust stream, thus making the exhaust flow laminar. At low speed, the aerospike is doing all the work. At supersonic speed, your baffle introduces a shock wave behind this, creating a high pressure combustion chamber. This reduces (or more properly wrecks) the efficiency of the aerospike, but introducing fuel to your high pressure pocket of air would probably make up for that.

I hope I'm describing it properly. No diagrams

Best of luck!

-Slashy

Your completely right about the aerospike approach, which is why I would not choose to go that direction. A traditional rocket bell would function much better for this application, as altitude increases, the exhaust plume tends to expand a predictable rate. What I think that I am not communicating properly, is that with a properly shaped engine bell, the supersonic boundary between the surface of the exhaust plume, and the incoming air is of a sufficient pressure that the plume is essentially a solid mass in comparison. Air compressed in this manner does not actually mix wit the exhaust plume until it is clear of the shroud, so it's no really an afterburner, which injects fuel directly into an exhaust plume.

So the design goal should be geometry that allows us to treat to rocket component and the ramjet component as two SEPARATE systems. Our shroud length should be less than the distance of the coherence of the exhausts supersonic boundary layer. This will ensure that the exhaust does NOT MIX until it is clear of the engine. For the following example, imagine a system where you have a conventional rocket, with several variable geometry ram jets around its axis.

Let us take a design that call for a final thrust of T(f). The rocket portion should be designed to produce up to .75 * T(f), the subsonic ramjet configuration shall be designed to reach at minimum .5 T(f), and in supersonic configuration, at minimum, .25 * T(f). The rocket shall be throttled at liftoff to maximum power, and remain there until enough velocity is achieved to ignite the subsonic ram jet configuration. The main engine is throttled such that it is producing equal thrust to the ramjet(s). Thus during this phase, the Isp of the system can be given by using the ratio of thrust that each engine is contributing to the system, to find the weighted average. So for our example, a rocket with several radial ram jets, we can take an average ram jet Isp of 2000s, and average that with a typical rocket Isp of 250s, for a total system Isp of 1125s during subsonic flight. When the rocket achieves supersonic flight, the ram jet geometry is changed such that the engine can operate at those speeds. Because this requires changing the ratio of inlet area, to diffuser area from greater than 1, to less than one, this generally causes the specific impulse of the supersonic mode to be lower in a variable geometry design. However, at this point the Isp of the rocket has usually increased. So when we take, for example, a slice of the Isp average at high altitude and speed, the ram jet would be operating at say, 1500s and the rocket at 300s. However, in our example, the trust ratio has changed in favor of the rocket. So recalculating, weighting the average by proportion of contributed thrust, we still arrive at an Isp of 600s for supersonic operation.

But we don't simply have to strap variable geometry ram jets to a rocket to gain this kind of efficiency. What I think you fail to understand, is that air passing near a stream of supersonic exhaust gas on the same vector can not enter the stream. It might as well be a solid. It will eventually lose pressure and velocity and allow air to mix, however this does not occur close to the engine. Thus if we design a shroud that is less than the distance before the air starts mixing with the exhaust, we essentially have a variable size SOLID at our disposal. This is why the ram jet component and the rocket component can be treated just like our previous example of literally having to different kinds of engines on the same vehicle.

See the following diagram below, keeping this in mind. Also keep in mind that as altitude increases, the exhaust from any given engine bell geometry tends to increase in diameter. We use this effect to allow for a change in intake area, to combustion area, inside the ram jet shroud.

[sgt_flyer]

don't forget you'll need at one point (if you're going with liquid fuels) to drive your turbopumps (for both cooling and engine feeding). the current way to drive turbopumps remains using a gas generator + turbine, which will eat some of your ISP / and add a bit of weight to your system.

[GoSlash27]

Ahh... I see what you're getting at.

You're effectively using the original rocket exhaust as an obstruction in order to squeeze the ambient air into a combustion chamber. How do you propose to fight the reversion? I'd expect the ambient air to create a high pressure pocket at the inlet, forcing the slipstream to flow around your chamber rather than through it.

Regards,

-Slashy

[saabstory88]

I am trying to trust that the ram jet design handbooks I have been reading are correct in their math. You do actually divert a large amount of air in real sub, and supersonic ram jet designs, around the engine, setting up the geometry such that only the right amount of air is allowed into the intake. It's all a matter of forward inlet area, versus rear of inlet frontal area. That ratio determined your flow into the engine.

I am working on some data sets to try to determine how much efficiency can actually be gained on top of a standard hybrid rocket. I'm working on the assumption of a 90 second total boost time, with a desired final velocity of 1200 m/s, and an altitude of 40-50km. I picked this number, because it's pretty much the starting point you would need to take a reasonably sized (m < 600kg) hybrid rocket body second stage close to orbital speeds. Speaking to a second stage, the fractional mass ratio would allow you to maybe, MAYBE, put a cube sat on top of this thing.

[GoSlash27]

My understanding with ramjets is that that's not the way that they work. They actually create a shock wave and it's the shock wave that compresses the intake air.

*scratching head*

[Nothalogh]

Sorry for the necrobump, but I just had an idea.

What about MIPCC style water injection into the air-augmenter?

[SciMan]

Sounds like it would work, but intuition says that wouldn't be as good as something that can burn in an oxygenated atmosphere.

[Buzz light fear]

On 7/6/2014 at 1:13 PM, saabstory88 said:

 

You could just use an engine that doesn't burn it's fuel completely like a H2 and O2 fueled engine and then the leftover fuel could be burnt in the ramjet adapters

Edited July 15, 2017 by Buzz light fear

[magnemoe]

https://www.nextbigfuture.com/2017/07/usa-and-china-getting-close-to-flying-hypersonic-demo-airplanes.html
This point out an ram/ scram jet with an rocket engine inside, it will simplify things a lot over an pure engine.