SP-22 Constant Flow engine

[Bruno Ogorelec]

Guys, I have to confess to something: I do not regard the pulsejet as such as a serious jet engine.

It can serve as one in dire need, but frankly, the only thing it really does right is produce a great amount of hot gas very simply and cheaply. Everything else about it is problematic for one reason or another.

I have long become convinced that, if we are to use the pulsejet for propulsion, it will be part of a more complex system, which will address some of the glaring shortcomings. Yet, the system must not be too complex, lest we kill the best things about the pulsejet – low cost and accessibility. I hope I have arrived at one of the many possible solutions that can make the pulsejet more usable than it currently is.

One of the problems with the pulsejet is pulsation. OK, I know a pulsejet needs to pulsate in order to work, but pulsation is a pesky thing, too. Among other things, it produces vibration. With a big engine, this can be a very serious problem. A 100-lb thrust pulsejet will deliver a 100-lb punch to the airframe it is mounted on approximately 100 times a second. Not good for the airframe. Even less good for the pilot.

Also, the pulsejet has a problem of the absence of inertia. It can quickly rev up and down because there is no heavy mass in motion within the engine. This is nice, but it also means that it cannot sustain running through even a single bad cycle. It just cannot misfire-but-work, the way other engines can. One bang goes wrong and the engine stops dead. Instantly. There’s no flywheel to carry it through a brief bad patch. It just stops. Not nice if you’re flying.

The Constant Flow engine (my internal designation is SP-22) addresses the above problems.

Thrust augmentation lies at the heart of this engine. The toroidal central augmenter duct has two tangential intakes and two tangential outlets. Two pulsejet combustors blow into this system. If they fire alternately, the two combustors will generate a constant flow in the ring duct. As one exhaust stream abates, the other will rise and vice versa. The flow will always be in the same direction and of the same intensity, given the same throttle setting on both engines.

The output through the two tangential nozzles will be equal. The nozzles rotate around the duct axis, so that thrust can easily be vectored around 360 degrees with no loss in thrust.

Should one engine quit, the other will continue producing symmetrical thrust on both nozzles, only it will be a little less than half the original thrust. Not ideal, but incomparably better than losing all power instantly and irreversibly.

The engine intakes are coupled to a similar toroidal ring manifold with tangential intakes and outlets. This means that the constant flow will also be maintained in the intakes. It also means that there will be a constant ram pressure in the intake manifold.

Namely, the fresh air sucked into the manifold will not stop when combustion in a combustor starts, but will simply continue swirling in the ring. Each engine will suck from the constantly swirling air in the manifold as much air as it needs at the moment it needs it. Unlike the conventional pulsejets, it will not be sucking on motionless air even when stationary, but will all ways have moving air blowing into the intake. Constant small ram pressure, even in stationary operation.

But, what about blow-back through the intake, you may ask. Well, the answer is simple. I intend to use engines with no blow-back of gas. There are valveless pulsejets that do not blow back, or do so only very modestly – weakly enough that even a very small amount of ram pressure will stop it. My Swirl Can (SP-06) engine belongs to that species. So do Rossco’s inverted intake engine and his most recent invention, of which I must not say a word. The intake of a Reynst pot is yet another irreversible pulsejet intake duct. Even an ‘ordinary’ Schubert can serve with an intake long enough.

So, no blowback. Oh, there’s a pressure shock that travels through the intake, but it is not coupled to notable mass movement and it simply dissipates harmlessly in the ring manifold, acting perpendicular to the flow.

I have thus addressed the issue of pulsation, keeping it in the two places where it is needed – the combustor, where it increases the mean effective pressure in the cycle and thus the engine efficiency, and the augmenter intake, where it increases the pumping efficiency. Pulsation is not needed elsewhere (indeed, it can be harmful) and it does not happen elsewhere.

I have provided constant ram pressure in the intake manifold, which can only benefit efficiency, not to mention usefully broadening the span of conditions under which the engine will work at all.

This layout also provides constant, non-pulsating thrust through two vectored nozzles. Partial engine failure does not kill the engine but only reduces its thrust by half, while maintaining symmetry.

The plan is to encase the flat assembly within a streamlined elliptical-section shell, which is also elliptical in plan, something like a very fat and stubby Spitfire wing, with vectoring nozzles sticking out of the tips. The ejector action of the thrust augmenters will produce a constant low pressure inside this cowling. Ai intakes can be designed into the cowling at strategic places, so that there is constant movement of cooling air through the inside while the engine is working.

It’s a kind of a much fancier version of the Pegg-Makowski twin-pulsejet pod, but I think this is much more effective.

The color sketch shows the engine parts packed rather loosely, for the ease of drawing, but the whole thing can be packaged much more tightly with some forethought and tinkering.

It may look complex, like a passionate embrace of two octopuses, but it is not really complicated at all. There are two simple pulsejets, a ring of steel tube with four outlets, and a bigger ring of carbon fiber duct with four outlets. That’s it. Nothing to it.

You can do a lot of stuff with this pancake. We have talked about surface-skimming ground effect craft, for instance. Well, put one of those engine packs out in front, like a canard, and it will provide both thrust and augmented lift.

Perch it on top of an aircraft fuselage and it will provide twin-engine thrust. Use two in tandem and, who knows, you might have vertical takeoff with transition to horizontal flight, Harrier-style, ha-ha-ha-ha-ha…. If they are REALLY powerful.

Put one on top of your backpack and you have an instant jet-powered flying belt.

Mount one below a gyroplane rotor (use a multi-blade rotor for efficiency) and use vertically rotated nozzles to spin the rotor to high enough speed and then rotate them backwards for takeoff and cruise thrust.

I am sure you can come up with additional ideas. Have in mind that you can rotate the nozzles independently to create all kinds of power vectors…

Have fun.

[larry cottrill]

Bruno -

Congratulations - that is really very cool. The only thing I see at all problematic is the potential resonance of those long, long intake sections. But, once the length of these is adjusted properly, that certainly ought to work.

You ought to be able to find a lot of motorcycle (and probably many other) exhaust sections that would help make a small working model of this.

Very nice!

L Cottrill

[Bruno Ogorelec]

The only thing I see at all problematic is the potential resonance of those long, long intake sections. But, once the length of these is adjusted properly, that certainly ought to work.

Thanks, Larry.

The intakes need not be very long. The sketch shows them as long because it was easier to draw that way. Imagine a much larger diameter intake torus (ring) and the intakes suddenly become shorter. Or, move the combustors so that combustion chambers are closer to the centerline. Etc. etc. You can shuffle the parts around a lot without changing the basic nature of the layout much.

One particular way to make the whole thing more compact, quiet and possibly even more efficient would be to pack the combustors within the pale green central duct. I just didn't want to go too far in those component reshuffles before introducing the basic concept to the public. It looks complex enough as it is.

If I catch some time, I may make a simple sketch showing the combustors secreted within the augmenter.

[multispool]

Hi Bruno,

Intriquing design!
I wonder how the pulse jets could be arranged to run out of phase with each other?
Maybe the pressure variation within the intake torroid might auto syncronise them to run out of phase!
Don't be too hard on the pulse jets quiting capabilities! Gas tubines only need a bubble of air in the fuel line to cause flame-out!

[Bruno Ogorelec]

I wonder how the pulse jets could be arranged to run out of phase with each other?

Thanks, Multispool. Your nickname indicates that you know what you are talking about, heh-heh-heh...

Experiments conducted at the University of Calgary about twenty years ago or so showed that you can make two pulsejets run out of phase by coupling their inlets, or by coupling their tailpipes or both. It has not been a big problem. I am betting that the coupling of intakes with a ring manifold will do the trick.

In fact, the tailpipes are also coupled in a manner of speaking, even though there's a decoupling at the gap between the tailpipe end and the augmenter inlet. If the two combustors synchronize their operation out of phase. that decoupling will also lock in, of perhaps not as strongly as the intakes.

It has long been my conceit that the two identical pulsejets with identical throttle settings will tend to synchronize out of phase when running at a proper distance from each other even without physical contact, simply because doing so will allow them to run at their energy minimum.

Just like two identical grandfather clocks will eventually synchronize even if they are situated in the opposed corners of the same room, if left to themselves.

[Bruno Ogorelec]

I have meanwhile edited the main color sketch and uploaded the new version. The old one had an extraneous line on the central ducting in the side view, suggesting a wrong layout. It is correct now.

Here's a sketch of the principle of the intake ring manifold, which may not be quite clear from the sketch of the entire engine and/or my written explanation. It also shows the tangential intakes on the SP-06 'Swirl Can' combustors much better.

[multispool]

Your nickname indicates that you know what you are talking about,

I'm not sure about that, just trying to sound intellegent!
Main amateur interest is miniature gas turbines.

[pezman]

I did a lot of work on synchronization in chaotic systems not so long ago. Synchronization of dynamical systems is absolutely fascinating and the mathematics are often very elegant. If anyone is interested, look up "synchronization manifold" in google. A closely related topic is the Ott-Grebogi-Yorke (OGY) control method. The mathematics might initially seem daunting, but when you get it, all the complexity melts away and you're left gawking at how simple and elegant these systems can be.

There is normally a range of mutual coupling strengths over which two independent systems will synchronize -- i.e. synchronization will not occur if coupling is too strong and it will not occur if coupling is too weak.

Oddly enough, the coupling characteristics of a two component system, i.e. the coupling strength at which "capture" begins and at which it ends, is often enough to characterize synchronised systems that consist of more than two components. For example, it can tell the range of couping strengths that will successfully synchronize "N" components, exactly how many components can be synchronized in the limit and so forth.

At any rate, neat idea (as expected, given the source).

[Mark]

Hi Bruno,

Intriquing design!
I wonder how the pulse jets could be arranged to run out of phase with each other?
Maybe the pressure variation within the intake torroid might auto syncronise them to run out of phase!
Don't be too hard on the pulse jets quiting capabilities! Gas tubines only need a bubble of air in the fuel line to cause flame-out!

And my little Logan with a fat silicone fuel tubing will feed on the methanol flow like a duck quacking wildly as it is running about. It nearly dies out and then speeds up to full grease and then all over the board with a spongy fuel line. It was the most funny sound I've heard a pulsejet make. It could miss, cough, sputter, rev, whine, all in a blink. It was comic.
Mark

[Mike Everman]

Nice drawings, Bruno. Are the "roundabouts" less for synchronization, and more for thrust distribution in event of one failing?
Granted, it's out of scale, but I have doubts that you can pull off the length of intake required for that end of it. Unless you force air in.

[skyfrog]

Congrats Bruno, You finally got the configuration settled down. Hope to see some engines running soon.

[Bruno Ogorelec]

Nice drawings, Bruno. Are the "roundabouts" less for synchronization, and more for thrust distribution in event of one failing?
Granted, it's out of scale, but I have doubts that you can pull off the length of intake required for that end of it. Unless you force air in.

A good question.

The first priority is thrust symmetry in single engine operation.

Flying with a single engine is just not realistic, given that we won't really be flying without having some 200 lbs of thrust at our disposal. I can't see anyone flying with a single 200-lb thrust pulsejet. It's got to be at least two engines with 100 lbs each (or a greater number of smaller engines). A twin-engine layout inevitably raises the issues of symmetry in single engine operation. Well, this pancake offers what’s needed.

The second priority is constant ram pressure. Mike, I think you are underestimating the ability of a pulsejet to pull air. Remember Reynst’s experiments with pulling against negative ambient pressure and his idea of having a drive turbine in the intake, driven by fresh air?

I would start with the intake proper (the part between the combustion chamber and the ring plenum) of approximately the same length as that required for the engine to run conventionally. The ring plenum would be of larger section than the intake proper. Given that I expect continuous operation to create continuous pressure in the plenum, we can consider plenum as the ambient.

At least a large part of the wave scenario will thus follow the situation you have with the pulsejet uncoupled from the assembly. As the working temperatures and pressures are likely to be somewhat higher in the coupled engine, the intake proper will probably have to be a bit longer to compensate. It will be a matter of some tinkering to arrive at an optimum length, but I really do not expect this to be critical.

You know that you can get the engine working in sub-optimal conditions if you blow into the intake. If the stagnation pressure at the combustor inlet is indeed above atmospheric, as I expect it to be, the limit parameters for the engine functioning will be broader than with the engine uncoupled.

The third priority is not a priority at all. It is synchronization of the two combustors. I do expect them to synchronize, but I was not guided by that requirement. The end layout is probably good for synchronization and if it happens, great – the noise will be lower etc. etc. but I will not cry into my coffee if it does not happen.