Has anyone instrumented a valved pulse jet while running it from lean to rich to see what changes occur in flight? Autos use things like temperature, mass air flow, and oxygen sensors to control mixture, but with just full throttle running, maybe a pulse jet will only need one or two sensors?
The idea is that if it is relatively easy to measure the mixture ratio, it may be relatively easy to control in an rc airplane. In rc flight, ram air pressure will lean out the mixture, altitude increases will enrich, and high speeds cool the pipe, probably changing something! The usual solution is to block the front of the engine to stop the ram air pressure from leaning out the mixture, but this looses the extra power the engine could put out. The best solution would be to control the mixture in flight. Control line guys use careful fuel tank positioning to counter the lean out, but in rc flight this solution isn’t possible.
The simplest sensor may be just air speed, adding extra fuel as speed increases, starting from a good ground running mixture. It would take some experiments to work out how much extra fuel at each airspeed, but relatively simple to implement. Anyone done something along these lines?
Thanks
Philip
Mixture ratio measurements
Moderator: Mike Everman
Re: Mixture ratio measurements
The usual means of throttling is adjusting the fuel flow. With a fixed air inlet, the mixture is self-determining. It has been noted before that maximum thrust favors the rich side of the range, just before flameout.
A few people have been able to build pulsejets with conventional carburetors attached and in these, presumably, the mixture could be better controlled over a throttle range by jet needle profiles and multiple jets.
I don't think there is a reason why both fuel and air can't be independently varied by a device, except the complexity involved. Weight of the physical metering and control system would be a consideration in an RC plane. Note that the physical size of an air inlet in a jet engine is considerably larger than a model IC engine, and so valving it in some way probably requires bigger and heavier hardware. There also may be other adverse effects on pulsejet operation of particular inlet restrictions.
If it's a subject that interests you, go for it. I'm sure a lot of folks would like to see any sensor results, even if they weren't planning carburetion controls.
A few people have been able to build pulsejets with conventional carburetors attached and in these, presumably, the mixture could be better controlled over a throttle range by jet needle profiles and multiple jets.
I don't think there is a reason why both fuel and air can't be independently varied by a device, except the complexity involved. Weight of the physical metering and control system would be a consideration in an RC plane. Note that the physical size of an air inlet in a jet engine is considerably larger than a model IC engine, and so valving it in some way probably requires bigger and heavier hardware. There also may be other adverse effects on pulsejet operation of particular inlet restrictions.
If it's a subject that interests you, go for it. I'm sure a lot of folks would like to see any sensor results, even if they weren't planning carburetion controls.
No problem is too small or trivial if we can really do something about it.
Richard Feynman
Richard Feynman
Re: Mixture ratio measurements
Actually, I was only interested in keeping an optimum mixture during flight, not adding a throttle. The thought was to adjust the fuel flow to match the air, not the other way around. Maximum air through the engine, then fuel to match. A couple of possible ways to do this would be a servo controlled needle valve, or by controlling the fuel pressure with a pump or some kind of pressure valve. The big question is how to measure what the mixture is, to then know which way to adjust it?
Philip
Philip
Re: Mixture ratio measurements
Well in that case, I really like your idea of adjusting the fuel flow with the air speed.
I like its simplicity, and the fact that you wouldn't actually need a servo, just a paddle and a linkage to the needle valve.
The earliest ASIs in real aircraft were a simple spring loaded pivoting arm with a paddle. The arm moved along a flat quadrant which was calibrated in air speed units. The whole thing was mounted on one of the wing struts so the pilot looked to the side to get his airspeed.
Anyway in a really high speed model, where ram effects might be a problem, the paddle area wouldn't have to be large to proportionally actuate a flow valve. It's not like a loaded control surface after all. The "paddle" might actually be a feature on the model, depending on the type. Maybe a pivoting windscreen or sliding canopy cover. A sliding scoop, cowl, landing gear wheel cover, bomb, etc.
A paddle and linkage would be lighter than a servo/sensor/control processor and the battery requirements to drive it.
I like its simplicity, and the fact that you wouldn't actually need a servo, just a paddle and a linkage to the needle valve.
The earliest ASIs in real aircraft were a simple spring loaded pivoting arm with a paddle. The arm moved along a flat quadrant which was calibrated in air speed units. The whole thing was mounted on one of the wing struts so the pilot looked to the side to get his airspeed.
Anyway in a really high speed model, where ram effects might be a problem, the paddle area wouldn't have to be large to proportionally actuate a flow valve. It's not like a loaded control surface after all. The "paddle" might actually be a feature on the model, depending on the type. Maybe a pivoting windscreen or sliding canopy cover. A sliding scoop, cowl, landing gear wheel cover, bomb, etc.
A paddle and linkage would be lighter than a servo/sensor/control processor and the battery requirements to drive it.
No problem is too small or trivial if we can really do something about it.
Richard Feynman
Richard Feynman
Re: Mixture ratio measurements
I have been looking into oxygen sensors as the most obvious way to measure the mixture. It looks like only one brand and type will work, the Bosch LSU wide band sensor. Most oxygen sensors are narrow band, they only measure right around stoichiometric, and I understand that pulse jets run on the rich side of stoichiometric, so a wide band sensor is needed. Too bad it only come in the giant heavy size, not the new mini motorcycle size. The sensor working temperature is 750 C with 900 C max, which sounds about right for the pipe temperature, dull red to bright red.
Running the wide band sensor is also trickier than a normal narrow band sensor, you have to make some feedback controls, then take those correction currents as the output signal. They also have heater circuits to make you car start working quicker, but I think that pulse jets heat up quick enough, they start glowing in about 4 second?
I am thinking now that an O2 sensor would be great for calibrating some kind of airspeed : mixture device, but then taking out the O2 sensor and running with just and airspeed sensor, since airspeed is the main variable, (assuming I have the right jetting on the ground, and you don't fly too high).
Philip
Running the wide band sensor is also trickier than a normal narrow band sensor, you have to make some feedback controls, then take those correction currents as the output signal. They also have heater circuits to make you car start working quicker, but I think that pulse jets heat up quick enough, they start glowing in about 4 second?
I am thinking now that an O2 sensor would be great for calibrating some kind of airspeed : mixture device, but then taking out the O2 sensor and running with just and airspeed sensor, since airspeed is the main variable, (assuming I have the right jetting on the ground, and you don't fly too high).
Philip