The rapid proliferation of uncrewed aircraft systems (UAS) has introduced new and pressing safety concerns for commercial aviation. Recreational UAS users in particular may be unaware of airspace regulations, increasing the risk of an airborne collision and subsequent engine ingestion. Because UAS components, such as lithium-polymer batteries and electric motors, are significantly denser and stiffer than birds or ice, existing aviation certification standards cannot be directly applied. To address this, the Federal Aviation Administration (FAA) sponsored the Task A43 research programme, conducted by The Ohio State University (OSU) and the National Institute for Aviation Research (NIAR). The project’s primary goal was to execute a live UAS engine ingestion test to validate computational modelling approaches used in previous airborne collision studies.
The live ingestion experiment The physical test was conducted at the Naval Air Warfare Center (NAWC) facility in China Lake, California. The research team selected a flightworthy CFM56-7B high-bypass turbofan engine, which is exclusively used on the Boeing 737 next-generation airliner, making it highly representative of modern commercial fleets. The chosen projectile was a DJI Phantom 3 standard quadcopter, selected because its key rigid components (battery, camera, and motors) are similar to newer models and because a high-fidelity computational model of this specific UAS had already been experimentally validated. The drone had a mass of 1.216kg (2.68lb).
To simulate a severe takeoff collision, the engine was operated at a fan rotational speed of 5,175 RPM. The UAS was launched into the engine at a relative translational speed of 92.6 m/s (180 knots). The target aim point was roughly 75% of the radial span of the fan blades, a location known to cause maximum fan damage while reducing the probability of core ingestion.
Computational modelling and validation A core objective of Task A43 was to evaluate how well computational simulations, performed using LS-DYNA software, could predict the real-world damage sustained during a UAS ingestion. The researchers developed a specific finite element model of the CFM56-7B fan assembly and compared its simulated ingestion results against the live test data. Furthermore, they compared the physical test against an “open representative fan assembly model” developed during previous research, which mimics the structural features of typical high-bypass engines without relying on proprietary commercial designs.
Data collection during the live test relied heavily on high-speed cameras, digital image correlation (DIC), and strain gauges mounted on the fan blades. Although lighting issues limited some of the DIC resolution, the cameras successfully captured the UAS’s orientation, velocity, and trajectory immediately prior to impact.
Damage severity and findings The live experiment resulted in significant damage to multiple fan blades. The physical results and the computational simulations aligned remarkably well, both categorising the outcome as a severity level 3 event. A severity level 3 classification indicates significant damage—such as material loss on the leading edges and visible cracking above the mid-span of the blades—but implies that the imbalance remains within the engine certification envelope, akin to a single blade-out event.
In both the physical experiment and the CFM56-7B simulation, the UAS was entirely obliterated upon impact. While the physical test exhibited a fireball explosion that the LS-DYNA software cannot computationally replicate, the kinematics of the collision and the specific blades impacted matched almost exactly. Furthermore, the steady-state imbalances, evaluated by measuring the shift in the centre of mass of the blades post-impact, were highly consistent between the physical engine model and the generic representative model.
Conclusions and industry impact The Task A43 report successfully validated the computational modelling methodology used to simulate UAS engine ingestions. Crucially, the research proved that the open representative fan assembly model behaves similarly to an actual in-service CFM56-7B engine under collision conditions.
This conclusion provides the aviation and UAS industries with a vital, non-proprietary tool for future safety testing. By relying on this experimentally validated representative model, aircraft engine manufacturers and UAS developers can safely and efficiently study foreign object ingestions, improve computational parameters, and mitigate the risks posed by the growing number of drones in the sky.
Related
Discover more from sUAS News
Subscribe to get the latest posts sent to your email.