What is it about?
Pulsejets are among the simplest jet engines. They have no compressor, no turbine and no rotating machinery. Air enters through a small spring-like flap (a reed valve), mixes with fuel and burns in a chamber. The burning raises the pressure, which pushes hot gas out of a long exhaust pipe as a pulse of thrust. The pressure then drops, the valve flaps open again, and the cycle repeats. The engine "breathes" on its own through resonant, acoustic pressure waves, firing many times per second. Because they are so simple, light and cheap to build, pulsejets are attractive for propelling high-speed drones and other small unmanned aircraft. However, the flow inside and behind these engines is very hard to predict. In this paper we improved our computer model of a valved pulsejet by adding a new sub-model of the inlet and reed valve. It accounts for how the valve moves and for the inertia of the air rushing through the inlet, without the huge cost of a full fluid–structure simulation. We then compared the simulation with experiments on the same engine. These included measurements of combustion chamber pressure, valve motion, air intake, thrust, and laser-based particle imaging velocimetry (PIV) of the exhaust. The PIV captured the powerful smoke-ring-like vortex that each pulse shoots out of the tailpipe.
Featured Image
Photo by Haci on Unsplash
Why is it important?
Earlier simulations treated the valve as if air flow responded instantly to pressure. As a result, they underpredicted both the air intake and the thrust. Our new valve model shows that the air and the valve have "momentum." Peak air intake occurs about one-ninth of a cycle after the lowest chamber pressure, and air keeps flowing in even after the chamber pressure rises above ambient. Capturing this effect raised the predicted air intake by about 50%. The predicted thrust improved from 28% error to just 1.7% (19.5 N predicted versus 19.1 N measured). The simulated exhaust velocities, vorticity levels and vortex ring positions also matched the experimental data well. This matters because pulsejets offer a route to inexpensive, lightweight propulsion for high-speed drones, where heavy and costly turbine machinery is a major drawback. Real drones also change speed constantly, so designers need a model that responds dynamically to changing conditions at the valve. A model that does this is a necessary first step. A validated simulation also lets engineers study the exhaust vortex ring, which is key to the large thrust gains possible when a pulsejet is paired with an ejector duct. Engineers can then optimize these engines on the computer rather than through costly trial and error.
Perspectives
I have long been fascinated by the pulsejet because it shows how much can be achieved with so little. A tube, a fluttering valve and the physics of pressure waves together produce real thrust with almost no moving parts. Yet that simplicity hides a remarkably complex, unsteady flow. For me this paper was satisfying because it closed a gap we had seen in our earlier work. The simulations were good, but they consistently fell short on thrust and air intake. Tracing that shortfall to the inertia of the valve and the incoming air, and then seeing the corrected model land within a couple of percent of the measured thrust, was a rewarding moment. This work was also a genuine collaboration between our group at North Carolina State University and Dan Paxson at NASA Glenn Research Center. That collaboration made it possible to test the simulations directly against detailed measurements on the very same engine. I hope the paper encourages others to see pulsejets not as historical curiosities but as a serious, low-cost option for drone propulsion and pulsed combustion research. I also hope it shows the value of always checking computational models carefully against experiment.
Andrey V Kuznetsov
North Carolina State University
Read the Original
This page is a summary of: Comparison Between Numerically Simulated and Experimentally Measured Flowfield Quantities Behind a Pulsejet, Flow Turbulence and Combustion, May 2010, Springer Science + Business Media,
DOI: 10.1007/s10494-010-9247-6.
You can read the full text:
Contributors
The following have contributed to this page
